PŘÍRODOVĚDECKÁ FAKULTA UNIVERZITY PALACKÉHO V OLOMOUCI KATEDRA ZOOLOGIE A ANTROPOLOGIE HABILITAČNÍ PRÁCE

PŘÍRODOVĚDECKÁ FAKULTA UNIVERZITY PALACKÉHO V OLOMOUCI KATEDRA ZOOLOGIE A ANTROPOLOGIE

HABILITAČNÍ PRÁCE Komentář a soubor publikací předložených k habilitačnímu řízení v oboru zoologie

Tomáš Grim Olomouc 2006

© Tomáš Grim 2006 © Academia 2005 © Academy of Sciences of the Czech Republic 1996, 1997 © Blackwell Publishers 2001, 2002, 2006 © British Trust for Ornithology 2004 © Cargo Publishers 2002 © Slovak Academy of Sciences 2001, 2006 © Springer-Verlag 2001 © The American Ornithologists’ Union 2005 © The Linnean Society of London 2005 © The Royal Society 2003 © The Wilson Ornithological Society 2006 © Vesmír 1999, 2000, 2001, 2003, 2006

2

Obsah Poděkování

4

Úvod • Komentář k odborným publikacím* • Komentář k populárně-vědeckým publikacím

7 7 11

Summary

13

Seznam vlastních prací

17

Jednotlivé práce

21

Na texty zařazené do habilitační práce je odkazováno pořadovým číslem [v hranaté závorce], které odpovídá číslování v seznamu prací na s. 17.

*

3

Poděkování Rád bych využil této možnosti, kdy nejsem svázán prostorovými nároky redaktorů odborných časopisů, a vyjádřil svoje poděkování v rozsahu, který v odborném článku nepřipadá v úvahu. Na prvním místě samozřejmě děkuji své mamince za její podporu a trpělivost s mým „lítáním po světě“ a mými dalšími úlety. Dále pak prof. J. Ševčíkovi za finanční podporu mého „sabbatical leave“ na Novém Zélandu a také za šibeniční termín pro odevzdání habilitační práce, který se pozitivně projevil snad i v tom, že je úvod stručný a řadu článků jsem musel dokončit o několik let dříve než by tomu bylo za „normálního“ provozu, kdy většinu času spolknou přednášky a cvičení. Svému vedoucímu prof. S. Burešovi jsem vděčný za jeho přátelský přístup a jeho toleranci k mému „rajzování“ (nejen v zájmu vylepšení výuky zoogeografie). Konkrétní osoby jsou uvedeny v abecedním pořadí. Rád bych tedy poděkoval alespoň v přibližném pořadí důležitosti ... za počáteční impuls Předně bych rád zdůraznil, že jsem samozřejmě nejvíce zavázán všem, na jejichž myšlenkách jsem mohl parazitovat. Nutným předpokladem takovéto činnosti je ovšem vůbec začít myslet a vědět, o čem myslet. V tomto směru jsem velmi vděčný Lumíru Gvoždíkovi za to, že mi na podzim roku 1997 (kdy jsem jako „sebevědomý a zkušený mladý ornitolog“ nastoupil na postgraduální studium na Katedře zoologie UP) ukázal, že o moderní biologii nevím vůbec nic; doporučením Dawkinsova zásadního dílka The Selfish Gene přesměroval můj další odborný vývoj na tu zdaleka nejzajímavější kolej. Tuto událost a s ní synchronní první setkání s Martinem Konvičkou a Janem Zrzavým v kavárně Caesar 21. října 1997 nelze popsat jinak než jako prozření či osvícení, které naprosto změnilo můj život a náhled prakticky na vše důležité. Následná setkávání s Martinem, Honzou a dalšími zajímavými lidmi (David Storch, Stanislav Komárek, Václav Bělohradský, Zdeněk Kratochvíl atd.) v (nejen) olomouckých restauračních podnicích a nekonečné debaty u tuctů piv a káv do (po)zavíracích hodin rozhodujícím způsobem ovlivnily názory prezentované v odborných i populárně-vědeckých článcích zahrnutých v této habilitační práci. Především Honzovi, Martinovi, Davidovi a Standovi jsem za jejich inspirující názory a debaty neskonale vděčný. ... za pomoc spoluautorů mých článků Velké poděkování musí pochopitelně směřovat i všem mým spoluautorům textů odborných i neodborných z důvodů, které snad není třeba zmiňovat: Nicolas F. Britton, Miroslav Čapek, Alena Dvorská, Martin Fellner, Nigel R. Franks, Marcel Honza, Miloslav Jirků, Oddmund Kleven, Martin Konvička, Petr Kovařík, Miloš Krist, Tomáš Kuras, Beata Matysioková, Oldřich Mikulica, Arne Moksnes, Vojtěch Mrlík, Vít Novák, Robert Planqué, Petr Procházka, Vladimír Remeš, Eivin Røskaft, Radim Šumbera, Ivan H. Tuf, Zdeněk Vermouzek, Milan Veselý, Karel Weidinger a Jan Zrzavý. ... za kritické připomínky Své odborné články bych nikdy nemohl publikovat v alespoň tak solidních časopisech, v jakých se nakonec ocitly, nebýt často drsné (ale o to vítanější) kritiky mých rukopisů od řady kolegů: Nick B. Davies (Cambridge, UK), Lumír Gvoždík (Studenec, ČR), Mark E. Hauber (Auckland, Nový Zéland), Marcel Honza (Brno, ČR), Miloš Krist (Olomouc, ČR), Arnon Lotem (Tel Aviv, Izrael), Anders P. Møller (Paříž, Francie), Václav Pavel (Olomouc, ČR), Jaroslav Picman (Ottawa, Kanada), Tomás Redondo (Granada, Španělsko), Vladimír Remeš (Olomouc, ČR), Eivin Røskaft (Trondheim, Norsko), Spencer G. Sealy (Manitoba, Kanada), Manuel Soler (Granada, Španělsko), Bård G. Stokke (Trondheim, Norsko) a Emil Tkadlec (Olomouc, ČR). Velmi inspirující diskuze mi ochotně poskytli Nick B. Davies (Cambridge, UK), Mark E. Hauber (Auckland, Nový Zéland), Øistein Holen (Oslo, Norsko), Rebecca Kilner (Cambridge, UK), Naomi Langmore (Canberra, Austrálie), Tore Slagsvold, Lars E. Johannessen, Bo T. Hansen, Øistein Holen, Oddmund Kleven a Nils C. Stenseth (Oslo, Norsko). Jejich připomínky výrazně vylepšily především rukopis o evoluci diskriminace parazitických mláďat [2]. Pozdní příchod Beate Schönert na naši schůzku v norském Bergenu mi poskytl nečekaný prostor pro „okamžik 4

prozření“, kdy mě napadl (doslova) „rarer enemy effect“ jako vysvětlení vzácnosti diskriminace juvenilních parazitů. V seznamu nejsou explicitně uvedeni anonymní recenzenti, za to si ovšem mohou sami. Přesto jim děkuji. ... za vytvoření inspirující pracovní atmosféry Můj pracovní pobyt v norském Trondheimu by neproběhl tak hladce a příjemně nebýt ochotné a přátelské pomoci Eivina Røskafta, Arne Moksnese a Bårda Stokkeho. Krátkodobý, ale o to intenzivnější pobyt v Oslo zorganizovali Tore Slagsvold, Lars E. Johannessen, Bo T. Hansen, Øistein Holen, Oddmund Kleven a Nils C. Stenseth. Největší dík ale patří Marku E. Hauberovi – nebýt jeho trpělivosti a ochoty diskutovat do nekonečna mé rukopisy a nakonec je i převést do angličtiny srozumitelné nejen ne-rodilým mluvčím, nemohl by být můj pobyt v Aucklandu nikdy tak produktivní. Přes zálibu v cestování trávím stále nejvíce času v blízkosti rodné hroudy, především v druhém patře budovy č. 26 na olomoucké třídě Svobody. Dlouhá období stáze mezi cestovními výjezdy jsou snesitelnější díky dlouhodobému mutualismu s řadou kolegů, především pak s Laďou Remešem, I. Hadem T. a Milanem Veselým. ... za redakční práci Rád bych poděkoval redaktorům časopisů ABC, Cargo, Koktejl, Mladá fronta DNES, Moravský ornitolog, Naším krajem, Nika, Ochrana přírody, Psychologie dnes, Ptačí svět, Ptáci kolem nás, Rock & Pop, Veronica, Vesmír, VTM, Věda a život, Zpravodaj SMP, Zprávy ČSO a Živa. Dále bych rád poděkoval redaktorům novin Brněnský večerník, Lidová demokracie, Lidové noviny, Rovnost a Svobodné slovo. Všichni mě učili, jak psát tak, aby se dal výsledek alespoň trošku číst. Paní Věra Amelová (nakladatelství Paseka a Mladá fronta) mi byla nápomocna radou nejen při překladu Dawkinsova „Slepýše“, ale i při recenzování překladů Sobeckého genu a Červené královny. Obzvláštní díky si zaslouží paní Pavla Loucká (redakce časopisu Vesmír): srdečně ji děkuji za inspirující hovory nad sérií káv v její pražské domovské redakci a za to, kolikrát mě decentně upozornila, že neumím česky (tuto větu zjevně nečetla). Daně Cambellové patří dík za poangličtění některých czenglishových verzí mých rukopisů a také anglického Summary této práce. ... za důvěru ze strany editorů odborných časopisů a grantových agentur Upřímně řečeno, větší radost než přijetí některých mých článků k publikaci mi daly žádosti o recenzování rukopisů od editorů odborných zahraničních časopisů Animal Behaviour, Auk, Behavioral Ecology, Behavioral Ecology and Sociobiology, Evolution, Journal of Avian Biology, Philosophical Transactions of the Royal Society London, Trends in Ecology and Evolution a Wilson Journal of Ornithology. Děkuji za jejich ochotu podstoupit takové riziko a to často i opakovaně. Stejně tak mne potěšily žádosti o posouzení grantových přihlášek od agentury GAČR a National Geographic Society a lektorování rukopisů pro nakladatelství Mladá fronta a Triton. ... za morální podporu při psaní recenzí Svým spolužákům z Brna a Olomouce, kteří pilně četli moje první „Vesmírné“ recenze, děkuji za povzbuzující přezdívku „brněnský řezník“ a návrh, abych provozoval v časopise Vesmír pravidelnou recenzní rubriku „Řeznický koutek“. Tato intelektuální podpora mne inspirovala k dalším recenzím (pro někoho spíše hromadným popravám) neschopných překladatelů, autorů mylně se domnívajících, že ví, o čem píší, a nakonec i nezodpovědných „odpovědných redaktorů“. Nutno ovšem přiznat, že tyto dnes již historické recenze mohly působit neobvykle jen díky prachbídné úrovni většiny „recenzí“ publikovaných v našich populárně-vědeckých časopisech a tradičnímu českému patolízalskému zvyku poplácávat po rameni místo toho, co by bylo většinou na místě, totiž otevřené kritiky. Stejně tak děkuji těm, kteří se do této „řeznické“

5

práce neštítili jít se mnou, především Martinovi Konvičkovi, I. H. Tufovi, Milanovi Veselému a Honzovi Zrzavému. ... za statistickou výpomoc Velmi jsem zavázán Emilu Tkadlecovi, jednak za výpomoc statistickou a jednak za seznámení s pracemi George C. Williamse a dalších skvělých mozků. ... za pomoc v terénu Bez pomoci řady kolegů a diplomantů v terénu by zdaleka nebylo možné shromáždit materiál v takovém množství, v jakém je zpracován v různých mých odborných článcích. Dík si zaslouží Miroslav Čapek, Alena Dvorská, Marcel Honza, Karel Janko, Oddmund Kleven, Beata Matysioková, Oldřich Mikulica, Arne Moksnes, Vojtěch Mrlík, Vít Novák, Petr Procházka, Ingar Øien, Eivin Røskaft, Geir Rudolfsen, Peter Samaš a Zuzana Strachoňová. Zvláštní (ve více významech toho slova) poděkování patří Marcelu Honzovi. Především svou první a zároveň téměř poslední radou („Támhle je rybník a kolem v rákosí jsou někde hnízda“) v první den, kdy začal můj terénní výzkum v horkém létě 1994, mě přesvědčil, že nejlepší způsob, jak se něco naučit, je spadnout do vody a pokud možno i plavat (mí diplomanti prominou:-). Denně od 30. ledna 1997 se pak přesvědčuji o pravdivosti jeho dvou tehdejších zásadních bonmotů: „Věda se žene vpřed *****skými kroky“ & „Research is research – and research must be done“. ... a hlavně nejmilejšímu mnohobuněčnému organismu ve známém vesmíru P. Š.

6

Úvod Předložená habilitační práce je souborem 25 odborných článků (z nichž 17 bylo publikováno, ostatní texty jsou v době předložení habilitační práce, tj. 5. května 2006, v recenzních řízeních) a dále 10 populárně vědeckých prací publikovaných v českých periodikách (stručný úvod k těmto pracím navazuje na s. 10). Komentář k odborným publikacím Odborné práce jsou založeny na dlouhodobém studiu řady modelových ptačích druhů: kukačka obecná Cuculus canorus, rákosník obecný Acrocephalus scirpaceus, rákosník velký Acrocephalus arundinaceus, kos černý Turdus merula, drozd zpěvný Turdus philomelos, pěnice černohlavá Sylvia atricapilla, lejsek bělokrký Ficedula albicollis. Rovněž studovaná témata jsou poměrně pestrá, od vztahů mezi hnízdními parazity a jejich hostiteli, přes potravní ekologii drobných pěvců, agresivní chování, rozpoznávání a diskriminaci podnětů, až po faktory ovlivňující publikační a citační úspěšnost vědeckých pracovníků či úlohu sexuální selekce pro evoluci zbarvení ptačího fenotypu. Pro přehlednost jsem se pokusil rozdělit odborné práce do čtyř tematických okruhů, které se týkají vztahů mezi mláďaty hnízdních parazitů a jejich hostiteli, dále potravní ekologie hostitelských a parazitických mláďat a souvisejících témat, hostitelské obrany proti dospělým parazitům a poslední skupinu tvoří výběr těžko zařaditelných prací týkajících se několika různých témat ptačí biologie. Kromě poslední skupiny článků jsou publikace uvnitř skupin řazeny v doporučovaném pořadí čtení. Největší prostor jsem věnoval tématu pro mě nejzajímavějšímu – vztahu mezi hostiteli (např. rákosníky) a parazitickými mláďaty kukačky obecné [1–10]. Toto téma je, zvláště v kontrastu se studiem adaptací na úrovni vajec, velmi zanedbáno [3] a jen minimum prací se této problematice věnovalo podrobněji. Soubor prací o parazitických mláďatech uvádějí dva přehledné články [1, 2]. První review se týká teorie mimikry nejen u mláďat, ale i vajec [1], druhé review pak kromě přehledu parazitických systémů, kde známe nebo předpokládáme odmítání parazitických mláďat, diskutuje i možná vysvětlení pro relativně málo běžný výskyt této hostitelské adaptace [2]. Následující rukopis [3] se věnuje publikačním a citačním aspektům výzkumu hnízdního parazitismu, které mohou souviset s předpokládanou vzácností adaptivních hostitelských reakcí k cizím mláďatům. Asi nejvýznamnějším zjištěním zařazeným do habilitační práce jsou texty o odmítání mláďat kukačky obecné hostitelem rákosníkem obecným. Na práci založenou na pozorovacích datech [4] navázalo experimentální testování mechanismů diskriminace parazitických mláďat [5]. Následujících pět prací [6–10] se zabývá převážně komunikací mezi parazitickými mláďaty a jejich pěstouny a následky této komunikace. Přestože bylo kukaččí mládě po desítky let uváděno jako klasický příklad nadnormálního podnětu (supernormal stimulus) ve všech učebnicích etologie a behaviorální ekologie, je naše práce [6] první, která se testováním této hypotézy rigorózně zabývala. Za podobně průkopnickou práci lze označit i následující článek [7], který se věnuje růstu a přežívání parazitických mláďat u využívaných a nevyužívaných hostitelů. Tato problematika byla obsáhle studována u severoamerického hnízdního parazita vlhovce hnědohlavého Molothrus ater, u evropských hostitelů kukačky je má práce [7] na toto téma vůbec první. Poslední tři práce bloku o mláďatech [8–10] jsou spíše kritického a diskuzního charakteru. Poukazují na chybné metodické přístupy a interpretace dat v práci jiných autorů, ale využívají i vlastní data, jejichž analýza vede k přehodnocení některých názorů a závěrů dřívějších prací. Potravně laděné práce se věnují jednak analýze potravy, kterou přinášeli hostitelé vlastním a parazitickým mláďatům [11–13], ekologickým faktorům, které mohou složení potravy ovlivňovat [11] a dále i možnému využití dat z potravních studií v ochraně přírody [14] a makroekologických analýzách [15]. První tři práce v bloku o obraně hostitelů proti adultním hnízdním parazitům [16–18] se věnují specifickému rozpoznávání těchto nepřátel potenciálními hostiteli a částečně i odmítáním parazitických vajec [17, 18]. Zajímavým a bez experimentálního přístupu těžko pozorovatelným fenoménem je i přílákávání „pomocníků“ při obraně hnízda jeho majiteli; tomuto se věnuje práce [19]. Z prací o obraně hnízda a rozpoznávání nepřátel je nejvýznamější práce první [16],

7

protože poukazuje na metodické problémy experimentálního přístupu a hodnocení dat, které se potenciálně týkají jakékoli práce o rozpoznávání a diskriminaci podnětů a má tedy velmi široký dopad. Umístění ptačího hnízda je v podstatě jedním z obranných mechanismů proti nalezení hnízda nepřáteli, tedy nejen predátory, ale i hnízdními parazity. Proto je zařazena i práce o vlivu umístění hnízda na riziko parazitace a možné souvislosti s rychlostí koevoluce parazit-hostitel [20]. Poslední blok odborných prací sestává z textů na další témata, kterými se v současné době zabývám. Práce [21] referuje o znovuobjevení velmi vzácného a ohroženého druhu lelka v jihoamerické Bolívii, což může mít význam pro ochranu bolívijských savan typu cerrados. Na téma ochrany přírody navazuje populárně-vědecký text v posledním pátém bloku [33, viz také 35]. Další práce se týkají ptačí ontogeneze [22], vlivu sociálního chování člověka na jeho pracovní výkon [23], faktorů ovlivňujících výběr hostitelů hnízdními parazity [24] a sexuální selekce [25]. Přehled hlavních poznatků a závěrů I. Parazitická mláďata: diskriminace hostitelem, mimikry, mechanismy “rodičovské” péče a růst • • •

•

•

•

•

Diskriminace* různých podnětů nemusí být vždy založena na jejich rozpoznávání (jak se všeobecně předpokládá) [4, 5]. Hostitelé kukačky obecné, tedy hnízdního parazita, jehož mláďata jsou vychovávána sama bez přítomnosti mláďat hostitele, mohou odmítat nejen parazitická vejce, ale i mláďata (což je v protikladu s teoretickými modely obrany hostitele proti hnízdním parazitům) [4]. Rákosník obecný dokáže diskriminovat parazitická mláďata, bez toho že by je rozpoznával, podle délky pobytu mláďat v hnízdě. Tato hostitelská reakce pravděpodobně mohla vzniknout v kontextu konfliktu rodič-potomek nezávisle na parazitaci kukačkou obecnou [5]. Hlavním faktorem zodpovědným za nízký výskyt odmítání parazitických mláďat hostitelem je pravděpodobně slabý selekční tlak na takové hostitelské adaptace (tzv. evoluční rovnováha). Všechna předešlá řešení záhadné vzácnosti proti-mláděcích adaptací u hostitelů, která byla dosud v literatuře navržena, jsou v rozporu s empirickými důkazy. Naopak hypotéza „vzácnějšího nepřítele“ (slabý selekční tlak na evoluci proti-mláděcích adaptací způsobený odmítáním parazitických vajec hostitelem) je v dobrém souladu s dostupnými daty a může teoreticky fungovat u jakéhokoli koevolučního systému parazithostitel [2]. Odmítání parazitických mláďat hostiteli však může být běžnější než se všeobecně soudí. „Vzácnost“ hostitelských adaptací proti mláďatům hnízdních parazitů může být částečně způsobena malou prozkoumaností reakcí hostitelů k cizím mláďatům („research & publication biases“)[3]. Podobnost mezi parazitickými a hostitelskými propagulemi (vejci, mláďaty) by se neměla zaměňovat za mimikry (jak se v literatuře často stává). Podobnosti vajec resp. mláďat parazita a hostitele mohou vznikat celou řadou evolučních, ekologických a behaviorálních procesů, přičemž jen malou část těchto procesů lze klasifikovat jako mimikry. Hypotézu o mimikry lze robustně testovat jen experimentálně a před přijmutím hypotézy o mimikry je třeba vyloučit alternativní vysvětlení (např. krypse, fylogenetická omezení apod.) [1]. Mládě kukačky obecné je pro své hostitele rákosníky obecné nadnormálním podnětem, dokáže tedy získat více rodičovské péče než mláďata hostitele za stejných podmínek. V důsledku vysokých potravních nároků parazitického mláděte dochází ke kvalitativním změnám složení potravy sbírané rákosníky a ke změnám velikostní distribuce potravních

* Diskriminace = rozdílná behaviorální reakce na dva či více různých podnětů, rozpoznávání = kognitivní proces schopnosti rozlišit dva či více podnětů, který může, ale nemusí vést k diskriminaci.

8

• •

•

•

položek. Tyto jevy mohou souviset se sníženou potravní selektivitou hostitelů krmících velké mládě parazita [6]. Kukaččí mláďata mohou dosahovat dramaticky odlišného růstu a přežívání v hnízdech různých blízce příbuzných druhů hostitelů (zástupci rodu Turdus), což může souviset s výběrem hostitelů kukačkou obecnou [7]. Oproti předešlým závěrům se ukázalo, že starší mláďata kukačky (> 15 dnů po vylíhnutí) stimulují své pěstouny ke krmení nejen vizuálně (plocha zobákové dutiny) a vokálně (frekvence žadonících hlasů), ale i behaviorálně (asymetrické mávání křídly, tj. zvedání jednoho z křídel nad úroveň těla a následná vibrace křídlem v této poloze) [8]. Žadonění pomocí křídel je pravděpodobně ancestrální způsob komunikace potravních potřeb mláďat svým rodičům (respektive pěstounům) u altriciálních ptáků (nebo přinejmenším u pěvců a kukaček). Hnízdní parazité využívají tohoto pre-existujícího komunikačního kanálu mezi potomky a rodiči [9]. Hypotéza virulence parazitických mláďat předpokládá, že parazitická mláďata mohou mít výhodu z přítomnosti mláďat hostitele (více mláďat přiláká více rodičovské péče, kterou pak mohou v rámci snůšky uzurpovat konkurenceschopnější parazitická mláďata na úkor méně kompetitivních mláďat hostitele). Nízká virulence některých parazitických mláďat (tj. jejich tolerance k přítomnosti mláďat hostitele) však nemusí být adaptivní a může být výsledkem fyzikálních omezení daných velikostí hnízda, velikostí mláďat hostitele či synchronizací líhnutí parazitických a hostitelských mláďat [10].

II. Parazitická a hostitelská mláďata: složení potravy •

•

• •

Potrava mláďat rákosníka obecného, rákosníka velkého (ohrožený druh v ČR) a kukačky obecné je převážně zastoupena hmyzem, zejména dvoukřídlým. Složení potravy je ovlivněno nejen mikrohabitatem (aktuální nabídkou v blízkosti hnízda), ale i přítomností parazitického mláděte kukačky v hnízdě [6, 11–14]. Přestože jsou pestřenky učebnicovým příkladem ochranných Batesiánských mimikry, tvoří významnou složku potravy rákosníků. Je možné, že mnohé domnělé příklady mimikry jsou pouze zdánlivé a vyplývají z nemimetických podobností pestřenek a vos či včel pro lidského pozorovatele [14]. Potravní studie založené na analýze potravních vzorků mohou být významným zdrojem faunistických údajů o vzácných, ohrožených či tradičními entomologickými metodami špatně zachytitelných druhů [14]. Faunistické studie by se mohly v budoucnu stát důležitým zdrojem primárních dat o distribuci různých taxonů pro makroekologické, biogeografické a ochranářské studie [15].

III. Obrana hostitelů proti dospělým parazitům • • •

• •

Výsledky studií rozpoznávání nepřítele (např. dospělých parazitů u hnízd hostitelů) mohou být významně ovlivněny výběrem kontrolních atrap pro prezentace u hnízd hostitelů [16]. Výběr behaviorálních proměnných pro výpočet kompozitních charakteristik (analýza hlavních komponent, PCA) chování ptáků při obraně hnízda může významně ovlivnit závěry takových studií [16]. Při analýze a interpretaci reakcí hostitele na přítomnost dospělých parazitů u hnízda je nutno brát v úvahu reakci na kontrolní atrapy (tzv. „background aggression“). Jinými slovy, anti-parazitická agresivita není celková úroveň reakce na atrapu parazita, ale od celkové reakce je třeba odečíst „background aggression“, jinak můžeme dojít k zavádějícím závěrům [16]. Kos černý, drozd zpěvný a rákosník obecný se brání proti parazitismu např. odmítáním parazitických vajec a agresivitou vůči dospělému parazitovi u hnízda. Jejich rozpoznávací schopnosti vajec i dospělých parazitů jsou ale omezené [17, 18]. Pěnice černohlavá je velmi agresivní vůči kukačce u hnízda, což může částečně vysvětlit, proč není pěnice černohlavá v současné době využívána jako hostitel kukačkou. Naopak

9

•

•

•

reakce kosa černého a drozda zpěvného k parazitickým vejcím a dospělcům nemohou vysvětlit, proč se jim kukačka vyhýbá [16, 17]. Chování rákosníků obecných na jejich hnízdech dopoledne a odpoledne naznačuje, že odpolední kladení kukačkou není adaptace proti zvýšené přítomnosti hostitelů na hnízdech v dopoledních hodinách. Navíc k rozhodnutí odmítnout parazitické vejce dochází se zpožděním po parazitaci [18]. Agresivní a nápadná obrana hnízda některými ptáky (např. pěnice černohlavá) může přilákat ptáky z okolí. Tento jev lze ovšem vysvětlit jako vedlejší produkt proximátních faktorů (nápadnost obrany), který nemá pro hostitele zjevný adaptivní význam (ultimátní rovina vysvětlení, viz také [27]). Naopak přilákaní ptáci mohou získávat důležité informace o přítomnosti nebezpečí v okolí svých hnízd [19]. Jednou z důležitých strategií obrany hostitele proti hnízdním parazitům je ukrytí hnízda. Nenáhodná distribuce hostitelů s ohledem na jejich schopnost obrany proti parazitismu (naivní jedinci akceptující parazitismus vs. zkušení jedinci odmítající parazitismus) však může vést k nenáhodně zvýšenému riziku parazitace té části populace, která se proti parazitismu brání málo či vůbec. Tento jev může zpomalovat rychlost koevoluce mezi parazity a hostiteli, především evoluci mimikry [20].

IV. Ostatní •

•

•

•

•

10

Lelek bělokřídlý (Eleothreptus candicans), vzácný a ohrožený neotropický druh, může být rozšířenější, než se dříve předpokládalo, a kromě Brazílie a Paraguaye může mít životaschopné populace i v dalších oblastech, např. v Bolívii. Toto zjištění může mít význam pro záchranu tohoto ochranářsky zajímavého druhu [21]. Vejce kukačky obecné jsou proti rozbití skořápky mechanicky odolnější než podobně velká vejce hostitelů (rákosník velký). Výhoda z této adaptace proti narušení vejce hostiteli (rozklování a vyhození) však s sebou nese i nevýhodu: kukaččí embryo musí vynaložit mnohem více energie na vylíhnutí ve srovnání s podobně velkým embryem hostitele rákosníka velkého. S tím mohou souviset i některé morfologické rozdíly mezi mláďaty parazita a hostitele (např. umístění a velikost vaječného zubu). Závěry této práce je však třeba potvrdit analýzou srovnávacího materiálu získaného od neparazitických druhů kukaček [22]. Přestože jsou faktory ovlivňující úspěšnost při publikování a citování ve středu zájmu současných biologů, žádná práce dosud neuvažovala vliv sociálního chování vědců na jejich publikační a citační produktivitu. Na vzorku českých behaviorálních ekologů a evolučních biologů, kteří se zabývají modelovou skupinou ptáků, se ukázalo, že investice do sociálních aktivit (měřená objemem piva zkonzumovaného ve volném čase) negativně ovlivňuje publikační i citační výkon respondentů. Princip „trade-off“ (tj. kompromis mezi investicemi do různých aktivit) lze tedy aplikovat i na sociální chování člověka a použít jej pro vysvětlení publikačních a citačních patternů (publication & citation biases) [23]. Nevyužívání některých potenciálních hostitelů hnízdními parazity lze vysvětlit celou řadou ekologických a behaviorálních faktorů, např. složením potravy dodávané mláďatům, odmítáním parazitických vajec či mláďat, morfologií hnízda, přílišnou agresivitou vůči dospělcům parazita atd. Absenci parazitace konkrétních hostitelů většinou nelze vysvětlit jedním faktorem, ale jako interakci různých faktorů. Parazité se mohou vyhýbat různým navzájem příbuzným druhům hostitelů z často velmi odlišných důvodů [7, 16–18, 24]. Modré až modro-zelené zbarvení vajec některých ptáků, např. lejska bělokrkého, je způsobenu biliverdinem, který má antioxidační funkce. Z teorie sexuální selekce vyplývá, že samice mohou investovat biliverdin do vaječných skořápek jako signál kvality svým partnerům. Ti by v reakci na tento samičí handikepový signál měli podle hypotézy samičího signálování zvyšovat svou rodičovskou investici. U lejska bělokrkého se sice zdá, že modrá vejce souvisejí s fenotypickou kvalitou samice, ale samci na tento potenciální signál nereagují změnou svých rodičovských investic [25].

Komentář k populárně-vědeckým publikacím Případného čtenáře této práce asi překvapí oddíl populárně-vědeckých publikací, který obvykle nebývá součástí habilitačních prací. Důvodů k zařazení tohoto oddílu bylo více: (i) popularizační činnost by měla být nedílnou součástí práce a tedy i odborného profilu každého badatele (tedy nejen pracovníka AVČR, ale samozřejmě i vysokoškolského pedagoga), (ii) některé z těchto zařazených prací jsou nepochybně podnětnější a přínosnější než některé z čistě vědeckých prací a (iii) některé „pouze“ populárně-vědecké texty stály nesrovnatelně více námahy a vyžadovaly výrazně širší teoretický „background“ než mnohé mé práce odborné. Populárně-vědecké práce tedy mohou o publikačních a pedagogických schopnostech autora vypovídat často i více než práce čistě odborné. Především je třeba zdůraznit, že publikování populárně-vědeckých článků v časopisech a novinách je nepsanou povinností každého výzkumného pracovníka – je-li veškerý výzkum plně hrazen z daní prostých občanů, pak je každý vědec morálně zavázán podat svým mecenášům (tedy občanům) zprávu o tom, za co jejich prostředky utratil on, případně další badatelé u nás či v zahraničí. Ostatně popularizace základního výzkumu (který k ničemu praktickému z definice není) je jediným výstupem této činnosti, který lze označit za „výstup praktický“. Při výběru prací do tohoto přehledu jsem se řídil třemi kritérii – reprezentativnost, popularita a praxe. Prvním kritériem mám na mysli snahu podat co nejvýstižnější přehled témat, kterým se věnuji, druhým pak pozitivní ohlas mezi čtenářstvem. Posledním kritériem bylo praktické využití konkrétních populárně-vědeckých textů: např. článek o explanačních rovinách v evoluční biologii a behaviorální ekologii [27] je povinnou literaturou pro posluchače kurzů „Systém a fylogeneze strunatců“ a „Etoekologie“ na PřF UP, řada článků o exotických biotopech [včetně 35] je doporučenou literaturou pro posluchače kurzu „Zoogeografie“ na PřF UP, hromadná recenze překladů populárně-vědeckých knih s ilustrací různých typů překladatelských chyb [26] je zase využívána jako jeden z výukových materiálů pro posluchače kurzu „Teorie překladu“ na Katedře anglistiky a amerikanistiky UP (doc. PhDr. D. Knittlová, CSc. in verb.). Do habilitačního spisu jsem zařadil většinou práce týkající se mého nejbližšího oboru, tedy behaviorální ekologie. Jde jednak o práce teoretické [27, 28], dále texty zabývající se historií a kontroverzemi v rámci zmíněného výzkumného proudu [29], práce popularizující výsledky mého vlastního bádání [32] a výzkumu jiných autorů [30]. Ve své popularizační činnosti se snažím neomezovat na práce ornitologické [32], o čemž svědčí např. rozbor problematických aspektů memetiky jakožto výzkumu kulturní evoluce [31] nebo článek o evoluci potravních adaptací u masožravých rostlin láčkovek [34]. Zásadnímu tématu o koncepčních rozdílech mezi přírodními a sociálními/humanitními vědami je věnován paradigmatický (slovy dr. J. Sádla) text [28]. Za pravděpodobně nejdůležitější populárněvědeckou publikaci mezi vybranými články lze považovat text o prioritách ochrany přírody [33]. Tato práce upozorňuje na propastný rozdíl mezi mediálním obrazem ochrany přírody a jejími skutečnými prioritami, o nichž se nemluví a o kterých mnoho ochranářských aktivistů a profesionálů dokonce ani neví. Vzhledem k aktuálnosti a závažnosti tématu se jedná o zřejmě jediný text v předložené habilitační práci, který může mít praktické dopady. Z desatera vybraných prací jsou tři recenze [26, 28, 29], což není náhoda. Domnívám se – a ve svých recenzích jsem se pokusil ukázat –, že recenze je literární útvar většinou českých autorů využívaný nedostatečně a chápaný mylně jako synonymum pro anotaci. Většina „recenzí“ publikovaných v českých populárně-vědeckých (a často i odborných!) periodikách totiž jsou pouhými anotacemi: přehledy obsahu (často dokonce jen jako stručně komentované názvy kapitol!) recenzovaného díla doplněnými údaji o ceně a dostupnosti dané knihy. To je velmi smutné, poněvadž recenze by měla především kriticky zhodnotit význam dané práce, zařadit ji do kontextu ostatních autorových prací, srovnat ji s tématicky spřízněnými díly v našem i cizích jazycích a vyzdvihnout přednosti i nedostatky recenzované práce. Tato základní kritéria nesplňují ani vzdáleně téměř žádné texty publikované pod hlavičkou „Recenze“ v českých populárně-vědeckých časopisech. Doufám, že se mně i mým spoluautorům podařilo – přes počáteční někdy až šokovanou reakci části čtenářstva – přispět k tomu, že dnes začínají převládat recenze kritické, které nepodlézají zavedeným „autoritám“ a „zasloužilým badatelům“,

11

ale jasně a nekompromisně poukazují na často velmi mrzkou kvalitu českých učebnic, překladů i původních děl českých autorů. Jak ostatně poznamenal šéfredaktor časopisu Vesmír Mgr. Ivan Boháček, který otiskl první z mých ostřeji laděných recenzí: „Přečíst si paskvil a nevarovat před ním potenciálního a neznalého čtenáře, je zločinem proti tomuto čtenáři“. Právě z tohoto důvodu do habilitačního spisu zařazuji i některé z populárně-vědních článků, neboť se přímo týkají mé pedagogické činnosti, ať už tím, že jde o výukové materiály využívané mnou či dalšími pedagogy [26, 27, 30, 33, 35], nebo tím, že jde o recenze učebnic a skript [26, 29]. Tyto práce [26–35] tedy mohou mít přímé praktické využití na rozdíl od prací sepsaných na základě výsledků základního výzkumu [1–20].

12

Summary This thesis is to be submitted in partial fulfilment for the requirements of lectureship (i.e. assistant professorship) at the Faculty of Sciences, Palacký University, Czech Republic. This thesis covers various aspects of avian behavioural ecology, namely brood parasite-host coevolution, diet ecology of birds, sexual selection and other issues. The studied model species include the brood parasitic common cuckoo Cuculus canorus, its used hosts: the reed warbler Acrocephalus scirpaceus, the great reed warbler Acrocephalus arundinaceus, potential but currently not used hosts: the blackbird Turdus merula, the song thrush Turdus philomelos, and the blackcap Sylvia atricapilla. Aggressive behaviour and sexual selection was also studied in the collared flycatcher Ficedula albicollis. This thesis is presented as a set of 25 scientific publications* in English (15 published in IF journals, 2 published in non-IF international peer-reviewed journals and 8 manuscripts submitted to IF journals at the time of submission of this thesis, i.e. 5th May 2006) and 10 popular-scientific articles in Czech. The latter were included (i) to give a Czech reader and the lectureship committee a better idea of author’s scope of work and interest, (ii) because writing popular-scientific articles is a natural and important part of scientific work (all research is paid for by taxes and the general public has a right to know how its money is being spent on scientific research), (iii) some of the popular-scientific papers included are used as teaching materials at the Faculty of Sciences of Palacký University, Department of Zoology [27, 30, 33, 35] and the Faculty of Philosophy, Department of English and American Studies [26]. The papers are grouped into five subsets and arranged in the order of suggested reading (except for the last V. set of popular-scientific writings which are presented in alphabetical order). I hope that the papers speak for themselves, thus I will provide a brief summary only of the main findings and most important conclusions: I. Parasitic chicks: host discrimination, mimicry, mechanisms of “parental” care and growth (papers 1–10) • •

•

• •

•

*

Discrimination of various cues may not be based on recognition of those cues (as traditionally accepted in the literature) (4, 5). Reed warblers, the most frequently used and most studied host of the cuckoo, can discriminate against parasitic chicks without recognising them. The suggested mechanism of discrimination is based on the length of parental care at the nest with no respect to the nest content itself (4, 5). The major factor behind the observed rarity of parasitic chick discrimination (in contrast to frequent adaptation against parasitic eggs) is the host’s own behaviour – the rejection of parasitic eggs. A simple threshold model of parasite-host coevolution leads to “the rarer enemy effect” which provides an equilibrium explanation for the apparent rarity of host discrimination of alien chicks. This model is in concordance with the available evidence. In contrast, previous explanations suggested in the literature are at odds with empirical data or are unworkable in principle [2]. The apparent rarity of chick discrimination could be, at least partly, explained by research, publication and citation biases [3]. It is crucial to differentiate between the mimetic and non-mimetic similarities of parasite and host progeny (eggs, nestlings). The hypothesis on mimicry cannot be supported by a comparative description (as was frequently done previously). It must be tested experimentally and alternative explanations (e.g. crypsis, learning etc.) for the similarity of the traits under study must be examined carefully [1]. Although the common cuckoo chick is traditionally accepted as a textbook example of the supernormal stimulus, the first study to test the hypothesis rigorously was carried out in 2001 [6]. We supported the hypothesis by showing that a cuckoo chick gets more food Papers are referred to by their arabic numerals in [parentheses]. See page 17. 13

• •

• •

than a host chick of the same size which is alone in the nest (i.e. controlling for the chick size and brood size). More interestingly, this supernormal food consumption leads, in old cuckoo chicks, to predictable qualitative changes in diet composition and distribution of prey item sizes [6]. Cuckoo chicks show dramatically different growth rates and survival in the nests of two closely related potential, but avoided hosts [7]. Although a previous study suggested that reed warbler parents use mainly or exclusively gape area and calling rates of chicks when deciding how much food to bring to the nest, I showed that this rule cannot apply in old and large parasitic chicks. I suggest that parasitic chicks use wing raising and shaking to elicit higher provisioning rates when gape area and calling rates no longer provide appropriate information on the hunger of the parasitic chick [8]. Wing-shake begging may be an ancestral trait of altricial birds (at least of passerines and cuckoos). Parasitic chicks may exploit pre-existing parental responsiveness to this begging behaviour [9]. The hypothesis of parasitic chick virulence suggests that parasitic chicks may benefit from being tolerant to host chicks. The presence of nest-mates may increase provisioning per nest but the parasite may monopolize more than its fair share due to its better competitive abilities. Although the hypothesis may hold in some parasites (e.g. brownheaded cowbirds Molothrus ater), I show that it cannot work in the common cuckoo. Also, the existence of other host-parasite systems with “tolerant” parasitic chicks are explained more parsimoniously by constraints on chick virulence (e.g. size asymmetries between parasitic and host chicks, nest design, host parental strategies) than by benefits of low virulence [10].

II. Parasitic and host chicks: diet ecology (papers 11–15) • • •

•

Both parasitic and host chicks are fed with approximately similar food but there are also important differences in both the quality and size of the food delivered [11–13]. Hoverflies are textbook examples of protective Batesian mimicry. Despite this they form an important part of diet of reed warblers [14]. Avian foraging studies may provide important faunistic data on rare species and taxa which are difficult to collect using traditional entomological methods. For example, the diet of reed warbler and cuckoo chicks from the Czech Republic contained several very rare species including a newly discovered species for that country [14]. Studies of bird diets may become a source of data for macroecological, biogeographical and conservation studies in the future [15].

III. Host defences against adult parasites (papers 16–20) • •

• •

14

An absence of discrimination in enemy recognition studies may reflect a methodological artefact resulting from a lack of great distinction between the test (e.g. parasite) and control dummies. This can lead to erroneous inferences about coevolution [16]. Inclusion of some behavioural variables into a composite measure of nest defense (PCA) can confound the results of such studies. In other words, it is important to consider the biological (not statistical) significance of the particular behavioural responses of tested individuals in the particular discrimination context [16]. The blackcap is highly aggressive toward the brood parasite, recognizes the parasitic cuckoo as a special enemy and this may partly explain why it is avoided by the cuckoo [16]. Reed warblers do not recognize the cuckoo egg immediately after being parasitized and the absence of differences in their behaviour between morning and afternoon does not support previous explanations for the afternoon laying in the cuckoo (i.e. lower aggressiveness and/or presence of hosts at the nest) [18].

•

•

Conspicuous nest defence by blackcaps attracted a wide array of various species of birds into the vicinity of the tested nests. However, this pattern is best explained as a byproduct of proximate factors (i.e. conspicuousness of blackcap behaviour) and seems to be selectively neutral for that species. In contrast, it may be beneficial for the attracted birds themselves (e.g. obtaining information on potential dangers in the general area close to their nests) [19]. The best defence against parasitism is to avoid being parasitized in the first place. However, nest concealment may vary non-randomly between acceptors and rejecters of parasitism (e.g. young naive vs. old experienced individuals). This may result in a nonrandomly higher parasitism of acceptors than rejecters which may slow down hostparasite coevolution rates (e.g. quality of mimicry of parasitic eggs) [20].

IV. Miscellaneous scientific publications (papers 21–25) •

•

•

•

•

The white-winged nightjar Eleothreptus candicans, a rare and endangered neotropical savannah species, may have a larger range than previously thought. Our record of this species from Bolivia may be important in establishing conservation programmes for the species [21]. Parasitic cuckoos have evolved structurally more resistant eggshells. However, this adaptation bears the cost of a more complicated hatching process in the parasitic embryo. Some morphological traits of cuckoo hatchlings (e.g. the size and position of egg tooth) may be adaptations for hatching from structurally strong eggs. However, this hypothesis needs to be tested comparatively after obtaining data on the hatchling morphology of nonparasitic cuckoos [22]. Although publication and citation biases have recently received increasing attention, no study to date has considered human social behaviour as a potentially confounding factor. I showed that publication and citation success of behavioural ecologists and evolutionary biologists studying birds in the Czech Republic is highly significantly and negatively influenced by their social activities which are measured by annual beer consumption [23]. A low frequency of parasitism of some potential cuckoo hosts can be explained by many factors (e.g. diet composition, breeding synchrony, egg rejection etc.). Absence of parasitism in closely related thrushes of the genus Turdus can be explained only by interactions between various factors and closely related hosts species may sometimes be avoided for very different reasons [24]. Blue-green egg colour was hypothesized to be a post-mating sexual signal of females to increase parental investment by their partners. Although blue-green egg colour of female collared flycatchers correlates with their phenotypic quality, males of this species do not seem to adjust their parental care (nest defence, provisioning rates) to the level of this potential signal.

V. Popular-scientific works (papers 26–35) •

I selected these articles so as to be roughly representative of my popularisation work. I have written about the specific problems of translation of popular-scientific books from English to Czech [26], about proximate and ultimate explanations in behavioural ecology [27], relationships between social and natural sciences [28] and evolution of my own discipline from ethological roots to current behavioural ecology [29]. Paper [30] is about the role of mafia-like relationships in nature (from cuckoos to cytoplasmatic incompatibility in Wolbachia) and gives an example of my writing about results of other researchers. In contrast, paper [32] is an example of popularizing my own results, in this case discrimination of parasitic chicks by reed warbler hosts. The other papers have focussed on various subjects, e.g. memetics as a study of cultural evolution (from a critical point of view) [31], the evolution of carnivorous adaptations in pitcher-plants [34] or zoogeography of the largest inundated savannah in the world, the Pantanal [35]. From

15

a practical point of view, the most important paper is about the huge discrepancy between the picture of conservation priorities and the state of natural environments painted by media and the biological “reality” [33]. Because of my interest in travelling and birdwatching in tropical countries I was able to illustrate various phenomena discussed in these and other articles with my own photographs. I am also happy to mention that some of these popular-scientific writings are used in education at our university (see above). Most of these (and many others) papers have also received a very positive response from readers, have lead to sometimes quite heated discussions both in person and in the media and have been covered by newspapers.

16

I. Parazitická mláďata: diskriminace hostitelem, mimikry, mechanismy „rodičovské“ péče a růst 1.

Grim T. 2005: Mimicry vs. similarity: which resemblances between brood parasites and their hosts are mimetic and which are not? Biological Journal of the Linnean Society 84(1): 69–78.

2.

Grim T. 2006: The evolution of nestling discrimination by hosts of parasitic birds: why is rejection so rare? Evolutionary Ecology Research (in press).

3.

Grim T.: Equal rights for chick brood parasites. (subm.)

4.

Grim T., Kleven O. & Mikulica O. 2003: Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Proceedings of the Royal Society of London B 270(1): S73–S75.

5.

Grim T.: Experimental evidence for chick discrimination without recognition in a common cuckoo host. (subm.)

6.

Grim T. & Honza M. 2001: Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behavioral Ecology and Sociobiology 49(4): 322–329.

7.

Grim T. 2006: Cuckoo growth performance in parasitized and unused hosts: not only host size matters. Behavioral Ecology and Sociobiology (in press).

8.

Grim T.: Signals of need in parent-offspring communication: why gaping and calling alone fail to explain provisioning rates? (subm.)

9.

Grim T.: Wing-patch begging, wing-shaking and evolutionary origins of nestling begging strategies. (subm.)

10.

Grim T.: Low virulence of brood parasitic chicks: adaptation or constraint? Ornithological Science (in press).

II. Parazitická a hostitelská mláďata: potravní ekologie 11.

Grim T. & Honza M. 1996: Effect of habitat on the diet of reed warbler (Acrocephalus scirpaceus) nestlings. Folia Zoologica 45(1): 31–34.

12.

Grim T. & Honza M. 1997: Differences in parental care of reed warbler (Acrocephalus scirpaceus) in its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zoologica 46(2): 135–142.

13.

Grim T. 1999: Potrava mláďat rákosníka velkého (Acrocephalus arundinaceus). Sylvia 35(2): 93–99.

14.

Grim T. 2006: An exceptionally high diversity of hoverflies (Syrphidae) in the food of the reed warbler (Acrocephalus scirpaceus). Biologia 61(1): 235–239.

15.

Grim T. 2006: Avian foraging studies: an overlooked source of distribution data for macroecological and conservation studies. Diversity and Distributions 12: 1–3.

17

III. Obrana hostitelů proti dospělým parazitům 16.

Grim T. 2005: Host recognition of brood parasites: implications for methodology in studies of enemy recognition. Auk 122(2): 530–543.

17.

Grim T. & Honza M. 2001: Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to experimental parasitism by the common cuckoo Cuculus canorus. Biologia 56(5): 549–556.

18.

Honza M., Grim T., Čapek M., Moksnes A. & Røskaft E. 2004: Nest defence, enemy recognition and nest inspection behaviour of experimentally parasitised reed warblers Acrocephalus scirpaceus. Bird Study 51(3): 256–263.

19.

Grim T.: Does conspicuous nest defence by the blackcap attract nest defence “helpers”? (subm.).

20.

Grim T. 2002: Why is mimicry in cuckoo eggs sometimes so poor? Journal of Avian Biology 33(3): 302–305.

IV. Ostatní odborné práce 21.

Grim T. & Šumbera 2006: A new record of the endangered white-winged nightjar (Eleothreptus candicans) from Beni, Bolivia. The Wilson Journal of Ornithology 118(1): 109– 112.

22.

Honza M., Picman J., Grim T., Novák V., Čapek M. & Mrlík V. 2001: How to hatch from an egg of great structural strength. A study of the common cuckoo. Journal of Avian Biology 32(3): 249–255.

23.

Grim T.: Publication and citation biases: a possible role of social activities? (subm.)

24.

Grim T., Røskaft E., Moksnes A., Stokke B. G., Honza M., Moskat C., Kleven O. & Rudolfsen G.: Constraints on host choice in parasitic birds: why are thrushes not parasitized by cuckoos? (subm.)

25.

Krist M. & Grim T.: Are blue eggs sexually selected signal of female collared flycatchers? A cross-fostering experiment. (subm.)

V. Populárně-vědecké práce 26.

Grim T. 1999: O holé kryse krtčí a jiné zvířeně. To by se překladateli stát nemělo. Vesmír 78(8): 464–467.

27.

Grim T. 2000: Paralelní vysvětlení. Proč a jak se ptát „proč“ a „jak“. Vesmír 79(2): 92–93.

28.

Grim T. 2000: Poučení z krizového vývoje společenských věd. Vesmír 79(9): 524–527.

29.

Grim T. 2001: Etologie versus sociobiologie? Vesmír 80(4): 231–233.

30.

Grim T. 2001: Mafiánské kukačky a tyranští mravenci. Kukačky nutí hostitele vychovávat parazitická mláďata. Vesmír 80(9): 488–490.

18

31.

Grim T. 2002: Limity memetiky. Cargo 2002(1, 2): 94–100.

32.

Grim T. 2003: Dobrá zpráva pro ornitology – špatná pro kukačky. Čím se prozrazují mláďata hnízdních parazitů? Vesmír 82(8): 437–441.

33.

Grim T. 2006: Kde jsou ochranářské prority? Medializace kontra ochrana přírody. Vesmír 85(3): 140–147.

34.

Grim T. & Jirků, M. 2003: Jak fungují „pekelné konve“. Adaptace masožravých láčkovek. Vesmír 82(10): 559–563.

35.

Šumbera R. & Grim T. 2005: Pantanal − na otevřené scéně. Živa 53(2): 93–96.

19

20

1. Grim T. 2005: Mimicry vs. similarity: which resemblances between brood parasites and 1. and which are not? their hosts are mimetic Grim T. 2005: Mimicry vs. similarity: resemblances between brood parasites and Biological Journal of which the Linnean Society 84(1): 69–78. their hosts are mimetic and which are not? Biological Journal of the Linnean Society 84(1): 69–78.

Blackwell Science, LtdOxford, UKBIJBiological Journal of the Linnean Society0024-4066The Linnean Society of London, 2005? 2005 841 6978 Original Article PARASITIC MIMICRY T. GRIM

Biological Journal of the Linnean Society, 2005, 84, 69–78. With 1 figure

Mimicry vs. similarity: which resemblances between brood parasites and their hosts are mimetic and which are not? TOMÁSˇ GRIM* Department of Zoology, Palack y¢ University t rˇ. Svobody 26, CZ-771 46 Olomouc, Czech Republic Received 30 January 2004; accepted for publication 1 May 2004

Mimicry is one of the most conspicuous and puzzling phenomena in nature. The best-known examples come from insects and brood parasitic birds. Unfortunately, the term ‘mimicry’ is used indiscriminately and inconsistently in the brood parasitic literature despite the obvious fact that similarities of eggs, nestlings and adults of brood parasites to their hosts could result from many different processes (phylogenetic constraint, predation, intraspecific arms-races, vocal imitation, exploitation of pre-existing preferences, etc.). In this note I wish to plead for a more careful use of the term. I review various processes leading to a similarity between propagules (both eggs and nestlings) of brood parasites and their hosts and stress that: (1) mimetic and non-mimetic similarities should be differentiated, (2) a mere similarity of host and parasite propagules provides no evidence for mimicry, (3) mimicry is more usefully understood as a (coevolutionary) process rather than an appearance, and (4) mimicry terminology should reflect the process which led to mimetic similarity. Accepting the mimicry hypothesis requires both the experimental approach and rejection of alternative hypotheses explaining similarities of host and parasite propagules. © 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78.

ADDITIONAL KEYWORDS: adaptation – brood parasitism – coevolution – constraint – convergence – cowbird – cuckoo – pre-existing preferences.

INTRODUCTION Motto: ‘The possibility of mimicry in eggs must be treated with caution, as pure coincidence in their colouration is so general a phenomenon. . . . this state of affairs warns us to be very wary about attributing a given resemblance to mimicry.’ Swynnerton (1916, p. 553).

From mind-boggling similarity among various phylogenetically unrelated butterflies to very generalized ‘prey’ presented by angler fish, mimicry has attracted human interest for a long time (Wickler, 1968; Komárek, 1998, 2003). Such resemblances result from coevolution (Janzen, 1980) or sequential evolution (Futuyma, 1998) and have been extensively studied among insects (for a review see, e.g. Wickler, 1968) and also brood parasites and their hosts (Fig. 1) (for reviews of brood parasitic systems see Rothstein & Robinson, 1998; Davies, 2000). Insect mimicry gener-

*E-mail: [email protected]

ally provides protection against predation or attracts pollinators (Wickler, 1968; Vane-Wright, 1976), whereas coevolutionary mimicry in brood parasites is a counter-adaptation against host antiparasitic response (Rothstein, 1990). With respect to avian parasite–host coevolution attention has been focused mainly on mimicry in parasitic eggs and their rejection by hosts (e.g. Davies & Brooke, 1989; Moksnes et al., 1990). Apparent evidence for nestling mimicry in brood parasites received much less attention (for the most comprehensive review see Redondo, 1993). Is every similarity between parasitic and host propagules an example of mimicry? Various authors realized long ago that this is not the case: for example, eggs of parasite and host can be similar simply because they share the same environment, where they suffer predation from visually orientated predators. If so, then, similarity is cryptic (non-mimetic) and results from convergent evolution (Harrison, 1968; Mason & Rothstein, 1987). In this review I will show

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

69

70

T. GRIM

Number of papers on brood parasitism

90 80 70 60 50 40 30 20 10 0 1980

1985

1990

1995

2000

2005

Figure 1. Number of papers on brood parasitism from 1980 to 2003. Second-order polynomial regression of year against the number of papers on intra- and interspecific brood parasitism in birds: R2 = 0.92, F2,21 = 113.1, P < 0.0001; second-order regression coefficient is significant: t = 2.46, P = 0.02. Data from the Web of Science.

that the issue is even more complex than was previously thought. When discussing various traits in the context of coevolutionary theories it should be clear if these traits are: (1) specific adaptations and counteradaptations (i.e. result of coevolution between parasites and hosts), (2) adaptations resulting from other (noncoevolutionary) selection pressures, or just (3) byproducts of some other – perhaps adaptive – traits (for discussions see Janzen, 1980; Vane-Wright, 1980; Ryan, 1990; Grim, 2002). The following analysis is based on these crucial differences and it is argued that mimicry terminology should reflect the process that led to the given mimetic similarity. I suggest that the term ‘nestling mimicry’ and less so ‘egg mimicry’ is usually applied indiscriminately to any similarity between parasitic and host chicks or eggs, respectively (see, e.g. Jourdain, 1925; Lack, 1968; Mundy, 1973; Wyllie, 1981; Davies & Brooke, 1988; Redondo & Arias de Reyna, 1988b; Redondo, 1993; Hughes, 1997; Johnsgard, 1997; Davies, Kilner & Noble, 1998; Gill, 1998; and references in these papers). However, the similarity could result from various proximate processes (as I will show) and only in some of these cases is the similarity mimetic within the generally accepted definition of mimicry: mimicry involves the mimic (e.g. parasitic chick) simulating signal properties of the model (e.g. host chick) which are perceived as signals of interest by a signal-receiver (e.g. fosterer), such that the mimic gains in fitness as a result of the signal-receiver identifying it as an example of the model (Vane-Wright, 1976, 1980; see

also Wickler, 1968). In other words, mimicry is the result of selection imposed by the signal-receiver (Wickler, 1968; also termed the operator by VaneWright, 1976; or dupe by Pasteur, 1972). Thus, mimicry is in most cases (but see later) a typical coevolutionary phenomenon, i.e. it arises from a reciprocal interaction between two or more evolutionary lineages (i.e. species or group of individuals within species, e.g. males or females – see intraspecific mimicry), with each party selecting for changes in the other (Dawkins & Krebs, 1979; Janzen, 1980). As similarity may also result from other processes, many cases of similarity between parasitic and host chicks traditionally reported in the literature as ‘mimicry’ are more usefully described as non-mimetic. The main aims of this note are to: (1) review various processes producing similarities between parasitic and host propagules, and (2) clarify the definition of parasitic coevolutionary mimicry (i.e. similarity resulting from coevolution between parasites and their hosts, but not from other evolutionary or behavioural processes; see also Rothstein, 1971; Rothstein, 1990: 485). I argue that to support the coevolutionary mimicry hypothesis one must show experimentally that a particular host rejects at least some alien eggs/ nestlings while it accepts natural parasitic eggs/nestlings more often than it does more dissimilar models or natural alien eggs/nestlings. On the one hand, during the last three decades dozens of hosts victimized by brood parasites have been tested with non-mimetic, mimetic and conspecific eggs (models or natural eggs) for their egg discrimination abilities (e.g. Rothstein, 1975; Davies & Brooke, 1989; Moksnes et al., 1990; Grim & Honza, 2001b). On the other hand, very few hosts have been tested experimentally for nestling discrimination abilities (e.g. Davies & Brooke, 1988; Langmore, Hunt & Kilner, 2003). Thus it is quite possible that ‘the rarity of chick mimicry in parasitic birds’ (Rothstein & Robinson, 1998) is just a pseudo-problem: an artefact of low research in this area (see also Grim, Kleven & Mikulica, 2003). Hopefully, clarification of the term ‘mimicry’ from the proximate point of view (i.e. ‘what process led to the similarity?’) might foster research interest in this area and could contribute to an improved design for experiments to test nestlingrelated adaptation in hosts and their parasites.

EGGS: MIMICRY OR SIMILARITY? Similarity of parasitic and host eggs might result from various processes (Table 1). (i) Phylogenetic constraints. When a host and its parasite are closely related (e.g. parasitic honey-guides, Indicatoridae, and their barbet, Capitonidae and

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

PARASITIC MIMICRY Table 1. Review of hypotheses that could explain a similarity of host and parasite eggs or nestlings Hypothesis

Eggs

Nestlings

(i) (ii) (iii) (iv) (v) (vi) (vii) (viii) (ix)

+ + + + + + – – –

+ + – + – + + + +

Phylogenetic constraints Random matching Spatial autocorrelation Nest predation Egg replacement by cuckoos Host discrimination Non-random matching Pre-existing preferences Vocal imitation

Applicability of the particular hypothesis indicated by ‘+’. Only hypotheses in bold explain mimetic similarity of parasite and host propagules. Other hypotheses explain nonmimetic similarities. All hypotheses are discussed in detail in the text.

woodpecker, Picidae, hosts, or parasitic viduines, Viduinae, and host estrildids, Estrildidae), the similarity of egg appearance is probably the result of common descent (Payne, 1967; Lack, 1968; Sorenson & Payne, 2001). A similar case can be made for facultative parasites: both black-billed cuckoos Coccyzus erythropthalmus and yellow-billed cuckoos Co. americanus, that occasionally parasitize each other, various passerines or conspecifics, have bluish-green eggs (Hughes, 1997; Lorenzana & Sealy, 2002). Obviously, no selection for similarity (genuine mimicry) in the context of intraspecific parasitism is needed. This also holds for the initial stages of intrageneric parasitism. However, although the initial similarity would be due to phylogenetic constraint, it may be maintained later by stabilizing selection. For example, in the case of Müllerian mimicry the mimetic process can act to prevent divergence of the aposematic pattern that might otherwise occur (Mallet & Joron 1999; Beatty, Beirinckx & Sherratt, 2004). Nevertheless, the case of Co. erythropthalmus and Co. americanus, discussed here, the stabilizing selection for similar egg coloration must be extremely weak as the frequency of intrageneric parasitism is very low. Furthermore, stabilizing selection is an unlikely cause of similarity of eggs of parasites and their hosts which breed in holes (barbets, woodpeckers) or dark-domed nests where physical constraint (low illumination) prevents visual discrimination (tactile discrimination has been documented only once: Mason & Rothstein, 1986). (ii) Random matching. Eggs of the brown-headed cowbird Molothrus ater sometimes resemble eggs of acceptor hosts (Rothstein & Robinson, 1998). This similarity cannot be caused by host response to parasit-

71

ism (acceptors cannot select for mimicry and female cowbirds are not host specialists, thus even host rejection cannot lead to the evolution of egg mimicry). M. ater parasitize large numbers of passerines (> 200; Rothstein, 1990), which of course show limited interspecific variation in the appearance of their eggs. Consequently, an occasional similarity between parasitic eggs (‘a general passerine type’; Lack, 1968) and host eggs might be an inevitable result of parasitism on a large pool of host species (Rothstein, 1990). Chance or accidental similarity clearly has nothing to do with ‘mimicry’ (Vane-Wright, 1976, 1981). (iii) Spatial autocorrelation in the diet of hosts and parasites. Cherry & Bennett (2001) hypothesized that similar diet or other environmental similarities could influence the coloration of both host and parasite eggs in the same way (the same explanation was first suggested by Baldamus, 1853, cited in Jourdain, 1925). Thus the similarity would not be the result of selection for colour matching. The results of Cherry & Bennett (2001) are in line with the hypothesis but this requires further testing. (iv) Nest predation. Conspicuous parasitic eggs might increase the risk of predation of hosts nests (Harrison, 1968; see also Mason & Rothstein, 1987). Thus, predation could select for an inconspicuous appearance in both host and parasite eggs; both types of eggs would consequently converge on the same coloration. Therefore, the similarity would result from convergent evolution (not coevolution), would be an example of crypsis (not mimicry) and could not be accepted as an adaptation (but as a by-product of selection for crypsis in the same environment; see also Vane-Wright, 1980). However, egg colour does not influence nest predation rates in open-nesting passerines (Weidinger, 2001) and there is also no support for this hypothesis in the most common host of the common cuckoo Cuculus canorus – the reed warbler Acrocephalus scirpaceus (Davies & Brooke, 1988). (v) Egg replacement by competing female cuckoos. Some already parasitized host nests are visited by a second cuckoo, which can remove the first parasitic egg. Such selective egg removal is advantageous as only one cuckoo chick can be raised per host nest due to parasitic nestling eviction behaviour and as a rule the first egg laid hatches first. In this case the model is the host’s egg, the mimic is the egg of the first cuckoo and the operator is the second-arriving cuckoo. The resulting mimicry can be considered as an example of class D mimicry (see Vane-Wright, 1976: 46). This is interesting as Vane-Wright (1976) considered this category, where operator and mimic belong to the same species while the model belongs to a different species, as ‘logically empty with respect to purely bio-

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

72

T. GRIM

logical systems’. Egg predation by female parasites laying in already parasitized nests was hypothesized to explain similarity between parasitic and host eggs in Horsfield’s bronze-cuckoo Chrysococcyx basalis (Brooker & Brooker, 1989, 1990), but is unlikely to be an important force behind the evolution of egg mimicry in Cu. canorus (Davies & Brooke, 1988). (vi) Host discrimination. Dissimilar parasitic eggs can be rejected by a host, which selects for egg mimicry in parasites (Davies & Brooke, 1988, 1989; Moksnes et al., 1990). The match between parasitic and host eggs is genuine mimicry and increases with increasing rejection rate (Brooke & Davies, 1988). Clearly, Cases (i) (ii) and (iii) have nothing to do with coevolution at all, Case (iv) is crypsis, and only Cases (v) and (vi) are examples of mimicry. Thus, the mere similarity of host and parasite eggs provides no evidence for mimicry. Furthermore, it is fundamental to differentiate Cases (v) and (vi), as only the latter process can be described within the framework of coevolution between brood parasites and their hosts as an interspecific asymmetric arms-race (Dawkins & Krebs, 1979). A nice example of Case (v) comes from splendid fairy-wrens (Malurus splendens) that accept even strongly non-mimetic eggs (Brooker & Brooker, 1989). However, parasitic eggs laid in their nests by Ch. basalis are highly ‘mimetic’ (see plate 3c in Davies, 2000) probably as a result of egg replacement by competing female cuckoos (Brooker & Brooker, 1990). This similarity is an example of mimicry, but it is fundamentally different from mimicry which evolved by host rejection of dissimilar eggs – mimicry in Case (vi) results from an interspecific arms-race while mimicry in Case (v) results from a conspecific arms-race (see Dawkins & Krebs, 1979). However, this hypothesis needs stronger evidence than that provided so far by Brooker & Brooker (1989, 1990).

NESTLINGS: MIMETIC OR JUST SIMILAR? In the literature on brood parasites, mimicry is the most frequently cited example of an adaptation evolved by parasites as a response to host discrimination (e.g. Rothstein & Robinson, 1998; Davies, 2000). Applying the same criteria (mimetic similarity is a coevolutionary counter-adaptation against host antiparasitic behaviour) for eggs and nestlings shows that the match between parasitic and host chicks can also be a consequence of various processes (Table 1), and many similarities are non-mimetic. (i) Phylogenetic constraints. Closely related species show similar traits not because they were selected to be so but just because they are closely related. I am aware of no published case of this phenomenon in

obligate parasitic birds. In facultative parasites, palatal structures are indistinguishable in closely related Co. americanus and Co. erythropthalmus (Nolan, 1975) that sometimes parasitize conspecifics and each other (Hughes, 1997; Lorenzana & Sealy, 2002). In an extreme case, phylogenetic inertia works in all conspecific parasites – parasitic and host nestlings are identical from the start. Chicks are similar but clearly nonmimetic. (ii) Random matching. Many cases of similarity between parasitic and host nestlings (Redondo, 1993) are unlikely to withstand close scrutiny. For example, nestling Cu. canorus begging vocalizations closely resemble (‘mimic’) those of a A. scirpaceus brood (Davies et al., 1998). This similarity seems like perfect mimicry (see sonograms in Davies et al., 1998 and Fraga, 1998). However, evidence for mimicry cannot be obtained by a comparative description: experiments are needed as parasitic coevolutionary mimicry by definition results from host (operator) discrimination (Rothstein, 1990: 485; see also Rothstein, 1971). The vocal similarity between Cu. canorus and A. scirpaceus nestlings is not mimicry as: (1) A. scirpaceus readily feeds nestlings of several other species introduced into their nests (Davies & Brooke, 1988, 1989; Davies et al., 1998), (2) Cu. canorus begging ‘mimics’ the host chicks’ begging only during a part of the nestling period when the dietary requirements of the parasite match those of the host brood and ‘mimicry’ disappears later when begging increases in old Cu. canorus, and (3) the begging call structure in Cu. canorus nestlings does not vary according to Cu. canorus host races (Butchart et al., 2003). Furthermore, various species of passerines sometimes feed a single parasitic chick despite the fact that various ‘fosterers’ have different nestling begging calls (Sealy & Lorenzana, 1997), indicating that very rough (and clearly non-mimetic) similarities of begging calls can lead to the feeding of alien chicks. The similarities of begging call rates between cuckoo and fosterers’ chicks of different host species (Butchart et al., 2003) can be explained by the fact that begging behaviour of altricial nestlings reflects offspring past experience (Kedar et al., 2000; Rodríguez-Gironés, Zuniga & Redondo, 2002). In other words, a Cu. canorus chick that is fed less often than it needs (because the host feeding rules require a higher rate of begging to provide the required level of provisioning) would increase its begging to match the call rate shown by a host’s own brood with similar food needs. A learning response by chicks, i.e. a change in the rate but not structure of begging calls, to parental provisioning provides evidence against the cuckoo chick mimicry hypothesis. Increase in begging call rates in response to social context is probably universal among altricial

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

PARASITIC MIMICRY nestlings, as shown by Rodríguez-Gironés et al. (2002), and thus cannot be considered as a specific adaptation against host antiparasitic defence. I suggest that the structural similarity between Cu. canorus and A. scirpaceus calls (Davies et al., 1998) (but not other host species studied so far) may be a consequence of: (1) a relatively low variation of nestling begging calls among altricial birds in general, and (2) parasitism of many host species – the greater the host species pool, the higher the diversity of their begging calls and, consequently, the higher the probability of random matching between parasitic and host begging calls (this is analogous to random matching in parasitic eggs; see Case (ii) below and Rothstein, 1990). Nestlings of the great spotted cuckoo Clamator glandarius are sometimes rated as mimetic in their appearance (e.g. Davies & Brooke, 1988). Host magpies Pica pica are able to discriminate against (not feed and even actively kill) parasitic nestlings, but only under experimental cross-fostering of a parasite to a previously unparasitized nest (Soler et al., 1995b). Yet P. pica discrimination works only at fledgling stage (where no ‘mimicry’ exists), and not at nestling stage (where ‘mimicry’ has been supposed; Davies & Brooke, 1988), probably because P. pica learn to recognize their offspring as those that hatched in their nests (Soler et al., 1995b). Thus there is a potential for discrimination but it cannot be used by P. pica against parasite chicks, as, under natural conditions, parasitic chicks are present in the nest before P. pica fosterers start to learn the appearance of their offspring. Moreover, under natural conditions magpies preferentially feed supernormal parasites (Redondo, 1993; Soler et al., 1995a; see also discussion in Grim & Honza, 2001a). All these data indicate strongly that the similarity between Cl. glandarius and P. pica nestlings is not mimicry, but is probably attributable to a low variance in altricial nestling appearance (to the human eye almost any altricial nestlings are similar enough to indicate the possibility of mimicry). On the other hand, vocal similarity between Cl. glandarius and its host is probably explained by a learning response in parasitic chicks (Redondo & Arias de Reyna, 1988b; see Case (ix) later in this section). Clear evidence for the vocal mimicry hypothesis could only be obtained by demonstrating that hosts discriminate against chicks with dissimilar begging calls. (iii) Spatial autocorrelation. This hypothesis most likely cannot be applied to nestlings (for details see the section on eggs). (iv) Nest predation. In principle, this hypothesis may apply to nestlings (see the logic of the argument in the section on eggs). However, I am aware of no evidence in favour of the hypothesis that predation selects for cryptic nestling plumage in altricial birds.

73

(v) Egg replacement by cuckoos. Female cuckoos can replace only eggs, thus the hypothesis can be tested only at the egg stage. (vi) Host discrimination. Some parasitic nestlings resemble host progeny in appearance (viduines: Payne, Woods & Payne, 2001; screaming cowbird Molothrus rufoaxillaris: Fraga, 1998) or vocalizations (Ch. basalis: Langmore et al., 2003) and experiments have shown that dissimilar chicks are penalized by hosts. These examples can be accepted as parasitic chick mimicry. (vii) Non-random matching – parallel evolutionary modifications in response to common factors. This hypothesis considers only vocal mimicry; see also Case (ix). Redondo & Arias de Reyna (1988a), Briskie, Martin & Martin (1999) and Haskell (1999) showed that similar environmental factors (e.g. nest-type dependent predation, habitat structure) may lead to convergence in the design of nestling begging calls. Thus, a similar environment could result in the evolution of similar begging calls in parasitic and host nestlings without any discrimination by fosterers (see also McLean & Waas, 1987). (viii) Pre-existing host preferences. When interpreting any trait within an evolutionary framework it is crucial to differentiate between its evolved original function and the current effect of the particular trait (e.g. Ryan, 1990; Futuyma, 1998). Therefore, it should be noted that no special discrimination of parasitic chicks (analogous to a discrimination of parasitic eggs) evolved by hosts in response to parasitism is needed for the evolution of similarity of parasitic and host nestlings. I suggest that if fosterers have innate preferences for certain nestling traits (e.g. red gape, Götmark & Ahlström, 1997), similarity of parasitic and host nestlings may result (e.g. both host and parasite nestlings with redder gapes would be fed more and have better survival than would those with gapes of different and less preferred colour; consequently parasite and host chicks would converge in their appearance). The resemblance resulting from pre-existing preferences should be explicitly differentiated from coevolutionary mimicry as ‘adaptations by the parasite should be called counterdefenses only if they evolved in response to host defenses’ (Rothstein, 1990) – and coevolutionary mimicry is a counterdefence by definition. Thus, a similarity resulting from pre-existing host preferences (sensory exploitation, Ryan, 1990) should be distinguished from genuine coevolutionary mimicry (see above) because it would not be the result of antiparasitic behaviour (pre-existing host preference is clearly not an antiparasitic defence) and a counteraction on the part of a parasite – the same preferences would be applied to both kinds of nestling.

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

74

T. GRIM

Under this scenario the similarity between a host and a parasite is not the result of coevolution (Janzen, 1980) but sequential evolution (Futuyma, 1998). Although adaptive (as a discrimination mechanism against a parasite), the preference is not an antiparasitic adaptation but an incidental consequence of the host’s pre-existing cognitive machinery (see also Ryan, 1990). Pre-existing preference may perhaps be moulded by selection to establish host ability to reject parasitic nestlings, but such a trait would then be better included under the ‘host discrimination’ label (preexisting preference then should be considered as preadaptation for discriminative behaviour). The ‘preference hypothesis’ can be tested against the ‘discrimination hypothesis’ mainly by phylogenetic comparative methods. The preference hypothesis predicts that the host parental preference was established before the evolution of nestling traits, i.e. the same preferences will be found in phylogenetically close species that are not parasitized (see also Ryan, 1990). A teleonomical approach (see Williams, 1966) may also be helpful – under the preference hypothesis it is expected that similarity of host and parasite traits will only be rough (e.g. red mouth colour). Complex mouth patterns seem to be redundant (from the point of view of the signalling theory) outside the context of interspecific recognition. To deliver food successfully, parents do not need to see any complex gape patterns; they only need conspicuous signals. There is no reason why a conspicuous signal should be complex. Thus, species– specific complex traits (e.g. detailed mouth patterns in Estrildidae and their Vidua parasites) could hardly be expected to evolve under some non-specific pre-existing preference and indicate specific host adaptation and specific parasite counter-adaptation (genuine coevolutionary mimicry). However, a teleonomical approach is only an auxiliary criterion which cannot replace phylogenetic comparative methods. It should be noted that pre-existing host preferences could also lead to divergence in appearance of host and parasite eggs – rufous bush chats Cercotrichas galactotes accept more model eggs with contrasting spots than they do mimetic eggs. Thus, host preferences could select attractive but non-mimetic colour patterns on parasitic eggs (Alvarez, 1999). (ix) Vocal imitation by parasitic nestling. Courtney (1967) and Mundy (1973) hypothesized that non-evicting brood parasitic nestlings (which are raised with their hosts’ young) could imitate nestling calls of their host chicks, and all parasites (both evicting and nonevicting) could imitate calls of fosterers which induce young to beg. I am aware of no clear evidence in favour of this hypothesis (but see Redondo & Arias de Reyna, 1988b); however, it provides a plausible and testable

explanation for similarities in begging calls of parasites and their hosts. Such a mechanism, if found, could be accepted as a case of mimicry only if the particular host rejected non-learning nestlings. If a researcher finds that nestlings of other non-parasitic species phylogenetically related to the particular parasite also imitate calls of their parents, the mimicry hypothesis should be rejected. Imitation would then be parsimoniously explained by phylogenetic constraint as a preadaptation ‘for’ later parasitism; see Case (i) . Only an improvement in an ability to imitate (in comparison with closely related non-parasitic species) would be accepted as mimicry. To sum up, Cases (i), (ii), (iv) and (vii) are not examples of mimicry; only processes described under Cases (vi), (viii) and (ix) could lead to mimetic similarities between host and parasite chicks. Vocal imitation by parasitic chicks should be accepted as an example of mimicry only if it was obviously selected by host discrimination against non-imitating chicks. Furthermore, coevolutionary mimicry (selected by host discrimination) and sequential evolution mimicry (selected by host pre-existing preferences) should be differentiated. It should be noted that two explanations – Cases (i) and (iv) – may not be independent. The low diversity of altricial begging displays (frequency and structure of calls, gape coloration) may not result from phylogenetic constraints (i) and can be caused by pre-existing biases of parents that feed only chicks with particular characteristics (iv). This hypothesis (‘generality of preexisting parental biases leads to convergence of begging displays across various avian taxa’) can be tested by phylogenetic comparative methods. This review has shown that: (1) mimicry in both eggs and nestlings is driven by several different processes, and (2) many similarities between host and parasitic young could be non-mimetic. The comparison of parasite and host traits can only indicate a possibility of mimicry, and focus research efforts on particular host–parasite systems. However, there is a problem with this in that very dissimilar chicks may in fact be mimetic (see later). The question, ‘is the similarity of host and parasitic chicks mimetic or not?’ cannot be solved without experimental exposure of hosts to dissimilar nestlings of other species crossfostered to their nests. When testing chick mimicry hypotheses, nestlings of some non-parasitic species should be cross-fostered into: (1) the tested host nests, and (2) other nests of their own species (to control for possible confounding effects of cross-fostering). The species chosen for cross-fostering experiments should feed its nestlings a similar diet to that of the host species and its size should be similar to that of the host species (to eliminate the possibility of malnutrition in

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

PARASITIC MIMICRY cross-fostered chicks caused by limited feeding ability of hosts). It is predicted that if the similarity of parasitic and host nestlings is mimetic then cross-fostered nestlings of a non-parasitic species will show lower survival and/or decreased growth rates in the nests of the host in comparison with their survival in nests of their conspecifics. After establishing host nestling discrimination ability, experimental manipulation of particular nestling traits can be employed to determine cues for chick recognition and discrimination (see, e.g. Soler et al., 1995b; see also Rothstein, 1982).

CONFUSION RESULTING FROM INCONSISTENCIES IN MIMICRY ASSESSMENT An inclusion in the literature of a particular parasite under the label ‘mimetic nestlings’ was based predominantly on subjective assessment of similarity of parasite to host nestlings (one does not need to give references here as the reader can see this in almost any randomly chosen paper which includes the words ‘nestling’ and ‘mimicry’; see also Introduction). Not surprisingly, such a subjective approach leads to inconsistencies in attributing mimicry. For example, the cuckoo finch Anomalospiza imberbis and Jacobin cuckoo Clamator jacobinus are included in a mimetic category of nestlings by Davies & Brooke (1988: table XVI), whereas they are treated as non-mimetic by Davies (2000: 23 and 111). The confusing nature of this comparative approach is clearly illustrated in the paper by Hughes (1997). The author concluded that eggs of Coccyzus cuckoos are mimetic because ‘blue-green eggs … fully or nearly match the eggs of over 70% of their reported host species, a proportion significantly greater than if hosts were being selected at random from the potential host pool’ (Hughes, 1997: 1380). Although such a conclusion is appealing, Lorenzana & Sealy (2002) documented an absence of differential response by hosts to nonmimetic (ancestral white-type) and supposedly mimetic (blue) eggs and falsified the mimicry hypothesis. Furthermore, Lorenzana & Sealy (2002) claimed that only 33% of hosts lay eggs that match Coccyzus eggs. These confusions and inconsistencies in approach by various authors point towards two major problems related to the issue of mimicry. These will be dealt with in the next two sections.

A PROBLEM WITH HUMAN STANDARDS One problem with a description of parasitic eggs or nestlings as mimetic or non-mimetic is that the eggs are assessed not by the eyes of a host but by the eyes of a human researcher (which are irrelevant to the

75

evolution of similarity between host and parasitic propagules; Dittrich et al., 1993). This problem is especially serious when researchers choose the design (appearance) of mimetic vs. non-mimetic model eggs in egg recognition experiments. There are three different levels of the problem. (1) Birds (but not humans) are sensitive to ultraviolet light and some parasitic eggs that appear nonmimetic in visible light are highly similar to host eggs in UV-light (Cherry & Bennett, 2001). (2) Considering only visible light cues, one can imagine that human perception and discrimination abilities are either better or worse than those of the relevant host. In the former case, dissimilarity of parasitic and host eggs for the human eye does not imply that parasitic eggs are non-mimetic – they can be mimetic in the sense that they are sufficiently similar to host eggs to fool a host (which is potentially able to reject at least some alien eggs) to accept them. On the other hand, if human perception is worse than that of a host, a close similarity of parasitic and host propagules to the human eye would clearly not imply mimicry (this possibility is indicated by the fact that some hosts reject up to 100% eggs judged as being mimetic by human observers; Moksnes & Røskaft, 1992). (3) Finally, there is an additional problem that does not relate to a quality of the human senses but to the fact that we are predominantly visual organisms. According to human judgement a pinkishyellow morph of the shining bronze-cuckoo Ch. lucidus chick is mimetic of the chicks of its host, the superb fairy-wren M. cyaneus, while the black morph is very different from the host’s nestlings. However, the pale morph is always rejected while the dark morph is often accepted (Langmore et al., 2003). The reason is that a different sensory modality (vocal cues) is used for discrimination by hosts. This indicates that human standards are unimportant for evolution (Dittrich et al., 1993). Which of many measurable parameters – egg size, shape, colour, spotting, shine – is more important for our judgement of mimicry? Is a parasitic egg that is the same colour but a different size from a host egg more (or less) mimetic than is a parasitic egg that differs in colour but matches in size? Thus, I believe that to talk about mimicry one should not compare egg parameters by eye, but be required to show that: (1) a particular host shows at least some rejection of model eggs, and (2) it accepts natural parasitic eggs more frequently than it does more dissimilar models or natural alien eggs. To determine which model is more dissimilar, the dis-

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

76

T. GRIM

crimination modality or cue (e.g. size, shape, colour) must first be established. The same principle holds for nestlings.

A PROBLEM WITH CONTINUOUS VARIATION AND DISCRETE CATEGORIES Another problem is our human need and necessity to divide continuous variables into discrete categories. This holds both for parasitic adaptations (e.g. continuous egg appearance in colour, size, spotting, is described as ‘mimetic’ or ‘non-mimetic’) and host evolved responses (e.g. continuous variation in the frequency of rejection of parasitic eggs is reduced to label of ‘acceptor’ or ‘rejecter’). Any such categorizing is inherently dependent on the individual tested and the particular circumstances and could lead to confusion when discussing results of various studies. I stress that ‘acceptor’ or ‘rejecter’ could only be a label for a particular interaction, not for an individual or species (see also Vane-Wright, 1981). Thus it is crucial to realize that the question, ‘Is A. scirpaceus an acceptor or a rejecter?’ when it ejects or deserts about 40% of Cu. canorus eggs (in my study area in the Czech Republic) has no objective answer.

CONCLUSIONS In an important note Janzen (1980) called for the retention of the usefulness of ‘coevolution’ by removing it from synonymy of usage with ‘interaction’, ‘symbiosis’, ‘mutualism’ and ‘animal–plant interaction’. For example, the dietary needs of a particular mammal could possibly have coevolved with fruit traits. But the dietary needs could also have evolved long ago before the mammal met the plant in its new habitat and started to provision on fruits that fulfilled its already established needs. Thus, the hypothesis of ‘coevolution’ needs stronger evidence than merely congruence in traits between, for example, mammal-dispersed seed and the mammal (Janzen, 1980). By the same logic, the hypothesis of ‘mimicry’ needs stronger evidence than a congruence in traits (appearance) between parasite and host eggs or nestlings. Researchers should focus more on the process behind the similarity than on the similar appearance itself. By definition, mimicry is an adaptation evolved by selection pressure from signal-receivers (Vane-Wright, 1980). Similarities resulting from other forces (e.g. phylogenetic constraint, predation) should not be dubbed ‘mimicry’. Thus, it should be stressed that ‘to resemble’ does not mean ‘to mimic’ and ‘similarity’ does not necessarily mean ‘mimicry’. When the appearance of eggs and nestlings changes over evolutionary time as a result of host discrimination, then it is more useful to understand mimicry not

as similarity of appearance but as a process – a coevolutionary process in fact (egg mimicry is either explicitly or implicitly understood as such in the brood parasitic literature, but the same cannot be said for nestling mimicry). To accept mimicry as simply a similarity of appearance (the approach frequently adopted so far) leads to two major problems: the subjectivity of human standards, and the continuous variance of mimetic or similar traits. Poor similarity (to the human eye) may be mimetic, whereas apparently close similarity may have nothing to do with mimicry (see also Vane-Wright, 1981; Dittrich et al., 1993; Cherry & Bennett, 2001). If a host (1) rejects alien nestlings, and (2) accepts parasitic ones, then these parasitic chicks are clearly mimetic even if they do not show any similarity to the human eye. To support the coevolutionary mimicry hypothesis (for both parasitic eggs and chicks) one should demonstrate that a host rejects dissimilar parasitic propagules as the mimicry is a defence against rejection evolved during coevolution between parasite and host or results from host pre-existing preferences. The use of the word ‘mimicry’ in other contexts devalues an otherwise very useful term. Differentiation between mimetic and non-mimetic similarities and various sorts of the latter is beneficial as such a process-based terminology would reflect different evolutionary dynamics of particular (non)-mimetic systems. Mimicry is a subtle concept and should not be used indiscriminately.

ACKNOWLEDGEMENTS I thank N. B. Davies, M. E. Hauber, Ø. H. Holen, M. Krist, V. Remesˇ , E. Røskaft and B. G. Stokke for valuable comments on various versions of the manuscript. Suggestions by two anonymous referees greatly improved the paper. I am especially indebted to R. I. Vane-Wright for his detailed comments and improvement to my English. Finally, I thank all whose ideas I have parasitized.

REFERENCES Alvarez F. 1999. Attractive non-mimetic stimuli in cuckoo Cuculus canorus eggs. Ibis 141: 142–144. Beatty CD, Beirinckx K, Sherratt TN. 2004. The evolution of Müllerian mimicry in multispecies communities. Nature 431: 63–67. Briskie JV, Martin PR, Martin TE. 1999. Nest predation and the evolution of nestling begging calls. Proceedings of the Royal Society of London B 266: 2153–2159. Brooke ML, Davies NB. 1988. Egg mimicry by cuckoos Cuculus canorus in relation to discrimination by hosts. Nature 335: 630–632. Brooker MG, Brooker LC. 1989. The comparative breeding

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

PARASITIC MIMICRY behaviour of two sympatric cuckoos, Horsfield’s bronzecuckoo Chrysococcyx basalis and the shining bronze-cuckoo C. lucidus, in Western Australia: a new model for the evolution of egg morphology and host specificity in avian brood parasites. Ibis 131: 528–547. Brooker MG, Brooker LC. 1990. Why are cuckoos host specific? Oikos 57: 301–309. Butchart SHM, Kilner RM, Fuisz T, Davies NB. 2003. Differences in the nestling begging calls of hosts and host-races of the common cuckoo, Cuculus canorus. Animal Behaviour 65: 345–354. Cherry MI, Bennett ATD. 2001. Egg colour matching in an African cuckoo, as revealed by ultraviolet-visible reflectance spectrophotometry. Proceedings of the Royal Society of London B 268: 556–571. Courtney J. 1967. The juvenile food-begging call of some fledgling cuckoos – vocal mimicry or vocal duplication by natural selection? Emu 67: 154–157. Davies NB. 2000. Cuckoos, cowbirds and other cheats. London: Academic Press. Davies NB, Brooke ML. 1988. Cuckoos versus reed warblers: adaptations and counteradaptations. Animal Behaviour 36: 262–284. Davies NB, Brooke ML. 1989. An experimental study of coevolution between the cuckoo, Cuculus canorus, and its hosts. II. Host egg markings, chick discrimination and general discussion. Journal of Animal Ecology 58: 225–236. Davies NB, Kilner RM, Noble DG. 1998. Nestling cuckoos, Cuculus canorus, exploit hosts with begging calls that mimic a brood. Proceedings of the Royal Society of London B 265: 673–678. Dawkins R, Krebs JR. 1979. Arms races between and within species. Proceedings of the Royal Society of London B 205: 489–511. Dittrich W, Gilbert F, Green P, McGregor P, Grewcock D. 1993. Imperfect mimicry: a pigeon’s perspective. Proceedings of the Royal Society of London B 251: 195–200. Fraga RM. 1998. Interactions of the parasitic screaming and shiny cowbirds (Molothrus rufoaxillaris and M. bonariensis) with a shared host, the bay-winged cowbird (M. badius). In: Rothstein SI, Robinson SK, eds. Parasitic birds and their hosts. Oxford: Oxford University Press, 173–193. Futuyma DJ. 1998. Evolutionary biology, 3rd edn. Sunderland: Sinauer Associates. Gill BJ. 1998. Behavior end ecology of the shining cuckoo, Chrysococcyx lucidus. In: Rothstein SI, Robinson SK, eds. Parasitic birds and their hosts. Oxford: Oxford University Press, 143–151. Götmark F, Ahlström M. 1997. Parental preference for red mouth of chicks in a songbird. Proceedings of the Royal Society of London B 264: 959–962. Grim T. 2002. Why is mimicry in cuckoo eggs sometimes so poor? Journal of Avian Biology 33: 302–305. Grim T, Honza M. 2001a. Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behavioral Ecology and Sociobiology 49: 322–329. Grim T, Honza M. 2001b. Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to exper-

77

imental parasitism by the common cuckoo Cuculus canorus. Biologia 56: 549–556. Grim T, Kleven O, Mikulica O. 2003. Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Proceedings of the Royal Society of London B 270: S73–S75. Harrison CJO. 1968. Egg mimicry in British cuckoos. Bird Study 15: 22–28. Haskell DG. 1999. The effect of predation on begging-call evolution in nestling wood warblers. Animal Behaviour 57: 893– 901. Hughes JM. 1997. Taxonomic significance of host-egg mimicry by facultative brood parasites of the avian genus Coccyzus (Cuculidae). Canadian Journal of Zoology 75: 1380–1386. Janzen DH. 1980. When is it coevolution? Evolution 34: 611– 612. Johnsgard PA. 1997. The avian brood parasites. deception at the nest. New York: Oxford University Press. Jourdain FCR. 1925. A study on parasitism in the cuckoos. Proceedings of the Zoological Society, London 44: 639–667. Kedar H, Rodríguez-Gironés MA, Yedvab S, Winkler DW, Lotem A. 2000. Experimental evidence for offspring learning in parent- offspring communication. Proceedings of the Royal Society of London B 267: 1723–1727. Komárek S. 1998. Mimicry, aposematism and related phenomena in animals and plants. Bibliography, 1800–1990. Prague: Vesmír. Komárek S. 2003. Mimicry, aposematism and related phenomena – mimetism in nature and the history of its study. München: LINCOM. Lack D. 1968. Ecological adaptations for breeding in birds. London: Methuen. Langmore NE, Hunt S, Kilner RM. 2003. Escalation of a coevolutionary arms race through host rejection of brood parasitic young. Nature 422: 157–160. Lorenzana JC, Sealy SG. 2002. Did blue eggs of Black-billed (Coccyzus erythropthalmus) and Yellow-billed (C. americanus) cuckoos evolve to counter host discrimination? Auk 119: 851–854. Mallet J, Joron M. 1999. Evolution of diversity in warning color and mimicry: polymorphisms, shifting balance, and speciation. Annual Review of Ecology and Systematics 30: 201–223. Mason P, Rothstein SI. 1986. Coevolution and avian brood parasitism: cowbird eggs show evolutionary response to host discrimination. Evolution 40: 1207–1214. Mason P, Rothstein SI. 1987. Crypsis versus mimicry and the color of shiny cowbird eggs. American Naturalist 130: 161–167. McLean IG, Waas JR. 1987. Do cuckoo chicks mimic the begging calls of their hosts? Animal Behaviour 35: 1896–1898. Moksnes A, Røskaft E. 1992. Responses of some rare cuckoo hosts to mimetic model cuckoo eggs and to foreign conspecific eggs. Ornis Scandinavica 32: 17–23. Moksnes A, Røskaft E, Braa AT, Korsnes L, Lampe HT, Pedersen HC. 1990. Behavioural responses of potential hosts towards artificial Cuckoo eggs and dummies. Behaviour 116: 64–89.

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

78

T. GRIM

Mundy PJ. 1973. Vocal mimicry of their hosts by nestlings of the great spotted cuckoo and striped crested cuckoo. Ibis 115: 602–604. Nolan V. 1975. External differences between newly hatched cuckoos (Coccyzus americanus and C. erythropthalmus). Condor 77: 341. Pasteur G. 1972. Le mimétisme. Que sais-je? Paris: PUF. Payne RB. 1967. Interspecific communication signals in parasitic birds. American Naturalist 101: 363–375. Payne RB, Woods JL, Payne LL. 2001. Parental care in estrildid finches: experimental tests of a model of Vidua brood parasitism. Animal Behaviour 62: 473–483. Redondo T. 1993. Exploitation of host mechanisms for parental care by avian brood parasites. Etología 3: 235–297. Redondo T, Arias de Reyna L. 1988a. Locatability of begging calls in nestling altricial birds. Animal Behaviour 36: 653–661. Redondo T, Arias de Reyna L. 1988b. Vocal mimicry of hosts by great spotted cuckoo Clamator glandarius: further evidence. Ibis 130: 540–544. Rodríguez-Gironés MA, Zuniga JM, Redondo T. 2002. Feeding experience and relative size modify the begging strategies of nestlings. Behavioral Ecology 13: 782–785. Rothstein SI. 1971. Observation and experiment in the analysis of interactions between brood parasites and their hosts. American Naturalist 105: 71–74. Rothstein SI. 1975. An experimental and teleonomic investigation of avian brood parasitism. Condor 77: 250–271. Rothstein SI. 1982. Mechanisms of avian egg recognition: which egg parameters elicit responses by rejecter species? Behavioral Ecology and Sociobiology 11: 229–239. Rothstein SI. 1990. A model system for coevolution: avian brood parasitism. Annual Review of Ecology and Systematics 21: 481–508. Rothstein SI, Robinson SK, eds. 1998. Parasitic birds and their hosts. Oxford: Oxford University Press. Ryan MJ. 1990. Sexual selection, sensory systems and sen-

sory exploitation. In: Futuyma D, Antonovicz J, eds. Oxford surveys in evolutionary biology, Vol. 7. Oxford: Oxford University Press, 157–195. Sealy SG, Lorenzana JC. 1997. Feeding of nestling and fledgling brood parasites by individuals other than the foster parents: a review. Canadian Journal of Zoology 75: 1739– 1752. Soler M, Martinez JG, Soler JJ, Møller AP. 1995a. Preferential allocation of food by magpies Pica pica to great spotted cuckoo Clamator glandarius chicks. Behavioral Ecology and Sociobiology 37: 7–13. Soler M, Soler JJ, Martinez JJ, Møller AP. 1995b. Chick recognition and acceptance: a weakness in magpies exploited by the parasitic great spotted cuckoo. Behavioral Ecology and Sociobiology 37: 243–248. Sorenson MD, Payne RB. 2001. A single ancient origin of brood parasitism in African finches: implications for hostparasite coevolution. Evolution 55: 2550–2567. Swynnerton CFM. 1916. On the coloration of the mouths and eggs of birds. II. On the coloration of eggs. Ibis 4: 529– 606. Vane-Wright RI. 1976. A unified classification of mimetic resemblances. Biological Journal of the Linnean Society 8: 25–56. Vane-Wright RI. 1980. On the definition of mimicry. Biological Journal of the Linnean Society 13: 1–6. Vane-Wright RI. 1981. Only connect. Biological Journal of the Linnean Society 16: 33–40. Weidinger K. 2001. Does egg colour affect predation rate on open passerine nests? Behavioral Ecology and Sociobiology 49: 456–464. Wickler W. 1968. Mimicry in plants and animals. London: Weidenfeld and Nicolson. Williams GC. 1966. Adaptation and natural selection. Princeton: Princeton University Press. Wyllie I. 1981. The cuckoo. London: Batsford.

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 84, 69–78

2. Grim T. 2006: The evolution of nestling discrimination by hosts of parasitic birds: why is 2. so rare? rejection Grim T. 2006: The evolution of nestlingEcology discrimination of parasitic birds: why is Evolutionary Researchby (inhosts press). rejection so rare? Evolutionary Ecology Research (in press).

The evolution of nestling discrimination by hosts of parasitic birds: why is rejection so rare? Tomáš Grim1,2 School of Biological Sciences University of Auckland PB 92019 Auckland New Zealand 1

Department of Zoology Palacky University tr. Svobody 26 CZ-771 46 Olomouc Czech Republic e-mail: [email protected] (correspondence) 2

ABSTRACT Question: Why do hosts of parasitic birds defend against parasitic eggs but not nestlings? Data incorporated: Reported cases of parasitic chick discrimination or mimicry and all previously published explanations for the rarity of these phenomena. Analysis method: Contrasting the predictions of previous hypotheses and fitting available data from both parasitic and non-parasitic birds to assess the relative validity of each explanation. Results: None of the previously suggested hypotheses appears to provide a general explanation for the scarcity of chick discrimination. Various cognitive and behavioral traits potentially usable for discrimination of parasitic chicks are present in virtually all avian taxa, including host lineages, yet these traits are not used to reject parasites. Thus, low selection pressure imposed by rare parasites is the most likely general explanation for the absence of these adaptations in the context of brood parasitism. Based on this I predict that nestling discrimination and mimicry should predominantly evolve in hosts that are forced to accept parasite eggs because of the close match between parasitic and host eggs. This is likely to occur due to egg mimicry or phylogenetic and physical constraints. I demonstrate that available evidence is in line with this rare parasite hypothesis. Conclusion: A host’s own behaviour may play a crucial role in retarding the escalation of the arms-race to the nestling stage. Keywords: arms-race, brood parasitism, coevolution, discrimination, evolutionary lag, evolutionary equilibrium, recognition

1

INTRODUCTION The apparent absence of adaptations in response to strong selection pressures is puzzling. One widely discussed example is the lack of nestling discrimination by most hosts of parasitic birds – both scientists and laypersons wonder at the picture of a miniature passerine feeding a huge cuckoo chick (Rothstein and Robinson, 1998; Davies, 2000). On the one hand, many hosts of parasitic birds evolved very fine discrimination against parasite adults (Sealy et al., 1998) and eggs (Davies, 2000). On the other hand, parasitic nestling discrimination has been documented only a few times (Soler et al., 1995a; Fraga, 1998; Lichtenstein, 2001a; Payne et al., 2001; Langmore et al., 2003; Grim et al., 2003). This rarity of nestling discrimination is even more enigmatic because other phenotypic changes and behavioral traits involved in arms-races (Dawkins and Krebs, 1979) between brood parasites and their hosts provide textbook examples of adaptations and coevolution (e.g. egg mimicry by cuckoo host races: Alcock, 1998; Futuyma, 1998). In this review, I summarize current evidence for nestling discrimination by host parents and discuss all proposed explanations for the “rarity of chick discrimination” enigma. I will argue that the previous explanations are applicable only under limited circumstances and will propose a general hypothesis that applies to all host–brood parasite systems. AVAILABLE EVIDENCE FOR CHICK DISCRIMINATION As “recognition” I define the internal process which can (but need not, see Sherman et al., 1997; Soler et al., 1999; Mateo, 2002) lead to behavioral discrimination. By “discrimination” and “rejection” I mean the behavioral acts of differential response to two stimuli (Beecher, 1991). Further, I differentiate between “nestling rejection” which always results in nestling death (e.g. nest desertion or chick removal) and “nestling discrimination” which does not (e.g. differential parental allocation of food within parasitized brood). While nestling rejection is a specific co-evolved response to parasitism, nestling discrimination may also result from inability of parasitic chicks to communicate their state of hunger effectively to fosterers. The clear-cut cases of chick rejection are scarce (Table 1). Langmore et al. (2003) reported that Australian superb fairy-wrens Malurus cyaneus desert 40% of nests with the Horsfield’s bronze cuckoo Chrysococcyx basalis chick and all nests with the shining bronzecuckoo Ch. lucidus. This host behavior selected for nestling vocal mimicry in Horsfield’s bronze-cuckoos (Langmore et al., 2003). African estrildids are parasitized by mimetic Vidua finches and discriminate against alien nestlings other than that of their particular mimetic parasite (Nicolai, 1964; Payne et al., 2001; Schuetz, 2005). South American bay-winged cowbirds Agelaioides badius discriminate fledglings of the generalist parasite the shiny cowbird Molothrus bonariensis but rear chicks of the specialist parasite the screaming cowbird M. rufoaxillaris (Fraga, 1998; Lichtenstein 2001b). Shiny cowbird chicks grow normally in the nests of this host but are provisioned very poorly after fledging and usually do not survive more than 24 hours outside host nests. In contrast, screaming cowbird fledglings are provisioned fully and appear very similar to host bay-wing fledglings. This indicates that similarity of bay-winged and screaming cowbird chicks is mimetic (Grim, 2005). The low success of shiny cowbird chicks – almost 70% of chicks starve to death – in nests of rufous-bellied thrushes Turdus rufiventris also involves active parental discrimination to favour own versus parasitic chicks (Lichtenstein, 2001a). Levaillant’s cuckoo Clamator levaillantii has a begging call resembling the vocalizations of host, the arrow-marked babbler Turdoides jardineii, young (Mundy, 1973). Fosterers sometimes attack the cuckoo but not host young in fledged broods but when the cuckoo begins to beg, the hosts stop the attacks and start to feed the parasite (Redondo, 1993; Payne, 2005 and references therein). These behavioral interactions clearly indicate that the babbler host is able to discriminate against the parasite and the Levaillant’s cuckoo chick is able to effectively counter the discrimination with calls mimicking those of the host’s own young. Reed warblers Acrocephalus scirpaceus desert 15% of older common cuckoo Cuculus canorus chicks in a study area in the Czech Republic (Grim et al., 2003). Reed warblers discriminate parasitic cuckoo chicks by restricting their parental care at the nest to ~14 days which is enough to rear their own nestlings (that fledge when 10–11 days old) but not enough to rear the cuckoo (18–21 days). In this system discrimination does not involve 2

chick recognition (Grim et al., 2003) and the cue triggering this “discrimination without recognition” behaviour is the duration of parental care (Grim unpubl. data). Neither the intensity of this care nor the presence of a single chick in the nest could explain desertions (Grim unpubl. data). Davies and Brooke (1989) cross-fostered chaffinch Fringilla coelebs chicks to dunnock Prunella modularis nests. Chicks either disappeared or grew subnormally while host young survived well. This may indicate some sort of parental discrimination against alien chicks (see later for discussion). Indirect but suggestive evidence for parasitic chick discrimination and chick mimicry was found for host grey warblers Gerygone igata vs. parasitic shining bronze-cuckoos (McLean and Waas, 1987; Gill, 1998) and host house crows Corvus splendens vs. parasitic koel Eudynamys scolopacea (Dewar, 1907; Payne, 2005 and references therein). Parasitic chick appearance and begging varies geographically with the appearance of host nestlings in these systems. Additionally, the begging calls of the diederik cuckoo Chrysococcyx caprius covary with begging calls of its various hosts (Reed, 1968) suggesting a response to host discrimination. McLean and Waas (1987) also described noticeable similarity of begging calls of host Mohoua spp. vs. the parasitic long-tailed cuckoo Urodynamis taitensis chicks. Jourdain (1925) and Lack (1968) made a case for chick mimicry in the Jacobin cuckoo Clamator jacobinus and the great spotted cuckoo Clamator glandarius (see also Mundy 1973). Accordingly, in the Jacobin cuckoo Gaston (1976) reported that host “babblers began to lose interest in the young cuckoo in the last few days before it left the nest” (p. 334) and mentioned cases when cuckoo fledgling was ignored by its fosterers. For the great spotted cuckoo Redondo (1993) and Soler et al. (1995a) reported experimentally induced rejection of parasitic chicks by magpie Pica pica hosts. However, discrimination of foreign chicks does not occur under natural conditions and therefore cannot select for any chick mimicry (for a discussion see Grim, 2005). Giant cowbird Scaphidura (Molothrus) oryzivora nestlings are similar to nestlings of their oropendola (Psarocolius spp.) and cacique (Cacicus spp.) hosts and the morphological similarity disappears after parasite chicks gain independence from fosterers which is suggestive of mimicry (Redondo 1993). Unfortunately, hosts’ responses to parasite chicks are unknown. Geographical variation in rictal flange color of brown-headed cowbird Molothrus ater nestlings was hypothesized to reflect chick rejection and/or discrimination by hosts (Rothstein 1978). However, there is no evidence that any hosts of that parasite adjust their provisioning rates according to the rictal flange color (Ortega 1998). Thus, the geographical variation among subspecies of the brown-headed cowbird more likely resulted from nonadaptive evolutionary processes, namely founder effects or random drift. Davies and Brooke (1988) included the cuckoo finch Anomalospiza imberbis in their list of mimetic parasites (their table XVI). However, there is no resemblance between the parasitic cuckoo finch young and host cisticola (Cisticola spp.) and prinia (Prinia spp.) offspring (Davies 2000, p. 23). For more details on chick discrimination and mimicry in general see Redondo (1993) and Grim (2005). In general, host behavior towards parasitic chicks received very poor attention in observational or experimental studies. Absence of evidence is not evidence of absence, thus it is possible that chick discrimination and rejection are more common than available evidence indicates. However, there are theoretical reasons to expect host defenses against alien nestlings to be rarer than adaptations against alien eggs. I will discuss these reasons in the framework of evolutionary lag, proximate constraints, host exploitation and evolutionary equilibrium hypotheses (Table 2). THE EVOLUTIONARY LAG HYPOTHESIS (ELH) The ELH suggests that no rejecter mutations are present in the host gene pool or that such mutations have not yet spread (Rothstein, 1982; Hosoi and Rothstein, 2000). Failure to respond to selection Mutations are by definition random with respect to possible adaptive improvement (Futuyma, 1998: p. 282). In general, natural selection requires heritable variation correlated

3

with fitness variation, and there is no guarantee that such variation exists. The absence of such variation may even lead to extinction (Futuyma, 1998: p. 713). Some acceptor hosts of the brown-headed cowbird have been in contact with the parasite for a long time and show various pre-adaptations (large nests, large bills, eggs dissimilar to those of cowbirds) known to facilitate the evolution of egg rejection. Yet, phylogenetically close species have evolved egg rejection (Peer and Sealy, 2004). In the absence of any apparent constraints (but see Hauber et al., 2004) on the evolution of egg rejection we are left with only one explanation: chance plays an important role in evolution (Rothstein, 1975, 1982). Kemal and Rothstein (1988) supported this idea by showing that acceptors are able to differentiate between broken and intact eggs and to eject broken eggs, but they are unable to use this behavior in the context of brood parasitism even when heavily parasitized (Hauber, 2003). Similarly, dunnocks are able to remove dead chicks from their nests but accept both cuckoo eggs and nestlings (Davies and Brooke, 1989; but see later). What is probably lacking is a mutational change in their decision-making mechanisms (see also Sherman et al., 1997; Hauber and Sherman, 2001). Lack of pre-adaptations The lack of any adaptation may be in principle explained by a lack of suitable preadaptations. The quantitative change in an existing behavior is more probable than the emergence of a qualitatively new behavioral pattern (Dawkins, 1982; Hosoi and Rothstein, 2000). For example, nest desertion in response to brood parasitism is easy to evolve because it requires only a small change in triggering of pre-existing desertion behavior that evolved in response to various nest intruders and predators (Hosoi and Rothstein, 2000). Similarly the evolution of egg ejection is a logical extension of nest sanitation behavior (Rothstein, 1975, Moskát et al., 2003). In contrast, Redondo (1993) argued that nestling rejection could not be expected to evolve from sanitation behavior because parents only remove dead nestlings from their nest – ejection of living nestlings would be strongly selected against when signs indicate that chicks are healthy. However, ejection of living eggs must also be strongly selected against when there are signs that eggs are “healthy” (unbroken). There is no reason to assume that parents can remove eggs if they carry new stimuli (cracks when they are broken or different color when they are parasitic) and simultaneously claim that they cannot do the same with nestlings. In fact removal of living nestlings in passerines was observed (Robertson and Stutchbury, 1988; Møller, 1992). Thus, not only nest sanitation behavior (contra Redondo, 1993) but also brood reduction (Clotfelter et al., 2000), offspring desertion (Székely et al., 1996) and infanticide (Crook and Shields, 1985; Møller, 1992) could provide behavioral preadaptations for adaptive response to parasite nestlings. Most importantly, currently there are no methods available to test whether there has been insufficient time or frequency of beneficial mutations to improve host pre-adaptations. Thus, the ELH can be considered as an explanation of the last resort (Davies, 2000; Hosoi and Rothstein, 2000). Therefore I will pay more attention to other, testable explanations. THE PROXIMATE CONSTRAINTS HYPOTHESIS (PCH) Various physical and cognitive constraints may affect evolution of chick discrimination (Table 2). For other constraints that apply to both egg and chick rejection see Table 1 in Grim (2002). Physical constraints Theoretically, a host might reject a parasitic egg but accept a parasitic chick because the chick is much heavier than an egg to be ejected (for physical constraints on egg rejection see Rohwer and Spaw, 1988; Røskaft et al., 1993). Such a behavior may be adaptive in hosts of non-evicting parasites where at least some host chicks survive to fledging despite the competition for food with a parasitic chick (Davies, 2000). However, even small birds can remove both relatively large (e.g. Davies and Brooke, 1989) and living (e.g. Robertson and Stutchbury, 1988; Møller, 1992) nestlings of other pairs from their nests. Some fosterers may refuse to feed experimentally cross-fostered nestlings and even peck them to death (Holcomb, 1979; Redondo, 1993; Soler et al., 1995a).

4

One can argue that chick ejection by hosts of the common cuckoo would provide little benefits as cuckoo chicks evict host clutch or brood. However, a substantial delay between hatching and the start of eviction of the hosts eggs/nestlings by the cuckoo chick (mean 20.5 hours in the nests of the reed warbler; M. Honza, K. Voslajerova, unpubl. data) leaves enough time to recognize and eject a parasitic chick. Thus, the main problem for a host is presumably not the physical removal of parasitic chick per se, but the recognition of parasitic nestlings. Learning constraints Egg recognition in some bird species may be learned by an imprinting-like process (Lotem et al., 1992, 1995; see also Victoria, 1972; Rothstein, 1974). Such behavior would be adaptive even when the host is parasitized during its first breeding attempt (Lotem, 1993). In contrast, imprinting on nestling appearance is not adaptive in hosts of evicting parasites because the only object for host to imprint on would be the parasite (Lotem, 1993; but may be adaptive in hosts of non-evicting parasites, see Lawes and Marthews, 2003). Nevertheless, in various birds first-year naïve breeders are already able to reject alien eggs which indicates that imprinting is not an universal basis for egg recognition (Davies and Brooke, 1988; Stokke et al., 1999; Marchetti, 2000; Soler et al., 2000; Amundsen et al., 2002; Stokke et al., 2004; see also Mark and Stutchbury, 1994 for recognition of adult parasites by first-time breeding females). Thus, Lotem’s hypothesis can be viewed as an argument about cognitive constraints in a few host species where parents learn chick appearance. Redondo (1993) stated that in species that reject alien eggs there are no obvious reasons why hosts should not use “an already existing set of egg recognition mechanisms” to reject parasite chicks. However, the validity of this argument is limited for two reasons. First, in hosts where discrimination is learned naïve breeders tend to accept more parasitic eggs than old breeders who tend to reject alien eggs (Lotem et al., 1992, 1995). Therefore, old hosts that have “an already existing set of egg recognition mechanisms” (Redondo, 1993) would have only low chance to face the young cuckoo because the parasite was probably destroyed at the egg stage. In contrast, naïve breeders that perform poorly in egg discrimination should also perform poorly in nestling discrimination (under assumption that egg and nestling discrimination mechanisms are similar; Redondo, 1993). Thus, situations when the parasitic nestling could potentially be rejected should be very rare – both in naïve and experienced breeders. Consequently, the selection for chick mimicry would be very weak despite the existence of host egg discrimination abilities. To make things worse for the host species, there are good reasons to expect that young naïve breeders may be parasitised more frequently than old experienced individuals (Brooker and Brooker, 1996; Grim, 2002). Second, in hosts with innate discrimination (i.e. independent of experience), different sensory modalities may serve recognition of eggs vs. nestlings. Eggs may be recognized by visual or tactile cues (Langmore et al., 2005), while nestling discrimination is more likely based on vocal cues (Langmore et al., 2003). Thus, “an already existing set of egg recognition mechanisms” (Redondo, 1993) would be again unavailable for chick discrimination. If both egg and chick discrimination were based on the same cues (e.g. visual) then even a naïve host would have a low chance to face the parasite chick due to the temporal order of its innate defences (i.e., the parasite would be killed at the egg stage thus leaving nothing to be discriminated at the chick stage). Chick variability constraint McLean and Maloney (1998) suggested that low interspecific variability of physical appearance of altricial nestlings would lead to unacceptably high recognition costs in chick rejecters. However, birds are capable of well-developed discrimination of individual conspecifics (e.g. their offspring; Beecher et al., 1981) based on intraspecific variation, not to speak of interspecific discrimination. Even when such discrimination takes place at a later developmental stage (fledging), differences between two host offspring at any stage are surely much smaller than differences between parasitic and host young (one exception – chicks of Vidua parasites are well within the intraspecific variability of the host species chicks’ traits – is the result of previous host selection that led to excellent mimicry;

5

Sorenson and Payne, 2001). It is highly unlikely that birds are capable of cognitive feats enabling discrimination of conspecifics, but not heterospecifics.

Appearance stability constraint Davies and Brooke (1988) argued that eggs look the same throughout incubation while nestlings’ phenotypes change during growth making chick discrimination hard. However, at least in some species eggs may change their appearance during laying and incubation – they get soiled and become less transparent (e.g. Turdus spp.; own observations; see also Fig. 1 in Mason and Rothstein, 1986). Parents’ cognitive ability to respond to minor differences in continually changing nestling morphology and vocalizations when allocating food within a brood (Wright and Leonard, 2002) also casts doubt on the “changing chick appearance” argument. Simultaneous comparison constraint The idea that discrimination is easier if there is a model present for comparison (Dawkins, 1982) was supported by the finding that host nestling discrimination was observed mainly in non-evicting parasites (Davies, 2000; but see Table 1), where hosts can compare their own nestlings with the alien nestlings. However, to discriminate effectively, birds do not need any comparative material: hosts of both evicting and non-evicting parasitic birds are able to reject the entire parasite clutch (exchanged by experimenters for an original one) even without any of their own eggs present (e.g. Victoria, 1972; Rothstein, 1974; Lotem et al., 1995; Lahti and Lahti, 2002). More importantly, estrildids (hosts of non-evicting Vidua finches) can discriminate against a whole brood of another species with no conspecific young for comparison (Payne et al., 2001). In addition, there are cognitive systems that could well work in the context of parasitic chick discrimination without any host own nestlings present for comparison: (1) discrimination can be innate (parent-offspring recognition: Sherman et al., 1997; Langmore et al., 2003; mate recognition: Slagsvold et al., 2002; predator recognition: Mark and Stutchbury, 1994; Veen et al., 2000), (2) discrimination can be based on individuals’ own phenotype (self-referent phenotype matching: Hauber and Sherman, 2001; Mateo, 2002), (3) a host could theoretically learn begging vocalizations of its nestmates when it is nestling itself; later when it starts to breed on his/her own the host could compare this learned template (Sherman et al., 1997) with chick vocalizations and reject any chick(s) with mis-matching begging calls, (4) increasing evidence shows that songbirds use olfaction in fine scale discrimination (e.g. Petit et al., 2002) and thus a host could also imprint on its nestmates’ scent as a recognition cue, (5) a future host can also learn from the appearance of its nestmates (Soler and Soler, 1999) but only after its eyes open several days after hatching. Because appearances of the hatchlings and chicks may change quickly over several days in some species, it is questionable whether such a mechanism could be used against newly hatched parasitic nestlings. However, discrimination based on the learning of nestmate’s appearance could work against older parasitic chicks, (6) both hosts of evicting and non-evicting parasites may employ passwordrecognition (Hauber et al., 2001). Here, innate predisposition to recognize a specific cue (“password”) triggers learning of other traits of the password-giver’s phenotype. According to this scenario even host nestlings that are reared alongside parasite nestling(s) will be able to identify and learn about conspecific traits without mistakenly incorporating parasitespecific traits into their recognition template; see above hypothesis (3) to (5). Most importantly, observations of naïve first-time breeders rejecting alien eggs (see above) strongly point to the possibility that the recognition of host’s progeny can be experience-independent, supporting hypothesis (1). This is in line with increasing awareness of the importance of non-imprinting recognition mechanisms in studies of recognition in birds (Göth and Hauber, 2004).

6

To sum up, both behavioral and cognitive constraints can hardly provide any general explanation for the paucity of chick discrimination. THE HOST EXPLOITATION HYPOTHESIS (HEH) According to the HEH the main problem for hosts is that they recognize their offspring based on the very same traits that signal offspring food requirements (Redondo, 1993). Signals provided by parasitic nestlings are misinterpreted by hosts as signals of high quality offspring (Redondo, 1993). Discrimination of alien nestlings would require gross change of a host’s cognitive system to make a food allocation system (“this nestling is hungrier”) different from a nestling discrimination system (“this nestling is mine”). The validity of HEH has been questioned by data from magpies raising great spotted cuckoos. Magpies recognize their own nestlings as those present in the nest (Soler et al., 1995a), but are also able to discriminate against cuckoo nestlings introduced to nonparasitized broods at the end of the nestling period. This clearly shows that magpies can discriminate among nestlings according to both their hunger level and their appearance – they can ignore alien nestlings despite their intense begging. These observations indicate that “food allocation” and “nestling recognition” are two separate decision making mechanisms. Thus, both Redondo’s host exploitation model and Lotem’s imprinting model (see above) are essentially arguments about cognitive constraints. As cognitive constraints are easily modified by selection (Sherman et al., 1997), these models have only limited applicability for evolutionary hypotheses. Similarly to the HEH the supernormal stimulus hypothesis (Dawkins and Krebs, 1979) suggested that cuckoo intense begging manipulates host behavior so that “the host can no more resist the supernormal manipulative power of the cuckoo nestling than the junkie can resist his fix” (Dawkins and Krebs, 1979). While a cuckoo chick provides hosts with the supernormal stimulus (Grim and Honza, 2001), this supernormal behavior is not an adaptation against host discrimination but serves to compensate for the subnormal visual component of the begging signal (Kilner et al., 1999). THE EVOLUTIONARY EQUILIBRIUM HYPOTHESIS (EEH) The EEH states that rejection costs outweigh or balance parasitism costs causing a stable equilibrium maintained by stabilizing selection (Rohwer and Spaw, 1988; Lotem et al., 1992, 1995; Brooker and Brooker, 1996; Lawes and Marthews, 2003). Hosts can accept parasitism when parasites punish rejecters (Soler et al., 1995b, 1999) or when parasitism provides benefits (e.g. parasitic chicks may remove ectoparasites from host chicks; Smith, 1968). Raising a parasitic chick may also be a handicap indicating parental qualities of parasitized individuals (Zahavi and Zahavi, 1997). The handicap idea could hardly work in cuckoo or cowbird hosts as these have typically short life-spans and costs of raising a parasitic chick are too high, but tests of this hypothesis are not yet available. More realistic scenarios under the EEH are asymmetry in benefits of egg vs. chick discrimination and a low selection for anti-parasitic adaptations at the chick stage. Asymmetry of benefits of egg vs. chick discrimination Selection for host anti-parasitic adaptations is stronger at the egg stage than at the nestling stage because egg rejection provides greater benefits (Davies and Brooke, 1988). After a cuckoo chick hatches it may be too late in the breeding season for renesting (Dawkins and Krebs, 1979; Davies and Brooke, 1988). This may hold for some species breeding only during a short time window of northern summer (Moksnes et al., 1993). However, time delay between egg rejection and potential hatchling rejection by desertion (which is the most costly alternative) would be only about 10 days in reed warblers. This would probably have a minimal effect on differences in overall breeding success of egg vs. nestling rejecters in the reed warbler. Costs of breeding later in the season (average loss of 0.20 eggs and 0.38 young for 10 days delay in breeding due to declining breeding success during the season; Øien et al., 1998) would be off-set by strongly declining probability of nest predation at warbler nests (see Fig. 1 in Davies and Brooke, 1988). Even no re-nesting would provide benefits for deserters: no costs of prolonged care for the parasite and increased probability of survival due to lack of breeding (Nur, 1984). Therefore, selection should favour antiparasitic behavior even after host offspring were evicted by the parasite. If costly nest

7

desertion at the egg stage frequently evolved as a response to parasitism (Davies and Brooke, 1989; Hosoi and Rothstein, 2000), there is no reason why nest desertion at the nestling stage should not be favored as well. Thus, asymmetry of benefits between egg and nestling stages is unlikely to explain alone the absence of nestling rejection in cuckoo hosts. The rarer enemy hypothesis: rare eggs vs. rarer chicks In general, intensity of selection is manifested in the cost of not responding to a selection pressure and the probability of facing this pressure. Even the most deleterious event for an animal provides negligible selection if it is sufficiently rare. This might provide the clue for understanding the rarity of parasite nestling discrimination. As egg discrimination provides both greater benefits and is less costly (in terms of cognitive costs and costs of recognition errors) compared to nestling discrimination (Davies and Brooke, 1988), it should be selected more strongly and spread more quickly after a naïve host starts to be exploited by the parasite (however, this alone cannot explain the rarity of chick rejection; see above). Thus, the egg discrimination threshold (i.e. the frequency of encountering a parasite at the egg stage needed for positive selection of host defenses against eggs; see also Sherman et al., 1997) is clearly lower than the chick discrimination threshold (i.e. the frequency of encountering a parasite at the chick stage needed for positive selection of host defenses against chicks) (Fig. 1). After the parasite starts to exploit a particular host (Fig. 1a), the parasitism rate must reach some level (Fig. 1b) to select for host defenses, i.e. benefits of anti-parasitic response must exceed costs of defense. Increase in host defenses decreases fitness of the parasite. This could lead to parasitism rates sufficiently low to select against host defenses (Fig. 1c; Soler et al., 1998; Takasu, 1998). This in turn lowers selection pressure on the parasite and parasitism rates increase (Fig. 1d; Brooker et al., 1990). However, the selection pressure is always stronger on the evolution of parasite adaptations than on host counter-adaptations (Dawkins and Krebs, 1979; Servedio and Lande, 2003). Therefore, the parasite can win the battle against the host at the egg stage in the long term by evolving “perfect” mimicry, i.e. when similarity of parasite and host eggs does not leave any signature cues (Beecher, 1991) available for recognition. Now parasitism rate – unconstrained by host defenses – should increase (Fig. 1e). Only then may the parasitism rate overcome the chick discrimination threshold and the evolution of nestling discrimination is expected (Fig. 1f). If mutation is random, then even novel hosts may defend against parasitism not with egg rejection but with nestling rejection. If egg rejection later arises, it will spread more quickly than nestling rejection. This may decrease effective parasitism rate at the chick stage below the chick discrimination threshold. Importantly, the evolution of egg rejection is not prohibited by the increasing frequency of nestling rejection, i.e. egg rejection constrains nestling rejection both at individual and population levels but not vice versa. Selection acting on nestling discrimination in an egg rejecter is frequency-dependent on the rate of egg rejection. Parasite egg rejection renders few opportunities that favour nestling recognition in comparison to egg recognition. Cuckoo eggs are “rare enemies” (Dawkins and Krebs, 1979). However, cuckoo nestlings are even “rarer enemies”. Some parasitic eggs are rejected by hosts (e.g. reed warblers, 38%), some accepted eggs are depredated with the host clutch (23%), or ejected by another cuckoo (4%), and some are infertile, laid too late in the breeding cycle of the host, or do not hatch for some other reason (2%) (Øien et al., 1998; Øien et al. pers. comm.). All these processes cumulatively relax selection on the evolution of nestling discrimination. Even egg acceptors face fewer chicks than eggs due to egg predation. The rarer enemy hypothesis predicts that unless a host stops rejecting parasite eggs, it is unlikely that selection pressure at the nestling stage will reach a level necessary for positive selection of costly anti-nestling adaptations. In strong rejecters (e.g. Phylloscopus warblers; Moksnes and Røskaft, 1992) there is virtually no selection for nestling discrimination. In contrast, almost all hosts listed in Table 1 are probably 100% acceptors of natural parasitic eggs, which supports the hypothesis. The only egg rejecter reported to discriminate parasitic chicks is the reed warbler (Table 1). Reed warblers from the population where chick discrimination was reported reject 8

~40% of naturally laid parasitic eggs (Øien et al., 1998), but the high parasitism rates in this particular population (~15%) suggest the defense is not very effective. However, the warbler adopts a strategy of delayed chick discrimination, probably to ensure itself that it has a cuckoo in the nest. No costs of recognition errors were found so far: reed warblers were reported never to desert one-warbler-nestling broods (Davies, 2000, Grim and Honza, 2001). The advantage of error-free defense bears the cost of long care for the parasite before desertion, which should slow down the spread of chick rejection. Further, this study population is most likely parasitized by cuckoos only for a short time (Honza et al., 2004) which may also explain coexistence of two antiparasitic strategies (see Planqué et al., 2002). The dunnock is a 100% acceptor of cuckoo eggs. In turn, 3 out of 7 chaffinch chicks transferred to dunnock nests by Davies and Brooke (1989) disappeared within 3 days and even surviving chicks grew subnormally. In contrast, the reed bunting Emberiza schoeniclus accepts 0% of parasitic eggs and reed warbler chicks transferred to two reed bunting nests grew well (Davies and Brooke, 1989). These differences are expected under the rarer enemy effect. Because of the small sample sizes, the dunnock clearly warrants more detailed study of its responses to alien nestlings. Chicks of all bronze-cuckoos (Table 1) evict host offspring, which casts doubt on the traditional view that simultaneous comparison of own and alien nestling is necessary for discrimination (Davies, 2000). Some authors (e.g. Redondo, 1993; Gill, 1998) considered discrimination against foreign nestlings in a species that accepts foreign eggs to be puzzling. However, as egg rejection keeps effective parasitism rates at the nestling stage at levels that might not allow positive selection for nestling discrimination, there are good reasons to expect exactly this pattern. At a general level, almost all hosts of the common cuckoo show a relatively high rejection rate of parasitic eggs (Davies and Brooke, 1989) which could – together with extremely low parasitism rates – explain the absence of nestling discrimination in these hosts. Hosts of the brown-headed cowbird are either close to 100% egg rejecters (selection for nestling rejection is then nil) or they accept almost all parasite eggs which is probably a result of a recent colonization by the parasite (Rothstein and Robinson, 1998) so we cannot expect to observe nestling rejection either. A recent mathematical model (Planqué et al., 2002) is in line with the rarer enemy effect (see also Britton et al. 2006). The authors assume that host defensive strategy is an innate trait (see also May and Robinson, 1985; Takasu, 1998; Servedio and Lande, 2003). Planqué et al. (2002) showed that a chick rejection strategy cannot invade a host population where individuals eject parasitic eggs. These conclusions represent a mathematical support for the verbal model presented in the present paper. Further, egg rejection is more likely to spread as a first defense in the host gene pool. Finally, the model also indicates that a seemingly beneficial mixed strategy of rejection of both eggs and nestlings is less likely to establish itself in competition with pure egg rejection than with a strategy of pure chick rejection (Planqué et al., 2002). I stress that the verbal rarer enemy model (this study) and the mathematical model by Planqué et al. (2002) are not identical to each other. Most importantly, Planqué et al. (2002) did not explicitly predict that chick discrimination should evolve mainly in egg acceptors nor did discuss any empirical evidence in favour of their hypothesis. I emphasize that the importance of egg rejection is a major constraint on the evolution of chick rejection because this constraint may work in any hosts of cuckoos, cowbirds and other brood parasites without any limitations. Most importantly, both behavioral (e.g. food allocation rules, nest sanitation, infanticide, brood reduction) and cognitive (e.g. innate mate recognition, self-referent phenotype matching) adaptations potentially useful for discrimination of parasitic chicks evolved in virtually all birds. Theoretically, there is no reason to expect that only species exploited by brood parasites are “special and unlucky” to be devoid of these preadaptations. Empirically, various such preadaptations (e.g. food allocation rules) are in fact found in host species that accept parasite nestlings (see also Rothstein, 1982; Kemal and Rothstein, 1988). THE ARMS-RACE HYPOTHESIS REVISITED Davies and Brooke (1989) described an arms-race between the cuckoo and its host as a sequence of evolutionary events. Parasitism of naïve host selects for rejection of non-

9

mimetic eggs, this selects for mimicry in parasite’s eggs, and the process can result in host extinction, host switching by the parasite or parasite extinction. If the host is freed from parasitism and egg recognition is costly, the host can revert to become an egg acceptor (Davies and Brooke, 1989). However, when egg rejection has insignificant costs it may be retained for long evolutionary periods (Rothstein, 2001). Additional stages should be added. Where the parasite has successfully evolved perfect egg mimicry, effective selection for nestling discrimination by a host can start, because this selection is no longer inhibited by egg rejection. Evolution of nestling discrimination by a host can be followed by evolution of chick mimicry, or the parasite can go extinct or switch hosts. If a parasite evolves perfect chick mimicry it could drive its host extinct. Some of the parasite-host systems from Table 1 fit a slightly different scenario. The evolution of egg discrimination in some hosts is physically constrained by dark interiors of their nests (Fraga, 1998; Gill, 1998) or by a close similarity of parasite and host eggs due to the phylogenetic constraints (Sorenson and Payne, 2001). However, the both constraints would affect the evolution of nestling discrimination same as evolved parasitic egg mimicry. FUTURE DIRECTIONS More experimental cross-fostering studies are urgently needed to shed more light on the prevalence of chick discrimination and rejection among hosts of parasitic birds. The rarer enemy hypothesis suggests that acceptors of natural parasitic eggs should be tested first (Table 1). It is predicted that these hosts should feed less, refuse to feed or even desert alien chicks that differ phenotypically (e.g. in mouth markings, begging calls and behavior) from their own nestlings. In these cross-fostering experiments I recommend that (i) naturally non-parasitic species should be used as potential objects of discrimination by hosts because natural parasites may already be mimetic and thus avoid being discriminated against (Schuetz, 2005), and (ii) these species should feed similar types and amounts of food to nestlings and be of similar body size to the host species. (iii) Cross-fostering of these “parasitic” chicks to nests of their own conspecifics would control for possible confounding effects of the cross-fostering process itself. Parasites of discriminating hosts may have already evolved chick mimicry. Therefore traits of chicks of obligate brood parasites should be manipulated experimentally to reveal host discriminating abilities (Redondo, 1993; Schuetz, 2005). The host discrimination hypothesis would be supported if significant differences in host parental care to own and alien nestlings (and/or growth performance of these, Grim 2006) were found. The parasitic chick mimicry hypothesis would be supported if the host discriminated against alien experimental chicks but discriminated less or not at all against chicks of its natural parasite. Further, comparative methods may show whether hosts of brood parasites evolved more restrictive feeding regimes than closely related non-parasitized species. I predict that egg-accepting hosts of brood parasites (which are under stronger selection for chick discrimination than egg-rejecting hosts, under otherwise equal conditions) should be less responsive to begging strategies that deviate from those of their own chicks than eggrejecting hosts or non-hosts. ACKNOWLEDGEMENTS I am grateful to N.F. Britton, N.B. Davies, N.R. Franks, B.J. Gill, B.T. Hansen, M.E. Hauber, Ø. Holen, L.E. Johannessen, M.R. Kilner, O. Kleven, M. Konvicka, M. Krist, N. Langmore, A. Lotem, A. Moksnes, A.P. Møller, R. Planqué, T. Redondo, V. Remes, E. Røskaft, S.G. Sealy, T. Slagsvold, N.C. Stenseth, B.G. Stokke, D. Storch, F. Takasu, E. Tkadlec, K. Ueda and A. Zahavi for helpful comments and discussions on various earlier drafts of the MS. I thank T. Caraco for his editorial comments. During never-ending work on this paper I was supported by grants from the Research Council of Norway, the Czech Ministry of Education (grants No. 153100012 and MSM6198959212) and the Grant Agency of the Czech Republic (206/03/D234). REFERENCES Alcock, J. 1998. Animal Behavior. An Evolutionary Approach. 6th ed. Sunderland MA:

10

Sinauer. Amundsen, T., Brobakken, P.T., Moksnes, A. and Røskaft, E. 2002. Rejection of common cuckoo Cuculus canorus eggs in relation to female age in the bluethroat Luscinia svecica. J. Avian Biol., 33: 366–370. Beecher, M.D. 1991. Successes and failures of parent-offspring recognition in animals. In Kin recognition (P.G. Hepper, ed.), pp. 94–124. Cambridge: Cambridge University Press. Beecher, M.D., Beecher, I.M. and Hahn, S. 1981. Parent-offspring recognition in bank swallows (Riparia riparia): II. Development and acoustic basis. Anim. Behav., 29: 95– 101. Britton, N.F., Planqué, R. and Franks, N.R. 2006. Evolution of defence portfolios in exploiter-victim systems. B. Math. Biol. (in press). Brooker, L.C., Brooker, M.G. and Brooker, A.M.H. 1990. An alternative population/genetics model for the evolution of egg mimesis and egg crypsis in cuckoos. J. Theor. Biol., 146: 123–143. Brooker, M.G. and Brooker, L.C. 1996. Acceptance by the splendid fairy-wren of parasitism by Horsfield’s bronze-cuckoo: further evidence for evolutionary equilibrium in brood parasitism. Behav. Ecol., 7: 395–407. Clotfelter, E.D., Whittingham, L.A. and Dunn, P.O. 2000. Laying order, hatching asynchrony and nestling body mass in tree swallows Tachycineta bicolor. J. Avian Biol., 31: 329–334. Crook, J.R. and Shields, W.M. 1985. Sexually selected infanticide by adult male barn swallows. Anim. Behav., 33: 754–761. Cunningham, E.J.A. and Lewis, S. 2005. Parasitism of maternal investment selects for increased clutch size and brood reduction in a host. Behav. Ecol., 17: 126–131. Davies, N.B. 2000. Cuckoos, Cowbirds and Other Cheats. London: T & A.D. Poyser. Davies, N.B. and Brooke, M.L. 1988. Cuckoos versus reed warblers: adaptations and counteradaptations. Anim. Behav., 36: 262–284. Davies, N.B. and Brooke, M.L. 1989. An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. II. Host egg markings, chick discrimination and general discussion. J. Anim. Ecol., 58: 225–236. Dawkins, R. 1982. The Extended Phenotype. Oxford: Oxford University Press. Dawkins, R. and Krebs, J.R. 1979. Arms races between and within species. Proc. R. Soc. Lond. B, 205: 489–511. Dewar, D. 1907. An inquiry into the parasitic habits of the Indian koel. J. Bombay Nat. Hist. Soc., 17: 765–782. Fraga, R.M. 1998. Interactions of the parasitic screaming and shiny cowbirds (Molothrus rufoaxillaris and M. bonariensis) with a shared host, the bay-winged cowbird (M. badius). In Parasitic Birds and Their Hosts. (S.I. Rothstein and S.K. Robinson, eds.), pp. 173–193. New York: Oxford University Press. Futuyma, D.J. 1998. Evolutionary Biology. 3rd ed. Sunderland MA: Sinauer. Gaston, A.J. 1976. Brood parasitism by the pied crested cuckoo Clamator jacobinus. J. Anim. Ecol., 45: 331–345. Gill, B.J. 1998. Behavior end ecology of the shining cuckoo, Chrysococcyx lucidus. In Parasitic birds and their hosts (S.I. Rothstein and S.K. Robinson, eds.), pp. 143–151. New York: Oxford University Press. Göth, A. and Hauber, M.E. 2004. Ecological approaches to species recognition in birds through studies of model and non-model species. Ann. Zool. Fennici, 41: 823–842. Grim, T. 2002. Why is mimicry in cuckoo eggs sometimes so poor? J. Avian Biol., 33: 302– 305. Grim, T. 2005. Mimicry vs. similarity: which resemblances between brood parasites and their hosts are mimetic and which are not? Biol. J. Linn. Soc., 84: 69–78. Grim, T. 2006. Cuckoo growth performance in parasitized and unused hosts: not only host size matters. Behav. Ecol. Sociobiol. (in press) Grim, T. and Honza, M. 2001. Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behav. Ecol. Sociobiol., 49: 322–329.

11

Grim, T., Kleven, O. and Mikulica, O. 2003. Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Proc. R. Soc. Lond. B, 270: S73–S75. Hauber, M.E. 2003. Egg-caping is a cost paid by hosts of interspecific brood parasites. Auk, 120: 860–865. Hauber, M.E., Russo, S.A. and Sherman, P.W. 2001. Password for species recognition in a brood-parasitic bird. Proc. R. Soc. Lond. B, 268: 1041–1048. Hauber M.E. and Sherman P.W. 2001. Self-referent phenotype matching: theoretical considerations and empirical evidence. Trends Neurosci., 24: 609–616. Hauber, M.E., Yeh, P.J. and Roberts, J.O.L. 2004. Patterns and coevolutionary consequences of repeated brood parasitism. Proc. R. Soc. Lond. B, 271: S317–320. Holcomb, L.C. 1979. Response to foster nestlings by red-winged blackbirds at different reproductive stages. Wilson Bull., 91: 434–440. Honza, M., Grim, T., Capek, M., Moksnes, A. and Røskaft, E. 2004. Nest defence, enemy recognition and nest inspection behaviour of experimentally parasitized reed warblers Acrocephalus scirpaceus. Bird Study, 51: 256–263. Hosoi, S.A. and Rothstein, S.I. 2000. Nest desertion and cowbird parasitism: evidence for evolved responses and evolutionary lag. Anim. Behav., 59: 823–840. Jourdain, F.C.R. 1925. A study of parasitism in the cuckoos. Proc. Zool. Soc. Lond., 44: 639–667. Kemal, R.E. and Rothstein, S.I. 1988. Mechanisms of avian egg recognition: adaptive responses to eggs with broken shells. Anim. Behav., 36: 175–183. Kilner, R.M., Noble, D.G. and Davies, N.B. 1999. Signals of need in parent-offspring communication and their exploitation by the common cuckoo. Nature, 397: 667– 672. Lack, D. 1968. Ecological Adaptations for Breeding in Birds. London: Methuen. Lahti, D.C. and Lahti, A.R. 2002. How precise is egg discrimination in weaverbirds? Anim. Behav., 63: 1135–1142. Langmore, N.E., Hunt, S. and Kilner, R.M. 2003. Escalation of a coevolutionary arms race through host rejection of brood parasitic young. Nature, 422: 157–160. Langmore, N.E., Kilner, R.M., Butchart, S.H.M., Maurer, G., Davies, N.B., Cockburn, A., Macgregor, N., Peters, A., Magrath, M.J.L. and Dowling, D.K. 2005. The evolution of egg rejection by cuckoo hosts in Australia and Europe. Behav. Ecol., 16: 686–692. Lawes, M.J. and Marthews, T.R. 2003. When will rejection of parasite nestlings by hosts of nonevicting avian brood parasites be favored? A misimprinting-equilibrium model. Behav. Ecol., 14: 757–770. Lichtenstein, G. 1998. Parasitism by shiny cowbirds of rufous-bellied thrushes. Condor, 100: 680–687. Lichtenstein, G. 2001a. Low success of shiny cowbird chicks parasitizing rufous-bellied thrushes: chick-chick competition or parental discrimination? Anim. Behav., 61: 401–413. Lichtenstein, G. 2001b. Selfish begging by screaming cowbirds, a mimetic brood parasite of the bay-winged cowbird. Anim. Behav., 61: 1151–1158. Lotem, A. 1993. Learning to recognize nestlings is maladaptive for cuckoo Cuculus canorus hosts. Nature, 362: 743–745. Lotem, A., Nakamura, H. and Zahavi, A. 1992. Rejection of cuckoo eggs in relation to host age: a possible evolutionary equilibrium. Behav. Ecol., 3: 128–132. Lotem, A., Nakamura, H. and Zahavi, A. 1995. Constraints on egg discrimination and cuckoo-host co-evolution. Anim. Behav., 49: 1185–1209. Marchetti, K. 2000. Egg rejection in a passerine bird: size does matter. Anim. Behav., 59: 877–883. Mark, D. and Stutchbury, B.J. 1994. Response of forest-interior songbird to the threat of cowbird parasitism. Anim. Behav., 47: 275–280. Mason, P. and Rothstein, S.I. 1986. Coevolution and avian brood parasitism: cowbird eggs show evolutionary response to host discrimination. Evolution, 40: 1207–1214. Mateo, J.M. 2002. Kin-recognition abilities and nepotism as a function of sociality. Proc. R. Soc. Lond. B, 269: 721–727.

12

May, R.M. and Robinson, S.K. 1985. Population dynamics of avian brood parasitism. Am. Nat., 126: 475–494. McLean, I.G. and Maloney, R.F. 1998. Brood parasitism, recognition and response: the options. In Parasitic Birds and Their Hosts (S.I. Rothstein and S.K. Robinson, eds.), pp. 255–272. New York: Oxford University Press. McLean, I.G. and Waas, J.R. 1987. Do cuckoo chicks mimic the begging calls of their hosts? Anim. Behav., 35: 1896–1898. Moksnes, A. and Røskaft, E. 1992. Responses of some rare cuckoo hosts to mimetic model cuckoo eggs and to foreign conspecific eggs. Ornis Scand., 23: 17–23. Moksnes, A., Røskaft, E. and Korsnes L. 1993. Rejection of cuckoo (Cuculus canorus) eggs by meadow pipits (Anthus pratensis). Behav. Ecol., 4: 120–127. Moskát, C., Székely, T., Kisbenedek, T., Karcza. Z. and Bártol, I. 2003. The importance of nest cleaning in egg rejection behaviour of great reed warblers Acrocephalus arundinaceus. J. Avian Biol., 34: 16–19. Møller, A.P. 1992. Sexual selection in the monogamous barn swallow (Hirundo rustica). II. Mechanisms of sexual selection. J. Evol. Biol., 5: 603–624. Mundy, P.J. 1973. Vocal mimicry of their hosts by nestlings of the great spotted cuckoo and striped crested cuckoo. Ibis, 115: 602–604. Nicolai, J. 1964. Der Brutparasitismus der Viduinae als ethologisches Problem. Z. Tierpsychol., 21: 129–204. Nicolai, J. 1974. Mimicry in parasitic birds. Sci. Am., 231: 92–98. Nur, N. 1984. The consequences of brood size for breeding blue tits. I. Adult survival, weight change and the cost of reproduction. J. Anim. Ecol., 53: 479–496. Ortega, C. 1998. Cowbirds and other brood parasites. Tucson: The University of Arizona Press. Øien, I.J., Moksnes, A., Røskaft, E. and Honza, M. 1998. Costs of cuckoo Cuculus canorus parasitism to reed warblers Acrocephalus scirpaceus. J. Avian Biol., 29: 209–215. Payne, R.B., Woods, J.L. and Payne, L.L. 2001. Parental care in estrildid finches: experimental tests of a model of Vidua brood parasitism. Anim. Behav., 62: 473– 483. Peer, B.D. and Sealy, S.G. 2004. Correlates of egg rejection in hosts of the brown-headed cowbird. Condor, 106: 580–599. Petit, C., Hossaert-McKey, M., Perret, P., Blondel, J. and Lambrechts, M.M. 2002. Blue tits use selected plants and olfaction to maintain an aromatic environment for nestlings. Ecol. Lett., 5: 585–589. Planqué, R., Britton, N.F., Franks, N.R. and Peletier, M.A. 2002. The adaptiveness of defence strategies against cuckoo parasitism. B. Math. Biol., 64: 1045–1068. Redondo, T. 1993. Exploitation of host mechanisms for parental care by avian brood parasites. Etología, 3: 235–297. Robertson, R.J. and Stutchbury, B.J. 1988. Experimental evidence for sexually selected infanticide in tree swallows. Anim. Behav., 36: 749–753. Rohwer, S. and Spaw, C.D. 1988. Evolutionary lag versus bill-size constraints: a comparative study of the acceptance of cowbird eggs by old hosts. Evol. Ecol., 2: 27– 36. Rothstein, S.I. 1974. Mechanisms of avian egg recognition: possible learned and innate factors. Auk, 91: 796–807. Rothstein, S.I. 1975. An experimental and teleonomic investigation of avian brood parasitism. Condor, 77: 250–271. Rothstein, S.I. 1978. Geographic variation in nestling coloration of parasitic cowbirds. Auk, 95: 152–160. Rothstein, S.I. 1982. Successes and failures in avian egg and nestling recognition with comments on the utility of optimal reasoning. Am. Zool., 22: 547–560. Rothstein, S.I. 2001. Relic behaviours, coevolution and the retention versus loss of host defences after episodes of avian brood parasitism. Anim. Behav., 61: 95–107. Rothstein, S.I. and Robinson, S.K., eds. 1998. Parasitic Birds and Their Hosts: Studies in Coevolution. New York: Oxford University Press.

13

Røskaft, E., Rohwer, S. and Spaw, C.D. 1993. Cost of puncture ejection compared with costs of rearing cowbird chicks for northern orioles. Ornis Scand., 24: 28–32. Schuetz, J.G. 2005. Reduced growth but not survival of chicks with altered gape patterns: implications for the evolution of nestling similarity in a parasitic finch. Anim. Behav., 70: 839–848. Sealy, S.G., Neudorf, D.L., Hobson, K.A. Gill, S.A. 1998. Nest defense by potential hosts of the Brown-headed Cowbird: methodological approaches, benefits of defense, and coevolution. In Parasitic Birds and Their Hosts (S.I. Rothstein and S.K. Robinson, eds.), pp. 194–211. New York: Oxford University Press. Servedio, M.R. and Lande, R. 2003. Coevolution of an avian host and its parasitic cuckoo. Evolution, 57: 1164–1175. Sherman, P.W., Reeve, H.K. and Pfenning, D.W. 1997. Recognition systems. In Behavioural Ecology (J. R. Krebs and N.B. Davies, eds.), pp. 69–96. Oxford: Blackwell Scientific. Slagsvold, T., Hansen, B.T., Johannessen, L.E. and Lifjeld, J. T. 2002. Mate choice and imprinting in birds studied by cross-fostering in the wild. Proc. R. Soc. Lond. B, 269: 1449–1455. Smith, N.G. 1968. The advantage of being parasitized. Nature, 219: 690–694. Soler, J.J., Sorci, G., Soler, M. and Møller, A.P. 1999. Change in host rejection behavior mediated by the predatory behavior of its brood parasite. Behav. Ecol., 10: 275–280. Soler, M., Palomino, J.J., Martin-Vivaldi, M. and Soler, J.J. 2000. Lack of consistency in the response of rufous-tailed scrub robins Cercotrichas galactotes towards parasitic common cuckoo eggs. Ibis, 142: 139–158. Soler, M. and Soler, J.J. 1999. Innate versus learned recognition of conspecifics in great spotted cuckoos Clamator glandarius. Anim. Cogn., 2: 97–102. Soler, M., Soler, J.J., Martinez, J.G. and Møller, A.P. 1995a. Chick recognition and acceptance – a weakness in magpies exploited by the parasitic great spotted cuckoo. Behav. Ecol. Sociobiol., 37: 243–248. Soler, M., Soler, J.J., Martinez, J.G. and Møller, A.P. 1995b. Magpie host manipulation by great spotted cuckoos: evidence for an avian mafia? Evolution, 49: 770–775. Soler, M., Soler, J.J., Martinez, J.G., Perez-Contreras, T. and Møller, A.P. 1998. Microevolutionary change and population dynamics of a brood parasite and its primary host: the intermittent arms race hypothesis. Oecologia, 117: 381–390. Sorenson, M.D. and Payne, R.B. 2001. A single ancient origin of brood parasitism in African finches: implications for host-parasite coevolution. Evolution, 55: 2550– 2567. Steyn, P. 1973. Some notes on the breeding biology of the striped cuckoo. Ostrich, 44: 163– 169. Stokke, B.G., Moksnes, A., Røskaft, E., Rudolfsen, G. and Honza, M. 1999. Rejection of artificial cuckoo (Cuculus canorus) eggs in relation to variation in egg appearance among reed warblers (Acrocephalus scirpaceus). Proc. R. Soc. Lond. B, 266: 1483– 1488. Stokke, B.G., Rudolfsen, G., Moksnes, A. and Røskaft E. 2004. Rejection of conspecific eggs in chaffinches: the effect of age and clutch characteristics. Ethology, 110: 459–470. Székely, T., Webb, J.N., Houston, A.I. and McNamara, J.M. 1996. An evolutionary approach to offspring desertion in birds. Current Ornithol., 13: 271–331. Takasu, F. 1998. Why do all host species not show defense against avian brood parasitism: Evolutionary lag or equilibrium? Am. Nat., 151: 193–205. Veen, T., Richardson, D.S., Blaakmeer, K. and Komdeur, J. 2000. Experimental evidence for innate predator recognition in the Seychelles warbler. Proc. R. Soc. Lond. B, 267: 2253–2258. Victoria, J.K. 1972. Clutch characteristics and egg discriminative ability of the African village weaver (Ploceus cucullatus). Ibis, 114: 367–376. Webster, M.S. 1994: Interspecific brood parasitism of Montezuma oropendolas by giant cowbirds: parasitism or mutualism? Condor, 96, 794–798. Wright, J. and Leonard, M.L. (eds). 2002. The Evolution of Begging. Dordrecht: Kluwer Academic Publishers. Zahavi, A. and Zahavi, A. 1997. The Handicap Principle. Oxford: Oxford University Press.

14

Table 1. Overview of host-parasite systems with reported cases of chick rejection/discrimination or mimicry. Systems are listed according to quality of evidence from direct to circumstantial. “0 (?)” = the available evidence suggests that hosts are either pure acceptors or reject natural parasitic eggs only very weakly. See text for details. Host

Parasite

Natural egg rejection rate (%) 0

Evictor parasite

References

Malurus cyaneus Estrildidae spp. Agelaioides (Molothrus) badius Turdus rufiventris Turdoides jardineii Acrocephalus scirpaceus Prunella modularis Gerygone igata

Chrysococcyx basalis, Ch. lucidus Viduinae spp.* Molothrus rufoaxillaris, M. bonariensis

+

0

–

0

–

Langmore et al., 2003 Nicolai, 1964, 1974; Payne et al., 2001 Fraga, 1998; Lichtenstein 2001b

Molothrus bonariensis

0

–

Clamator levaillantii

0 (?)**

–

Cuculus canorus

38

+

Cuculus canorus***

0

+

Chrysococcyx lucidus

0

+

Eudynamys scolopacea

0

– (India)

Davies and Brooke, 1989 McLean and Waas, 1987; Gill, 1998; McLean and Maloney, 1998 Dewar, 1907

Chrysococcyx caprius

0 (?) **

+

Reed, 1968

Urodynamis taitensis

0 (?)

+

Turdoides spp. (India)

Clamator jacobinus

0 (?)**

–

Psarocolius spp., Cacicus spp.

Scaphidura (Molothrus) oryzivora

****

–

McLean and Waas, 1987 Jourdain, 1925; Lack, 1968; Gaston, 1976 Smith 1968; Webster, 1994; Ortega 1998; Cunningham and Lewis, 2005

Corvus splendens Ploceidae spp. Mohoua spp.

Lichtenstein, 1998, 2001a Mundy, 1973; Steyn, 1973 Grim et al., 2003

* Includes ~15 host–parasite species pairs where mimicry independently evolved (Davies, 2000). ** Parasitic eggs are reported to be a very good match for host eggs (Davies, 2000; Payne, 2005) thus rendering egg rejection unlikely. *** Cross-fostering study with Fringilla coelebs chicks (see text for details). **** Giant cowbird hosts’ responses to parasite eggs are strongly specific for a particular host breeding colony (Smith 1968) and the parasite evolved highly mimetic eggs (Fig. 1 in Smith 1968; but for detailed discussion see Ortega 1998, pp. 108–113).

15

Table 2. Overview of hypotheses on nestling discrimination in hosts of parasitic birds. Selected papers on the particular hypothesis in the References column include also those arguing against particular hypotheses. See text for details. Evolutionary lag hypothesis (ELH) Failure to respond to selection Lack of pre-adaptations The proximate constraints hypothesis (PCH) Physical insufficiency Cost of misimprinting Variability of chick appearance Instability of chick appearance Absence of comparative material Host exploitation hypothesis (HEH) Exploitation of pre-existing adaptive behaviour Supernormal stimulus Evolutionary equilibrium hypothesis (EEH) Mafia Beneficial parasitism Handicap principle Low benefits after eviction of host brood Low parasitism rate Parasitic egg rejection by a host

16

References Rothstein, 1975, Rothstein, 1988 Redondo, 1993

1982;

Kemal

and

this study (cf. Rohwer and Spaw, 1988) Lotem 1993; Lawes and Marthews, 2003 Davies and Brooke, 1989; McLean and Maloney, 1998 Davies and Brooke, 1989 Davies and Brooke, 1989

Redondo, 1993 Dawkins and Krebs, 1979; Soler et al., 1995a; Grim and Honza, 2001

Soler et al. 1995b, 1999 Smith, 1968; Webster, 1994 Zahavi and Zahavi, 1997 Davies and Brooke, 1989; Lotem, 1993 this study this study

Fig. 1. Hypothetical changes in the effective parasitism rates with respect to two thresholds that selection pressure must overcome in order to select for egg- or nestling-related adaptations in a host. For details, see text (section “The rarer enemy hypothesis”).

17

3. Grim T.: Equal rights for chick brood parasites. 3. (subm.) Grim T.: Equal rights for chick brood parasites. (subm.)

Equal rights for chick brood parasites Tomáš Grim1 Department of zoology Palacký University tř. Svobody, 26 CZ-771 46 Olomouc Czech Republic e-mail: [email protected] 1

Coevolution is one of the major issues embraced by evolutionary biology. A widely known example of coevolution is brood parasitism, i.e. evolutionary interaction between avian parasites (e.g. cuckoos) and hosts (e.g. small songbirds) that lose fitness by raising parasitic progeny. General textbooks on evolutionary biology (e.g. Futuyma 1998), ecology (e.g. Begon et al. 1996) and behavioural ecology (e.g. Krebs & Davies 1993, Manning & Dawkins 1998, Alcock, 2005) include brood parasitism among premier examples for coevolution and hostparasite interactions. All these textbooks focus primarily on host defenses against parasitism and parasite counter-defenses. However, only egg discrimination and egg mimicry are discussed in detail while chick discrimination and chick mimicry are sparsely mentioned (Alcock 2005). This is not surprising because studies of host responses to parasitic chicks seem to be uncommon (reviewed in Redondo 1993 and Grim 2006a, see also Grim 2005a). Moreover, a reading of most brood parasitism papers gives the impression that hosts may defend against parasitism solely by egg rejection (for supporting data see below). However, hosts defend against parasitism at several stages: by nest defense (Grim 2005b), egg rejection (Davies & Brooke 1989) and chick rejection (Langmore et al. 2003). I believe that a more balanced view of co-evolutionary arms-races is desirable and future papers dealing with brood parasitism should mention all three possible host defense strategies. As a quick look at brood parasitism studies reveals we know a lot about parasitic eggs and very little about parasitic chicks (Rothstein & Robinson 1998, Davies 2000, Payne 2005). There are both research and citation aspects of this unequal treatment of eggs vs. chicks. Paying attention to research, publication and citation biases is an important part of scientific work (e.g. Windsor 1997, Møller & Jennions 2002, Leimu & Koricheva 2005, Wong & Kokko 2005) and the study of brood parasitism should be no exception. Although the unequal proportion of studies on eggs vs. chicks exists in the study of all avian brood parasite-host systems, I will deal with the issue using the studies of the well known brood parasite, the common cuckoo (Cuculus canorus; hereafter “cuckoo”), to illustrate the magnitude of the issue. Research: number and sample sizes of egg vs. chick studies An overview of brood parasitism papers on the cuckoo included in the Web of Science (papers published from, 1985 to, 2005, search phrase: “brood parasitism or cuckoo”) showed that only a minority of studies of hosts of the cuckoo paid any attention to parasitic chicks (Figure 1a). Further, there is a large difference in cumulative sample sizes of the egg vs. chick studies (Figure 1a). Finally, ten times more cuckoo host species were experimentally tested for their responses to parasitic eggs than to parasitic chicks (Figure 1b). I did not include studies based on museum collections (which would greatly increase sample size for egg studies). Noticeably, almost all studies of parasitic eggs were designed to test for host discrimination abilities while only two (out of total of 16) studies of parasitic chicks were designed to test for chick discrimination. The extent of this research inequality is unlikely to be explained by a lower availability of chicks than eggs for experiments. Although predation continually decreases sample sizes for chicks until fledging, it probably does not do so 20 times which would be necessary to explain the huge sample size differences (Fig. 1). Further, one does not need parasitic chicks themselves to study host discrimination abilities (see Grim 2005 and 2006a for discussion). In fact cross-fostering of chicks of other non-parasitic species is essential to

1

understand both initial stages of interspecific parasitism (Slagsvold 1998) and evolved host responses (Davies & Brooke 1988, 1989). Further, a large amount of work has been done on chicks of open-nesting passerines, many of which suffer high predation and serve as hosts for interspecific brood parasites in studies of parent-offspring interactions (reviewed in Wright & Leonard 2002). Here chick availability for research (i.e. sample size) seems not to be a constraint. In addition, some studies of brood parasitic chicks achieved strong conclusive results despite being based on very limited sample sizes (e.g. Dearborn 1998 n=6 in some analyses, Tanaka & Ueda 2005 n=6, Grim 2006b n=6 per host species). This directly rejects the explanation that low research effort on host discrimination of parasitic chicks can be explained by sample size limitations. Citations: representation of eggs vs. chicks in references Superimposed on this research inequality is the uneven citation index of egg vs. chick papers. The citation inequality is seemingly not apparent when one compares citation rate of egg vs. chick papers (2.6 vs. 2.5 citations per year on average). However, this comparison is more obscuring than revealing. In fact, papers that experimentally tested for host chick discrimination (Davies & Brooke 1988, 1989) also tested for egg discrimination and are cited as sources for information on egg discrimination but not on chick discrimination. Additionally, other studies on chicks are concerned with growth and begging behaviours and were not intended to study host chick discrimination. Therefore I searched for terms “chick”, “nestling”, “discrimination” and “rejection” in the text of PDF files of recent (last 5 years, up to October 2005) field studies of the cuckoo. The studies dealing with parasitic chicks cite papers on parasitic eggs without exception (16 out of 16 in the cuckoo literature). In a striking contrast, the studies dealing with parasitic eggs do not cite any papers on parasitic chicks typically – only one out of 30 field studies on parasitic eggs in the cuckoo published during the last five years cited a paper on chick discrimination. Most papers on parasitic eggs in general do not mention even a possibility that host defenses could extend beyond the egg stage – thus, they overlook not only Cuculus canorus papers but also those on other host-parasite systems (e.g. Nicolai 1964, Redondo 1993, Fraga 1998, Lichtenstein 2001). This citing inequality may give an impression that chick discrimination is even non-existent (see e.g. Winfree 1999). This is, of course, not so (Grim 2006a). Clearly, I do not claim that this inequality reflects some deliberate intent to squelch chick discrimination studies. Possibly this reflects a natural feed-back between number of already published studies on some issue and number of new research projects on the very same issue. This commentary, of course, is not meant to devalue egg discrimination studies in any way. However, appreciation of the chick discrimination issue would perhaps lead to a more balanced view of host-parasite coevolution (see also Windsor 1997). Rarity of chick discrimination and mimicry: reality or myth? The supposed rarity (e.g. Johnsgard 1997) or even non-existence (e.g. Winfree 1999) of parasitic chick discrimination and mimicry reported in literature does not follow from available data (Grim 2006a). In fact, it is studies of chick discrimination which are rare (Figure 1). As for chick discrimination itself we simply do not know whether it is rare or relatively common as only five cuckoo host species were studied in this respect – in contrast to 54 host species tested with parasitic eggs – and sample sizes of these chick studies are very limited indeed (Figure 1a). To my best knowledge, this research inequality was invoked never before as a potential confounding factor that may lead to general conviction of the rarity of chick discrimination. The most telling observation in support of this argument is that the very first groundbreaking studies of host responses to cuckoo eggs (Davies & Brooke 1988, 1989) also tested host responses to cross-fostered alien chicks with the results suggestive of parasitic chick discrimination by some hosts (see discussion in Grim 2006a). Strikingly, a plethora of egg discrimination experimental studies followed during next two decades but very few chick discrimination studies were conducted. One study that paid attention to entire nestling period up to fledging in the cuckoo (Grim et al. 2003) found that hosts (reed warblers, Acrocephalus scirpaceus) in fact show chick discrimination. There is now experimental evidence that the behaviour of reed warblers is indeed defence against parasitism by cuckoos (Grim subm).

2

Additional factors may increase general impression of rarity of chick discrimination and mimicry. For example, chick mimicry is usually judged from similarity in appearance of parasitic and host chicks and host discrimination is inferred when a similarity in parasitic and host chick phenotypes is found (for reviews see Redondo 1993 and Grim 2005). However, visually mimetic parasitic chicks can be rejected while visually non-mimetic parasitic chicks can be accepted by hosts as mimicry may be limited to vocal signals only (Langmore et al. 2003). In my opinion, one of the major problems in the study of brood parasitism currently is the ruling paradigm that chick discrimination is a rare escalation of the arms-race which according to theory is not expected to evolve (e.g. Lotem 1993; but see Langmore et al. 2003). The current research focus on the egg stage in parasite-host interactions provides an information feedback that again and again strengthens our impression that host defend against parasitism only when faced with alien eggs (for an exception see Stokke et al. 2005). As Amotz Zahavi noted in a similar context “A major disadvantage of a dominant theory that is accepted by everyone around you is that observers in the field have a strong tendency to overlook findings that do not fit in with the theory. [...] exceptions either go unreported, or, if reported, are not considered important in discussions of the findings.” (Zahavi 2003, p. 862). Research priorities The minority of brood parasitism studies that investigated host responses to chicks of several species of brood parasites (for review see Grim 2006a, for case studies e.g. Redondo 1993, Fraga 1998, Lichtenstein 2001, Payne et al. 2001, Langmore et al. 2003, Grim et al. 2003, Schuetz 2005) are typically not considered in descriptions of host-parasite armsraces (but see Stokke et al. 2005). This is unfortunate as it might discourage students of brood parasitism to invest their research effort in the study of host responses to chicks. Why should the study of host responses to parasitic chicks be rewarding? Chick rejection is an additional line of host defenses against parasitism which can take place when egg discrimination does not work (Langmore et al. 2003) or cannot work – e.g. when perfect egg mimicry or similarity of parasitic and host eggs resulting from phylogenetic constraints prevents evolution of egg discrimination (Grim 2005). There are good theoretical reasons – and suggestive empirical evidence – to expect that we should find chick discrimination mainly in egg acceptors (Planqué et al. 2002, Grim 2006a). Clearly, if chick discrimination was find to be more prevalent in such hosts it would profoundly change our view of coevolutionary dynamics between brood parasites and their hosts. Additionally, interactions between parasitic chicks and their hosts already proved to be important model systems for the study of more general evolutionary issues, e.g. communication of chick hunger and parental feeding rules (Kilner et al. 1999, Hauber & Montenegro 2002) and altruism directed to unrelated individuals (Kilner et al. 2004). In my view research on host behavior to parasitic chicks should become a priority in the study of brood parasitism in the years to come. This should not occur at the expense of studies of egg discrimination mechanisms but instead lead to a broadening of the current research focus. This would hopefully allow for a more balanced understanding of hostparasite interactions in particular and coevolution in general. REFERENCES Alcock, J. 2005: Animal behavior: an evolutionary approach. 8th ed. – Sinauer Associates, Sunderland Massachusetts. Begon, M., Harper, J. L. & Townsend, C. R. 1996: Ecology. 3rd ed. – Blackwell, Oxford. Davies, N. B. 2000: Cuckoos, cowbirds and other cheats. – T. & A.D. Poyser, London Davies, N. B., & Brooke, M. de L. 1988: Cuckoos versus reed warblers: adaptations and counteradaptations. – Animal Behaviour 36: 262–284. Davies, N. B., & Brooke, M. de L. 1989: An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. II. Host egg markings, chick discrimination and general discussion. – Journal of Animal Ecology 58: 225–236. Dearborn, D. C. 1998: Begging behavior and food acquisition by brown-headed cowbird nestlings. – Behavioral Ecology and Sociobiology 43: 259–270. Futuyma, D. J. 1998: Evolutionary Biology. 3rd ed. Sinauer Associates, Sunderland, Massachusetts.

3

Grim, T. 2005a: Mimicry vs. similarity: which resemblances between brood parasites and their hosts are mimetic and which are not? – Biological Journal of the Linnean Society 84: 69–78. Grim, T. 2005b: Host recognition of brood parasites: Implications for methodology in studies of enemy recognition. – Auk 122: 530–543. Grim 2006a: The evolution of nestling discrimination by hosts of parasitic birds: why is rejection so rare? – Evolutionary Ecology Research (in press). Grim 2006b: Cuckoo growth performance in parasitized and unused hosts: not only host size matters. – Behavioral Ecology and Sociobiology (in press). Grim, T. subm: Experimental evidence for chick discrimination without recognition in a common cuckoo host. Grim, T., Kleven, O. & Mikulica, O. 2003: Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? – Proceedings of the Royal Society London B 270: S73–S75. Hauber, M. E. & Montenegro K. 2002: What are the costs of raising a brood parasite? Comparing host parental care at parasitized and non-parasitized broods. – Etología 10: 1–9. Kilner, R. M., Madden, J. R. & Hauber, M. E. 2004: Brood parasitic cowbird nestlings use host young to procure resources. – Science 305: 877–879. Kilner, R. M., Noble, D. G. & Davies, N. B. 1999: Signals of need in parent-offspring communication and their exploitation by the common cuckoo. – Nature 397: 667– 672. Krebs, J. R. & Davies, N. B. 1993: An introduction to behavioural ecology. – Blackwell, Oxford. Langmore, N. E., Hunt, S. & Kilner, R. M. 2003: Escalation of a coevolutionary arms race through host rejection of brood parasitic young. – Nature 422: 157–160. Leimu, R., & Koricheva, J. 2005: What determines the citation frequency of ecological papers? – Trends in Ecology and Evolution 20: 28–32. Lichtenstein, G. 2001: Low success of shiny cowbird chicks parasitizing rufous-bellied thrushes: chick-chick competition or parental discrimination? – Animal Behaviour 61: 401–413. Lotem, A. 1993: Learning to recognize nestlings is maladaptive for cuckoo Cuculus canorus hosts. – Nature 362: 743–745. Manning, A., & Dawkins, M. S. 1998: An introduction to animal behaviour. – Cambridge University Press, Cambridge. Møller, A. P., & Jennions, M. D. 2002: Testing and adjusting for publication bias. – Trends in Ecology and Evolution 16: 580–586. Nicolai, J. 1964: Der Brutparasitismus der Viduinae als ethologisches Problem. – Zeitschrift für Tierpsychologie 21: 129–204. Payne, R. B. 2005. The cuckoos. Oxford, University Press Oxford. Payne, R. B., Woods, J. L. & Payne, L. L. 2001: Parental care in estrildid finches: experimental tests of a model of Vidua brood parasitism. – Animal Behaviour 62: 473–483. Planqué, R., Britton, N. F., Franks, N. R. & Peletier, M. A. 2002: The adaptiveness of defence strategies against cuckoo parasitism. – Bulletin of Mathematical Biology 64: 1045–1068. Redondo, T. 1993: Exploitation of host mechanisms for parental care by avian brood parasites. – Etología 3: 235–297. Rothstein, S. I., & Robinson, S. K. 1998: Parasitic birds and their hosts: studies in coevolution. – New York, Oxford University Press. Schuetz, J. G. 2005: Low survival of parasite chicks may result from their imperfect adaptation to hosts rather than expression of defenses against parasitism. – Evolution 59: 2017–2024. Slagsvold, T. 1998: On the origin and rarity of interspecific nest parasitism in birds. – American Naturalist 152: 264–272. Stokke, B. G., Moksnes, A. & Røskaft, E. 2005: The enigma of imperfect adaptations in hosts of avian brood parasites. – Ornithological Science 4: 17–29. Tanaka, K. D. & Ueda, K. 2005: Horsfield's hawk-cuckoo nestlings simulate multiple gapes for begging. – Science 308: 653–653.

4

Windsor, D. A. 1997: Equal rights for parasites. – Perspectives in Biology and Medicine 40: 222–229. Winfree, R. 1999: Cuckoos, cowbirds, and the persistence of brood parasitism. – Trends in Ecology and Evolution 14: 338–343. Wong, B. B. M., & Kokko, H. 2005: Is science as global as we think? – Trends in Ecology and Evolution 20: 745–476. Wright, J., & Leonard, M. L. 2002: The evolution of begging. – Dordrecht, Kluwer Academic Publishers. Zahavi, A. 2003: Indirect selection and individual selection in sociobiology: my personal views on theories of social behaviour. – Animal Behaviour 65: 859–863.

5

Figure 1 Research effort devoted to eggs and nestlings in studies of the Cuckoo and its hosts. a) Total number of nests with eggs vs. nestlings observed or experimentally manipulated, superscripts show number of studies. Experiments refer to egg rejection experiments and studies based on cross-fostering or any manipulation of parasitic and hosts chicks to test hypotheses on brood parasitism and parent-offspring interactions. Observations refer to natural parasitism rates and studies reporting growth or fledging success of parasitic chicks. b) Total numbers of studied host species. Data are from 73 field studies published from 1985 until October 2005 (source: Web of Science; search parameters: “brood parasitism or cuckoo”). Only studies based on field research were included. Figure 1a

Sample size

4000

40

7

20

eggs

chicks

eggs

10

3000 2000 1000 0

Experiments

chicks

Observations

No. of species studied

Figure 1b

60 50 40 30 20 10 0 eggs

chicks

Experiments

6

eggs

chicks

Observations

4. Grim T., Kleven O. & Mikulica O. 2003: Nestling discrimination without recognition: a possible defence mechanism for4. hosts towards cuckoo parasitism? Grim T., Kleven O. & Mikulica 2003: Nestling discrimination recognition: a Proceedings of the O. Royal Society of London B 270(1):without S73–S75. possible defence mechanism for hosts towards cuckoo parasitism? Proceedings of the Royal Society of London B 270(1): S73–S75.

never been shown to reject parasitic nestlings (Davies & Brooke 1989). A simple theoretical model (Lotem 1993) showed that an adaptive host response to parasitic eggs can be based on learning the appearance of eggs in a host’s nest. However, learned nestling discrimination is too costly to be favoured by selection when a parasite evicts the host’s offspring (e.g. European cuckoo, Cuculus canorus) (but see Langmore et al. 2003). Thus, it would be logical to expect that a more adaptive alternative for hosts would be to use a discrimination mechanism that does not involve learning and/or recognition. During our research on interactions between the European cuckoo and its most common host, the reed warbler (Acrocephalus scirpaceus), we obtained tentative support for the hypothesis that such antiparasitic behaviour could work in this host.

Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Toma´sˇ Grim1*, Oddmund Kleven2 and Oldrˇich Mikulica1

2. MATERIAL AND METHODS We studied interactions between the reed warbler and the cuckoo during the period 1984–2002, in the southeastern region of the Czech Republic (47°40⬘ N, 16°48⬘ E). Reed warblers were parasitized by the cuckoo at a relatively high rate (15.0%; Øien et al. 1998). Detailed descriptions of the study area and field procedures are presented elsewhere (Øien et al. 1998; Kleven et al. 1999; Grim & Honza 2001). We investigated parental care by reed warblers in terms of feeding frequencies (number of feedings per hour) and feeding rates (amount of dried food in milligrams delivered per hour). Food samples were collected with a neck-collar (see Grim & Honza 2001). We observed 109 cuckoo nestlings.

1

Laboratory of Ornithology, Palacky´ University, Trˇ. Svobody 26, CZ-771 46 Olomouc, Czech Republic 2 Zoological Museum, Natural History Museums and Botanical Garden, University of Oslo, PO Box 1172 Blindern, N-0318 Oslo, Norway * Author for correspondence ([email protected]). Recd 20.02.03; Accptd 12.03.03; Online 14.04.03

One of the great evolutionary puzzles is why hosts of parasitic birds discriminate finely against alien eggs, but almost never discriminate against parasitic chicks. A theoretical model has shown that an adaptive host response to alien eggs can be based on learning. However, learned nestling discrimination is too costly to be favoured by selection in hosts of evicting parasites, such as the European cuckoo (Cuculus canorus). Indeed, parasitic chick rejection has never been reported for any European cuckoo host species. As learned nestling discrimination is maladaptive, one can expect that a viable alternative for hosts would be to use discrimination mechanisms not involving learning and/or recognition. We suggest that hosts may starve and desert cuckoo chicks that require higher amounts of food than an average host brood at fledging (i.e. feeding rates to a parasite are outside the normal range of host behaviour in unparasitized nests). Our observations of the reed warbler (Acrocephalus scirpaceus) at parasitized nests indicate that such behaviour could possibly work in this host species.

3. RESULTS Hosts increased their feeding rates to cuckoo chicks significantly from hatching (day 0) until day 11 post-hatch (11 days is the fledging period of host chicks) (linear regression: R 2 = 0.66, F1,24 = 46.22, p ⬍ 0.0001). A cuckoo chick at 11 days of age receives 1.4 times more food than an average host brood (3.3 chicks) of the same age (feeding rates (mean ± s.d.) to one chick: reed warbler = 47.8 ± 22.2 mg h⫺1, n = 14; cuckoo = 217.1 ± 54.8 mg h⫺1, n = 11). Reed warblers did not increase feeding rates from day 12 to day 15 (linear regression: R 2 = 0.13, F1,13 = 1.84, p = 0.20). However, at four nests (out of 57) the reed warbler pairs decreased their feeding frequencies to old (more than 12 days old) cuckoo nestlings (nestlings were fed at normal frequencies when 11–12 days old, with 13–26 feedings per hour) and finally stopped feeding them (hereafter defined as desertion) despite a high begging activity in parasitic nestlings. Both parents were always active at the nest (parents were not colour-banded, but we assumed that birds regularly visiting the particular nests are its owners), i.e. starvation could not be caused by the death of one of the parents and insufficient provisioning by the second mate. Moreover, at one of these nests fosterers removed nest material from the parasitized nest and began to build a new nest nearby (ca. 2 m away), while the cuckoo chick was still begging. Nestlings died when 14.8 ± 1.0 (mean ± s.d.) days old. Another four cuckoo nestlings died in their nests at the same age (14.3 ± 1.0 days). Nest predation was ruled out as nestlings showed no sign of injuries. All nestlings that were found dead grew normally until day 11 and only later decreased their growth in comparison with successfully fledged chicks (table 1). Moreover, we found one 15-day-old nestling in another nest deserted by its foster parents and two same-age nestlings dead below their nests. In the two latter cases, chicks could simply fall out of the nest because of high activity and are not included in the desertion rate, which was 15.8% (nine out

Keywords: brood parasitism; Cuculus canorus; rejection; coevolution; nestling discrimination; recognition

1. INTRODUCTION Coevolution between brood parasites and their hosts has resulted in some of the best examples of adaptations, for example, egg mimicry in bird parasites (Moksnes & Røskaft 1995) and egg recognition and rejection in their hosts (Davies & Brooke 1989). On the other hand, parasite–host systems provide examples of some behaviours perceived by researchers as clearly maladaptive, for example, discrimination of parasitic nestlings is extremely rare (Nicolai 1974; Fraga 1998; Lichtenstein 2001) despite fine host ability to discriminate against sometimes well-mimetic parasitic eggs (Redondo 1993). Noticeably, in general, all hosts of the European cuckoo show at least some rejection of parasitic eggs, but a host species has Proc. R. Soc. Lond. B (Suppl.) 270, S73–S75 (2003) DOI 10.1098/rsbl.2003.0017

S73

 2003 The Royal Society

S74 T. Grim and others

Nestling discrimination and cuckoo parasitism

Table 1. Mass (g) of cuckoo nestlings that successfully fledged or were deserted by reed warbler hosts. (Only chicks found within 24 h after hatching (day 0) are included. Growth was measured as (i) mass (g) at a particular age, and (ii) the slope of a regression line of mass (g) against age for each individual nestling. Mean ± s.e. and p-values for Mann– Whitney tests are shown.) age (days) 0 11 14 slope for days 0–11 slope from day 12 onwards

fledged (n = 23) 2.9 ± 0.1 53.4 ± 1.0 63.2 ± 1.1 4.7 ± 0.1 1.9 ± 0.3

of 57 nestlings that survived until fledging or desertion). We did not observe any desertions of unparasitized broods and there were no weight declines in any of the host broods studied (n = 57). 4. DISCUSSION We propose that hosts ‘discriminated’ against parasitic nestlings by not being willing to increase their parental effort above the level designed by selection for the needs of their own nestlings. This is indicated by the fact that cuckoos were fed with increasing amounts of food only until the age of 11 days, which is the time when the host’s own young usually fledge. Later, feeding frequencies levelled off (fledged cuckoos) or decreased (deserted cuckoos). Feeding rates to fledged cuckoos in our study were very similar to those reported by Kilner et al. (1999). Cuckoo chicks in reed warbler nests both have a longer nestling period (18 versus 11 days for host chicks) and require more food than an average-sized host brood when 8 days old or older. Both these factors (time and amount of care) could potentially be used as cues by hosts to stop further investment. Moreover, reed warblers feeding a cuckoo chick that overgrew their average-sized brood at fledging reduced their selectivity of foraging behaviour. This was indicated by a decreasing average size of food items delivered to large parasitic nestlings (Grim & Honza 2001). This could result from physiological changes (e.g. exhaustion) serving as proximate cues for restricting parental care to large cuckoo nestlings (Holen et al. 2001). The hypothesis that increased parental effort can lead to desertion has been supported experimentally in the puffin (Fratercula arctica) (Johnsen et al. 1994). Our observations are consistent with a hypothesis suggesting that if a mutant host parent never increases its feeding rate above a fixed threshold, it would effectively starve the super-demanding parasite and at the same time satisfy the needs of its own offspring (Holen et al. 2001). The host desertion behaviour could be a by-product of physiological constraints. However, if there was any genetically determined variation in the individuals’ willingness or ability to care for their brood then selection would favour individuals not willing or able to provide care above the level required by an average host brood. This hypothesis predicts that hosts in areas sympatric with the parasite should provide more constrained parental care (length and/or amount of care) than those in allopatry. The same is predicted for among-species differences—regular cuckoo hosts should be less willing to care for broods for prolonged periods than species unsuitable as fosterers. Proc. R. Soc. Lond. B (Suppl.)

deserted (n = 3)

p

2.8 ± 0.2 53.7 ± 2.3 48.3 ± 2.7 4.8 ± 0.3 –4.9 ± 0.7

0.629 0.873 0.007 0.873 0.006

Observations that host behaviour, under natural conditions, can lead to parasitic chicks’ death are very rare— they are reported only for three host–parasite systems (Fraga 1998; Lichtenstein 2001; Langmore et al. 2003). All other cases of nestling discrimination in parasitic birds were elicited under unnatural experimental conditions (Redondo 1993; Soler et al. 1995), or are indirectly inferred from the similarity between parasitic and host nestlings (reviewed in Redondo 1993). In other cases hosts attacked parasitic fledglings, but these successfully prevented hosts from discrimination by intense begging (see Redondo 1993). Estrildid hosts have an ability to discriminate against their Vidua parasites (Nicolai 1974), but under natural conditions, parasites prevent rejection by mimicking host nestlings. Cuckoo nestlings are unlikely to respond to the selection pressure, resulting from host desertion, increasing their growth and shortening their nestling period. Age at fledging is probably genetically fixed in the cuckoo, as parasitic chicks fledge at the same age in both reed warbler and great reed warbler (Acrocephalus arundinaceus) hosts, despite the fact that they grow more slowly and weigh significantly less in nests of the former (Kleven et al. 1999). Reed warbler cuckoos are probably growing at the maximum rates allowed by the host’s foraging abilities. Nestling desertion by reed warblers is probably not a result of host nestling recognition ability. Reed warblers readily feed nestlings of several other species introduced into their nests (Davies & Brooke 1988, 1989; Davies et al. 1998). Discrimination may work simply because the cuckoo nestling period is much longer than that of the host. Deserting a cuckoo chick before fledging is more costly than egg ejection or desertion at earlier stages of the nesting cycle. However, it is less costly than rearing a parasite until independence. Hosts deserting cuckoo nestlings would gain important advantages: the benefit of potential re-nesting and no costs of prolonged (four extra weeks) care for the parasite. Deserters could also benefit from a better survival in comparison to acceptors (tradeoff between current and future reproduction). The fact that a cuckoo nestling is probably unable to respond to this host behaviour by accelerating its growth due to developmental constraints (see above) might potentially also affect the coevolutionary process. Importantly, our observations indicate that hosts are not forced to accept a cuckoo chick due to its supernormal begging (as previously believed; Redondo 1993) and can respond to cuckoos in a way that releases them from prolonged care for a parasite. Further, reed warbler chick dis-

Nestling discrimination and cuckoo parasitism T. Grim and others S75

crimination cannot lead to the evolution of parasitic chick mimicry because discrimination is not based on recognition (similarity of parasitic and host nestlings would not prevent a host from discrimination). Nestling desertion (16%) is relatively rarer than egg rejection (38%; Øien et al. 1998) in our study population. This can be due to the relatively low benefits of chick rejection (in comparison to egg rejection), low selection pressure from cuckoos (because of egg rejection by hosts and high predation, less than 5% of hosts experience 11-day-old or older cuckoo chicks) or being an evolutionary novelty. However, we cannot exclude that a similar behaviour is also present or even common in other host species, as virtually no studies have focused on the cuckoo fledging period in any host species so far. Our observations indicate that more research should focus on host responses to parasitic nestlings during the critical period before fledging (see Redondo 1993). The behaviour of reed warblers towards old cuckoo chicks indicates that some hosts could discriminate against parasitic nestlings even without recognizing them. This gives an interesting impetus for future research on this fascinating issue. Acknowledgements The authors are grateful to all who contributed to the fieldwork. N. B. Davies, Ø. H. Holen, L. E. Johannessen, N. E. Langmore, J. T. Lifjeld, A. Lotem, A. Moksnes, A. P. Møller, T. Redondo, T. Slagsvold, M. Soler, B. G. Stokke, E. Tkadlec and two anonymous referees provided valuable comments on the manuscript. Davies, N. B. & Brooke, M. L. 1988 Cuckoos versus reed warblers: adaptations and counter-adaptations. Anim. Behav. 36, 262–284. Davies, N. B. & Brooke, M. L. 1989 An experimental study of coevolution between the cuckoo, Cuculus canorus, and its hosts. II. Host egg markings, chick discrimination and general discussion. J. Anim. Ecol. 58, 225–236.

Proc. R. Soc. Lond. B (Suppl.)

Davies, N. B., Kilner, R. M. & Noble, D. G. 1998 Nestling cuckoos, Cuculus canoruss, exploit hosts with begging calls that mimic a brood. Proc. R. Soc. Lond. B 265, 673–678. (DOI 10.1098/rspb. 1998.0346.) Fraga, R. M. 1998 Interactions of the parasitic screaming and shiny cowbirds (Molothrus rufoaxillaris and M. bonariensis) with a shared host, the bay-winged cowbird (M. badius). In Parasitic birds and their hosts (ed. S. I. Rothstein & S. K. Robinson), pp. 173–193. Oxford University Press. Grim, T. & Honza, M. 2001 Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behav. Ecol. Sociobiol. 49, 322–329. Holen, H. Ø., Saetre, G. P., Slagsvold, T. & Stenseth, N. Chr. 2001 Parasites and supernormal manipulation. Proc. R. Soc. Lond. B 268, 2551–2558. (DOI 10.1098/rspb.2001.1818.) Johnsen, I., Erikstad, K. E. & Saether, B.-E. 1994 Regulation of parental investment in a longlived seabird, the puffin Fratercula arctica—an experiment. Oikos 71, 273–278. Kilner, R., Noble, D. G. & Davies, N. B. 1999 Signals of need in parent–offspring communication and their exploitation by the common cuckoo. Nature 397, 667–672. Kleven, O., Moksnes, A., Røskaft, E. & Honza, M. 1999 Host species affects the growth rate of cuckoo (Cuculus canorus) chicks. Behav. Ecol. Sociobiol. 47, 41–46. Langmore, N. E., Hunt, S. & Kilner, R. M. 2003 Escalation of a coevolutionary arms race through host rejection of brood parasitic young. Nature 422, 157–160. Lichtenstein, G. 2001 Low success of shiny cowbird chicks parasitizing rufous-bellied thrushes: chick–chick competition or parental discrimination? Anim. Behav. 61, 401–413. Lotem, A. 1993 Learning to recognize nestlings is maladaptive for cuckoo Cuculus canorus hosts. Nature 362, 743–745. Moksnes, A. & Røskaft, E. 1995 Egg-morphs and host preference in the common cuckoo (Cuculus canorus): an analysis of cuckoo and host eggs from European museum collections. J. Zool. 236, 625–648. Nicolai, J. 1974 Mimicry in parasitic birds. Sci. Am. 231, 92–98. Øien, I. J., Moksnes, A., Røskaft, E. & Honza, M. 1998 Costs of cuckoo Cuculus canorus parasitism to reed warblers Acrocephalus scirpaceus. J. Avian Biol. 29, 209–215. Redondo, T. 1993 Exploitation of host mechanisms for parental care by avian brood parasites. Etologı´a 3, 235–297. Soler, M., Soler, J. J., Martinez, J. G. & Møller, A. P. 1995 Chick recognition and acceptance—a weakness in magpies exploited by the parasitic great spotted cuckoo. Behav. Ecol. Sociobiol. 37, 243–248.

5. Grim T.: Experimental evidence for chick discrimination without recognition in a 5. common cuckoo host. Grim T.: Experimental evidence for chick discrimination without recognition in a (subm.) common cuckoo host. (subm.)

Experimental evidence for chick discrimination without recognition in a common cuckoo host Tomáš Grim1 1 Department of Zoology Palacky University tr. Svobody 26 CZ-771 46 Olomouc Czech Republic e-mail: [email protected]

ABSTRACT Recognition is considered a critical mechanism for discriminatory behaviour in animals. Theoretically, chick recognition and discrimination are not predicted to evolve in hosts of brood parasites that evict nestmates. Yet, an earlier study showed that host reed warblers (Acrocephalus scirpaceus) of an evicting parasite, the common cuckoo (Cuculus canorus), can avoid the costs of prolonged care for parasitic young by deserting the cuckoo chick before it is able to fledge. Desertion was not based on specific recognition of parasite because hosts accept any chicks cross-fostered into their nests. Thus, the mechanism of this adaptive host response remains enigmatic. Here I show experimentally that the cue triggering this “discrimination without recognition” behaviour is the length of parental care. Neither the intensity of this care nor the presence of a single-chick in the nest could explain desertions. The proposed mechanism of discrimination strikingly differs from those found in other parasite-host systems because hosts do not need any internal recognition template of the parasite’s appearance to effectively discriminate. Hosts respond similarly to foreign chicks, whether heterospecific or experimental conspecifics, suggesting that this desertion response originally evolved in the context of parent-offspring conflict over the length of the nestling period. I also demonstrate that this discriminatory mechanism is non-costly in terms of recognition errors. Comparative data strongly suggest that parasites cannot counter-evolve any adaptation to mitigate effects of this host defence. These findings have crucial implications for the process and end-result of host-parasite arms-race and understanding of cognitive basis of discriminatory mechanisms in general. Keywords: recognition

brood

parasitism;

discrimination;

mechanism;

parent-offspring

conflict;

1

1. INTRODUCTION Recognition is a crucial underlying mechanism of any discriminatory behavioural pattern, from associating with kin or selecting suitable food types to habitat choice and recognition of enemies. Not surprisingly, both direct (e.g. innate) and indirect (e.g. learned in a contextdependent manner) recognition are central to many questions in behavioural ecology (Sherman et al. 1997). One of the most thoroughly studied model systems for recognition is brood parasitism (Davies 2000). Recognition and discrimination of parasitic eggs received major attention (Davies & Brooke 1989; Lotem et al. 1995; for review see Davies 2000) while discrimination of alien chicks so far was rarely tested experimentally (Davies & Brooke 1989; Soler et al. 1995; Lichtenstein 2001; Payne et al. 2001; Schuetz 2005; for review see Redondo 1993 and Grim 2006a). Several examples of well developed chick mimicry suggest that host discrimination has arisen during the coevolution between a few hosts and parasites (Redondo 1993; Grim 2005, 2006a). In general, hosts can reject natural or experimental parasitic chicks by nest desertion (Langmore et al. 2003; Grim et al. 2003), refusal to feed them (Lichtenstein 2001; Payne et al. 2001) or by directly attacking, killing and/or ejecting them from the nest (Redondo 1993; Soler et al. 1995). However, the proximate mechanisms of host’s differential responses towards foreign chicks under natural conditions are generally unknown. The only thoroughly studied case is the discrimination based on the structure of begging calls in parasitic bronze-cuckoos (Chrysococcyx spp.) by superb fairy-wrens (Malurus cyaneus) in Australia (Langmore et al. 2003). A theoretical model (Lotem 1993) predicted the absence of chick discrimination in hosts of evicting parasites. Surprisingly, parasitic chick discrimination now has been reported for one of the best known evicting brood parasites, the common cuckoo (Cuculus canorus). About 15% of the cuckoo chicks were deserted by host reed warblers (Acrocephalus scirpaceus) (Grim et al. 2003). That reed warblers are victimized by an evicting parasite need not be in a discrepancy with Lotem’s (1993) model. While the model assumed learned chick recognition, Grim et al. (2003) predicted that adaptive host responses to parasitic chicks need not be based on learning or even recognition. Accordingly, parent reed warblers may in principle discriminate against cuckoo chicks by refusing to provide care for longer periods and/or with increased investments than that needed for successful fledging of their own chicks (see also Holen et al. 2001). Previous study (Grim et al. 2003) provided only correlative evidence in favour of this “discrimination without recognition” hypothesis. Therefore, I tested the idea experimentally in the same population where parasitic chick desertion was observed. If warbler responses to parasitic chicks were not based on learning or the recognition of the parasite then they should work against any parasite-like brood, i.e. any individual(s) requiring longer parental care than hosts’ own chicks under normal conditions. I used a cross-fostering experiment to alter those normal conditions. I manipulated brood sizes to create single- and four-chick broods and used crossfostering of different aged broods to force parents to care for nestlings for prolonged or shortened periods than normally with varying amounts of parental investment. The parental care period was defined as a time from the hour of hatching (of the first chick in four-chick broods) till the hour of fledging (of the last chick in four-chick broods) of the brood. In experimental nests hatching refers to the original brood while fledging refers to the new cross-fostered brood. Thus, the term parental care period refers to nests (i.e. the time parents at the particular nest spent with care for own chicks + chicks cross-fostered to them) while the term nestling period refers to broods (i.e. time chick(s) spent in their original natal nest + in the new nest where they were cross-fostered to; for details on experimental procedure and definition of other terms see §2(b) below). In general I predicted that parents should be less willing to care for prolonged nests. This could be manifested in shortened nestling periods or even in higher chick mortality resulting from lower parental care. In an extreme case parents could desert prolonged nests. On the other hand the parental decisions to care for chicks for a relatively fixed period could result in longer nestling periods at shortened nests. Here, chicks would not be “under pressure” to leave a nest. In shortened nests also no starvation or desertion was expected. Under the “discrimination without recognition” scenario reed warblers could use two types of information for the decision to desert a parasitic cuckoo chick (Grim et al. 2003). Under the “parental fatigue” hypothesis, parents can respond to the amount of parental care elicited by the brood and decide to desert a cuckoo chick when the investment into it is significantly higher than that required for raising a typical host brood. Physiological

2

exhaustion could work as a proximate cue triggering the desertion. Under the “time limit” hypothesis, parents can respond to the parental period length irrespective of the level of parental investment. The parental fatigue hypothesis predicts that desertion rate and chick mortality should be higher and nestling period should be shorter in prolonged four-chick broods than in all other treatment-groups due to increased costs of parental care at prolonged four-chick broods. In contrast, the time limit hypothesis predicts increased mortality, desertion rate and shorter nestling periods in both four and single-chick broods in the prolonged treatment-group in comparison to all other groups but no differences between four and single-chick broods. The single-chick broods allow the testing of a third hypothesis. Under the “single chick” hypothesis, parents could desert the cuckoo chick because it is alone in the nest (Langmore et al. 2003). According to this hypothesis the mortality and desertion rate should be higher in single-chick broods which should also fledge earlier. Under all hypotheses I expected that only a portion of tested host pairs will respond to prolonged or shortened nestling periods as not all host pairs desert cuckoo chicks in the study area (Grim et al. 2003). 2. METHODS (a) Study area and organism The study was conducted in 2002–2005 on the Luzice fish pond system in the south-eastern part of the Czech Republic (47° 40' N, 16° 48' E). A detailed description of the study area and standard field procedures has been presented elsewhere (Grim & Honza 1997, 2001). The study host species, the reed warbler, has an average nestling period of 11–12 days while the duration of parental care for the nests with successfully hatched cuckoos lasts 18 days (range: 17–21 days; Grim 2006b). The reed warbler shows weak hatching asynchrony but is not a brood reducer under normal conditions (Cramp 1992). Host adults were not individually marked and I assumed that birds regularly visiting a particular nests and its vicinity are its owners, especially when these birds responded by alarm calls to a human intruder (see Grim et al. 2003). All time variables (nestling periods etc.) were measured in hours and then converted to days. When data fitted normal distributions I used parametric tests (the use of nonparametric tests gave qualitatively same results). All values shown are means±SE. (b) Experimental manipulations of nestling periods First, I manipulated brood sizes to obtain four- and single-chick broods (see also Grim & Honza 2001) to vary the level of parental expenditure on raising chicks (average brood size of warblers in the study area is 3.2, Grim & Honza 1997). Second, both for four- and single-chick broods I established four treatment-groups (see also Nilsson & Svensson 1993; Johnsen et al. 1994): 1) unmanipulated control (nests were subjected to the same procedures as all other nests except of cross-fostering), 2) manipulated control (brood exchanged for the same age brood from another nest at the age of 3 days), 3) shortened brood/nest (an older brood that was moved to a nest originally containing a younger one) and 4) prolonged brood/nest (a younger brood that was crossfostered to a nest originally containing an older one). Because I found no differences between unmanipulated and manipulated four-chick control broods (n = 13 and 5 successful nests; Mann-Whitney tests: the nestling period length: p = 0.69, the length of parental care at the nest: p = 0.46, number of fledged chicks: p = 0.83) I pooled the data (hereafter control group). Manipulated control nests did not differ in the ages of chicks (age difference in days = 0.40±0.08, range: 0.25–0.50, n = 3 nest pairs). Matched-pairs of prolonged–shortened broods differed 4–5 days on average in their ages. Both four-chick (the age difference = 4.36±0.37, range = 2.04–7.34, n = 14 nest pairs) and single-chick broods (the age difference = 4.66±0.32, range = 3.46–6.84, n = 11 nest pairs) were subject to similar experimental changes in brood ages (t25 = –0.61, p = 0.55). The age when broods were cross-fostered did not significantly differ between single-chick (7.16±0.25) and four-chick (7.76±0.21) broods (t25 = –1.86, p = 0.08; data for a shortened brood from each pair). I varied the age differences between matched-pairs to test the prediction that the longer the prolongation of the care at the nest the higher the probability of desertion or chick mortality will be.

3

From the chick’s point of view shortened broods were transferred to nests where parents “expected” to care for chick(s) for longer than their new chicks needed, as their original chick(s) were younger. In contrast, prolonged broods were moved to nests where parents were “ready” to care for chick(s) for a shorter time than their new chicks needed, as their original chick(s) were closer to fledging. The interaction between brood size and the change of the length of care that was required by a particular nest, allowed disentangling the possible effects of (i) the length of care and (ii) the amount of care which would be impossible to achieve with naturally parasitized nests only. Prolonged four- and single-chick broods greatly (~4-fold) differ in the amount of care (e.g. mass of food) needed but not in the length of care needed. Prolonged four-chick broods need more care than a normal four-chick brood. However, the prolonged single-chick brood obviously needs much lower amount of food (~less than twice lower) than the average host brood under normal conditions. At the time when the hatching of a clutch or fledging was expected the nest was checked three times per day. The fledging time was calculated as a midpoint hour between two nest checks when chick(s) left the nest. Nestlings in four-chick broods were marked with a small colour dot on the nape with non-toxic colours for individual recognition. This enabled to check nests with minimal disturbance and without a need to handle chicks in the period shortly before fledging. The prolonged nests were inevitably visited more times than shortened nests. However, the standard field procedures lead to the difference of only four visits between those two treatment-groups. This could hardly have any effect on both desertion rate and nestling periods between the two groups as (i) reed warblers are tolerant to human presence in our study area (for supporting data see Honza et al. 2004), (ii) despite much more frequent visits to nests in an earlier study no adverse effect on chicks growth and no desertions were detected in that study (Grim & Honza 2001) and (iii) length of the nestling period did not differ between control nests checked personally and nests checked by long term monitoring with time-laps video which were not visited for two days before fledging (U13,6 = 1.40, p = 0.16). (c) Data analyses In the main analysis I fitted regression models with experimental treatment (shortened, control, prolonged), brood size (four, one) and their interaction as effect variables and the parental care period or nestling period as a response. I repeated this crucial analysis with individual nestlings as units of analysis. I fitted general linear mixed model (PROC MIXED in SAS; normal error distribution, parameters estimated by REML, degrees of freedom calculated using Kenward-Roger method) with the same predictors and the brood identity as a random effect. When I repeated all analyses presented in the Results separately for the four- and single-chick brood data sets I obtained qualitatively identical results. Therefore I pooled the data when testing the effect of age differences in matched-pairs broods on nestling period lengths and fledging success. Sample sizes for fledging success (total n = 87) and nestling period lengths (total n = 69) differ in some groups as at some nests the exact nestling period could not be reliably estimated. (d) Ethical note The aim of this study was to study experimentally brood desertion. This has inevitably leaded to suffering (starving and death) of some nestlings. However, I tried to obtain lowest sample sizes that allowed for meaningful comparisons among experimental treatments and the application of robust statistical tests. I did not continue to increase the sample sizes after four desertions were observed which conforms to the above mentioned rule. At one of the deserted four-chick nests one chick was still alive and begging (I transferred it to another nest with similar aged brood where the chick was accepted by fosterers and survived; this recipient nest was excluded from all analyses). The experiments were done under license from The Central Commission for Animal Welfare of the Czech Republic (No. 065/2002–V2) and in accordance with the laws and ethical guidelines of the Czech Republic.

4

3. RESULTS (a) The parental care and nestling periods Parents from shortened groups cared for nests for shorter whereas those from prolonged groups cared for longer time than those from control groups (figure 1; F2,65 = 214.80, p < 0.0001). Four-chick nests were on average attended slightly longer than single-chick nests (F1,65 = 4.30, p = 0.04). No interaction between experimental treatments and brood sizes was observed (F2,63 = 0.20, p = 0.82) and this interaction was eliminated from the final model. As with parental care periods, the variation in nestling periods was best explained using the same predictors of both treatment and brood size and the interaction between treatments and brood sizes was not significant: F2,63 = 0.93, p = 0.40. Nestlings from shortened broods spent more whereas nestlings from prolonged broods spent less time in the nest than those from control group (figure 2; F2,65 = 13.21, p < 0.0001). The nestling period was longer in four- than single-chick broods across treatments (F2,65 = 8.68, p = 0.0045). Analyses of nestling periods at the level of individual chicks (with brood identity as a random effect) gave qualitatively similar results. Nestling periods significantly decreased from shortened through control to prolonged broods (GLMM: F2,59 = 10.29, p < 0.0001) but the brood size effect became marginally non-significant (F1,95.2 = 3.55, p = 0.0625). The interaction between treatments and brood sizes was non-significant (F2,94.4 = 1.79, p = 0.17). This test is conservative as it artificially decreases the length of nestling periods in four- but not single-chick broods in comparison to the total nestling period from hatching of the first till fledging of the last chick (average length vs. total length of nestling period: paired t-test: t69 = 5.70, p < 0.0001). The length of the period that nests were prolonged (+days) or shortened (–days) significantly and negatively correlated with the nestling period (rs = –0.56, n = 69, p < 0.0001). However, when only prolonged nests were analysed, the relationship disappeared (rs = –0.05, n = 17, p = 0.84). This means that not all individuals in the population follow the same decision rules when forced to care for nest for prolonged periods. This suggests there is a high interindividual variability in host responses to “parasitic” chicks in the study population corresponding to similarly high intrapopulation variability in responses to parasitic eggs in hosts of parasitic birds in general (Davies 2000). (b) Brood desertions Desertions occurred solely in prolonged broods at rate 18.2% (n = 22). Both parents were observed in the close vicinity of deserted nests but they did not feed nestlings. Thus, the nestlings’ death did not result from insufficient provisioning by a widowed member of parental pair or predation – the nestlings were without any bites or peck marks (see also Grim et al. 2003). Contrary to the parental fatigue hypothesis, the desertion rate was not higher in prolonged four- (2 out of 12) than single-chick broods (2 out of 10; one tailed Fisher’s exact test, p = 0.75). Latencies from cross-fostering to desertion were similar in single- (5.21±0.13 days) and four-chick broods (5.02±0.44 days). The parental care for deserted nests lasted slightly shorter for deserted single- (12.67±0.42) than four-chick broods (13.27±0.56). The age of chicks when deserted was slightly higher in single- (8.19±0.52) than four-chick broods (6.50±1.17). The small number of desertions do not allow for meaningful statistical testing in these comparisons. Deserted nests did not differ from successful prolonged nests in the age of the new brood (U4,17 = –0.40, p = 0.69), the recipient brood (U4,17 = 0.13, p = 0.89) or the difference between these broods (U4,17 = –0.22, p = 0.82). However, desertions occurred significantly later than was the normal parental care period at control nests (U4,27 = 2.74, p = 0.006; the normal length of parental care was not significantly different between control four and single-chick broods: U18,9 = 1.70, p = 0.09). Therefore the probability of brood desertion significantly increased with increasing prolongation of parental care when all data were included in the analyses (nominal logistic regression: 2 = 10.80, n = 87, p = 0.001). The total length of parental care for deserted nests was lower than that for prolonged but successful nests (U4,17 = –2.69, p = 0.007).

5

(c) Fledging success and comparison with cuckoos Additionally, there was some chick mortality within broods in prolonged and control fourchick broods. The number of fledged chicks differed significantly among the three four-chick brood treatments (ANOVA: R2 = 0.27, F2,47 = 8.32, p = 0.0009). The shortened and control groups did not differ from each other (4.0±0.00 vs. 3.84±0.09 fledglings per brood), but prolonged broods showed significantly lower fledging success (2.83±0.44) than both control and shortened groups (Tukey-Kramer HSD, p = 0.004). The difference remained significant even when the two deserted nests were excluded from the analyses (ANOVA: R2 = 0.20, F2,45 = 5.35, p = 0.0085). Again the prolonged group (3.40±0.27) differed significantly from the other two groups (Tukey-Kramer HSD, p = 0.048) which in turn did not differ from each other. Because of desertions and partial brood mortality, the fledging success (percentage of hatched chicks that fledged, predation excluded) was significantly lower in prolonged broods in comparison to both control and shortened broods in the pooled four- and singlechick broods data set (figure 3; ANOVA: R2 = 0.22, F2,87 = 11.54, p < 0.0001; Tukey-Kramer HSD: p = 0.0005). The control and shortened broods did not differ from each other. Excluding deserted nests did not change the results qualitatively (ANOVA: R2 = 0.10, F2,83 = 4.69, p = 0.011; Tukey-Kramer HSD: p = 0.046). Fledging success significantly decreased with increasing prolongation of the care at the nest (rs = –0.41, n = 87, p = 0.0025). The effect was significant even when excluding deserted nests (rs = –0.29, n = 83, p = 0.046). The total desertion rate of prolonged warbler broods (18.2%) did not differ significantly from that of cuckoo chicks in the same study area (15.8%, n = 57; 2 = 0.07, p = 0.80). However, warbler broods were deserted at younger ages (12.97±0.33 days) than cuckoo chicks (14.56±0.29 days; U4,57 = –2.43, p = 0.015). 4. DISCUSSION I tested experimentally predictions of three hypotheses that were specifically designed to untangle the proximate mechanisms of chick desertion behaviour in the common cuckoo host, the reed warbler. I observed that prolonged broods suffered higher rates of breeding failure than control and shortened broods. Latencies from cross-fosterings to desertions at deserted prolonged nests were 5–6 days strongly indicating that the desertions occurred not as a response to a sudden change of chick appearance (if it was so then desertions should have occured immediately or shortly after the cross-fosterings). Moreover, virtually the same level of change in chick appearance at shortened nests did not elicit any desertions (see also Davies & Brooke 1989). Chicks in prolonged broods also fledged at an earlier age. Experimentally induced changes in fledging success and nestling periods did not differ between four- and single-chick broods. These results provide strong support for the time limit hypothesis. The temporal pattern of desertion of cuckoo chicks also supports the time limit hypothesis (see §4(a) below). In contrast, I found no support for the alternative hypotheses. Specifically, the slightly longer care observed in four- than single-chick nests is to the contrary of what was predicted by the parental fatigue hypothesis and is in agreement with the time limit hypothesis. Zero desertion rates in control single-chick broods are contrary to the single chick hypothesis and imply that brood size of one is not the cue triggering desertion of cuckoo chicks (see also Grim & Honza 2001). (a) Response to cuckoos vs. prolonged warbler broods Experimental findings of this study support the earlier hypothesis that reed warblers may discriminate against parasitic cuckoo chicks by restricting the parental care period to a time needed for successful fledging of their own young. The parental care period at successfully fledged control nests was always less than 13 days and average nestling period was always less than 12 days which is in line with earlier observations (Grim et al. 2003). In contrast, cuckoo chicks probably cannot fledge when younger than 16–17 days due to weak motor abilities – their feet are so feeble that chicks cannot grasp reed stems and move anywhere from the nest (own observations; see also Grim et al. 2003). If parents would not be willing to care for a chick in the nest for much longer than 12–13 days they would effectively discriminate against cuckoos. This prediction was in line with the observations that deserted cuckoo chicks died in reed warbler nests when 14–15 days old (Grim et al. 2003). Surprisingly, warbler chicks were deserted about two days earlier than cuckoo chicks. As cuckoo chicks provide their warbler hosts with a supernormal stimulus (Grim &

6

Honza 2001) they could perhaps delay the desertion response of their fosterers as predicted by the parental manipulation hypothesis (Redondo 1993). This seems to be a suitable explanation as other cuckoo species chicks can even save themselves from host physical aggression with intense begging (Redondo 1993; Grim 2006a). Alternatively or additionally, around the time of desertion cuckoo chicks are much bigger than warbler chicks (~50 vs. less than ~10g). Thus, they have more resources and better thermoregulation abilities (Hund & Prinzinger 1980) than small warbler chicks. This might help them to survive for longer after being deserted. The longer latencies until desertion at nests parasitized by cuckoos than at nests experimentally prolonged through conspecific “parasitism” also support the time limit hypothesis. If foster parents used the amount of parental expenditure as a cue for desertion (the parental fatigue hypothesis), then they would have to desert cuckoo chicks much earlier than they actually did. In fact they should desert parasitic chicks earlier than they deserted prolonged warbler broods as a cuckoo chick overgrows average host brood before the fledging period of warblers, at the age of 8–9 days (Grim 2006b). Already Grim et al. (2003) provided evidence against parental fatigue hypothesis (“cuckoo chicks … require more food than an-average sized host brood when 8 days old or older” p. 74), although they did not interpret the data in that way. The parental fatigue hypothesis cannot be reconciled with the long delay between the 8 day size threshold and the desertion of the cuckoo chick 6 days later. This is because during this period the cuckoo chick receives approximately the same amount of food as the average host brood from hatching till fledging (Grim & Honza 2001). (b) Discrimination without recognition and parent-offspring conflict The desertion rate of cuckoo chicks (Grim et al. 2003) and prolonged warbler broods (this study) did not differ statistically. This indicates that reed warbler parents used similar decision rules when faced with prolonged warbler broods and those parasitized by cuckoos. The probability of being deserted strongly increased with the increasing prolongation of nestling period in both single- and four-chick broods. However, within prolonged nests the deserted chicks were deserted prior to the end of the parental care period of successful prolonged nests. Still, the majority of prolonged broods were fledged successfully. Coupled with shorter nestling periods in prolonged broods this may suggest that most individuals in the study population are restrictive as for length of care at the nest which they are ready to provide. However, only some of them are ready to desert the nest if chicks do not fledge “in time”. These deserters are also more restrictive as regards the length of parental care at the nest. This inter-individual variance in responsiveness to prolonged care at non-parasitized nests might explain low overall observed desertion rate of cuckoo chicks (~15%). Such parental restrictiveness may have evolved in the context of parent-offspring conflict over the length of the nestling period (Trivers 1974) as in other birds studied so far (Johnsen et al. 1994). The absence of evidence for similar behaviour in other reed warbler populations may be due to methodological reasons (e.g., other authors rarely studied older cuckoo chicks, see discussion in Grim et al. 2003; Grim 2006a) or high predation rates in the study population (M. Honza, B. Matysiokova, T. Grim, unpubl. data) which may increase the benefits of forcing the brood to fledge as soon as possible. The extent of experimental manipulation (~4 days difference in ages of matched-pair broods) was much larger than observed differences in nestling periods between prolonged and shortened warbler broods (~1 day). This suggests that there are limits to parental manipulation resulting probably from ontogenetic constraints on chick growth (Starck & Ricklefs 1998). Observations of longer nestling periods in shortened than in control nests indicate that chicks fledge earlier during normal compared with less restrictive experimental conditions. This apparent case of parent-offspring conflict (Trivers 1974) deserves more attention as virtually no study so far has experimentally studied the parent-offspring conflict over the length of the nestling period in open-nesting passerines. (c) Implications of “discrimination without recognition” for coevolutionary processes Overall, both observational and experimental evidence suggest that reed warbler discrimination behavior is relatively rare and that the cue for discrimination without recognition is the duration of the parental care period. This may have serious implications for coevolutionary dynamics. Within the frame-work of arms-race it is expected that the evolution of host discrimination will be followed by the evolution of chick mimicry (Davies

7

2000; Stokke et al. 2005; Grim 2006a). Surprisingly, several lines of evidence suggest that this cannot be the case in the common cuckoo in central Europe. First, the length of nestling period appears to be genetically fixed in the cuckoo, as indicated by the data on growth patterns of cuckoo chicks in various host species (Grim 2006b). This entails that cuckoo chicks raised by reed warblers grow at their maximum possible rate allowed by host provisioning capacities (Grim 2006b) and cannot evolve any counter-adaptation against host discrimination without recognition based on the length of care at the nest. Second, this host behaviour, in turn, is probably not costly in terms of recognition errors as reed warblers did not mistakenly desert any of their not-prolonged broods (n = 65). Although the nonwillingness to prolonged care probably originated in the context of parent-offspring conflict it would be selected positively more strongly in a parasitized than in non-parasitized population. Under normal conditions when a host is not parasitized there is no need for such a restrictive care as chicks are always able to fledge within 13 days (figure 2). Therefore a desertion response simply makes no sense in a non-parasitized warbler population. Not surprisingly, I did not observe any cases of desertions at non-parasitized nests (this study; Grim & Honza 2001; Grim et al. 2003; Grim 2006b). In contrast, the temporally more restrict care leading to desertion would easily spread in the parasitized population as it would provide effective anti-parasite defence while being not costly when deserter would not be parasitized in a particular breeding attempt. Under such conditions, “deserters” should have higher fitness than “non-deserters” and desertion may be viewed as an anti-parasitic adaptation because it would be selected by parasitism pressure. The resulting spread of this adaptation in the long term may render such a host an unsuitable fosterer for cuckoos. This option has not previously been considered in debates on host selection by parasites. Only suitable diet composition, large population sizes, short nestling periods (Soler et al. 1999) and presence or absence of host brood reduction strategy (Soler 2002) are thought to be main factors facilitating parasitism by cuckoos. Thus, the results of the present study have also implications for host selection by brood parasites. The above presented scenario may also prove fruitful for theoretical models of host-parasite coevolution (see e.g. Planqué et al. 2002). These findings strikingly differ from those predicted by other studies on hostparasite discrimination in that hosts do not need to have an internal representation or recognition template of the parasite’s appearance to afford discrimination (Hauber & Sherman 2001). In turn, physiological or temporal decision rules, evolved perhaps in the absence of parasitism due to parent-offspring conflict, regarding when to terminate feeding a brood, whether conspecific or parasitic, alone are sufficient to implement effective antiparasite responses. This study could not be done without the help of co-workers who assisted with the collection of data: A. Dvorska, M. Honza, B. Matysiokova, P. Prochazka, P. Samas and Z. Skoumalova. I am grateful for comments by N. B. Davies, O. Kleven, M. Krist, V. Remes, and E. Tkadlec. I was supported by grants MSM6198959212 and GACR 206/03/D234.

8

REFERENCES Cramp, S. (ed) (1992) The birds of the Western Palearctic. Volume VI. Oxford: Oxford University Press. Davies, N. B. 2000 Cuckoos, cowbirds and other cheats. London: T. & A. D. Poyser. Davies, N. B. & Brooke, M. L. 1989 An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. II. Host egg markings, chick discrimination and general discussion. J. Anim. Ecol. 58, 225–236. Grim, T. 2005 Mimicry vs. similarity: which resemblances between brood parasites and their hosts are mimetic and which are not? Biol. J. Linn. Soc. 84, 69–78. Grim, T. 2006a The evolution of nestling discrimination by hosts of parasitic birds: why is rejection so rare? Evol. Ecol. Res. (in press). Grim, T. 2006b Cuckoo growth performance in parasitized and unused hosts: not only host size matters. Behav. Ecol. Sociobiol. (in press.). Grim, T. & Honza, M. 1997 Differences in parental care of reed warbler (Acrocephalus scirpaceus) to its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zool. 46, 135–142. Grim, T. & Honza, M. 2001 Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behav. Ecol. Sociobiol. 49, 322–329. Grim, T., Kleven, O. & Mikulica, O. 2003 Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Proc. R. Soc. Lond. B 270, S73–S75. Hauber, M. E. & Sherman, P.W. 2001 Self-referent phenotype matching: theoretical considerations and empirical evidence. Trends Neurosci. 24, 609–616. Holen, Ø. H., Saetre, G. P., Slagsvold, T. & Stenseth, N. C. 2001 Parasites and supernormal manipulation. Proc. R. Soc. Lond. B 268, 2551–2558. Honza, M., Grim, T., Capek, M., Moksnes, A. & Røskaft, E. 2004 Nest defence, enemy recognition and nest inspection behaviour of experimentally parasitized reed warblers Acrocephalus scirpaceus. Bird Study 51, 256–263. Hund, K. & Prinzinger, R. 1980 Zur Jugendentwicklung der Körpertemperature und des Körpergewichtes beim Kuckuck Cuculus canorus. Ökologie der Vogel 2, 130–131. Johnsen, I., Erikstad, K. E. & Saether, B-E. 1994 Regulation of parental investment in a long-lived seabird, the puffin Fratercula arctica – an experiment. Oikos 71, 273–278. Langmore, N. E., Hunt, S. & Kilner, R. M. 2003 Escalation of a coevolutionary arms race through host rejection of brood parasitic young. Nature 422, 157–160. Lichtenstein, G. 2001 Low success of shiny cowbird chicks parasitizing rufous-bellied thrushes: chick-chick competition or parental discrimination? Anim. Behav. 61, 401– 413. Lotem, A. 1993 Learning to recognize nestlings is maladaptive for cuckoo Cuculus canorus hosts. Nature 362, 743–745. Lotem, A., Nakamura, H. & Zahavi, A. 1995. Constraints on egg discrimination and cuckoohost co-evolution. Anim. Behav. 49, 1185–1209. Nilsson, J. A. & Svensson, M. 1993 Fledging in altricial birds: parental manipulation or sibling competition? Anim. Behav. 46, 379–386. Payne, R. B., Woods, J. L. & Payne, L. L. 2001 Parental care in estrildid finches: experimental tests of a model of Vidua brood parasitism. Anim. Behav. 62, 473–483. Planqué, R., Britton, N. F., Franks, N. R. & Peletier, M. A. 2002 The adaptiveness of defence strategies against cuckoo parasitism. B. Math. Biol. 64, 1045–1068. Redondo, T. 1993 Exploitation of host mechanisms for parental care by avian brood parasites. Etología 3, 235–297. Schuetz, J. G. 2005 Reduced growth but not survival of chicks with altered gape patterns: implications for the evolution of nestling similarity in a parasitic finch. Anim. Behav. 70, 839–848. Sherman, P. W., Reeve, H. K. & Pfenning, D. W. 1997 Recognition systems. In Behavioural Ecology (eds. J. R. Krebs & N. B. Davies), pp. 69–96. Oxford: Blackwell Scientific. Soler, M. 2002 Breeding strategy and begging intensity: influences on food delivery by parents and host selection by parasitic cuckoos. In The evolution of Begging (eds. J. Wright & M. L. Leonard), pp. 413–427. Dordrecht: Kluwer. Soler, J. J., Møller, A. P. & Soler, M. 1999 A comparative study of host selection in the European cuckoo. Oecologia 118, 265–276.

9

Soler, M., Soler, J. J., Martinez, J. G. & Møller, A. P. 1995 Chick recognition and acceptance: a weakness in magpies exploited by the parasitic great spotted cuckoo. Behav. Ecol. Sociobiol. 37, 243–248. Starck, J. M. & Ricklefs, R. E. (eds.) 1998 Avian growth and development. New York: Oxford University Press. Stokke, B. G., Moksnes A. & Røskaft E. 2005 The enigma of imperfect adaptations in hosts of avian brood parasites. Ornith. Sci. 4, 17–29. Trivers, R. L. 1974 Parent-offspring conflict. Am. Zool. 14, 249–264.

Figure 1. The effects of experimental treatments on parental care periods at nests that successfully fledged. Mean total lengths of time (+SE) that the four- (filled bars) or singlechick nests (open bars) were attended by parents plus fosterers. Superscripts show sample sizes. Figure 2. The effect of experimental treatments on nestling periods in broods that successfully fledged. Mean total lengths of time (+SE) that the four- (filled bars) or singlechick broods (open bars) spent in their original plus recipient nests. Sample sizes are same as in figure 1. Figure 3. Fledging success (percentage of fledged chicks) in four- and single-chick broods data sets pooled (excluding predated nests) in relation to experimental treatments. Mean+SE shown for shortened (n = 27), control (n = 38) and prolonged (n = 22) broods. Both comparisons of either including deserted nests (filled bars) or excluding them (the open bar) are presented.

Parental care period (days)

18.0

18 9

9 8

shortened

control

prolonged

16.0 14.0 12.0 10.0 8.0 6.0

Figure 1.

10

14 11

Nestling period (days)

13.0

12.0

11.0

10.0 shortened

control

prolonged

shortened

control

prolonged

Figure 2.

Fledging success (%)

100 90 80 70 60 50

Figure 3.

11

6. Grim T. & Honza M. 2001: Does supernormal stimulus influence parental behaviour of 6. the cuckoo’s host? Grim T. & Honza M. 2001: Does supernormal stimulus49(4): influence parental behaviour of Behavioral Ecology and Sociobiology 322–329. the cuckoo’s host? Behavioral Ecology and Sociobiology 49(4): 322–329.

Behav Ecol Sociobiol (2001) 49:322–329 DOI 10.1007/s002650000295

O R I G I N A L A RT I C L E

T. Grim · M. Honza

Does supernormal stimulus influence parental behaviour of the cuckoo’s host?

Received: 11 August 2000 / Revised: 12 September 2000 / Accepted: 26 August 2000 / Published online: 1 February 2001 © Springer-Verlag 2001

Abstract The supernormal stimulus hypothesis (SSH) states that a cuckoo chick should obtain more parental care than host young by means of exaggerated sensory signals. We tested the SSH by comparing parental care by reed warblers at parasitized and non-parasitized nests. A comparison of feeding rates to parasite and host chicks of the same size showed that parasitized nests received more food than non-parasitized ones with one host chick. There was an interesting relationship between average prey length and the mass of a cuckoo chick: prey length first increased with chick mass, but decreased after the cuckoo chick outgrew the average-sized host brood (three to four young at fledging). This might be expected if fosterers reduced the selectivity of their foraging behaviour when trying to satisfy the supernormal food demands of the parasitic chick. This suggestion is supported by the finding that the relationship between nestling mass and proportion of less economical small prey is inverse to the relationship between nestling mass and prey size. These results suggest that the parental behaviour of reed warblers is adjusted by selection to the needs of an average-sized brood. The overall proportion of insect orders was significantly different between the parasitic and host chicks. This result probably reflects the opportunistic foraging habits of the host. The qualitative difference (proportion of insect orders) between host and cuckoo nestling diets is partly a by-product of unequal length distribution of members of different taxonomic groups. The results of this study are consistent with the SSH.

Communicated by T. Czeschlik T. Grim (✉) Laboratory of Ornithology, Palack´y University, Trˇ. Svobody 26, 77146 Olomouc, Czech Republic e-mail: [email protected] M. Honza Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic, Kveˇtná 8, 60365 Brno, Czech Republic

Keywords Brood parasitism · Supernormal stimulus · Parental care · Feeding · Cuculus canorus

Introduction There are two main strategies to pass on genes to future generations: obligatory child-bearing and facultative species-specific child-caring (Dawkins 1989). Several species use the obviously beneficial strategy of exploiting the child-care behaviour of other species. The common cuckoo, Cuculus canorus, provides one of the bestknown examples. After hatching, the cuckoo nestling evicts the host young and exploits the parent-young communication system by tuning into the sensory predispositions of its fosterers (Kilner et al. 1999). The cuckoo chick has traditionally been reported as a compelling example of supernormal stimulus (e.g. Lack 1968; Dawkins and Krebs 1979; Wyllie 1981; Dawkins 1989; Alcock 1998; Manning and Dawkins 1998). The supernormal stimulus hypothesis (SSH) predicts that a cuckoo chick should provide exaggerated sensory signals for its fosterers and elicit a higher level of parental care than host young under similar conditions (Dawkins and Krebs 1979). Surprisingly, until the last few years, there was no reliable experimental confirmation of the SSH in the common cuckoo. By the end of the 1980s, the SSH had only been tested twice. The two papers (Davies and Brooke 1988; Brooke and Davies 1989) which reported having tested the supernormal effect of the cuckoo nestling in the nest of the reed warbler, Acrocephalus scirpaceus, rejected the SSH. The hypothesis was verified for the first time by Soler et al. (1995a) for the great spotted cuckoo, Clamator glandarius, parasitizing magpie, Pica pica, nests. Recently, the SSH was tested in a series of experiments in several main host species of the common cuckoo (Davies et al. 1998; Kilner and Davies 1999; Kilner et al. 1999). They showed that cuckoo chick begging calls imitate the calling of an entire brood of host young (Davies et al. 1998). Cuckoo vocal begging is a supernormal stimulus; however, the colour of

323

the nestling gape does not influence the feeding rate of either reed warblers or two other host species (Noble et al. 1999; personal observations in reed warblers). On the other hand, the success of brown-headed cowbird, Molothrus ater, nestlings in competition for food with yellow warbler, Dendroica petechia, young is not caused by the supernormal effect of the parasitic chick (Lichtenstein and Sealy 1998). In a preliminary study (Grim and Honza 1997), we studied the food of cuckoo and reed warbler nestlings. In the present study, we test the SSH. Cuckoo nestling attains sixfold higher mass than the reed warbler during its occupation of the host nest, although its nestling period is less than twice as long (Kleven et al. 1999). Cuckoo food demands should be higher than those of host nestlings. Therefore, a parasitic chick should obtain more food (mass expressed in grams) than a same-sized host chick. We also predicted that fosterers should modify their foraging behaviour to satisfy high parasite demands. Namely, they should became less selective and collect less economical food items they would not otherwise take. Lower selectivity could be reflected in the relationship between prey and chick size. Finally, a supernormal effect of the parasitic nestling, could, through the changed food selectivity of its fosterers, lead not only to quantitative (see above) but also qualitative differences (i.e. in dominances of diet groups) in the composition of food delivered to parasitic versus host nestlings.

Methods The field work was carried out from May to mid-July in 1996–1998 on two fish pond systems, 20 km apart, near the villages of Lednice and Luzˇ ice in the south-eastern part of the Czech Republic (47°40′ N, 16°48′ E), about 60 km south-east of the city of Brno. Both areas are ecologically very similar – ponds are situated in a flat agricultural lowland landscape and are surrounded by deciduous woods. All the nests used in this study were placed in Phragmites australis reed vegetation (Honza et al. 1998). Both the Lednice and Luzˇ ice study plots have a relatively high parasitism rate of cuckoos in the nests of reed warblers (Moksnes et al. 1993; Øien et al. 1998). We examined reed warbler parental care in terms of the amount of food brought to nestlings. Number and size of prey items delivered during one feeding bout are very variable. Feeding frequencies may not give an accurate picture of true nestling consumption – there can even be a negative relationship between the frequency of feeding and the actual amount of food delivered (Royama 1966). Therefore, we used the neck-collar method which enables an accurate analysis of the quantity of food allocated to nestlings and precise prey identification. Plastic-coated wire ligature placed around the nestling neck hinders the swallowing of food but is loose enough not to strangle the chick (Soler et al. 1995a). Neck-collars were applied for 1 h, and food delivered by parents was removed every 20 min because food accumulated in the gape could influence nestling behaviour and, consequently, feeding rate. Food sampling had no effect on nestling growth parameters (chick mass, length and width of bill; unpublished data). We measured several parameters of nestling size before the application of neck-collars. We weighed chicks to the nearest 0.1 g (reed warblers and small cuckoos) or 0.5 g (cuckoos above 10 g) with a Pesola spring balance. Gape length was measured as the maximum distance from the tip of the bill to the furthest point of the fleshy fold of the rictal flange. Gape width was defined as the

maximum distance between the sides of the closed bill. Both width and length was measured to the nearest 0.1 mm with a thin ruler. Gape area (mm2) was calculated from the length (L) and width (W) of the bill assuming a fully open bill according to the formula: W√[L2-(W/2)2]. We used feeding rate (mg of food delivered during 1 h to one nestling) as a parameter of parental care. Because confounding variables (e.g. the size and number of chicks) have an effect on food delivery to the nest (e.g. Moreno 1987) we had to select a parameter of chick size for comparison of the intensity of parental care between species. Feeding rate of both parasitic and host nestlings was significantly positively correlated with age, gape area and mass of chicks (all correlations were significant even after a sequential Bonferroni test, P<0.01). Testing differences in the feeding rates of parasitic versus non-parasitic nestlings according to age would have introduced an important confounding variable (chick size) into the comparison. At the time of hatching, cuckoo nestlings were already significantly bigger (heavier) than host nestlings (t=17.66, df=52, P<0.0001) and this difference accelerates strongly during the whole nestling period. Therefore, we had to choose between gape area and mass. The gape area and mass of the nestling are strongly correlated because both are a function of time. Growth of gape area in relation to an increase in mass is allometric and differs between young reed warblers and cuckoos. A host chick of the same size (mass) as a cuckoo nestling has a larger gape area than the parasitic chick and this difference accelerates during the nestling period – the difference between the slopes of linear regression lines tested by Student's t-test (Zar 1984) is significant (t=21.30, df=165, P<0.0001). Because chick dietary need is determined more by body size (a nestling feeds its body not its gape) we chose body mass as a measure of nestling size. We compared feeding rates in nestlings of both species with ANCOVA, controlling for chick size. Here, we only used data for nestlings from a mass range where data for both species were available (from 3 to 12 g). We only used nests with a single host chick because we were interested not in the cost of parasitism (i.e. parental effort of reed warblers to parasitic nestling vs their own entire brood) but in the supernormal stimulus (i.e. feeding rate to the parasitic chick vs its nonparasitic equivalent – one reed warbler nestling of the same size). Provisioning effort per nestling is affected by brood size in passerines (Rytkönen et al. 1996) and thus it is not possible to compare the feeding rate of the cuckoo chick with the average feeding rate per nestling from a brood of several nestlings because of the possible confounding effect of brood size. Therefore, brood sizes among non-parasitized nests were manipulated to obtain nests with only one chick. From naturally small broods of two to three nestlings one or two nestlings, respectively, were transferred to other nests with smaller than average broods of the same age to create broods with 1 or 4 nestlings. Only the former were used in the analysis. All the nests were situated in identical habitat (reed beds) with homogenous forest surroundings. An effort was made to take food samples simultaneously from both parasitized and non-parasitized nests. Altogether, we took food samples from 33 non-parasitized and 29 parasitized reed warbler nests. To avoid any possible effects from repeated sampling, we analysed only one (the first) sample taken from each nest. We used all available data in every analysis (i.e. sample sizes are not the same in all analyses). We found no significant effect of year, locality and other measured variables on the composition of food (unpublished data), and therefore compare entire data sets for both species unless otherwise stated. Data not normally distributed were transformed before analyses with parametric statistics. All transformed data sets were normally distributed after transformations. The term “sample” refers to the contents of one nestling crop. Cases when the chick was not fed were excluded from the analyses. Food samples (stored in 75% ethanol) from 1996 and 1997 were analysed further to examine both the qualitative and quantitative composition of the food and prey size (measured as length of body without appendages). To estimate biomass for calculating feeding rate, food samples from all years were dried to constant mass in an oven at 60°C for 48 h and then weighed on a precision balance to the nearest 0.0001 g.

324

Fig. 1 The relationship between nestling mass and amount of food delivered to it per hour was linear both for the reed warbler (open circles) and cuckoo (filled squares) (see text for details)

Results

Fig. 2 Frequency distribution [dominance=(number of items of respective order/total number of items)×100] of length of food items delivered to reed warbler (open bars, n=596; 25 samples) and cuckoo (filled bars, n=2,069; 23 samples) nestlings

Quantity of food (prediction 1) The feeding rates of both parasitic and host chicks were positively correlated with all measures of nestling size (see Methods). The relationship between nestling mass and feeding rate (both data log-transformed) was linear for both studied species (reed warbler: y=1.139x+0.411, R2=0.55, F1,27=31.10, P<0.0001; cuckoo y=0.724x+ 0.988, R2=0.64; F1,27=46.87, P<0.0001; Fig. 1). Reed warbler nestlings have disproportionately larger bills compared to same-sized cuckoos. Moreover, the reed warbler nestling gape area grew much faster than that of the cuckoo (see Methods). Therefore, we controlled for this difference by comparing chicks of the same mass (from 3 to 12 g). A comparison of feeding rates to reed warbler and cuckoo nestlings showed that parasitic nestlings obtained significantly higher amounts of food than same-sized host nestlings (ANCOVA, slopes: F1,34=1.00, P=0.324; elevations: F1,35=13.99, P<0.001). Length of food (prediction 2) In both years, there was a slight tendency for cuckoo chicks to be fed with smaller prey than reed warblers (5.4 vs 5.6 mm in 1996, 5.4 vs 6.0 mm in 1997). However, when nests were used as independent samples, the differences in average prey length were not statistically significant (Mann-Whitney test, 1996: U14,15=0.02, P=0.98; 1997: U 11,8=0.29, P=0.77). Prey longer than 14 mm were rarely taken, the longest prey being Ischnura elegans (Odonata) which appeared in the food of nestlings of both species. Length distribution of food items was not significantly different between species (Kolgomorov-Smirnov two-sample test: D=0.625, P=0.088; see also Fig. 2).

Fig. 3 Relationship between nestling mass and average length of food delivered by reed warbler fosterers to a cuckoo chick. Each point represents the average prey length in one sample taken from a cuckoo chick of known mass. A second-order polynomial regression (R2=0.50, F2,18=8.83, P=0.002) gives the best fit. Linear regression on the same data is not significant (R2=0.09, F1,20=1.81, P=0.19)

No significant linear relationship between chick size and mean prey length was found. However, for cuckoo (but not reed warbler) nestlings, a second-order polynomial regression significantly fitted the relationship between nestling mass and prey length (y=-0.005x2+ 0.335x+1.704; R2=0.50, F2,18=8.83, P=0.002; first-order regression coefficient: t=4.13, P<0.001; second-order regression coefficient: t=3.81, P<0.01; Fig. 3). The relationship between nestling mass and presence of small prey (0–2 mm) in food samples was the opposite {logistic regression: y=1/[1+exp(-0.005x2+0.328x-1.405)], R2=0.22, χ2=422.556, df=2, P<0.0001}. A similar significant effect was found for the presence of aphids {lo-

325

gistic regression: y=1/[1+exp(-0.004x2+0.205x+1.199)], R2=0.23, χ2=430.095, df=2, P<0.0001; all regression parameters in both regression equations were significant, P<0.0001}. These results indicate that the lower average size of prey delivered to older cuckoo nestlings was caused by an increase in the proportion of the smallest prey size category (0–2 mm) and specific prey types (small aphids). These relationships could be artefacts of uneven sampling of different-sized nestlings or could be caused by a changing food supply during the season (e.g. large size of nestlings later in the season might coincide with a higher proportion of aphids in the food supply). However, there was no seasonal trend in either prey length (linear regression: R2=0.04, F1,19=0.75, P=0.40; polynomial regression: R2=0.07, F2,18=0.70, P=0.51) or proportion of small prey (linear regression: R2=0.096, F1,19=2.02, P=0.17; polynomial regression: R2=0.11, F2,18=1.16, P=0.34) or aphids (linear regression: R2=0.01, F1,19=0.25, P=0.62; polynomial regression: R2=0.07, F2,18=0.63, P=0.55). Moreover, there was no relationship between the size of sampled nestlings and the date (linear regression: R2=0.06, F1,19=1.14, P=0.30; polynomial regression: R2=0.27, F2,18=3.32, P=0.06). Almost significant polynomial regression for the nestling size and date relationship is caused by the fact that more small nestlings were sampled during the second half of breeding season (this confirms that the relationship between chick size and prey size is not an artefact of uneven sampling). Thus, there was an increasing trend in prey length with age, which turned to a negative relationship between length of food and nestling size in older cuckoo chicks. The declining trend in prey length with nestling mass in large cuckoo chicks is consistent with prediction 2 (see Discussion). Quality of food (prediction 3) An analysis of food samples showed that both cuckoo and host chicks obtained a generally similar diet (Table 1). Nevertheless, the proportion of invertebrate orders in the diet of parasitic and non-parasitic nestlings was statistically significantly different (χ2=79.34, df=8, P<0.0001). Members of Diptera (especially Chironomidae) were the dominant part of the diet in both parasitized and non-parasitized nests followed by Sternorrhyncha (especially aphids) and Araneida. Within Diptera, the proportion of main groups was not significantly different between the cuckoo and reed warbler diet (Chironomidae: χ2=1.50, df=1, P=0.22; Syrphidae: χ2=1.91, df=1, P=0.18; Empididae: χ2=1.27, df=1, P=0.25). Regarding prediction 2, it is interesting that the proportion of smallest prey (aphids) was higher in the cuckoo diet both in 1996 (19.81 vs 8.25%) and 1997 (15.61 vs 11.03%). However, a comparison of aphid dominance (each nest used as an independent sample) showed that the differences were not statistically significant either in 1996 (Mann-Whitney test, U14,15=1.27, P=0.21) or 1997 (Mann-Whitney test, U11,8=0.56, P=0.58).

Table 1 Composition of food delivered to reed warbler and cuckoo nestlings. Total number of food items is 596 for reed warbler (25 samples) and 2,069 for cuckoo (23 samples) [D dominance=(number of items of respective order/total number of items)×100; F frequency=(number of samples in which items of respective order appeared/total number of samples×100]

Diptera Sternorrhyncha Araneida Auchenorrhyncha Coleoptera Heteroptera Gastropoda Hymenoptera Others

Reed warbler

Cuckoo

D (%)

F (%)

D (%)

F (%)

67.45 9.56 9.06 4.36 3.36 0.84 0.50 1.68 3.19

96.00 40.00 68.00 44.00 36.00 20.00 12.00 4.00 48.00

58.72 18.85 5.70 2.17 2.90 2.27 2.46 0.63 6.28

95.65 60.87 91.30 52.17 43.48 43.48 52.17 26.09 65.22

The overall qualitative difference in diet composition is partly explicable in light of the finding that there is a negative relationship between the length of food and chick mass in large cuckoos. If fosterers become less selective when they have a large cuckoo chick, they will feed smaller prey to parasitic nestlings. The distribution of size categories of insect bodies is clearly different among various taxonomic groups in the food supply – small prey are predominantly aphids. Thus, a primary difference in the length of food secondarily affects the qualitative composition of the diet of parasitic nestlings. These results show that (1) a cuckoo chick obtains more food than a same-sized reed warbler chick, (2) hosts presumably reduce the selectivity of foraging behaviour when feeding a large parasitic chick and (3) reduced selectivity leads not only to differences in the length of prey delivered but also, secondarily, to qualitative changes in food composition.

Discussion Quantity of food (prediction 1) Cuckoo nestlings parasitizing reed warbler fosterers were fed with larger amounts of food than host nestlings of the same size. In contrast, Kilner et al. (1999) comparing the number of feedings delivered to cuckoos and the broods of four reed warblers found no significant difference between the two species. The discrepancy between Kilner et al. (1999) and our results is probably explained by the different measure of parental investment used in the two studies: feedings delivered per hour in the former and feeding rate (mg food/h) in the latter. The amount of food obtained in one feeding is very variable (Royama 1966) and it is possible that highly variable feeding rates did not allow differences in food consumption between the nestlings of the species studied by Kilner et al. (1999) to be revealed. We measured the amount of food fed to nestlings directly and it is possible

326

that cuckoo nestlings in the study by Kilner et al. (1999) also obtained slightly more food than same-sized host chicks. The finding that brood size can affect per-individual feeding rate (Rytkönen et al. 1996) could provide another explanation for this discrepancy – Kilner et al. (1999) tested broods of four reed warblers while only one nestling was present in our experimental nests. By the end of the 1980s, the SSH had only been tested twice in the common cuckoo. In the first study, Davies and Brooke (1988) found that reed warblers showed no preference for a cuckoo chick when presented with a simultaneous choice between feeding their own versus parasitic chicks. It is possible that the absence of preference was due to the short duration of the test, small sample size or the learning ability of reed warblers (see discussion on the great spotted cuckoo below). Soler et al. (1995a) recalculated the data from Davies and Brooke (1988) and found that the cuckoo chick was fed on average 4.5 times per hour whilst reed warbler chicks received only 3.5 feedings. This recalculation supports the SSH but is not compelling for the reasons already mentioned. In the second study, reed warblers were able to increase their feeding rates when faced with experimentally enlarged broods (Brooke and Davies 1989). The authors asked why the cuckoo nestling did not use this spare feeding capacity. One of the suggested solutions (cuckoos are physiologically not able to grow faster) to this puzzle has already been falsified – cuckoo nestlings reared by the great reed warblers, A. arundinaceus, grew much faster than when fostered by reed warblers (Kleven et al. 1999). Suprisingly, this accelerated growth does not reduce the nestling period. Therefore, more probable than physiological constraint is the effect of signalling constraint, i.e. the cuckoo is not able to increase its begging call rate (Brooke and Davies 1989). Reed warbler parents integrate visual and vocal information from begging nestlings and adjust their feeding rate accordingly (Kilner et al. 1999). A begging cuckoo nestling compensates its subnormal visual display (small gape) by supernormal vocal begging. The cuckoo possibly cannot beg more frequently and use the ability of reed warblers to increase the intensity of their parental care. The great reed warbler may have different feeding rules than reed warblers and may feed the cuckoo nestling more than this smaller host even when the cuckoo nestling provides the same signals of need. This hypothesis needs testing. On the other hand, we think that the results of both our study and the data of Brooke and Davies (1989, Fig. 1d) show that the cuckoo chick actually uses this spare feeding capacity but only after it has outgrown the average reed warbler brood. Nevertheless, there must be a physiological limit to the growth rate of a young cuckoo. We have not tested this idea directly, but it is interesting to note that older cuckoo nestlings usually have full crops of food before the application of neck-collars. They were evidently not able to consume all the food delivered by the fosterers. A similar observation was obtained in great spotted cuckoo chicks (Redondo 1993). Moreover, Kilner and Davies (1999) have found that

common cuckoo nestlings consumed more food relative to their daily energy budget than blackbirds, Turdus merula, even though blackbirds grow faster than cuckoos. Length of food (prediction 2) There was a tendency for cuckoos to obtain smaller prey items than host young. In our previous study, we also found reed warblers provisioned parasitic chicks with smaller prey than their own nestlings (Grim and Honza 1997). If fosterers try to satisfy supernormal parasite food demands they should modify their foraging behaviour – they should become less selective and collect less economical food items. Thus, there could be a relationship between nestling size (mass) and prey size (average length of prey in sample), namely larger cuckoo chicks should be fed smaller prey than younger chicks because of the hypothesized reduced selectivity in host foraging behaviour. An analysis of the food delivered to cuckoos showed that for small nestlings there was a trend towards increasing length of food items. For larger cuckoos, the trend was the opposite (Fig. 3). No such relationship was found in the reed warbler. Average prey length is most strongly influenced by small diet items (they are usually numerous when present in a food sample), so we expected an increasing trend in dominance of the smallest food category (0–2 mm) and aphids to larger cuckoos. An increased dominance of small prey (aphids, juvenile spiders, Psocoptera) for older cuckoo nestlings (but not reed warbler chicks) was as expected if fosterers reduced their selectivity when feeding an older and more intensively begging parasitic chick. Foraging on small prey is considered less economical (e.g. Lifjeld 1988; McCarthy and Winkler 1999; Matyjasiak et al. 2000), which is clear for prey not distributed in groups. Foraging on aphids, which are typically social organisms, should not be less economical. Although a substantial proportion of the small prey in our samples was formed by aphids, we think that their collection, in contrast to that of larger prey types, is less economical for two reasons : (1) aphids are not always in compact groups (species Hyalopterus pruni in our study area) and (2) reed warblers often bring only one prey item in one feeding bout – even aphids are bought in very small loads of five to ten (personal observation; data not included in this study). More important is the fact that aphids formed 96.3% of the smallest prey (in the category 0–2 mm) in reed warbler nestling food, while only 62.3% of that in the food to the cuckoo chick. The category of smallest prey (0–2 mm) was formed not only by aphids, but also by e.g. small Psocoptera (included in “Others” in Table1) and juvenile spiders. Thus, the increase in the proportion of small prey delivered to a cuckoo nestling was partly caused by the rise in the percentage of aphids. However, prey items other than aphids had an important effect on the higher proportion of small prey in food delivered to cuckoo nestlings.

327

It is especially interesting to note that the curve (Fig. 3) begins to turn downwards when the mass of the cuckoo nestling reaches the equivalent of the total mass of an entire reed warbler brood (three to four nestlings on average, each weighing about 11 g). This result could indicate that reed warblers are adjusted by selection to feed an average-sized brood. Consequently, larger parental effort (for a big cuckoo chick or a larger-than-average brood) results in abnormal foraging behaviour, namely reduced food selectivity. However, this interesting result should be interpreted with caution even though the regressions are significant, because of the small sample size. Whether the fosterers really do change their foraging behaviour when working for a large cuckoo chick could be tested simply by a pair-wise experiment: to observe the foraging behaviour of fosterers and the composition of food delivered to small and big cuckoo chicks in two nests and then to cross-foster these chicks between the two nests and look for changes in food composition. We have found no published data for prey length for the cuckoo. In great tits, Parus major (Royama 1966), pied flycatchers, Ficedula hypoleuca (Lifjeld 1988) and tree swallows, Tachycineta bicolor (McCarthy and Winkler 1999), feeding frequency is negatively correlated with prey length, i.e. an increase in nestling hunger is accompanied with a decrease in parents' foraging selectivity. Fosterers trying to satiate a vigorously begging cuckoo nestling could be in a situation similar to parents who try to satisfy the need of (an experimentally enlarged) brood of their own (older) young. Thus, fosterers are forced to be less selective in prey choice and bring prey that is less profitable. Despite consuming smaller prey, cuckoo chicks obtained more food than same-sized warbler nestlings. Interestingly, Mayer (1971) found a higher percentage of aphids in the food of cuckoo chicks reared in great reed warbler nests (58.5% vs 8.1% according to number dominance). We found a similar difference in reed warblers caring for a cuckoo (Table 1). Reed warbler foraging behaviour is probably affected by environmental conditions in a similar manner – Bibby and Thomas (1985) reported that reed warbler nestlings obtained more small Homoptera in poor-quality habitat than in a high-quality environment (21.5 vs 13.6%). Interestingly, male barn swallows, Hirundo rustica, and female sand martins, Riparia riparia, handicapped by artificially elongated outermost tail feathers captured smaller insects than did controls (Møller 1989; Møller et al. 1995; Matyjasiak et al. 2000). Thus, an enlarged brood, old nestlings, poor environmental conditions, physical handicap and the presence of a cuckoo nestling in the nest can all impair foraging ability. Food quality (prediction 3) Information on the diet of cuckoo nestlings is scarce. Wyllie (1981) states without evidence that “nestling parasitic cuckoos are fed, of course, on whatever food each

particular host species normally brings to its own young”. Brooke and Davies (1989) analysed nestling faeces and provided data on the frequency of main prey types delivered both to cuckoo and reed warbler chicks. Large differences between their and our data probably stem from the opportunistic feeding nature of the host species (Davies and Green 1976) and the fact that a faecal analysis underestimates the content of soft prey while increasing the prevalence of more chitinous prey. However, both studies have shown that Diptera are the most prevalent component of the diet. Brooke and Davies (1989) reported a higher frequency of beetles in the faeces of cuckoo nestlings. We also found a higher frequency (but not dominance) of Coleoptera (Table 1) but these results are hardly comparable because of the different methods used in collecting samples. Nevertheless, Soler et al. (1995a) also found a similar trend in the great spotted cuckoo. The different qualitative composition of the diet fed to parasitic and host nestlings could, at least in part, be a by-product of uneven prey length distribution in different food groups (i.e. there are both small and big Diptera but no big aphids). Thus, the difference is primarily quantitative (in terms of food item length) and, secondarily, qualitative. Can the supernormal effect be suppressed by learning? The SSH has also been tested in the great spotted cuckoo. This species commonly parasitizes the magpie, P. pica, in southern Europe (Cramp 1985). The great spotted cuckoo chick does not show ejection behaviour and competes with the host brood by intense begging, thus drastically decreasing host reproductive success. Magpies allocate food preferentially to parasitic nestlings. This result corroborates the SSH (Soler et al. 1995a). On the other hand, magpies can discriminate a cuckoo chick if it is cross-fostered to their nest. This ability improves with the age of the alien chick – great spotted cuckoo chicks transferred to non-parasitized host nests at the end of the nestling period were neglected by fosterers at a higher rate than magpie nestlings cross-fostered to parasitized nests (Soler et al. 1995b). This result led Soler et al. (1995b) to the conclusion that the cuckoo nestling is not a supernormal stimulus. But magpies learn to recognize their young as those who hatch in their own nests. Therefore, the design of this experiment was asymmetrical: a young cuckoo moved to non-parasitized nest provided a novel stimulus but a magpie nestling cross-fostered to a parasitized nest did not. Therefore, we think that the great spotted cuckoo chick really does influence host parental behaviour by means of superstimulus (as shown in Soler et al. 1995a). At the same time, the supernormal effect of a cuckoo nestling can be reduced by the learning ability of magpies – but only under the unnatural conditions of experimental cross-fostering. Magpies cannot use their potential discriminatory ability under natural circumstances because they learn the appearance of

328

their young after they hatch. Therefore, the success of an alien chick depends on exploiting the obligatory reaction of magpies to feed all the chicks hatched in their nest (Soler et al. 1995a). It should be noted that the negative result of a choice experiment between own and parasitic nestlings in reed warblers (Davies and Brooke 1988) suggests that the situation is similar in the common cuckoo. Reed warbler parents possibly learn the appearance of their young during the nestling period and refuse to feed alien nestlings at the end of the nestling period. In conclusion, the supernormal effect of both the common and great spotted cuckoo nestlings can be suppressed by the discriminatory ability of fosterers acquired during the nestling period. However, this hypothesis needs experimental testing in the common cuckoo. It should be noted that our experimental design was also asymmetrical: unparasitized nests were manipulated to obtain nests with only one chick while parasitized nests remained unmanipulated. Perhaps it would be better to manipulate all experimental nests. However, in our study, parents were never confronted with foreign chicks: parents in parasitized nests were confronted with cuckoo nestlings that hatched in their nests and parents at unparasitized nests were confronted with reed warbler nestlings hatched in their nests (other nestlings were moved to other nests which were not used in this study). Thus we have studied parental behaviour under almost natural circumstances (e.g. under natural circumstances, brood size can be reduced to one by partial predation). Moreover, reed warblers are not able to discriminate between cuckoo and warbler nestlings; it is therefore highly improbable that our experimental design negatively influenced the results. Lichtenstein and Sealy (1998) have tested the SSH in the brown-headed cowbird, M. ater, parasitizing the yellow warbler, D. petechia. The cowbird nestling is fed more than its nestmates but not because of the exaggerated stimuli preferred by the fosterers. It receives a higher level of parental care because it is bigger and, therefore, able to reach higher than host nestlings (see also Dearborn 1998). After controlling for height, cowbird chicks were not more successful than yellow warbler young at gaining feedings even when they were bigger. Thus, Lichtenstein and Sealy (1998) have concluded that the breeding success of parasitic chicks is explicable on the basis of nestling competition and that brown-headed cowbird chicks do not exploit their host with a supernormal stimulus. Superstimulus as evolutionary artefact We hypothesize that the evolution of ejection behaviour followed an evolutionary period when the cuckoo chick was reared together with host nestlings. Therefore, the exaggerated begging of a parasitic chick (Davies et al. 1998) could have originated as a means of competition with the host nestlings. After the evolution of ejection behaviour, exaggerated begging probably remained

adaptive – chicks in open nests are at a high risk from predation, so shortening the nestling period (through higher intensity of parental care) is strongly adaptive. Moreover, the selection pressures against louder begging (intensity of predation is positively correlated with begging intensity: Redondo and Castro 1992) should have been reduced after evolution of the ejection behaviour because of the smaller brood size (i.e. reduced vocal output of the nest). Therefore, the supernormal stimulus of the cuckoo chick could be a “relic” from former competition with host nestlings while still remaining adaptive. This hypothesis could be tested using phylogenetic comparative methods when more information on the behaviour of ejecting and non-ejecting parasitic and non-parasitic cuckoos has been obtained. Both before and after the evolution of ejection behaviour in the cuckoo nestling, young cuckoos have exploited the pre-existing parental behaviour of their fosterers which is itself adaptive in the absence of parasitism (Redondo 1993). A supernormal stimulus could be the mechanism (in terms of proximate explanation) used for exploitation by young cuckoos. We conclude that the cuckoo chick, through supernormal begging, changes the foraging behaviour of the fosterer and, consequently, both the quantitative and qualitative composition of food delivered. Acknowledgements We are grateful to N.B. Davies, L. Gvozˇ dík, V. Pavel and E. Tkadlec for helpful comments on various versions of the manuscript. Suggestions from two anonymous referees greatly improved the paper. We thank S. Sweeney for improving the English. We also thank O. Kleven, O. Mikulica, A. Moksnes, I. Øien, E. Røskaft and G. Rudolfsen for their help in the field. We are indebted to I. Mísˇek who allowed us to use the precision balance in his laboratory. Financial support was received from Ministry of Education grant VS-96019 and from the Grant Agency of the Czech Republic (grant no. A6087801). This study has been carried out under permission given to M.H. and in accordance with the laws and ethical guidelines of the Czech Republic.

References Alcock J (1998) Animal behavior: an evolutionary approach, 6th edn. Sinauer, Sunderland Bibby CJ, Thomas DK (1985) Breeding and diets of the reed warbler at a rich and a poor site. Bird Study 32:19–31 Brooke M de L, Davies NB (1989) Provisioning of nestling cuckoos Cuculus canorus by reed warbler Acrocephalus scirpaceus hosts. Ibis 131:250–256 Cramp JS (ed) (1985) The birds of the Western Palearctic vol IV. Terns to woodpeckers. Oxford University Press, Oxford Davies NB, Green RE (1976) The development and ecological significance of feeding techniques in the reed warbler (Acrocephalus scirpaceus). Anim Behav 24:213–229 Davies NB, Brooke M de L (1988) Cuckoos versus reed warblers: adaptations and counteradaptations. Anim Behav 36:262–284 Davies NB, Kilner RM, Noble DG (1998) Nestling cuckoos, Cuculus canorus, exploit hosts with begging calls that mimic a brood. Proc R Soc Lond B 265:673–678 Dawkins R (1989) The selfish gene. Oxford University Press, Oxford Dawkins R, Krebs JR (1979) Arms races between and within species. Proc R Soc Lond B 205:489–511 Dearborn DC (1998) Begging behaviour and food acquisition by brown-headed cowbirds nestlings. Behav Ecol Sociobiol 43:259–270

329 Grim T, Honza M (1997) Differences in parental care of reed warbler (Acrocephalus scirpaceus) to its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zool 46:135–142 Honza M, Øien IJ, Moksnes A, Røskaft E (1998) Survival of reed warbler Acrocephalus scirpaceus clutches in relation to nest position. Bird Study 45:104–108 Kilner RM, Davies NB (1999) How selfish is a cuckoo chick? Anim Behav 58: 797–808 Kilner RM, Noble DG, Davies NB (1999) Signals of need in parent-offspring communication and their exploitation by the common cuckoo. Nature 397:667–672 Kleven O, Moksnes A, Røskaft E, Honza M (1999) Host species affects the growth rate of cuckoo (Cuculus canorus) chicks. Behav Ecol Sociobiol 47:41–46 Lack D (1968) Ecological adaptations for breeding in birds. Methuen, London Lichtenstein G, Sealy, SG (1998) Nestling competition, rather than supernormal stimulus, explains the success of parasitic brownheaded cowbird chicks in yellow warbler nests. Proc R Soc Lond B 265:249–254 Lifjeld JT (1988) Prey choice and nestling hunger: an experiment with pied flycatchers, Ficedula hypoleuca. Anim Behav 36:134–139 Manning A, Dawkins MS (1998) An introduction to animal behaviour. Cambridge University Press, Cambridge, UK Matyjasiak P, Jablonski PG, Olejniczak I, Boniecki P (2000) Imitating the initial evolutionary stage of a tail ornament. Evolution 54:704–711 Mayer J (1971) Ecological relationships of the great reed warbler (Acrocephalus arundinaceus L.) and the reed warbler (Acrocephalus scirpaceus Herm.) during breeding season (in Czech). PhD thesis, Masaryk University, Brno McCarthy JP, Winkler DW (1999) Foraging ecology and diet selectivity of tree swallows feeding nestlings. Condor 101:246– 254 Moksnes A, Røskaft E, Bicˇík V, Honza M, Øien IJ (1993) Cuckoo Cuculus canorus parasitism on Acrocephalus warblers in southern Moravia in the Czech Republic. J Ornithol 134:425– 434

Møller AP (1989) Viability cost of male tail ornaments in a swallow. Nature 339:132–135 Møller AP, de Lope F, López Caballero JM (1995) Foraging costs of a tail ornament: experimental evidence from two populations of barn swallows Hirundo rustica with different degrees of sexual size dimorphism. Behav Ecol Sociobiol 37:289–295 Moreno J (1987) Parental care in the wheatear Oenanthe oenanthe: effects of nestling age and brood size. Ornis Scand 18:291–301 Noble DG, Davies NB, Hartley IR, McRae SB (1999) The red gape of the nestling cuckoo (Cuculus canorus) is not a supernormal stimulus for three common hosts. Behaviour 136:759–777 Øien IJ, Moksnes A, Røskaft E, Honza M (1998) Costs of cuckoo Cuculus canorus parasitism to reed warblers Acrocephalus scirpaceus. J Avian Biol 29:209–215 Redondo T (1993) Exploitation of host mechanisms for parental care by avian brood parasites. Etología 3:235–297 Redondo T, Castro F (1992) The increase in risk of predation with begging activity in broods of magpies Pica pica. Ibis 134:180–187 Royama T (1966) Factors governing feeding rate, food requirement and brood size of nestling great tits (Parus major). Ibis 108:313–347 Rytkönen S, Koivula K, Orell M (1996) Patterns of per-brood and per-offspring provisioning efforts in the willow tit Parus montanus. J Avian Biol 27:21–30 Soler M, Martinez JG, Soler JJ, Møller AP (1995a) Preferential allocation of food by magpies Pica pica to great spotted cuckoo Clamator glandarius chicks. Behav Ecol Sociobiol 37:7–13 Soler M, Soler JJ, Martinez JG, Møller AP (1995b) Chick recognition and acceptance: a weakness in magpies exploited by the parasitic great spotted cuckoo. Behav Ecol Sociobiol 37:243– 248 Wyllie I (1981) The cuckoo. Batsford, London Zar JH (1984) Biostatistical analysis, 2nd edn. Prentice-Hall, Englewood Cliffs, NJ

7. Grim T. 2006: Cuckoo growth performance in parasitized and unused hosts: not only 7.matters. host size Grim T. 2006: Cuckoo growth performance parasitized(in and unused hosts: not only Behavioral Ecology and in Sociobiology press) host size matters. Behavioral Ecology and Sociobiology (in press)

Cuckoo growth performance in parasitized and unused hosts: not only host size matters Tomáš Grim1 1 Department of Zoology Palacký University tř. Svobody 26 CZ-771 46 Olomouc Czech Republic e-mail: [email protected]

Abstract The quality and quantity of food delivered to young are among the major determinants of fitness. A parental provisioning capacity is known to increase with body size. Therefore, brood parasitism provides an opportunity to test the effects of varying provisioning abilities of different sized hosts on parasitic chick growth and fledging success. Knowledge of growth patterns of common cuckoo, Cuculus canorus, chicks in nests of common hosts is very poor. Moreover, no study to date has focused on any currently unused hosts (i.e. suitable cuckoo host species in which parasitism is currently rare or absent). Here, I compare growth performance of cuckoo chicks in nests of a common host (the reed warbler, Acrocephalus scirpaceus) and two unparasitized hosts (the song thrush, T. philomelos, and the blackbird, Turdus merula). Parasitic chicks were sole occupants of observed nests, thus eliminating the confounding effect of competition with host chicks. Experiments revealed striking differences in parasitic chick growth in the two closely related Turdus hosts. Cuckoo chicks cross-fostered to song thrush nests grew very quickly and attained much higher mass at fledging than those in nests of their common reed warbler host. Alternatively, parasitic chicks in blackbird nests grew poorly and did not survive until fledging. I discuss these observations with respect to host selection by parasitic cuckoos. Key words Brood parasitism • Host selection • Parental care • Growth • Cuculus canorus

1

Introduction Both quality and amount of food delivered by parents, critically affects growth, future survival and reproduction of young (Martin 1987; Starck and Ricklefs 1998). There is ample empirical evidence indicating that increased parental provisioning leads to a higher fledging mass and consequently higher survival (e.g. Davies 1986; Magrath 1991; Lindén et al. 1992). Furthermore, parental provisioning capacity is positively correlated with parent body size (Saether 1994). Under natural conditions, nestlings cannot benefit by deciding to be fed by an alternative larger foster parent with greater provisioning ability. Therefore, heterospecific brood parasitism (Davies 2000) across different host species provides an opportunity to test the plasticity of nestling growth under various parental provisioning regimes and the possible effect on nestling survival (Kilner and Davies 1999; Kilpatrick 2002; Kilner 2003). We would expect, under equal conditions, to see higher parasitism rates of larger hosts with greater provisioning capacity (Wiley 1986; Kleven et al. 1999). This hypothesis assumes that a parasitic chick can evict its nestmates – potential competitors – and/or procure enough food in competition with host chicks (Kleven et al. 1999; Glassey and Forbes 2003). The hypothesis that host size positively affects parasitic chick growth and fledging success was supported in the study of two sympatric cuckoo, Cuculus canorus, hosts (Kleven et al. 1999) and three hosts of the shiny cowbird, Molothrus bonariensis (Wiley 1986). In contrast, Soler and Soler (1991) found that host species size did not influence parasitic great spotted cuckoo, Clamator glandarius, growth and Kilpatrick (2002) reported that brown-headed cowbird, M. ater, nestling growth was not strongly linearly related to adult host mass. However, Kilner (2003) and Kilner et al. (2004) found a curvilinear relationship between host size and cowbird chick survival (cowbirds are most likely to fledge when reared alongside host young of intermediate size). Despite the evidence presented by Kleven et al. (1999), the hypothesis is in striking contrast with the observed general pattern of natural parasitism rates in the common cuckoo: the most frequent cuckoo host is the miniature reed warbler, Acrocephalus scirpaceus, while large thrushes, Turdus spp., are almost never parasitized (Moksnes and Røskaft 1995) despite being an order of magnitude larger than warblers (12 g vs. 70–100 g; Cramp 1988). Of course, various other factors can be important in host selection by parasitic birds – some hosts may e.g. provide nestlings with food which is indigestible (e.g. seeds, fruits) to brood parasitic chicks (Glue and Morgan 1972), follow food allocation rules which disfavour parasitic chicks (Soler 2002), may be too aggressive to female cuckoos near their nests (Grim 2005a) or reject almost all parasitic eggs (Moksnes and Røskaft 1992). Nonetheless, it is puzzling that thrushes are not parasitized by cuckoos when they are considered suitable hosts with their egg rejection rates and aggression levels within the range of commonly parasitized cuckoo hosts (Davies and Brooke 1989; Grim and Honza 2001a; Moskát et al. 2003). In contrast, Moksnes et al. (1990) considered thrushes to be unsuitable hosts because young cuckoos could not evict host eggs from fieldfare, T. pilaris, nests and survive. A cuckoo chick is however, able to evict host eggs and chicks from song thrush, T. philomelos, and blackbird, T. merula, nests (T. Grim, M. Honza, O. Kleven, A. Moksnes, C. Moskát, E. Røskaft, unpublished data). To my knowledge there are no data published on chick growth and survival in nests of other unused hosts, including the European song thrush and blackbird which in comparison to fieldfares are much more common potential hosts. The growth of parasitic chicks has been studied in some host-parasite systems (see above), particularly in the brown headed cowbird and its hosts (reviewed in Kilpatrick 2002). On the other hand, cuckoo chick growth has only been observed in the dunnock, Prunella modularis, the robin, Erithacus rubecula (Werth 1947), the white wagtail, Motacilla alba (Werth 1947; Numerov 2003, p. 149), the redstart, Phoenicurus phoenicurus (Khayutin et al. 1982), the rufous bush chat, Cercotrichas galactotes (Alvarez 1994), the great reed warbler, A. arundinaceus, and the reed warbler (Kleven et al. 1999) (see also Wyllie 1981). All hosts mentioned above, are classified as current cuckoo hosts (Davies and Brooke 1989; Moksnes and Røskaft 1995) and no study thus far has considered the growth of cuckoo young in currently unused hosts. Such information is essential if we want to understand host selection and avoidance by parasitic birds (Rothstein and Robinson 1998; Strausberger 1998; Woolfenden et al. 2003).

2

The aims of the current study were to compare the growth and survival of cuckoo chicks in nests of a common host, the reed warbler, and in the nests of two potential host species not currently parasitized by the cuckoo (the song thrush, the blackbird). This was to experimentally test across species, the known correlative relationship between parent size and provisioning capacity (Saether 1994), and to improve our understanding of factors underlying host selection by brood parasitic birds. Methods I studied cuckoo growth in nests of reed warblers from late May to mid July 1994–2005 at two fish pond systems in South Moravia, Czech Republic: Lednice (48° 47' N, 16° 49' E) and Lužice (47° 40' N, 16° 48' E), both about 60 km south-east of the city of Brno. For details on field procedures see Grim and Honza (2001b). Cross-fostering experiments with song thrush and blackbird nests were done from 1998 to 2005 in several forest tracts adjacent to Lužice ponds. Adult masses of studied species averaged 12 g for the reed warbler (Cramp 1992), 70 g for the song thrush and 100 g for the blackbird (Cramp 1988). In comparison, the adult cuckoo weighed 120 g on average (Wyllie 1981). The lengths of incubation period averaged 11 days for the reed warbler (Cramp 1992), 13 days for the song thrush and 13 days for the blackbird (Cramp 1988). Cuckoo eggs hatch after 11–12 days of incubation (own observations). I analysed cuckoo chicks growth in 79 reed warbler nests, 6 song thrush nests and 6 blackbird nests (O. Kleven, A. Moksnes and E. Røskaft kindly provided their unpublished data on 2 cuckoo chicks transferred to blackbird nests, other chicks were opportunistically cross-fostered by me). The sample size for “reed warbler” cuckoos is the largest published dataset but the sample sizes for thrush hosts are relatively low. This is due to the difficulty in finding wellsynchronized nests of these alternative hosts, due mainly to relatively low breeding densities (especially the blackbird) and very high predation rates (both thrushes; T. Grim, unpublished data, see also Grim and Honza 2001a). In some years high predation on both warbler and thrush nests prevented any experiments. Moreover, parasitism rates generally declined throughout the study period leading to a low availability of chicks for crossfostering independently of predation. However, results were clear cut and the sample sizes are sufficient to evaluate some hypotheses under investigation. Both the song thrush and blackbird reject parasitic eggs in my study area (Grim and Honza 2001a). Therefore I transferred cuckoo chicks after they had hatched in reed warbler nests to prevent eggs being wasted due to the host rejection response. Nestling competition negatively influences nestling growth and survival (in nonparasitic broods: Martin 1987; in parasitic broods: Trine 1998; but see Kilner et al. 2004). In the present study this potentially confounding factor was eliminated by host eggs being evicted by cuckoos themselves or being removed by the author after the eviction behaviour ceased and the eggs did not hatch. Kleven et al. (1999) tested for the possible confounding effect of genetic differences between cuckoo females and their progeny who were parasitizing the reed warbler and the great reed warbler. This was carried out by cross-fostering cuckoo chicks hatched in reed warbler nests to great reed warbler nests and vice versa. They found that the host species contributed significantly to the observed differences in cuckoo growth in the two respective hosts, suggesting that parental genetic effects were of minor importance (Kleven et al. 1999). In the present study I used only chicks hatched in the nests of the reed warbler, thus it is unlikely that the growth differences observed would result from parental genetic effects. For each hosts I evaluated the length of the nestling cuckoo period (i.e. time from hatching to fledging in days) and several growth parameters (mass, tarsus length). Mass was measured with a portable electronic balance to the nearest 0.01 g and maximum tarsus length was measured with an electronic calliper to the nearest 0.01 mm (Sutherland et al. 2004). I measured nestlings every day when possible and an effort was made to measure a particular nestling at the same time of day. As some measurements were missing, sample sizes differ among tests. Further, predation decreased sample sizes from hatching through to fledging. In calculations I used the raw data, results presented here are rounded up to 0.1 g to give biologically significant information. Part of the data set for “reed warbler” cuckoos was

3

published by Grim et al. (2003), however, these data were analysed in a different way and tested a different hypothesis. All data for “song thrush” and “blackbird” cuckoos are novel. I used nonlinear regression (SAS Institute 2000) to fit the measures of body mass W to the logistic function W(t)=A/(1 + e(–K*(t–ti ))) where W(t) is mass at age t, A is the asymptotic mass, K is a measure of growth rate and ti is the inflection point on the growth curve (Starck and Ricklefs 1998). I fitted logistic growth curves for each individual nestling. Table 1 provides average values for the three growth parameters (i) for all nestlings that survived at least one week and (ii) for subset of successfully fledged nests. Further, I analysed video recordings of cuckoo chicks (from 9 till 19 days of age) in two song thrush nests (11 and 8 hours in total respectively) to get information on food delivered by hosts. For a tentative interspecific test of a possible effect of host size on parasitic chick growth, I collated data on adult mass (averaged for the two sexes) from Cramp (1988, 1992) and data on chick growth from Werth (1947), Khayutin et al. (1982), Alvarez (1994), Kleven et al. (1999) and Numerov (2003). To assess the possible effect of host species on chick growth I fitted general linear mixed model (normal error distribution, parameter estimated by REML) using PROC MIXED in SAS (SAS Institute 2000). Nest was included as a random effect. Denominator df were computed using Kenward–Roger method. I looked for a significant interaction between host species and chick age to test whether chicks grow differently in the different hosts nests. Because of the relatively smaller sample sizes for song thrush and blackbird broods I used nonparametric tests. Data are presented as means±SD. All tests are two-tailed. All analyses were done with SAS (SAS Institute 2000) and JMP software. Results Cuckoo chicks raised in nests of the reed warbler, song thrush and blackbird did not differ in their mass after hatching (Kruskal-Wallis ANOVA: χ2=0.25, df=2, P=0.88). At 3 days of age both mass and tarsus length did not differ among the three hosts (Kruskal-Wallis ANOVAs, P>0.05) although both measures were lowest for “blackbird” cuckoos. Measured masses at fledging in nests of the reed warbler (70.8±4.6, n=25) and the song thrush (95.3±6.4, n=3) were very similar to asymptotic masses (logistic growth curve based on successful nests; Table 1) indicating that data are robust. Naturally parasitised nests I found no correlation between fledging age and mass at fledging for “reed warbler” cuckoos (rs=0.22, n=25, P=0.29). The negative correlation between fledging age and tarsus length at fledging was highly significant despite a small sample size (rs=–0.97, n=5, P=0.0048). All chicks fledged from reed warbler nests between 17 and 21 days old, with most fledging at 18 days of age. Experimental nests Cross-fostering experiments with song thrushes and blackbirds led to strikingly different results. In comparison to the commonly parasitized reed warbler, the cuckoo chicks in song thrush nests gained approximately one third more mass at comparable ages (Fig. 1). These differences appeared after day 3 and were most apparent at fledging when “song thrush” cuckoos were significantly (1.35 times) heavier than “reed warbler” cuckoos (Mann-Whitney test: U25,3=2.75, P=0.006). Differences in tarsus length at fledging were less prominent but again significant (P=0.04). I did not find any differences in variance of logistic growth parameters (a, K, t) between “song thrush” and “reed warbler” cuckoos (O’Brien’s tests: all P>0.25). Three out of six cross-fostered “song thrush” cuckoo chicks successfully fledged at the same age (18.3±0.6) as those in reed warbler nests (18.0±1.1; Mann-Whitney test: U3,25=0.78, P=0.43). Of the other three, one was killed by a falling tree, one was predated on and one disappeared while the host clutch remained in the nest. The last nestling may have

4

vanished due to partial predation or may have died and the corpse was removed by the foster parents. All these chicks grew well until their disappearance. Growth rate (K) was lower in “song thrush” cuckoos compared to “reed warbler” cuckoos. However, growth increments (mass increases in grams) were higher in “song thrush” cuckoos as indicated by their higher fledging mass (Table 1). Although “song thrush” cuckoo chicks attained higher mass before fledging, they did not fledge earlier (see above). Although the sample size for blackbird experiments was also small (six transferred chicks), the results are again clear cut. No significant differences were detected until the age of three days (the blackbird cuckoos data were within the lower range observed at reed warbler and song thrush nests) but no cuckoo chick survived to 14 days of age. Fig. 1c clearly shows very poor growth of cuckoo nestlings in blackbird nests; chicks survived for 2, 3, 4, 6, 6 and 13 days respectively. All six “blackbird” cuckoos weighed less at their maximum age than average “reed warbler” cuckoos at the same age (Wilcoxon matchedpairs test: Z=2.99, P=0.0028). “Blackbird” cuckoo chicks had bloated bellies and suffered from diarrhoea. Two of these chicks were found dead in the nests and four were predated on but this happened after their growth was stagnating or even declining. Because of this I could not calculate parameters of logistic growth curves for “blackbird” cuckoos. There was no indication of desertion as foster parents were observed near the nests. General linear mixed model confirmed significant effect of host species on cuckoo chick growth (age vs. species interaction: F2,629 =84.78, P<0.0001). Diet composition Video recordings at two song thrush nests with cross-fostered cuckoo chicks showed 55 feedings. At both nests the food composition was dominated by earthworms (62% of feedings in pooled data set) including large specimens (up to 12 cm in length). Smaller diet items were insects (29%) and molluscs (9%). Cuckoo chicks routinely begged even after being fed with large food items (e.g. earthworms, caterpillars). Interspecific comparison The average fledging mass of cuckoos raised by 8 host species studied by Werth (1947), Khayutin et al. (1982), Alvarez (1994), Kleven et al. (1999), Numerov (2003) and Grim (this study) showed a slight and non-significant tendency to increase with respective host species adult mass (rs=0.21, P=0.61; Fig. 2). Exclusion of the data for dunnock (due to adverse weather conditions during growth of the studied cuckoo chick, see Werth 1947) or rufous bush chat (because of smaller size of studied cuckoo subspecies and unusually very early fledging, see Alvarez 1994) showed that results are qualitatively the same. Fledging age nonsignificantly increased with chick size (rs=0.61, P=0.11) and was not correlated with host body size (rs=0.21, P=0.61; Fig. 3). Quadratic equations fit to these data (host adult mass vs. cuckoo chick fledging mass and host adult mass vs. cuckoo chick fledging age) were nonsignificant even after data transformation to normality. Discussion This study showed striking differences in growth performance and survival of parasitic cuckoo chicks in nests of two closely related non-parasitized songbird species in comparison to a sympatric frequently parasitized host species. While “song thrush” cuckoo chicks grew faster and fledged heavier than “reed warbler” cuckoos the “blackbird” cuckoo chicks showed relatively poor growth and none of them fledged successfully. The observed growth parameters and the successful fledging of three “song thrush” cuckoos clearly rejects the hypothesis that the song thrush is an unsuitable host in respect to diet quality or parental chick feeding rules (cf. Moksnes et al. 1990; Kleven et al. 1999). Cuckoo chicks in song thrush nests attained on average an even higher mass (95.3 g) than cuckoo chicks in the nests of the great reed warbler (88.3 g; Kleven et al. 1999). However, this difference did not approach significance (P=0.13) perhaps due to a low sample size. Chicks cared for by reed warblers grew in accordance with data published by others (see Table 2 in Kleven et al. 1999). The large variation in growth data from my study indicate that fledging of cuckoo chicks is not triggered by some threshold mass, but rather may be caused by skeletal

5

growth (see significant correlation between fledging age and tarsus length) or by tissue (e.g. feather) development (see also Starck and Ricklefs 1998). Why did “blackbird” cuckoos grow and survive so poorly? Subnormal growth and survival of “blackbird” cuckoos could be caused by several factors. (i) High energetic costs of eviction of large host eggs may play a role – blackbird eggs are heavier than song thrush eggs (7.0 vs. 5.5 g respectively; Cramp 1988). However, a cuckoo chick is able to evict blackbird eggs from a blackbird nest (M. Honza, C. Moskát, pers. comm.) and cuckoo chicks grew poorly even after I removed unevicted host eggs. Mass growth of “reed warbler” cuckoo chicks, which were unable to reject song thrush eggs from song thrush nests attached to reed warbler nests in another series of experiments, was similar to that of “reed warbler” cuckoos under normal conditions (T. Grim, unpublished data). (ii) “Blackbird” cuckoos growth may also be hampered by disease. However, I find it unlikely that all six “blackbird” cuckoo chicks would suffer from a disease while all six “song thrush” cuckoo chicks would remain healthy. (iii) The hypothesis that foster parents refuse to feed chicks can be rejected as nestling mass increased. However, one chick grew quite well until 11 days of age but then its growth declined (Fig. 1c). This raises the possibility of discrimination without recognition when parents might detect alien chicks by their longer nestling periods than in their own chicks (Grim et al. 2003; T. Grim unpublished data; blackbird chicks are well able to fledge even when 9 days old under normal conditions, Cramp 1988). However, all “blackbird” cuckoos grew subnormally even after spending only a few days in blackbird nests and five out of six did not survive for more than a week which casts doubt on this explanation. (iv) In some “blackbird” cuckoo chicks their growth stagnated. This pattern is similar to a cuckoo chick hatched in a nest of the linnet, Carduelis cannabina, that grew normally three days post-hatch but later its mass stagnated at 8.5 g until it died at the age of six days (Alvarez 1994). This indicates a possible role of food composition as linnet feeds seeds to its chicks while all used cuckoo hosts are insectivorous. A priori I expected that cuckoo chicks would not be able to survive in nests of both the song thrush and the blackbird because of large proportion of non-insect prey delivered by these potential hosts to their nestlings (also A. Moksnes, E. Røskaft, pers. comm.). Song thrushes feed nestlings mainly with molluscs while blackbirds with earthworms (Cramp 1988). In a striking contrast both large molluscs and earthworms are never fed to nestlings by reed warblers and great reed warblers (Grim and Honza 1997; Grim 1999). However, video recording revealed that cuckoos in song thrush nests were frequently fed with large earthworms and this caused no problems to parasitic chicks; on the contrary they grew even better than those fed on purely insect diet by reed warblers (unfortunately, no data on “blackbird” cuckoos diet were available). Interestingly, in another study rufous bush chats fed relatively high amounts of plant diet (grapes, 16.9% in naturally parasitized nests) to parasitic cuckoo chicks without any obvious adverse effects on their growth (Martin-Galvez et al. 2005). This indicates that any a priori expectations on the effect of diet on cuckoo chick growth performance should be treated with caution and should be tested experimentally. (v) Alternatively, cuckoo chicks may be unable to communicate their hunger effectively thus eliciting insufficient provisioning by fosterers. This hypothesis cannot be currently evaluated due to a lack of data on parental provisioning rules in blackbirds. (vi) The most intriguing explanation which fits the observed interspecific differences in growth considers host food allocation rules in respect to breeding strategy. Soler (2001, 2002) suggested that clutch-adjuster species (i.e. those in which usually all nestlings survive to fledging) preferentially feed smaller chicks while brood reducer species (i.e. those in which usually some chicks starve) prefer to feed larger chicks provoking the starvation of smaller chicks. If the parasitic chick is larger than host chicks the parasite would benefit from parasitising brood reducers. However, if the parasite is smaller than a host chick then clutch-adjuster host is a good choice. Blackbird and song thrush hatchlings weigh 6–8 g (Cramp 1988) which is 2–3 times more than the cuckoo hatchling (Honza et al. 2001). This size advantage of host chicks is unlikely to be compensated for by cuckoo chicks through a slightly shorter (11–13 vs. 12–14 days) incubation period due to the very fast growth of thrush chicks and perhaps also

6

improper incubation of parasitic eggs (cuckoo eggs are considerably smaller than thrush eggs; Cramp 1988; Honza et al. 2001). The blackbird with its asynchronous hatching (Cramp 1988) is a brood reducer (Magrath 1992) while the song thrush shows synchronous hatching (Cramp 1988) and can be considered clutch adjuster. Thus, the Soler (2002) hypothesis is apparently in accordance with cuckoo chicks surviving in nests of the song thrush but not in those of the blackbird. However, in my experimental nests there were no host chicks left (see Methods) thus leaving no role for intra-brood competition and parental favouritism. Nevertheless, previous coevolution between the cuckoo and its various hosts, could adapt cuckoo chicks for growth in song thrush nests. Quick elimination of cuckoo chicks in blackbird nests through brood reduction might prevent any coevolutionary adaptation of cuckoo chicks to blackbird parental provisioning rules and/or food composition. Physiological constraints on growth Cuckoo chick growth was very quick during the first two weeks after hatching but after day 15 it levelled off (Fig. 1a,b). Interestingly, one “song thrush” cuckoo chick did not retard its growth as described above but increased its mass until fledging (see also Werth 1947). This observation and much faster growth of “song thrush” cuckoos in comparison to “reed warbler” cuckoos clearly rejects previous suggestions that cuckoo chicks grow in reed warbler nests at the maximum rate enabled by physiological constraints (see e.g. Brooke and Davies 1989; Alvarez 1994). However, “physiological constraint on growth” hypothesis was already falsified by growth data from cuckoos reared by great reed warblers (Kleven et al. 1999). In an interesting experiment Bussmann (1947) hand reared a young cuckoo. This ad libitum fed nestling reached 80.0 g on day 14 (day 13 in Bussmann’s growth marking) and 100.0 g when 18 days old (fledging time). These data are very close to those obtained in the current study from “song thrush” cuckoos (80.0 on day 14 and 95.0 g at fledging). This indicates that “song thrush” cuckoos indeed grow close to the upper limit of their physiological possibilities. This suggests that there is a limit on how cuckoo chicks can exploit host parental care. To gain more insight into this issue it would be interesting to study chick provisioning rules in large cuckoo hosts and test how they are exploited by parasitic chicks. Interspecific comparison The tentative interspecific comparison did not support the hypothesis that by parasitising larger hosts, the cuckoo chick benefits by higher fledging mass and/or shorter nestling period. Kilpatrick (2002) also found no significant effect of host size on growth of parasitic brown-headed cowbird chicks while Kilner (2003) and Kilner et al. (2004) found that cowbirds grow fastest in intermediate host sizes with 1–2 host nestmates. However, the sample size in my study is very small. It would be interesting to repeat this comparison after more data on cuckoo growth in various host species is available. Conclusions Results of this study indicate that host size may positively influence cuckoo chick growth (song thrush host) but other currently unidentified factors may play a role (blackbird host). The fact that “song thrush” cuckoo chicks grew well and fledged at a very high mass raises a question as to why song thrushes are not regularly parasitized by cuckoos. At present, one can only reject hypotheses that this host is too aggressive or rejects parasitic eggs at high rates (Grim and Honza 2001a). Alternatively, poor survival of “blackbird” cuckoos raises questions concerning the origins of blackbird behaviours that are so far interpreted as antiparasitic responses (Davies and Brooke 1989; Grim and Honza 2001a; Moskát et al. 2003). The biology of parasitic cuckoo chicks has so far received unwarrantedly little attention in general (Grim 2005b, 2006). This study indicates that cross-fostering of parasitic chicks to nests of currently non-parasitized hosts, yields important insights into host selection and growth plasticity in brood parasitic birds (see also Soler 2002). This gives strong impetus for future research in this unexplored area.

7

Acknowledgements Suggestions by three anonymous referees and editor substantially improved the paper. I would like to thank M. E. Hauber, O. Kleven and V. Remeš for their comments on earlier versions of the MS. P. Procházka kindly provided some video recordings. I am grateful to all who helped with the fieldwork: A. Dvorská, M. Honza, O. Kleven, B. Matysioková, A. Moksnes, P. Procházka, E. Røskaft, G. Rudolfsen and V. Šícha. When working on this paper I was supported by grants from the Research Council of Norway, the Czech Ministry of Education (grants No. 153100012 and No. MSM6198959212) and the Grant Agency of the Czech Republic (206/03/D234). No nest was abandoned after experiments were performed. The experiments were done under license from The Central Commission for Animal Welfare of the Czech Republic (No. 065/2002–V2) and in accordance with the laws and ethical guidelines of the Czech Republic. References Alvarez F (1994) Rates of weight increase of cuckoo (Cuculus canorus) and host (Cercotrichas galactotes) chicks. Ardeola 41:63–65 Brooke ML, Davies NB (1989) Provisioning of nestling cuckoos Cuculus canorus by reed warbler Acrocephalus scirpaceus hosts. Ibis 131:250–256 Bussmann J (1947) Wachstum und Jugendzeit eines Kuckucks. Ornithol. Beobachter 44:41–49 Cramp S. (ed) (1988) The birds of the Western Palearctic. Tyrant Flycatchers to Thrushes. Volume V. Oxford University Press, Oxford Cramp S. (ed) (1992) The birds of the Western Palearctic. Warblers. Volume VI. Oxford University Press, Oxford Davies NB (1986) Reproductive success of dunnocks, Prunella modularis, in a variable mating system. 1. Factors influencing provisioning rate, nestling weight and fledgling success. J Anim Ecol 55:123–138 Davies NB (2000) Cuckoos, cowbirds and other cheats. T and AD Poyser, London Davies NB, ML Brooke (1989) An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. I. Host egg discrimination. J Anim Ecol 58:207–224 Glassey B, Forbes S. (2003) Why brown-headed cowbirds do not influence red-winged blackbird parent behaviour. Anim Behav 65:1235–1246 Glue D, Morgan R (1972) Cuckoo hosts in British habitats. Bird Study 19:187–192 Grim T (1999) Food of great reed warbler (Acrocephalus arundinaceus) young. Sylvia 35:93– 99 (in Czech, with summary in English) Grim T (2005a) Host recognition of brood parasites: implications for methodology in studies of enemy recognition. Auk 122:530–543 Grim T (2005b) Mimicry vs. similarity: which resemblances between brood parasites and their hosts are mimetic and which are not? Biol J Linn Soc 84:69–78 Grim T (2006) The evolution of nestling discrimination by hosts of parasitic birds: why is rejection so rare? Evol Ecol Res (in press). Grim T, Honza M (1997) Differences in parental care of reed warbler (Acrocephalus scirpaceus) to its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zool 46:135–142 Grim T, Honza M (2001a) Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to experimental parasitism by the common cuckoo Cuculus canorus. Biologia 56:549–556 Grim T, Honza M (2001b) Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behav Ecol Sociobiol 49:322–329 Grim T, Kleven O, Mikulica O (2003) Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Proc R Soc Lond B 270:S73– S75 Honza M, Picman J, Grim T, Novak V, Capek M, Mrlik V (2001) How to hatch from an egg of great structural strength: a study of the common cuckoo. J Avian Biol 32:249–255 Khayutin SN, Dmitrieva LP, Tartygina NG, Aleksandrov LI (1982) The behaviour of a nestling of Cuculus canorus in the nest of Phoenicurus phoenicurus. Zool zhurnal 61:1063–1077 (in Russian with English summary) Kilner RM (2003) How selfish is a cowbird nestling? Anim Behav 66:569–576 Kilner RM, Davies NB (1999) How selfish is a cuckoo chick? Anim Behav 58:797–808

8

Kilner RM, Madden JR, Hauber ME (2004) Brood parasitic cowbird nestlings use host young to procure resources. Science 305:877–879 Kilpatrick AM (2002) Variation in growth of brown-headed cowbird (Molothrus ater) nestlings and energetic impacts on their host parents. Can J Zool 80:145–153 Kleven O, Moksnes A, Røskaft E, Honza M (1999) Host species affects the growth rate of cuckoo (Cuculus canorus) chicks. Behav Ecol Sociobiol 47:41–46 Lindén M, Gustafsson L, Pärt T (1992) Selection on fledging mass in the collared flycatcher and the great tit. Ecology 73:336–343 Magrath RD (1991) Nestling weight and juvenile survival in the blackbird Turdus merula. J Anim Ecol 60:335–351 Magrath RD (1992) Roles of egg mass and incubation pattern in establishment of hatching hierarchies in the blackbird (Turdus merula). Auk 109:474–487 Martin TE (1987) Food as a limit on breeding birds: a life-history perspective. Annu Rev Ecol Syst 18:453–487 Martin-Galvez D, Soler M, Soler JJ, Martin-Vivaldi M, Palomino JJ (2005) Food acquisition by common cuckoo chicks in rufous bush robin nests and the advantage of eviction behaviour. Animal Behav 70:1313–1321 Moksnes A, Røskaft E (1992) Responses of some rare cuckoo hosts to mimetic cuckoo eggs and foreign conspecific eggs. Ornis Scand 23:17–23 Moksnes A, Røskaft E (1995) Egg-morphs and host preference in the common cuckoo (Cuculus canorus): an analysis of cuckoo and host eggs from European museum collections. J Zool 236:625–648 Moksnes A, Røskaft E, Braa AT, Korsnes L, Lampe HM, Pedersen HC (1990) Behavioural responses of potential hosts towards artificial cuckoo eggs and dummies. Behaviour 116:64–89 Moskát C, Karcza Z, Csörgö T (2003) Egg rejection in European blackbirds (Turdus merula): the effect of mimicry. Ornis Fennica 80:86–91 Numerov AD (2003) Interspecific and intraspecific brood parasitism in birds. Federal State Unitary Enterprise Publish & Polygraf Corporation, Voronezh (in Russian with English summary) Rothstein SI, Robinson SK (1998) Parasitic birds and their hosts. Oxford University Press, New York Saether B-E (1994) Food provisioning in relation to reproductive strategy in altricial birds: a comparison of two hypotheses. Evolution 48:1397–1406 SAS Institute (2000) SAS online document, version 8. SAS Institute, Cary, N.C. Soler M (2001) Begging behaviour of nestlings and food delivery by parents: the importance of breeding strategy. Acta Ethol 4:59–63 Soler M (2002) Breeding strategy and begging intensity: influences on food delivery by parents and host selection by parasitic cuckoos. In: Wright J, Leonard ML (eds) The evolution of Begging. Kluwer, Dordrecht, pp 413–427 Soler M. and Soler JJ (1991): Growth and development of great spotted cuckoos and their magpie hosts. Condor 93:46–54 Starck JM, Ricklefs RE (eds) (1998) Avian growth and development. Evolution within the altricial-precocial spectrum. Oxford University Press, New York Strausberger BM (1998) Temporal patterns of host availability, brown-headed cowbird brood parasitism, and parasite egg mass. Oecologia 116:267–274 Sutherland WJ, Newton I, Green RE (2004) Bird ecology and conservation. A handbook of techniques. Oxford University Press, Oxford Trine CL (1998) Wood thrush population sinks and implications for the scale of regional conservation strategies. Conserv Biol 12:576–585 Werth I (1947) The growth of a young cuckoo. Br Birds 40:331–334 Wiley JW (1986) Growth of shiny cowbird and host chicks. Wilson Bull 98:126–131 Woolfenden BE, Gibbs HL, Sealy SG, McMaster DG (2003) Host use and fecundity of individual female brown-headed cowbirds. Anim Behav 66:95–106 Wyllie I (1981) The cuckoo. Batsford, London

9

Table 1 Basic growth parameters for cuckoo chicks raised by reed warblers (RW) and song thrushes (ST). A is the asymptotic mass, K is a measure of growth rate and ti is the inflection point on the growth curve (Starck and Ricklefs 1998). Results of separate analyses (Mann-Whitney tests) for all (n=65 and 5 for RW and ST respectively) vs. only successful (n=25 and 3 for RW and ST respectively) nests are shown. Values are means±SD with the sample sizes in parentheses. For explanation see Methods. Growth parameters Whole data set A K ti

RW

ST

Significance

64.3±13.5 0.429±0.101 8.1±1.4

97.4±2.6 0.356±0.039 9.6±0.7

0.0002 0.11 0.01

Fledged A K ti

71.6±5.8 0.372±0.085 9.0±0.9

97.05±2.91 0.357±0.045 9.6±0.8

0.0016 0.80 0.22

Fig. 1 Mass growth of cuckoo chicks in nests of reed warblers (a), song thrushes (b) and blackbirds (c). Day 1=day of hatching. Fig. 2 Cuckoo chicks fledging mass plotted against the host adult mass (sexes averaged). DU = dunnock, GRW = great reed warbler, RCB = rufous bush chat, RE = redstart, RO = robin, RW = reed warbler, ST = song thrush, WW = white wagtail. Fig. 3 Cuckoo chick fledging age plotted against host adult mass (sexes averaged). DU = dunnock, GRW = great reed warbler, RCB = rufous bush chat, RE = redstart, RO = robin, RW = reed warbler, ST = song thrush, WW = white wagtail.

10

Mass (g)

110 100 90 80 70 60 50 40 30 20 10 0 0

5

10

15

20

25

20

25

20

25

Age (days)

Mass (g)

Fig. 1 (a)

110 100 90 80 70 60 50 40 30 20 10 0 0

5

10

15

Age (days)

Mass (g)

Fig. 1 (b)

110 100 90 80 70 60 50 40 30 20 10 0 0

5

10

15

Age (days)

11

Fig. 1 (c)

Chick fledging mass (g)

120

RO WW

RE

100

ST GRW

80

DU RW

60

RBC

40 20 0 0

20

40

60

80

Host adult mass (g)

Fig. 2

Chick fledging age (days)

25

DU

WW

RO

20

RW

15

ST

GRW RE RBC

10 5 0 0

20

40

60

Host adult mass (g)

Fig. 3

12

80

8. Grim T.: Signals of need in parent-offspring communication: why gaping and calling 8. provisioning rates? alone fail to explain Grim T.: Signals of need in parent-offspring communication: why gaping and calling (subm.) alone fail to explain provisioning rates? (subm.)

Parent-offspring communication: why gaping and calling alone do not explain provisioning rates? Tomáš Grim1 1 Department of Zoology Palacky University tr. Svobody 26 CZ-771 46 Olomouc Czech Republic e-mail: [email protected]

Abstract Understanding the mechanisms of parental provisioning rules is critical for any study of begging, parent-offspring conflict and communication between relatives. In a very influential study Kilner et al. (1999, Nature 397: 667–672) showed that reed warbler (Acrocephalus scirpaceus) parents adjust their feeding frequencies according to the gape area displayed and the call rate of their chicks. In a crucial test of generality of this parental rule, the authors showed that warblers used exactly the same rules when feeding own and parasitic cuckoo (Cuculus canorus) chicks. However, only relatively young cuckoo chicks were considered in this analysis. Here I show that results of Kilner et al. (1999) do not hold for older cuckoo chicks because both gape area and call rates level off long before fledging while both provisioning rates and chick mass continue to rise till fledging. I propose that when calling and gaping no longer solicit sufficient provisioning, parasitic chicks add another component of the multiple signal to increase feeding rates – asymmetrical wing-shaking (i.e. the nestling raises one of its wings at a time at angle up to a maximum of ~90º above horizontal and slightly shakes the wing in that raised position). Comparative data support this hypothesis. These new analyses have profound implications for both experimental and theoretical studies of begging because visual cues, additional to gaping (e.g. wing-shaking), will have to be considered to understand avian provisioning rules and their exploitation by brood parasites.

1

INTRODUCTION Begging in avian chicks is at the core of studies of parent-offspring conflict, communication and signalling (Kilner and Johnstone, 1997; Wright and Leonard, 2002). One crucial question is what cues parents use and what rules they follow when deciding how to provision to their young. In a ground-breaking and very influential paper Kilner et al. (1999) showed that reed warblers (Acrocephalus scirpaceus) adjust their feeding frequencies according to the area of gapes and the rate of begging calls of the brood. The generality of this parental decision rule was established by a critical match between observed and predicted call rates of parasitic cuckoo (Cuculus canorus) chicks raised by warblers. The authors concluded that “reed warblers follow exactly the same integration rule when provisioning a single cuckoo in their nest as when feeding a brood of their own young” (p. 667). Cuckoo chicks have relatively small gapes and therefore they have to compensate for this subnormal visual stimulus with supernormal vocal stimulus (Kilner et al., 1999). This lends support to the conclusion that gape area and call rate are the most important or perhaps only cues for adjusting the intensity of parental care (Kilner et al., 1999). However, that study examined only relatively young cuckoo chicks (88% of chicks aged <11 days, n=25). This leaves open the question of how older cuckoo chicks solicit sufficient provisioning when they are much larger than the entire host brood at fledging, i.e. when parents are working far outside their normal feeding rates at non-parasitized nests (Grim and Honza, 2001; Grim et al., 2003). I will show that feeding frequencies to old cuckoo chicks cannot be predicted from the feeding frequency formula of Kilner et al. (1999), and that old cuckoo chicks use additional begging strategies to solicit sufficient provisioning by fosterers. Further, I will offer an explanation for discrepancy between observed and predicted call rates in younger cuckoos too. METHODS In the present study I used empirical data collected from 1994 to 2005 in two fish-pond areas in the Czech Republic (47° 40' N, 16° 48' E). A detailed description of the study areas and all field procedures is presented elsewhere (Grim and Honza, 1997, 2001). For the analyses I used both field and literature data. I retrieved literature data from figures in Kilner and Davies (1999), Kilner et al. (1999) and Butchart et al. (2003) to reanalyse relationships between gape area, call rate and the age of nestlings. To test if the data are better fitted by linear or polynomial regressions I used the forward selection procedure (Zar, 1999, p. 453). In all cases the second-order polynomial regression provided better fit than linear or higher-order polynomial regressions. Further, I investigated parental care by reed warblers measured as feeding frequencies (number of feedings per hour) and feeding rates (amount of dried food in mg delivered per hour). Food samples for the latter measure were collected with a neck-collar (for a detailed description of the method see Grim and Honza, 1997, 2001). I weighed chicks to the nearest 0.1 g and measured the length and width of the bill to the nearest 0.1 mm. Because formerly I used a different formula to calculate the gape area in the previous study (Grim and Honza, 2001) I recalculated all data according to the formula of Kilner et al. (1999) for the current study (i.e. gape area=bill width*bill length). To make a meaningful comparison of growth rates of an average host and cuckoo brood I used only weight measurements that were obtained during the period of exponential growth of chicks (Starck and Ricklefs, 1998). The exponential growth period of reed warblers was from the age of two days (hatching day = 0) till the age of seven days (own unpublished data). The cuckoo chick reaches the period of exponential growth later (Grim, 2006). Therefore, the comparable five-day period of exponential growth of cuckoo chicks was from the age of five days till the age of ten days. An alternative comparison of the same aged broods of the host and the parasite would make no sense as when being of the same age the two types of broods are in different periods of their development (Starck and Ricklefs, 1998) and the rate of parental provisioning is not determined by the chick age but by the chick mass and stage of development (Grim and Honza, 2001). I included only nests that were measured both at the younger and older age (i.e. I filtered out potential confounding effects of among brood variance in growth rates). Particular broods were measured and observed at the same time of day throughout their nestling periods, thus eliminating any day time effects on measurements.

2

In the comparisons between brood mass and feeding frequency/rate I fitted general linear mixed models with species, brood mass and their interaction as factors, brood identity as a random effect and feeding frequencies or rates as a response. I used Box-Cox transformation to achieve normality of the data. I studied 99 cuckoo nestlings and 145 warbler broods (74 four- and 71 single-chick broods). In comparison of feeding frequencies the results were identical when single-chick broods were excluded. In the comparison of feeding rates only data from single-chick broods were available. When being fed the old cuckoo chicks (> 16 days) frequently raise and shake one of their wings (see below). To test whether this asymmetrical wing-shaking is related to begging, I analysed in detail video recordings of old (16–19 days old) cuckoo chicks begging in the nests of the reed warbler (n=4 chicks, 6 hours) and the song thrush (Turdus philomelos; n=2 chicks, 19 hours; all chicks originally hatched in reed warbler nests). I recorded chick behaviour both during feeding and when parents were not present. I differentiated cases when cuckoo chicks raised a wing above their shoulder level (the wing was at least 45º above horizontal, but usually ~90 º above horizontal, and the underwing was clearly visible from host parent position) and cases when they shook the wing but did not raise it and the underwing was not revealed. Additionally, I analysed video-recordings of younger (0–15 days old) cuckoo chicks in warbler nests (n=31 chicks, 181 hours). I used all available data therefore the sample sizes differ among analyses. In all cases I used only directly obtained measurements and I did not calculate any data used in the analyses (cf. Kilner et al., 1999). RESULTS AND DISCUSSION Why young cuckoo chicks beg at higher call rates than expected? According to Kilner et al. (1999) reed warblers use exactly the same decision rule when provisioning a single cuckoo in their nest as when feeding a brood of their own young. However, there is reason to believe that cuckoo chicks do not behave exactly according to predictions of Kilner et al. (1999). Kilner et al.’s (1999) feeding frequency formula (p. 669) is based on an assumption that host and parasitised broods grow at the same rate. In reality cuckoo chicks grow much faster than a median host brood. Data from my study area show that host’s four-chick brood (n=33) gains an average of 23.8±3.0g of mass during five days of the period of exponential growth from the age of two (19.3±3.2g) till the age of seven days (43.0±2.3g). Instead a cuckoo chick (n=25) gains 23.9% more mass (29.5±4.7g) in a comparable five day period of its exponential growth from the age of five (19.5±3.4 g) till the age of ten days (49.1±6.2g). Cuckoos in England gain even more mass during the same period (34.4g, Wyllie, 1981). Warbler and cuckoo broods in my sample weighed the same at the start (t58=0.28, P=0.78) while their masses were highly significantly different after the five day period (t58=4.60, P<0.0001). The difference of ~6g between host and parasitic broods is not trivial as it amounts to ~25% higher increase of mass in the cuckoo chick in comparison to the host’s own brood. This suggests that the cuckoo chick should receive higher amounts of food (the feeding rate) than the host brood of the same mass. This suggestion is based on reasonable assumptions that (i) the effectiveness of digestive processes does not significantly differ between host and parasitic chicks (which is highly likely as host and parasite chicks are under identical selection pressure on digestive physiology due to being fed with the same insect diet; Grim and Honza, 1997, 2001) and (ii) thermoregulation costs do not vary with brood size so as to significantly affect growth rates (which is also likely as growth rates do not differ between warbler chicks raised alone and those raised in four chick broods; T. Grim and B. Matysiokova, unpublished data). Therefore, it may be expected that cuckoos in the exponential growth period should call at higher rates – to obtain higher amounts of food – than would be predicted by Kilner et al.’s (1999) formula. This is indeed so: Figure 6 in Kilner et al. (1999) shows noticeably higher (~30–200% higher) observed than predicted call rates in cuckoos 7–9 days old (gape area: 240–300 mm2). Thus, cuckoo chicks in the exponential phase of growth must beg at higher call rates than expected under Kilner et al. (1999) formula to achieve significantly faster growth than host broods (for similar results in an other host-parasite system see Soler et al. 1999). As cuckoo chicks grow faster they should receive significantly more food than similar sized host chicks/broods (see above). In contrast, Kilner et al. (1999) concluded that there is no difference in provisioning of the cuckoo chick and same-sized host brood. Grim and

3

Honza (2001) questioned the validity of this result by showing that the cuckoo chick obtains significantly higher mass of food than the same-sized host chick. Indeed, the feeding frequencies used by Kilner et al. (1999) may be a too rough measure to reveal subtle differences in provisioning of cuckoo and host chicks (Grim and Honza, 2001). Feeding frequencies explain only little variability in feeding rates which are clearly more precise and direct measure of parental care (Figure 1) because feeding frequencies may correlate with true consumption even negatively (see Royama, 1966; Sejberg et al., 2000). Therefore, I predicted that feeding frequencies may be unable to discriminate between provisioning of cuckoo and warbler chicks while feeding rates (biomass of food, see above) may be. Feeding frequencies increased with brood mass (Figure 2, GLMM: F1,359=344.72, P<0. 0001), but did not differ between host and parasite broods (F1,86.5=0.03, P=0.86) (interaction mass*species was trivial, F1,360=0.00, P=0.99, thus was removed). When I excluded all data from cuckoo chicks weighing more than the heaviest warbler brood (48g; 80 measurements, 21 chicks), the results were qualitatively identical. In contrast, feeding rates increased not only with brood mass (Figure 3, F1,130=65.21, P<0.0001) but were significantly higher in cuckoo chicks (F1,54=37.48, P<0.0001) (the interaction mass*species was insignificant: F1,135=1.15, P=0.29). After including data from cuckoo chicks larger than the largest host chick (12 g; 71 measurements, 24 chicks) the results were qualitatively similar. Figure 3 shows only data from the same range of host and parasitic chicks used in the former analysis from technical reasons: range of mass data for cuckoos (3.7–68.0 g) was ~7 times higher than that for warblers (2.3–11.5 g). Thus, feeding frequencies did not while feeding rates did reveal differences in provisioning of host own and parasitic chicks. It follows that the feeding rule formula of Kilner et al. (1999) may have used an interrelated, but weaker measure of parental care, than food amount and the latter may better predict call rates of young cuckoo chicks. How do old cuckoo chicks procure sufficient provisioning? A paradox The elegantly simple feeding frequency formula presented by Kilner et al. (1999) included only gape areas displayed and calling rates (p. 669). The importance of the two components for provisioning of cuckoos was age-dependent: gape area increased with age less steeply than begging calls (Figure 4 in Kilner et al., 1999). Moreover, during the first days posthatch cuckoo chicks gape silently (Kilner and Davies, 1999). From the age of 6 days onwards cuckoo chicks continually increase the rates of their calling to remarkable levels (Davies et al., 1998; Kilner and Davies, 1999; Kilner et al., 1999). However, the cuckoo chick of course cannot increase the rate of calling ad infinitum. The rate of calling levels off after cuckoo chicks are ~14–15 days old as cuckoos might be unable to increase already extremely high calling rates any more (V. Sicha, pers. comm.). Indeed, variation in cuckoo chicks’ begging call rate with age is better fitted by a second order polynomial regression (R2=0.84, F2,25=58.40, P<0.0001, first-order regression coefficient: t=10.81, P<0.0001; second-order regression coefficient: t=–5.07, P<0.0001) than by a linear regression (R2=0.66; data from Figure 2 in Kilner and Davies, 1999). Also data from another study on variation of cuckoo chicks’ begging call rate with gape area are better fitted by a second order polynomial regression (R2=0.84, F2,29=65.94, P<0. 0001, first-order regression coefficient: t=11.34, P<0.0001; second-order regression coefficient: t=–2.74, P<0.01) than by a linear regression (R2=0.78; data from Figure 6 in Kilner et al., 1999). When I used chick age (provided in caption of Figure 6 in Kilner et al., 1999) instead of gape area in the latter analysis, the results were again highly significant (R2=0.88, F2,29=97.85, P<0.0001, firstorder regression coefficient: t=13.93, P<0.0001; second-order regression coefficient: t=–6.47, P<0.0001). This analysis is more informative because gape area does not increase in old cuckoo chicks (see below). Moreover, the pattern of call rates levelling off at ~14–15 days of chick age is found consistently in cuckoo chicks raised by other host species. For instance, in cuckoos from meadow pipit (Anthus pratensis) nests, the relationship between age and call rate is better fitted by a second order polynomial regression (R2=0.80, F2,9=12.32, P=0.0075, first-order regression coefficient: t=4.01, P=0.007, second-order regression coefficient: t=–2.55, P=0.04) than by a linear regression (R2=0.59; data from Figure 4 in Butchart et al., 2003). Except for one measurement in a nest of the great reed warbler (Acrocephalus arundinaceus), cuckoo call rates do not increase above 150–160 calls per 6 seconds in any host species (Figure 4 in Butchart et al., 2003). Unfortunately, there are no data from great reed warbler–cuckoos

4

after 14 days of age and in dunnock–cuckoos the range of data is too narrow for any meaningful analyses. However, in reed warbler–cuckoos the data from Butchart et al. (2003) are again clearly non-linear. These data are apparently not the same as those presented in Kilner et al. (1999) or Kilner and Davies (1999), providing an independent sample. To sum up, the pattern of age-dependent changes in call rates is strongly non-linear in cuckoos: young chicks do not call at all, mid-aged chicks increase their calling rates and old chicks’ level off their calling rates. Furthermore, bill (and gape) growth slows down in older cuckoo nestlings: 14 day old cuckoo chicks have almost the same gape area as those 18 days old (Figure 2; mean±SD gape area at 14 days: 361.2±7.4 mm2, n=16 chicks, gape area at 18 days: 372.3±11.3 mm2, n=6 chicks; see also Figure 4b in Kilner et al., 1999). Thus, there is an increase of only ~3% of gape area during four days of growth which is obviously a biologically insignificant change because changes in the gape area play only a marginal role in eliciting parental care in reed warblers: “feeding frequency=0.0162 (gape area displayed (in mm2) + 0.178 (calls per 6 s) + 8.23” (Kilner et al., 1999, p. 669). Indeed, substituting the average gape areas at 14 and 18 days into Kilner et al.’s (1999) formula and keeping call rates the same shows that the observed increase of the gape area may lead to the increase of feeding frequency of only 0.18 feeds. In a striking contrast, mass increases 12% during the same period (Grim, 2006). In contrast to the levelling off in gape area growth and call rates in old cuckoo chicks feeding frequencies to old cuckoo chicks continue to rise till fledging (regression of feeding frequency against chick age, own data for 14–21 days: GLMM: F1,38.2=7.57, P=0.009; see also Figure 3 in Kilner et al., 1999). Cuckoo chicks also significantly increase their mass till fledging (Grim 2006; Kleven et al., 1999). This increase in provisioning of the old cuckoo chicks till fledging cannot be explained by (i) gape area growth (see above), (ii) proportion of time spent gaping (8–11 days old cuckoos already gape continuously during begging, Figure 3 in Kilner and Davies, 1999), (iii) gape colour (which apparently does not influence feeding rates in reed warblers, see Noble et al., 1999, Grim and Honza, 2001, moreover the colour of the cuckoo gape does not change after 3 days post-hatch, Wyllie, 1981), (iv) call rates (as calling rate does not increase in old chicks, see above), (v) large body size (which is alone insufficient to stimulate reed warblers to increase provisioning, Davies et al., 1998) or (vi) by loudness of the calls (Davies et al., 1998, Butchart et al., 2003) which does not increase in old cuckoo chicks (V. Sicha, pers. comm.). Kilner et al. (1999) studied chicks almost solely before the age of ~11 days (only three out of 25 data points are for older cuckoos and there are no data for cuckoos older than 15 days). For these younger cuckoo chicks the formula provided by Kilner et al. (1999) seems to predict feeding frequencies reasonably well (but see above). However, the formula obviously cannot explain provisioning of old cuckoo chicks. To sum up, progressively increasing feeding frequencies and mass gain of old cuckoos until fledging cannot be explained by increasing gaping or calling as both these signals reach their upper limit long (~4–7 days) before fledging. A likely solution to this paradox is that old cuckoo chicks stimulate their hosts with some other so far unconsidered additional behaviour as well. I will show that this is indeed so. The solution of the paradox Wyllie (1981) reported on 16 days old common cuckoo chicks raised by reed warbler hosts that “When the hosts arrive with food [the cuckoo chick] quivers one outstretched wing and vibrates its huge mouth“ (p. 157). Indeed, older common cuckoo nestlings (>16 days) raise and shake their wings in nests of both reed warbler and song thrush (Turdus philomelos) hosts (own data). When the host arrives at a nest a chick raises one of its wings above the shoulder and slightly shakes it in that raised position. These observations are nothing aberrant: begging of cuckoo nestlings or fledglings with one outstretched shaking wing is pictured in almost all standard literature sources (e.g. Cramp, 1985, p. 411; Davies, 2000, the cover; Glutz von Blotzheim and Bauer, 1980, p. 210; Johnsgard, 1997, p. 189) and is featured on photographs in popular literature as well (e.g. Malchevsky, 1987, p. 113; Wyllie, 1981, photo no. 33). In general, wing-shaking (or “flapping”) is considered one of the mechanisms of begging (Wright and Leonard, 2002). Wing-shaking by both wings is commonly assumed to function as a stimulus to induce parental feeding by begging nestlings of many passerines and non-passerines from various phylogenetically unrelated families (Budden and Wright,

5

2001; Kilner, 1995; Leonard and Horn, 1998; Redondo and Castro, 1992; Smith et al., 2005; D. W. Mock pers. comm.). Brood parasites are no exception (Hauber and Ramsey, 2003; Payne, 2005; Soler et al. 1999). Wing-shake begging is generally interpreted as a signalling of highest nutritional need of the chick (see references above). I hypothesize that cuckoo’s asymmetrical wing-shaking (i.e. raising and shaking of one wing at a time) provides a solution to the apparent discrepancy between the results of Kilner et al. (1999) and observed increase of feeding frequencies to old cuckoo chicks. Wingshaking, similarly to head- and gape-shaking, may visually enhance the signal to hosts just as they do in many other species studied so far (see references above). This hypothesis predicts that wing-shaking should appear when gaping and calling are no longer sufficient to solicit sufficient provisioning by hosts (i.e. when both gape growth and call rates level off). The use of wing-shaking should increase with age in old cuckoos and should be used mainly or only when the fosterer is arriving at the nest with food. As wing-shaking in common cuckoos is asymmetrical (i.e. only one wing is raised at a time), I predict that if the behaviour is related to begging the raised wing should be directed towards the approaching fosterer. As predicted, the raising and shaking of one wing were performed without exception when a host either approached the nest or sat in a position to feed the chick (119 out of 119 cases). I did not observe a single case of asymmetrical wing-shaking in the absence of hosts. In the absence of the host older chicks sometimes simultaneously raised both their wings above the body level but they only stretched them and never shook them, which is in a striking contrast to asymmetrical wing-shaking performed solely during begging. During parental visits the wing was raised high and the underwing was clearly seen in 50 out of 119 feedings (42.0%, n=6 chicks; in all other cases the shaking wing was lower and it was impossible to see the underwing). Younger chicks (<16 days) did not raise or shake their wings (2623 feedings, 181 hours, n=31 chicks). The difference between young and old chicks is highly significant (presence/absence of wing-shaking, each chick represented by a single data point: 2=32.80, df=1, P<0.0001). Shortly before fledging (age: 17–19 days) chicks shook a wing every time the host approached the nest. This is illustrated by a positive correlation between the age of old (15–19 days) cuckoo chicks and percentage of feedings when a wing was raised sufficiently high to see its underside (rs=0.93, P=0.002, n=7 samples from different chicks). Older chicks almost always (96% of cases, n=50 observed wing raisings in 6 chicks aged 16–19 days) raised and shook the wing on the side of the approaching parent and even changed e.g. the left wing for the right one when a parent approaching from the left finally arrived at the right side of the nest (and vice versa). The relationship between the direction of approaching parent and the direction of wing-shaking was highly non-random ( 2=30.54, df=1, P<0.0001). This clearly indicates that wing-shaking is directed at host parents. It would be interesting to test whether presence of wing-shaking varies with time since previous feeding (i.e. a rough measure of hunger; Tanaka and Ueda, 2005), although this does not seem to be the case in the common cuckoo as old chicks shook their wings during all hosts visits to the nest. Moreover, old chicks are fed every 2–3 minutes (feeding frequencies up to 54 feeds/hour, Figure 2) and thus there was no variance to be explained by the time since the nestling was last fed. However, it is yet to be determined whether the frequency or the amplitude of wing-shaking may indicate hunger levels in old cuckoo chicks. Regular video recordings unfortunately cannot be used because wing-shake frequencies in some birds are too close to the frame-sampling rates of regular video film (Hauber and Ramsey, 2003; own video-recordings). Thus, (i) increasing feeding frequencies and mass of cuckoo chicks after the age of 15 days, (ii) the inability of old cuckoo chicks to increase both gaping and calling stimulation of host parents and (iii) the start of wing-shaking at ~16 days of age are consistent with the hypothesis that wing-shaking is a begging strategy of cuckoo chicks employed when gaping and calling cease to be sufficient to solicit enough provisioning from host parents. This would also explain the fast increase of the use of the wing-shake strategy from 0% of cases of begging at age 15 days up to 100% of cases of begging at age 17 days of cuckoo chicks. The trick must work otherwise old cuckoo chicks could not be provisioned at increasing levels and gain weight after 15 days of age till fledging 3–6 days later; note that there is a good evidence against alternative explanations (see above). Wing raising and shaking during begging in older cuckoo chicks cannot be explained by some additional alternative

6

hypotheses, e.g. “wing-shaking may arise when chicks are old enough to balance or when their wings are big enough to display”, as cuckoo chicks can raise their wings within a day after hatching (see eviction of host eggs or chicks, Davies, 2000, p. 51) and their wings are big and conspicuous much earlier than they start to be used during begging (cuckoo chicks are well feathered at the age of nine days, Wyllie, 1981, p. 155). As I have shown above wing-shaking seems to provide a plausible explanation for the puzzling discrepancy between the results of Kilner et al. (1999) and the provisioning of older cuckoo chicks. Wing-shaking could be considered as a special sort of postural begging (e.g. Redondo and Castro, 1992). In non-parasitic birds or parasites reared along host chicks it may be unclear if parents respond to the posture itself, or some correlate of it (e.g. height). This confounding factor cannot apply to the common cuckoo because fosterers have no other chicks to compare with the cuckoo as regards the height or other correlates of postural begging. From the same reason the position in the nest (McRae et al., 1993) can be excluded outright as a possible confounding variable in the current study. All these lines of evidence strongly suggest that raised wing-shaking is a specific begging signal of old common cuckoo chicks to hosts. Indeed, even Kilner et al. (1999) mentioned (p. 671) that “the cuckoo may be unable to solicit a higher rate of feeding, perhaps because there is an upper limit to the rate at which calls can be produced or perceived”. This may be exactly the reason why the old cuckoo chicks start to employ additional begging strategy – wing-shaking. Kilner et al. (1999) were correct that “any cues other than gape area and call rate must play a minor role in determining host provisioning rate” (p. 670) but only within the limited variability of ages of cuckoo chicks they used for their analyses. Reed warbler chicks also raise and slightly shake their wings when being fed by their parents but only during a very brief period before fledging (within a day before leaving the nest) and continue wing-shaking during begging after fledging (own observations). Thus, wing-shaking by reed warbler chicks may explain parental responsiveness to the same signal produced by old cuckoo chicks. This suggests that while parental feeding rules may differ between warbler and cuckoo nestlings, the decision rules of parents feeding own and parasitic chicks may become more similar again at later chick stages. CONCLUSION These results extend upon the conclusions of Kilner et al. (1999) by incorporating information thus far undescribed regarding the fascinating interactions between old cuckoo chicks and their hosts. Reed warblers indeed integrate visual and vocal signals when deciding how much food to bring to the nest but the visual signals do not solely include gaping (Kilner et al., 1999) but also wing-shaking in old cuckoo chicks. Further, biologically significant differences in the growth rates in host and cuckoo broods explain the higher than expected call rates in cuckoos in the exponential phase of growth. Accordingly, the models of parental provisioning rules – for any species where wing-shake begging is present – should include not only gaping and calling but also additional stimuli, e.g. wing-shaking, to obtain a better insight into parent-offspring communication. ACKNOWLEDGEMENTS I thank N. B. Davies for confirming that my re-analysis of Kilner et al. results is correct and M. Leonard and D. W. Mock for comments on previous versions of the MS. I am grateful to A. P. Møller for bringing the non-linearity of data from Fig. 6 in Kilner et al. (1999) to my attention. B. Matysiokova, M. Honza and others helped with collecting some data. When working on this paper I was supported by grants MSMT 6198959212 and GACR 206/03/D234. The study was done under license from The Central Commission for Animal Welfare of the Czech Republic (No. 065/2002–V2) and in accordance with the laws and ethical guidelines of the Czech Republic.

7

REFERENCES Budden AE, Wright J, 2001. Falling on deaf ears: the adaptive significance of begging in the absence of a parent. Behav Ecol Sociobiol 49:474–481. Butchart SHM, Kilner RM, Fuisz T, Davies NB, 2003. Differences in the nestling begging calls of hosts and host-races of the common cuckoo, Cuculus canorus. Anim Behav 65:345–354. Cramp S, ed 1985. Birds of the Western Palearctic. Vol. IV. Oxford: Oxford University Press. Davies NB, 2000 Cuckoos, cowbirds and other cheats. London: T & AD Poyser. Davies NB, Kilner RM, Noble DG, 1998. Nestling cuckoos, Cuculus canorus, exploit hosts with begging calls that mimic a brood. Proc R Soc Lond B 265:673–678. Glutz von Blotzheim UN, Bauer KM, eds 1980. Handbuch der Vögel Mitteleuropas. Vol. 9. Columbiformes – Piciformes. Wiesbaden: Akademische Verlagsgesellschaft. Grim T, 2006. Cuckoo growth performance in parasitized and unused hosts: not only host size matters. Behav Ecol Sociobiol (in press) Grim T, Honza M, 1997. Differences in parental care of reed warbler (Acrocephalus scirpaceus) to its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zool 46:135–142. Grim T, Honza M, 2001. Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behav Ecol Sociobiol 49:322–329. Grim T, Kleven O, Mikulica O, 2003. Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Proc R Soc Lond B 270:S73– S75. Hauber ME, Ramsey CK, 2003. Honesty in host-parasite communication signals: the case for begging by fledgling brown-headed cowbirds Molothrus ater. J Avian Biol 34:339– 344. Johnsgard PA, 1997. The avian brood parasites. Oxford University Press, New York. Kilner R, 1995. When do canary parents respond to nestling signals of need? Proc R Soc Lond B 260:343–348. Kilner RM, Davies NB, 1999. How selfish is a cuckoo chick? Anim Behav 58:797–808. Kilner RM, Johnstone RA, 1997. Begging the question: are offspring solicitation behaviours signals of need? Trends Ecol Evol 12:11–15. Kilner RM, Noble DG, Davies NB, 1999. Signals of need in parent-offspring communication and their exploitation by the common cuckoo. Nature 397:667–672. Leonard ML, Horn AG, 1998. Need and nestmates affect begging in tree swallows. Behav Ecol Sociobiol 42:431–436. Malchevsky AS, 1987. The cuckoo and its mentors. Leningrad (in Russian). McRae SB, Weatherhead PJ, Montgomerie R, 1993. American robin nestlings compete by jockeying for position. Behav Ecol Sociobiol 33:101–106. Noble DG, Davies NB, Hartley IR, McRae SB, 1999. The red gape of the nestling cuckoo (Cuculus canorus) is not a supernormal stimulus for three common hosts. Behaviour 136:759–777. Payne RB, 2005. The cuckoos. Oxford: Oxford University Press. Redondo T, Castro F, 1992. Signalling of nutritional need by magpie nestlings. Ethology 92:193–204. Royama T, 1966. Factors governing feeding rate, food requirement and brood size of nestling great tits Parus major. Ibis 108:313–347. Sejberg D, Bensch S, Hasselquist D, 2000. Nestling provisioning in great reed warblers (Acrocephalus arundinaceus): do males bring larger prey to compensate for fewer nest visits? Behav Ecol Sociobiol 47:213–219. Smith TE, Leonard ML, Smith BD, 2005. Provisioning rules and chick competition in asynchronously hatching common terns (Sterna hirundo). Behav Ecol Sociobiol 58:456–465. Soler M, Soler JJ, Martinez JG, Moreno J, 1999. Begging behaviour and its energetic cost in great spotted cuckoo and magpie host chicks. Can J Zool 77:1794–1800. Starck JM, Ricklefs RE, eds 1998. Avian growth and development. Evolution within the altricial-precocial spectrum. Oxford University Press, New York. Tanaka KD, Ueda K, 2005. Horsfield's hawk-cuckoo nestlings simulate multiple gapes for begging. Science 308:653–653.

8

Wright J, Leonard ML, eds 2002. The evolution of begging. Dordrecht: Kluwer Academic Publishers. Wyllie I, 1981. The cuckoo. London: Batsford. Zar JH, 1999. Biostatistical analysis. New Jersey: Prentice-Hall.

Figure 1 The relationship between two measures of parental provisioning rates: feeding rate (mg of dried food delivered per hour) and feeding frequency (number of feedings per hour). The data are best fitted by a linear regression (R2=0.20, F1,28=6.57, P=0.02). Each data point shows two ways of describing a particular sample of parental care. Figure 2 The relationship between nestling mass and feeding frequency. Original untransformed data are plotted for cuckoos ({; 158 measurements, 32 chicks) and reed warblers (z; 211 measurements, 67 broods). Figure 3 The relationship between nestling mass and feeding rate. Original untransformed data are plotted for cuckoos ({; 34 measurements, 15 chicks) and reed warblers (z; 105 measurements, 33 broods). Figure 4 Gape area (mean±SD) growth of cuckoo chicks in reed warbler nests (292 measurements, 46 chicks). For each age at least four measurements are included. Gape areas are calculated according to Kilner et al. (1999) as the product of gape width and length.

Feeding frequency (feeds/hour)

35 30 25 20 15 10 5 0 0

100

200

300

400

Feeding rate (mg/hour)

Figure 1.

9

Feeding frequency (feeds/hour)

60 50 40 30 20 10 0 0

10

20

30

40

50

60

70

80

12

14

Nestling mass (g)

Figure 2.

140 Feeding rate (mg/hour)

120 100 80 60 40 20 0 0

2

4

6

8

10

Nestling mass (g)

Figure 3.

450

2

Gape area (mm )

400 350 300 250 200 150 100 50 0 0

5

10 Nestling age (days)

Figure 4.

10

15

20

9. Grim T.: Wing-patch begging, wing-shaking and evolutionary origins of nestling begging 9. strategies. Grim T.: Wing-patch begging, wing-shaking and evolutionary origins of nestling begging (subm.) strategies. (subm.)

Wing-patch begging, wing-shaking and the evolutionary origins of nestling begging strategies Tomáš Grim1 1 Department of Zoology Palacký University Tř. Svobody 26 771 46 Olomouc Czech Republic [email protected]

Abstract Avian chicks use various begging strategies when soliciting parental care. Recently, a novel begging strategy was found in the Horsfield’s hawk-cuckoo Cuculus fugax. Chicks of this brood parasitic species raise and shake their wings and display to fosterers a gape-coloured patch on the undersides of their wings. Although the gape-coloured wing-patch may be a unique trait of the Horsfield’s hawk-cuckoo, wing-shaking in the context of begging is virtually universal in both brood parasites and their hosts. A simple comparative analysis suggests that wing-shake begging is most likely an ancestral feature of cuckoos and perhaps all altricial birds. I suggest that Horsfield’s hawk-cuckoo chicks could have exploited the universal pre-existing host responsiveness to wing-shake begging. Evolution could have then further proceeded by making the wing-shaking more conspicuous through the addition of another stimulus – the unique colourful wing-patch. I also hypothesize that wing-shake begging may have evolved from pre-fledging restlessness and was secondarily used in courtship displays, threatening postures and distraction displays by adults. Further discussions of these hypotheses may facilitate research into the so far poorly known phylogenetic history of chick begging strategies.

1

Probably the most conspicuous behaviour of avian nestlings is begging. Chicks call, gape, shake, jostle, stretch, and quiver their wings to obtain sufficient food from parents (Kilner and Johnstone 1997). There is a high interspecific variance of both chick begging strategies and parental responsiveness to various aspects of begging signals (reviewed in Wright and Leonard 2002). While most begging strategies described are based on calling and gaping (Kilner and Johnstone 1997), a “distinct form of signalling used by Horsfield’s hawk-cuckoo nestlings to obtain sufficient food” was reported recently (Tanaka and Ueda 2005, Tanaka et al. 2005). Whereas other brood parasites, e.g. the common cuckoo Cuculus canorus, are fed according to the same rules as host, e.g. the reed warbler Acrocephalus scirpaceus, nestlings (Kilner et al. 1999), this is not so in the Horsfield’s hawk-cuckoo Cuculus fugax. Probably because of the high risk of predation Horsfield’s hawk-cuckoo chicks “rarely beg loudly” (Tanaka and Ueda 2005, Tanaka et al. 2005). Instead, when begging the Horsfield’s hawk-cuckoo chick “raises and shakes the wing when host parents deliver food to the nest” and displays a gape-coloured skin patch on the underside of its shaking wing (Tanaka et al. 2005; italics added). Tanaka and co-workers supported the hypothesis that these wing-patches simulate extra gapes by three lines of evidence: the frequency of the wing-patch display increased with increasing intervals between feedings (i.e. a surrogate measure of nestling hunger), experimentally decreasing the conspicuousness of the patch through dyeing decreased feeding rates from hosts, and the parents sometimes mistakenly „fed“ the wing-patch itself (Tanaka et al. 2005). Thus both observational and experimental evidence indicate that Horsfield’s hawk-cuckoos evolved gape-coloured wing-patches to solicit extra food from their hosts. If the wing-patch begging is to be functional, several components must work at once: (i) wing raising, (ii) wing-patch, (iii) parental attentiveness not only to chick’s gape but also to the wing (that bears the patch). Although fosterers sometimes (very rarely) misfeed the patch instead of the gape (Tanaka et al. 2005) this cannot explain why the patch arose in the first place – misfeedings are the consequence of the patch, thus misfeedings cannot explain the cause of patch existence. Therefore, parental attentiveness to the wing (and the patch) is non-trivial and requires an explanation per se. Obviously, the patch must have evolved from the state of no patch and could evolved through parental preferences for cuckoo chicks with more conspicuous patches. Under this scenario, parental attentiveness to wing movement must have preceded the evolutionary origin of the patch. Unsurprisingly, there is strong evidence that the behavioural component of wing-patch begging display (wing-raising and shaking) is not a unique trait of the Horsfield’s hawk-cuckoos and that parents of various species of birds pay attention to wing movements when feeding chicks. Further, some results of Tanaka and co-workers are open to alternative interpretations. Therefore I will separately discuss both behavioural (wing-raising and shaking) and morphological (wing-patch) components of this begging signal. Hereafter, wing-shake begging is defined as begging when the nestling raises one or both of its wings at angle up to a maximum of ~90º above horizontal and slightly shakes or quivers its wing(s) in that raised position (for two examples see a video supplemented to this paper and to Tanaka and Ueda 2005). The aims of this short note are to treat the following issues related to findings in Horsfield’s hawk-cuckoos: 1) Wing-shaking during begging is virtually universal in the cuckoo family and appears to be an ancestral trait in all altricial birds. 2) I discuss possible functions of conspicuous underwing-patch of the common cuckoo, a species known to increase its provisioning by wing-shake begging (Grim subm.), and present two alternative hypotheses that may explain the wing-patch evolution in the Horsfield’s hawk-cuckoo. 3) I suggest that pre-fledging restlessness might be an evolutionary precursor for wingshaking behaviour. 4) Horsfield’s hawk-cuckoos may have exploited hosts’ responsiveness to this common feature of avian begging and enhance this behavioural stimulus with a morphological trait, the colourful patch, as an additional stimulus to obtain higher feeding rates. Behavioural component: wing-shaking and begging in other bird taxa Horsfield’s hawk-cuckoos are not the only species that use wing-shaking when begging. In general, wing-shaking (or “flapping” or “quivering”) is considered one of the mechanisms of

2

begging and sibling competition in general (Wright and Leonard 2002, Mock 2004). Wingshaking by both wings is commonly used as a stimulus to induce parental feeding by begging nestlings of various passerines from several phylogenetically unrelated families, e.g. Eastern kingbirds Tyrannus tyrannus (Morehouse and Brewer 1968), magpies Pica pica (Redondo and Castro 1992), canaries Serinus canaria (Kilner 1995), tree swallows Tachycineta bicolor (Leonard and Horn 1998), southern grey shrikes Lanius meridionalis (Budden and Wright 2001) or rufous-bellied thrushes Turdus rufiventris (Lichtenstein 2001). In fact, it is more noteworthy if young of some species do not show this typical passerine behaviour than when they do (O’Brien and Dow 1979). Also non-passerine chicks raise and/or shake their wings when begging, e.g. downy woodpeckers Dendrocopos pubescens (Kilham 1962), Australian pelicans Pelecanus conspicillatus (Vestjens 1977), great egrets Egretta alba (Mock 2004), common terns Sterna hirundo (Smith et al. 2005) and black storks Ciconia nigra (R. Hampl, pers. comm.). Wing-shake begging is, as a standard, interpreted to signal the highest nutritional need of the chick (see all references above). Brood parasites are no exception to wing-shake begging, as brown-headed cowbirds Molothrus ater (Dearborn and Lichtenstein 2002, Hauber and Ramsey 2003), shiny cowbirds M. bonariensis (Lichtenstein 2001), great-spotted cuckoos Clamator glandarius (Cramp 1985, Soler et al. 1999), and common cuckoos (Glutz von Blotzheim and Bauer 1980, Wyllie 1981, Cramp 1985, Malchevsky 1987, Davies 2000, Grim subm.) too shake their wings when begging. Therefore the claim that “nestling brood parasitic Cuculus … do not flutter their wings [when begging], and do so only inconspicuously after they fledge” (Payne 2005, p. 92) is incorrect. Within the family Cuculidae there are many cases of described “flapping”, “fluttering” and “quivering” usually with “stretched” wings during begging in both parasitic and non-parasitic chicks. This behaviour was described in at least 23 species in the following genera: Guira, Crotophaga, Geococyx, Centropus, Coua, Phaenicophaeus, Clamator, Coccycua, Coccyzus, Eudynamys, Scythrops, Chrysococcyx, Hierococcyx* and Cuculus (Armstrong 1965, Wyllie 1981, Payne 2005). Thus, wing-shake begging is known from all the main cuckoo clades (Crotophaginae, Neomorphinae, Centropodinae, Couinae, and both Phaenicophaeni and Cuculini within Cuculinae). This phylogenetic distribution (Harvey and Pagel 1991) of wing-shake begging clearly shows that the trait is ancestral to the cuckoo clade. The presence of wing-shake begging in other clades of birds, e.g. Passeriformes, Pelecaniformes, Ciconiiformes, Charadriiformes and Piciformes (see above), suggests that this begging strategy is ancestral to all birds. The widespread taxonomic distribution of wing-shake begging in birds is crucial as it explains why the hosts of Horsfield’s hawk-cuckoos are responsive at all to this display shown by the parasite chicks. Similarly, wing-shake begging in magpie chicks explains why magpie fosterers are responsive to the same behaviour in the great spotted cuckoo (Soler et al. 1999) and wing-shake begging of reed warbler nestlings explains why reed warbler fosterers increase their provisioning in response to the common cuckoo chick’s wing-raising and shaking and why don’t they ignore the signal (Grim subm.). In other words, if there were no universal parental responsiveness to nestlings’ wing-shaking in passerines, the Horsfield’s hawk-cuckoo’s and other parasites’ chicks could hardly be successful in eliciting any parental care through any wing-related traits (both behavioural and morphological) as their fosterers would simply not recognize this as a signal of need. Parental attentiveness to wing-raising and shaking is a necessary prerequisite for subsequent evolution of wingpatches. An alternative scenario where hosts do not show any pre-existing parental preference for wing related traits and this preference evolves only in response to the cuckoo behaviour is meaningless as a host mutant paying any attention to this behaviour in parasitic chicks would have lower fitness than a host without that preference. Nestlings of some species raise and shake both wings when begging (e.g. Redondo and Castro 1992) while in other species, e.g. Horsfield’s hawk-cuckoos (Tanaka and Ueda 2005), common cuckoos (Wyllie 1981) or Alpine accentors Prunella collaris (Armstrong 1965), only one wing is raised at the time. This interspecific variability in the form of wingshaking has no bearing on the conclusion that wing-shaking is an ancestral trait at least in cuckoos and passerines as in all cases the behaviour is used solely during begging Note: Payne (2005) treats subspecies of the Horsfield’s hawk-cuckoo breeding in Northeastern Asia (including Japan) as a separate species, the rufous hawk-cuckoo Hierococcyx hyperythrus.

*

3

(references above). As in the majority of cuckoo species nestling behaviour is virtually unknown (Payne 2005), there are obvious opportunities for the fruitful research in the future. Morphological component: is there any role for underwing “patch” in the common cuckoo? Common cuckoo chicks raise and shake one of their wings every time they are fed when older than 15–16 days and there is good evidence that the wing-shaking increases provisioning rates by hosts (Grim subm.). Cuckoo chicks seem to need to increase the intensity of the begging display by adding wing-shaking to their other begging displays at an older age because of an increased need for food. In other words, for old cuckoo chicks gaping and calling alone may be insufficient to elicit enough feeding from host parents. This is supported by the finding that both gape growth and call rates level off 4–7 days before fledging while feeding frequencies and body mass increase up to fledging in common cuckoos raised by reed warblers. Moreover, there is good experimental evidence against alternative explanations for increasing provisioning rates in old cuckoo chicks (Grim subm. and references therein). The behavioural component of the signal in the common cuckoo is identical to that of Horsfield’s hawk-cuckoo chicks as is obvious from video on-line material supplemented to this note (compare with Tanaka and Ueda 2005 supplementary video online material; see also Fig. 1 in this paper and Fig. 1 in Tanaka and Ueda 2005). However, Horsfield’s hawk-cuckoo chicks do more than just raise and shake their wings – they stimulate their hosts to feed with the naked gape-coloured underwing-patch (Tanaka and Ueda 2005). The common cuckoo chick has no gape-coloured patch but the underside of the wing has visible naked skin in the distal ulnar and proximal carpometacarpal region (i.e., in the same part of the wing as the Horsfield’s hawk-cuckoo chick, cf. Fig. 1 in Tanaka and Ueda 2005). During later development (approx. at the age of 17 days) this skin area becomes covered with white lesser underwing coverts which are quite conspicuous during wing-shaking. Although the non-random relationship between begging and wing-“raising and shaking” (sensu Tanaka et al. 2005) is supported in the common cuckoo (Grim subm.), the existence of visible skin and white feathers may have nothing to do with parent-“offspring” communication. In young common cuckoos the naked skin cannot be observed by hosts as young cuckoos (<15 days) do not raise their wings. However, the white underwing in old cuckoo chicks is frequently observable and quite conspicuous (even for a human observer distant from the nest). Common cuckoo chicks may use this patch also after fledging during the long period when they are fed by hosts (Wyllie 1981). Alternatively, the wing-patch may be used in social intra-specific communication after the young cuckoo starts to interact with conspecifics during the subsequent breeding season (Johnsgard 1997) or the patch may have no function at all in adulthood (even though white lesser underwing coverts are retained in adults, Wyllie 1981, photo no. 6). Another explanation for white lesser coverts could be mimicry of raptorial hawks, as argued by some authors (Wyllie 1981). The hypothesis that this white patch in the common cuckoo stimulates hosts to increase feeding rates could be tested with similar dyeing experiment as that done by Tanaka and Ueda (2005). The underwing of the common cuckoo chick is (i) visible due to wing-raising and shaking, (ii) it has conspicuous colour and (iii) the parent will see it most frequently when feeding old cuckoo chick which have the highest feeding demands (see above). This suggests some signalling of hunger and it would be therefore worthwhile to test the function of the patch. Further, it would be interesting to examine whether the colour of the patch in the Horsfield’s hawk-cuckoo chicks is important in itself. The results of dyeing experiment, where dyeing with black colour decreased feeding rates (Tanaka and Ueda 2005), may be interpreted in two different ways. (i) Under “conspicuous wing-patch” hypothesis any conspicuous patch is good enough to elicit host feeding. This is essentially equal to question whether “does yellow patch work with a same efficiency as orange or red one?”. (ii) Under “gape-mimicking patch” hypothesis the particular colour is crucial for successful exploitation of hosts. Although the colour similarity of patch to gape in Horsfield’s hawkcuckoo chicks strongly suggests that the colour of the patch matters (supporting the latter hypothesis), the issue cannot be resolved without experimental changes in patch colour. Although so far the wing-patch in Horsfield’s hawk-cuckoo chicks seems to be a unique trait, it is feasible that similar traits might evolve in other species as (i) behavioural

4

component of this begging strategy is virtually universal (see above), (ii) this behavioural component is a necessary prerequisite for the subsequent evolution of the wing-patch and (iii) “the whole of the [wing] skin is not full covered with feathers ... in many altricial species” (Tanaka et al. 2005, p. 462). Thus there are both behavioural and morphological traits that could serve as pre-adaptations for the evolution of colourful wing-patches in the variety of taxa. Evolutionary origins: is pre-fledging restlessness an evolutionary precursor of wingshake begging? If wing-shake begging were to function in Horsfield’s hawk-cuckoo all its components, including the conspicuous wing-patch, wing-raising (and perhaps shaking) and parental responsiveness to it, must be present at once. In other words, wing-patch could not evolve without wing-raising and specific parental responsiveness to wing-raising (see above). It is of course highly unlikely that all components of such a complex system would suddenly all appear in one individual – this argument holds for any complex trait (see Dawkins 1989). Therefore we should look for evolutionary precursors that enabled the existence of wingpatch begging in Horsfield’s hawk-cuckoo and wing-shake begging in many other birds. Wing-shaking during begging is common in various altricial nestlings after fledging (Wright and Leonard 2002; T. Grim personal observations in several species of Pachycephalidae, Callaetidae, Zosteropidae, Eopsaltridae, Muscicapidae, Paridae, Sylviidae, Fringillidae and Turdidae), including brood parasitic birds (Icteridae; Hauber and Ramsey 2003). Further, various passerines, e.g. reed warblers, frequently stretch and/or shake their wings shortly before fledging when parents are not present at the nest and this is accompanied by stretching of legs and feather preening (T. Grim, B. Matysiokova, unpublished observations). These behaviours in the absence of parents are very similar to wing-shaking during feeding by parents. This suggests that wing-shaking and -stretching could be related to pre-fledging restlessness (i.e. suite of behaviours performed before fledging, including leg and wing stretching, preening etc.; Gill 1990, p. 385). Behavioural patterns of pre-fledging restlessness could – due to their conspicuous nature – serve as evolutionary precursors for the evolution of wing (patch) begging (see below). To sum up, wing-shaking in the presence of parents may serve to solicit feeding (Wright and Leonard 2002) while wing-shaking by older nestlings in the absence of parents is most likely a manifestation of pre-fledging restlessness. Moreover, wing-shake begging may be secondarily used in social interactions among adult conspecifics including courtship feeding (e.g. Armstrong 1965, own observations), male sexual displays (e.g. Frith 1982) or appeasement displays (e.g. McLean 1988, Lott 1999). Some cuckoos use wing raising and shaking in threat postures (Johnsgard 1997) and wing-shake begging was hypothesized to be an evolutionary precursor of distractions displays that include “injury-feigning of the wing quivering type” which, unsurprisingly, occur solely in species with altricial nestlings (Armstrong 1965, p. 97). Interestingly, asymmetrical wing-shaking (raising and shaking of only one wing at a time) was found in a wide variety of taxa, ranging from the ostrich Struthio camelus, through many shorebirds and crakes to passerines, e.g. reed buntings Emberiza schoeniclus and Alpine accentors (Armstrong 1965). However, symmetrical wingshaking (with both wings raised and/or shaken at a time) seems to be much more prevalent (see above). Obviously, even movement alone can attract attention of hosts and increase feeding rates (e.g. Redondo and Castro 1992, Kilner 1995, Smith et al. 2005, Grim subm.). This modified hypothesis needs further testing in Horsfield’s hawk-cuckoo chicks as Tanaka and Ueda (2005) and Tanaka et al. (2005) did not experimentally test for the possible effect of movement itself (e.g. by observing feeding rates to cuckoo chicks that would be experimentally restrained from wing-shaking). Further, Tanaka and co-workers did not test for a possible effect of chick age on the frequency of wing-raising and shaking. Data from the common cuckoo show that only old chicks of this species use wing-shake begging (Grim subm.). Certainly, Horsfield’s hawk-cuckoo chicks cannot use the wing-raising begging strategy immediately after hatching – all altricial hatchlings are helpless, wings are too small to bear any reasonably big colourful patch and thus any wing-shake begging is phenotypically constrained shortly after hatching. Therefore, the ontogeny of wing-patch (morphological development) and wing-raising and shaking (behavioural development) also deserves more attention. Finally, it is so far unclear what effects of the amplitude and the

5

frequency of wing-shaking are as both Tanaka and Ueda (2005) and Grim (subm.) studied only the presence or absence of wing-raising and shaking during particular feedings of nestlings. I stress that there are obvious differences between common cuckoo and Horsfield’s hawk-cuckoo wing-shake begging. On the one hand, both species raise only one wing at a time, keep it high above the horizontal (at angle of ~90º) and slightly shake it in that position. On the other hand, the common cuckoo chick lacks a gape-coloured wing-patch so typical for the Horsfield’s hawk-cuckoo. However, taking “wing-patch begging” by the latter species as a unique trait inevitably means that we cannot understand its evolutionary origin. Therefore it is crucial to study similar (and perhaps not identical) behaviours and morphological structures in (un)related taxa as exemplified above. Only through the use of phylogenetic comparative methods may we achieve more profound insights into how the wing-patch begging strategy originated. This is, of course, a standard approach used in all areas of evolutionary biology (Harvey and Pagel 1991). Future directions Based on above discussed evidence I hypothesize that the wing-shake begging is an ancestral trait in altricial birds (or at least in passerines and cuckoos) which may have originated as a part of the pre-fledging restlessness. In addition, wing-shaking could be selected positively because it may augment visual effect of gaping, i.e. wing-shaking is similar to head and gape shaking at least as for frequency. Although the similarity is crude it is not surprising that it successfully elicits higher feeding rates (references above) as most birds readily feed alien chicks with totally dissimilar begging calls, gape morphologies and begging postures (Sealy and Lorenzana 1997, see also discussion in Grim 2005). An ability to shake wings may signal good health and/or quality of the nestling (Wright and Leonard 2002). In some species selection could further increase the conspicuousness of the signal by converging wing-patch colour to gape colour (Tanaka and Ueda 2005, Tanaka et al. 2005) to which parents are already responsive (Alvarez 2004) or to any conspicuous coloration (Götmark and Ahlström 1997). This hypothetical scenario provides a framework for future research for the poorly known chick stage in brood parasites (Davies 2000, Grim 2005, 2006) and virtually unexplored phylogenetic history of chick begging strategies in all birds. Acknowledgements – I am grateful to M. Leonard, V. Remes, M. Soler and K. D. Tanaka for their comments on the MS. My work is supported by grants MSM6198959212 and GACR 206/03/D234. Supporting online material (video)

6

References Alvarez, F. 2004. The conspicuous gape of the nestling common cuckoo Cuculus canorus as a supernormal stimulus for rufous bush chat Cercotrichas galactotes hosts. – Ardea 92: 63–68. Armstrong, E. A. 1965. Bird display and behaviour. – Dover, New York. Budden, A. E. and Wright, J. 2001. Falling on deaf ears: the adaptive significance of begging in the absence of a parent. – Behav. Ecol. Sociobiol. 49: 474–481. Cramp, S. (ed.) 1985. Birds of the Western Palearctic. Vol. 4. – Oxford University Press, Oxford. Davies, N. B. 2000. Cuckoos, cowbirds and other cheats. – T. & A.D. Poyser, London. Dearborn, D. C. and Lichtenstein, G. 2002. Begging behaviour and host exploitation in parasitic cowbirds. – In: Wright, J. and Leonard, M. L. (eds). The evolution of begging: competition, cooperation and communication. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 361–387. Dawkins, R. 1989. The selfish gene. – Oxford University Press, Oxford. Frith, C. B. 1982. Displays of Count Raggi’s Bird-of-Paradise Paradisaea raggiana and congeneric species. – Emu 81: 193–201. Gill, F. B. 1990: Ornithology. – W. H. Freeman, New York. Glutz von Blotzheim, U. N. and Bauer, K. M. (eds) 1980. Handbuch der Vögel Mitteleuropas. Vol. 9. Columbiformes – Piciformes. – Akademische Verlagsgesellschaft, Wiesbaden. Götmark, F. and Ahlström, M. 1997. Parental preference for red mouth in a songbird. – Proc. R. Soc. Lond. B 264: 959–962. Grim, T. subm. Signals of need in parent-offspring communication: why gaping and calling alone fail to explain provisioning rates? Grim, T. 2005. Mimicry vs. similarity: which resemblances between brood parasites and their hosts are mimetic and which are not? – Biol. J. Linn. Soc. 84: 69–78. Grim, T. 2006. The evolution of nestling discrimination by hosts of parasitic birds: why is rejection so rare? – Evol. Ecol. Res. (in press). Grim, T. and Honza, M. 2001. Does supernormal stimulus influence parental behaviour of the cuckoo's host? – Behav. Ecol. Sociobiol. 49: 322–329. Grim, T., Kleven, O. and Mikulica, O. 2003. Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? – Proc. R. Soc. Lond. B 270: S73–S75. Harvey, P. H. and Pagel, D. M. 1991. The comparative method in evolutionary biology. – Oxford University Press, Oxford. Hauber, M. E. and Ramsey, C. K. 2003. Honesty in host-parasite communication signals: the case for begging by fledgling brown-headed cowbirds Molothrus ater. – J. Avian Biol. 34: 339–344. Johnsgard, P. A. 1997. The avian brood parasites. – Oxford University Press, New York. Kilham, L. 1962. Reproductive behaviour of downy woodpeckers. – Condor 64: 126–133. Kilner, R. 1995. When do canary parents respond to nestling signals of need? – Proc. R. Soc. Lond. B 260: 343–348. Kilner, R. M. and Johnstone, R. A. 1997. Begging the question: are offspring solicitation behaviours signals of need? – Trends Ecol. Evol. 12: 11–15. Kilner, R. M., Noble, D. G. and Davies, N. B. 1999. Signals of need in parent-offspring communication and their exploitation by the common cuckoo. – Nature 397: 667–672. Lichtenstein, G. 2001. Low success of shiny cowbird chicks parasitising rufous-bellied thrushes: chick-chick competition or parental manipulation? – Anim. Behav. 61: 401– 413. Leonard, M. L. and Horn, A. G. 1998. Need and nestmates affect begging in tree swallows. – Behav. Ecol. Sociobiol. 42: 431–436. Lott, D. F. 1991. Bronzy sunbirds tolerate intrusion on foraging territories by female goldenwinged sunbirds that perform begging display. – J. Field Ornithol. 62: 492–496. Malchevsky, A. S. 1987. The cuckoo and its mentors. – Leningrad (in Russian), 264 pp. McLean, I. G. 1988. Breeding behaviour of the long-tailed cuckoo on Little Barrier Island. – Notornis 35: 89–98. Mock, D. W. 2004: More than kin, less than kind. Belknap Press, Harvard. Morehouse, E. L. and Brewer, R. 1968. Feeding of nestling and fledgling Eastern kingbirds. – Auk 85: 44–54.

7

O’Brien, P. H. and Dow, D. D. 1979. Vocalizations of nestling noisy miners Manorina melanocephala. – Emu 79: 63–70. Payne, R. B. 2005. The cuckoos. – Oxford University Press, Oxford. Redondo, T. and Castro, F. 1992. Signalling of nutritional need by magpie nestlings. – Ethology 92: 193–204. Sealy, S. G. and Lorenzana, J. C. 1997. Feeding of nestling and fledgling brood parasites by individuals other than the foster parents: a review. Can. J. Zool. 75: 1739–1752. Smith, T. E., Leonard, M. L. and Smith, B. D. 2005. Provisioning rules and chick competition in asynchronously hatching common terns (Sterna hirundo). – Behav. Ecol. Sociobiol. 58: 456–465. Tanaka, K. D. and Ueda, K. 2005. Horsfield's hawk-cuckoo nestlings simulate multiple gapes for begging. – Science 308: 653–653. Tanaka, K. D., Morimoto, G. and Ueda, K. 2005. Yellow wing-patch of a nestling Horsfield's hawk cuckoo Cuculus fugax induces miscognition by hosts: mimicking a gape? – J. Avian Biol. 36: 461–464. Vestjens, W. J. M. 1977. Breeding behaviour and ecology of the Australian pelican, Pelecanus conspicillatus, in New South Wales. – Aust Wildl Res 4: 37–58. Wright, J. and Leonard, M. L. (eds) 2002. The evolution of begging: competition, cooperation and communication. Kluwer Academic Publishers, Dordrecht, The Netherlands. Wyllie, I. 1981. The Cuckoo. – Batsford, London. Fig. 1. Common cuckoo (Cuculus canorus) chick wing-shake begging in the nest of a rare host – the song thrush (Turdus philomelos). The chick is 17 days old.

8

10. Grim T. 2006: Low virulence of brood parasitic chicks: adaptation or constraint? 10. Ornithological Science (in press) Grim T. 2006: Low virulence of brood parasitic chicks: adaptation or constraint? Ornithological Science (in press)

Low virulence of brood parasitic chicks: adaptation or constraint? Tomáš Grim1 Department of Zoology Palacký University tr. Svobody 26 CZ-771 46 Olomouc Czech Republic e-mail: [email protected] 1

Chicks of some brood parasitic species severely decrease hosts’ fitness through various behavioural strategies (e.g. eviction of host eggs or chicks, direct killing by pecking, aggressive begging) leading to the death of host progeny (Davies 2000). However, nestlings of some other parasitic species are more tolerant and raised in the presence of host offspring. In contrast to previous theory, which focused on the benefits of parasitic intolerance to host chicks, Kilner (2005) importantly drew attention to the costs of virulence. Although the killing of host progeny may be beneficial in terms of avoiding the cost of sharing or competing for parental care with host offspring, it might have also associated costs: by killing hosts young the parasite also kills its „helpers“ in soliciting of parental care. Both comparative and experimental evidence suggests that Brown-headed Cowbird (Molothrus ater) chick may benefit from presence of some host nestlings because a larger brood elicits higher parental feeding frequencies and the cowbird chick is able to outcompete host offspring to monopolize these extra feeds (Kilner 2003; Kilner et al. 2004). In addition, Kilner (2005) attempted to explain patterns of virulence and tolerance to host young by parasitic chicks in a wider context, i.e. across other brood parasite-host systems. The idea of a trade-off between benefits and costs of virulence toward host chicks is brilliant, however, I will show that (i) most evidence presented by Kilner (2005) as supportive of the hypothesis may be explained more parsimoniously in other ways, (ii) the author did not consider a substantial body of evidence that goes against her hypothesis in at least some parasitic species, and (iii) to understand interspecific variation in parasitic chicks virulence it is necessary to consider not only trade-offs between costs and benefits of particular chick behaviours (Kilner 2005) but also physical and other constraints on the evolution of chick-killing strategies. Despite the following criticisms I believe that studies such as Kilner (2005) are important because they foster more research in an area so far neglected in the study of parasitic chicks (Grim 2006b). COSTS, BENEFITS AND CONSTRAINTS ON PARASITIC CHICK BEHAVIOUR Kilner importantly stresses that we have to consider both benefits and costs of parasitic chick tolerance towards host young. However, there is a third “side of the coin“ that must be considered too – the constraints which set limits on chick behaviour. For instance, why do Great Spotted Cuckoos (Clamator glandarius) not evict their nest-mates? Kilner correctly rejects the evolutionary lag explanation (p. 56) which is anyway an explanation of the last resort (Davies 2000). However, she does not consider the most obvious and non-adaptive explanation: Great Spotted Cuckoo chicks hatch in the nests of large hosts (magpies Pica pica and other corvids). The eviction of large host eggs and chicks may simply not be an option for the small parasitic chick in a deep corvid nest (Wyllie 1981, p. 151; Davies 2000, p. 98; Payne 2005, p. 147). The same holds for hole-nesters (Rutila et al. 2002). There is some evidence that the eviction behaviour in the nests of some hosts is costly in terms of time (Nakamura 1990), impaired growth (Kleven et al. 1999), the risk of suicide by self-eviction (Molnar 1944; Wyllie 1981) and may even result in starving to death when the cuckoo chick concentrates on trying to evict the host nestlings (Soler 2002, p. 421). Under such physical constraints, the energy invested in eviction in large nests could be wasted. Instead, the Great Spotted Cuckoo chick uses an alternative and very successful strategy – it eliminates its competitors through exaggerated begging and wasting the food that could be otherwise taken by its competitors (Redondo 1993; Redondo & Zuniga 2002).

1

Kilner further discussed two other species of cuckoos. Common Koel (Eudynamys scolopacea) chicks indeed sometimes do not evict host young. Noticeably, they “tolerate“ host’s young only in the nests of large body hosts (Corvus, Urocissa) in India while in Australia they parasitize smaller hosts and evict their nestmates (Davies 2000; Payne 2005). The mechanism of non-eviction in Indian subspecies may simply be an unsuccessful eviction (but this hypothesis needs testing). In addition, chicks of the Channel-billed Cuckoo (Scythrops novaehollandiae) can hardly be described as “benign“ (p. 56): “Nestling hosts usually disappear within a week. The nestling cuckoo may crowd and starve them without evicting them...“ (Payne 2005, p. 385). It seems that parasitic offspring in all three species discussed are „tolerant“ to host offspring most likely because physical constraints imposed by host/nest selection by adult parasites. In my view, an ancestral evicting mutant of the Great Spotted Cuckoo or the Channel-billed Cuckoo would simply not be favoured by selection in the environment of large hosts due to too large energetic costs of eviction (see also Kleven et al. 1999; Rutila et al. 2002; Soler 2002). Intraspecific variability in the eviction behaviour of the Common Koel correlated with host size is in line with this constraint hypothesis. However, to fully evaluate the alternatives, experimental evidence is needed about the potential growth and survival benefits of nestmate tolerance at the earliest nestling stages of Common Koels and Channel-billed Cuckoos. NO COSTS OF KILLING OF HOST YOUNG FOR COMMON CUCKOO CHICKS Kilner (2005, p. 57) argued that provisioning of Common Cuckoo (Cuculus canorus) chicks by Reed Warblers (Acrocephalus scirpaceus) is relatively slow for “a parasite that is unconstrained by kinship in its demands for food”. According to Kilner this low provisioning rate “is unlikely to be explained by a constraint on the part of the Reed Warblers” and thus may reflect a cost of virulence for the cuckoo chick that killed host young and thus lost their assistance in obtaining enough food. This explanation is falsified by four lines of evidence. (i) Indirect but suggestive evidence is that differences in the qualitative composition of diet delivered to cuckoo vs. host chicks probably reflect exhaustion on the part of Reed Warbler fosterers (Grim & Honza 1997, 2001). The raising of a Cuckoo chick until fledging seems to be highly costly and hosts would be most likely unable to rear both their young and the parasitic chick if Common Cuckoos were non-evicting parasites. (ii) The data on the growth of cuckoo chicks in other host species clearly reject the hypothesis that cuckoo chicks are somehow constrained and unable to grow faster in the Reed Warbler nests (cf. Kilner 2005, p. 57–58). Cuckoo chicks have a capacity to reach about one third more mass in nests of a regular host, the Great Reed Warbler (A. arundinaceus; Kleven et al. 1999), and a rarely used host, the Song Thrush (Turdus philomelos; Grim 2006a), at the same age as in the nests of Reed Warbler. This shows directly that the constraint is not on the part of the cuckoo but the Reed Warbler. Therefore, the slow growth of cuckoo chicks and their relatively “low” provisioning in the nests of Reed Warblers is not caused by cuckoos’ internal constraints but by the hosts’ inability (or nonwillingness, Grim et al. 2003) to increase their feeding rates in the long term. Both the decreased foraging selectivity (Grim & Honza 2001) and growth patterns of parasitic chicks (Kleven et al. 1999; Grim 2006a) suggest that there is no spare provisioning capacity in host Reed Warblers that could be further exploited by cuckoo chicks. Although Brooke & Davies (1989) suggested that Reed Warblers might increase their feeding rates above those for the normal brood their experiments were probably too short-term to make the test rigorous. (iii) The presence of a single gape in the nest (p. 57) is also unlikely to explain the low feeding rate because Reed Warblers “follow exactly the same integration rule when provisioning a single cuckoo in their nest as when feeding a brood of their own young” (Kilner et al. 1999, p. 667). In other words, number of open mouths does not influence provisioning rates by Reed Warblers. (iv) Both observational (Rutila et al. 2002) and experimental (Soler 2002; MartinGalvez et al. 2005) evidence clearly showed that cuckoo chicks are very poor competitors for parental care when accompanied by hosts own brood (either due to unsuccessful eviction of host progeny from too deep or hole nests or due to the experimental change in competitive environment). To sum up, tolerance towards host young could not be beneficial for the Common Cuckoo chick in theory (i–iii) and it is not beneficial in reality (iv). The high virulence of

2

cuckoo chicks is adaptive and clearly non-costly in terms of lost assistance of host chicks in obtaining sufficient amounts of food. CHICK REJECTION BY HOSTS IS NOT RARE In my opinion, Kilner’s re-interpretation of results of Langmore et al. (2003) is distracting: “by killing host young, the parasite abolishes the hosts’ inclination to deliver food at the nest entirely“ (p. 58). As cuckoo chick rejections by hosts in that study were not explained as a by-product of brood reduction or desertion of uneconomically small one chick broods (as experimentally shown by Langmore et al. 2003) then it follows that by evicting host progeny the parasitic chick cannot trigger host rejection response. Additionally, Langmore et al. (2003) explicitly argued (p. 159) that sole occupancy of the nest is not (contrary to Kilner 2005, p. 60) cue responsible for desertion of parasitised broods. Although two influential experimental papers by Lotem et al. (1992, 1995) and his theoretical model (Lotem 1993) lead to the generally accepted impression that hosts learn the appearance of their eggs, the majority of empirical studies rejected the hypothesis – in all other common cuckoo host species than the Great Reed Warblers from Lotem et al. studies there is no evidence for age effects (and hence learning) on the egg discrimination abilities (Davies & Brooke 1988; Stokke et al. 1999; Marchetti 2000; Soler et al. 2000; Amundsen et al. 2002; Stokke et al. 2004). Thus, the arguments presented in the section “2) Rejection by hosts” may apply only in few hosts that learn appearance of their eggs but not in the majority of hosts that show experience-independent egg discrimination. It is inappropriate that Kilner (2005, p. 59) presents results of Langmore et al. (2003) as a “serious challenge to Lotem’s (1993) hypothesis”. The model was based on an assumption that the host learns egg and chick appearance. This assumption is clearly not met in the bronze-cuckoo (Chrysococcyx spp.) – Superb Fairy-wren (Malurus cyaneus) system (Langmore et al. 2003). Lotem’s (1993) model “will be falsified if nestling discrimination is exhibited by hosts in which only the parasitic nestling remains in the nest, and if discrimination in this case is learned” (Lotem 1993, p. 744; emphasis added). This is not the case in Superb Fairy-wrens (see also Grim 2006b). The comparison of Vidua and Shiny Cowbird (Molothrus bonariensis) systems with bronze-cuckoo system seems to imply that only the bronze-cuckoo host “unambiguously reject parasitic nestlings in favour of constructing a new nest“ (p. 60). Literally, this is undisputable but it should not mean that only Superb Fairy-wrens reject alien chicks. Both Vidua and Shiny Cowbird chicks are reared along host chicks. In Vidua species there is good evidence that these parasites are mimetic and thus one cannot expect to observe their rejection under natural conditions (for review see Grim 2005, for a case study Schuetz 2005). In the former system, host estrildids sometimes clearly reject (by refusing to feed) some of cross-fostered alien non-mimetic chicks which then die and are removed by hosts while hosts own chicks remain in the nest (Payne et al. 2001). Thus, there is no reason for hosts for “constructing a new nest”. In the latter system, host Bay-winged Cowbirds (Agelaioides badius) refuse to feed non-mimetic Shiny Cowbird fledglings that probably die shortly after fledging outside the nest (Fraga 1998). Thus, hosts do not need to construct new nests after rejection of alien chicks in both systems. This does not have any bearing on the observations that both species unambiguously reject some parasitic chicks. Finally, also Reed Warblers in at least one frequently parasitised population fulfil the definition of chick rejection: they desert some old parasitic chicks which die and hosts sometimes start to build a new nest immediately after desertion of parasitized nests (Grim et al. 2003; T. Grim unpubl. data). Moreover, there are various other systems where chick discrimination was observed or is suspected due to circumstantial evidence (Redondo 1993; Grim 2006b). Perhaps most importantly, after taking into account the research effort in egg vs. chick studies there is no big difference in “rarity” of chick in comparison to egg rejection behaviour (Grim 2006b) contrary to the generally accepted view in the literature (see any paper or book on brood parasitism mentioning chick discrimination). Thus, adaptive host response to parasitic chicks is much more frequent than Kilner (2005) implies. Fairy-wren vs. bronze-cuckoos system is not the first system where chick discrimination was found nor is it the only such system (for review see Grim 2006b).

3

EFFECTS OF HOST BREEDING STRATEGY ON PARASITIC CHICK VIRULENCE Kilner (2005, p. 62) predicted that the chick-killing brood parasites should show larger body size than their fosterers. This hypothesis is based on possible costs and benefits for the parasitic chick but does not consider other important factors, namely (i) physical constraints imposed e.g. by nest design and (ii) host breeding strategy. If the Common Cuckoo would parasitize hosts, e.g. thrushes of the genus Turdus, whose chicks are larger than itself after hatching then the Cuckoo would not be virulent. This would be not because it would not need to monopolize all the parental care at the nest (as suggested by Kilner) but because it would hatch in a deep nest and would be accompanied by large and quickly growing host chicks. Virulent behaviour, e.g. eviction, would simply not be an option under such physical constraints. Survival of the Cuckoo would then depend on the host breeding strategy: a Cuckoo chick would survive in nests of clutch adjusters (who prefer to feed smaller chicks within the brood, Soler 2001, 2002) but it would die in nests of brood reducers (who disfavour smaller chicks, Soler 2001, 2002). There is some evidence in favour of this hypothesis (Soler 2002, Grim 2006a). Most Common Cuckoo hosts are small passerines seemingly supporting Kilner (2005) hypothesis that chick-killing should evolve in systems with high parasite/host body-size ratio. However, in the nests of smaller hosts, the Cuckoo would thrive in nests of brood reducers (because it would be the largest chick in the brood) while it would have problems to obtain sufficient feeding in nests of clutch adjusters where parents would be – from cuckoo’s point of view – „distracted“ by smallest chick(s) in the brood. Therefore, in the nests of small brood reducers there is no need to evolve energetically costly (Kleven et al. 1999) and risky (Wyllie 1981) chick-killing strategy of evicting behaviour – host preference for large chicks would work well enough taking into account the existence of parasitic adaptations such as a shorter incubation period and a more exaggerated begging behaviour. Most Cuckoo hosts show clutch adjusting strategy (Soler 2002) and the Cuckoo’s evicting behaviour seems to be an adaptation to obtain all the care otherwise provided for the whole host brood. But the reason for the cuckoo’s eviction behaviour is not the magnitude of the parasitic chick’s provisioning requirements (as predicted by Kilner 2005) but its inability to withstand the competition with host young (Molnar 1944; Rutila et al. 2002; Soler 2002; Martin-Galvez et al. 2005). Possible effects of parasite/host body size ratio on parasitic chicks virulence seems to depend more on host breeding strategies than on parasite chick provisioning demands (as hypothesised by Kilner 2005). CONCLUSION The exaggerated begging by non-evictor brood parasitic young and their fast growth frequently results in starvation and early death of host young (Redondo 1993), which is in contradiction with predicted (Kilner 2005) benefits from non-eviction of host young. In my opinion, both comparative and experimental data strongly suggest that parasitic young in most parasitic species do not get any benefits from the presence of host young (e.g., in the Cuculus cuckoos). If there were any benefits of tolerance to host young they are most likely too small to outweigh both the benefits of elimination of host young (Redondo 1993) and the costs of tolerance of host young (Rutila et al. 2002; Soler 2002; Martin-Galvez et al. 2005). Further, the prediction that “the chick-killing brood parasites should be substantially larger than their hosts, whereas more benign brood parasites and their hosts should be more closely matching in size” (p. 62) also stems from alternative hypotheses, namely “physical constraints on eviction” hypothesis (this paper) and “host breeding strategy” hypothesis (Soler 2002) which seem to explain observed patterns of parasitic chick virulence in some host-parasite systems more parsimoniously than Kilner (2005) “cost of killing host young” hypothesis. Therefore, together with the trade-off idea Kilner (2005) suggested, constraints could be crucial for our understanding of interspecific variance in parasitic chick virulence. I believe that idea-rich paper by Kilner (2005) and the present paper will inspire more research in the neglected chick stage of parasite-host coevolution (Grim 2006b) ACKNOWLEDGEMENTS I thank M. Soler, F. Takasu and others for comments on earlier versions of the MS. My work is supported by grants MSM6198959212 and GACR 206/03/D234.

4

REFERENCES Brooke ML & Davies NB (1989) Provisioning of nestling cuckoos Cuculus canorus by reed warbler Acrocephalus scirpaceus hosts. Ibis 131: 250–256. Davies NB (2000) Cuckoos, cowbirds and other cheats. T & AD Poyser, London. Fraga RM (1998) Interactions of the parasitic screaming and shiny cowbirds (Molothrus rufoaxillaris and M. bonariensis) with a shared host, the bay-winged cowbird (M. badius). In: Rothstein SI & Robinson SK (eds) Parasitic birds and their hosts. pp 173–193. Oxford University Press, New York. Grim T (2005) Mimicry vs. similarity: which resemblances between brood parasites and their hosts are mimetic and which are not? Biol J Linn Soc 84: 69–78. Grim (2006a): Cuckoo growth performance in parasitized and unused hosts: not only host size matters. Behav Ecol Sociobiol (in press). Grim T (2006b): The evolution of nestling discrimination by hosts of parasitic birds: why is rejection so rare? Evol Ecol Res (in press). Grim T & Honza M (1997) Differences in parental care of reed warbler (Acrocephalus scirpaceus) in its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zool 46: 135–142. Grim T & Honza M (2001) Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behav Ecol Sociobiol 49: 322–329. Grim T, Kleven O & Mikulica O (2003) Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Proc R Soc Lond B 270: S73–S75. Kilner RM (2003) How selfish is a cowbird nestling? Anim Behav 66: 569–576. Kilner RM (2005) The evolution of virulence in brood parasites. Ornith Sci 4: 55–64. Kilner RM, Madden JR & Hauber ME (2004) Brood parasitic cowbird nestlings use host young to procure resources. Science 305: 877–879. Kilner RM, Noble DG & Davies NB (1999) Signals of need in parent-offspring communication and their exploitation by the common cuckoo. Nature 397: 667–672. Kleven O, Moksnes A, Røskaft E & Honza M (1999) Host species affects the growth rate of cuckoo (Cuculus canorus) chicks. Behav Ecol Sociobiol 47: 41–46. Langmore NE, Hunt S & Kilner RM (2003) Escalation of a coevolutionary arms race through host rejection of brood parasitic young. Nature 422: 157–160. Lotem A (1993) Learning to recognize nestlings is maladaptive for cuckoo Cuculus canorus hosts. Nature 362: 743–745. Lotem A, Nakamura H & Zahavi A (1992) Rejection of cuckoo eggs in relation to host age – a possible evolutionary equilibrium. Behav Ecol 3: 128–132. Lotem A, Nakamura H & Zahavi A (1995) Constraints on egg discrimination and cuckoo– host co-evolution. Anim Behav 49: 1185–1209. Martin-Galvez D, Soler M, Soler JJ, Martin-Vivaldi M & Palomino JJ (2005) Food acquisition by common cuckoo chicks in rufous bush robin nests and the advantage of eviction behaviour. Anim Behav 70: 1313–1321. Molnar B (1944) Cuckoo in the Hungarian plain. Aquila 51: 100–112. Nakamura H (1990) Brood parasitism of the cuckoo Cuculus canorus in Japan and the start of new parasitism on the azure-winged Magpie Cyanopica cyana. Jap J Ornithol 39: 1–18. Payne RB (2005) The cuckoos. Oxford University Press, Oxford. Payne RB, Woods JL & Payne LL (2001) Parental care in estrildid finches: experimental tests of a model of Vidua brood parasitism. Anim Behav 62: 473–483. Redondo T (1993) Exploitation of host mechanisms for parental care by avian brood parasites. Etología 3: 235–297. Redondo T & Zuniga JM (2002) Dishonest begging and host manipulation by Clamator cuckoos. In: Wright J & Leonard ML (eds) The evolution of begging. pp 389–412. Kluwer, Dordrecht. Rutila J, Latja R & Koskela K (2002) The common cuckoo Cuculus canorus and its cavity nesting host, the redstart Phoenicurus phoenicurus: a peculiar cuckoo-host system? J Avian Biol 33: 414–419. Schuetz JG (2005) Low survival of parasite chicks may result from their imperfect adaptation to hosts rather than expression of defenses against parasitism. Evolution 59: 2017–2024.

5

Sejberg D, Bensch S & Hasselquist D (2000) Nestling provisioning in great reed warblers (Acrocephalus arundinaceus): do males bring larger prey to compensate for fewer nest visits? Behav Ecol Sociobiol 47: 213–219. Soler M (2001) Begging behaviour of nestlings and food delivery by parents: the importance of breeding strategy. Acta Ethol 4:59–63. Soler M (2002) Breeding strategy and begging intensity: influences on food delivery by parents and host selection by parasitic cuckoos. In: Wright J & Leonard ML (eds) The evolution of Begging. pp 413–427. Kluwer, Dordrecht. Soler M, Martinez JG, Soler JJ & Møller AP (1995b) Preferential allocation of food by magpies Pica pica to great spotted cuckoo Clamator glandarius chicks. Behav Ecol Sociobiol 37: 7–13. Wyllie I (1981) The cuckoo. Batsford, London.

6

11. Grim T. & Honza M. 1996: Effect of habitat on the diet of reed warbler (Acrocephalus 11.nestlings. scirpaceus) Grim T. & Honza M. 1996: Effect habitat on the31–34. diet of reed warbler (Acrocephalus Foliaof Zoologica 45(1): scirpaceus) nestlings. Folia Zoologica 45(1): 31–34.

12. Grim T. & Honza M. 1997: Differences in parental care of reed warbler (Acrocephalus scirpaceus) in its own nestlings and 12. parasitic cuckoo (Cuculus canorus) chicks. Grim T. & Honza M. 1997: Differences in parental care of reed warbler (Acrocephalus Folia Zoologica 46(2): 135–142. scirpaceus) in its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zoologica 46(2): 135–142.

13. Grim T. 1999: Potrava mláďat rákosníka velkého (Acrocephalus arundinaceus). 13. Sylvia 35(2): 93–99. Grim T. 1999: Potrava mláďat rákosníka velkého (Acrocephalus arundinaceus). Sylvia 35(2): 93–99.

14. Grim T. 2006: An exceptionally high diversity of hoverflies (Syrphidae) in the food of the 14. reed warbler (Acrocephalus scirpaceus). Grim T. 2006: An exceptionally high diversity hoverflies (Syrphidae) in the food of the Biologia 61(1):of235–239. reed warbler (Acrocephalus scirpaceus). Biologia 61(1): 235–239.

Biologia, Bratislava, 61/2: 235—239, 2006

235 Short Communication

An exceptionally high diversity of hoverflies (Syrphidae) in the food of the reed warbler (Acrocephalus scirpaceus) Tomáš Grim Department of Zoology, Palacký University, Tř. Svobody 26, CZ–77146 Olomouc, Czech Republic; e-mail: [email protected]

Abstract: Despite being considered a classical example of protective Batesian mimicry hoverflies (Syrphidae) are known to be preyed upon by various passerines. The aim of the present study was to examine in detail food brought by reed warblers Acrocephalus scirpaceus to their nests to better understand the importance of hoverflies in the diet of small passerines. Using neck collars, 273 food samples containing 8,545 food items delivered to reed warbler and parasitic common cuckoo Cuculus canorus nestlings in warbler nests were recorded. The study was conducted during three breeding seasons in South Moravia, Czech Republic. An unusually high diversity of hoverflies was found – 27 species, including Mesembrius peregrinus (critically endangered species in the Czech Republic) and Mallota cimbiciformis (endangered species) – a new taxon to the Czech Republic. This indicates that nestling diet analyses may provide not only information on avian foraging behaviour but also important faunistic data. Thus, without the detailed identification to species level of material from foraging behaviour studies valuable scientific information may be lost. Overall dominance of Syrphidae was 3.7%, the most common species being Episyrphus balteatus (55.7%, n = 318). However, this number seriously underestimates the importance of hoverflies in the diet of reed warblers as hoverflies are one of the largest prey taken by warblers. Both larvae and pupae were rare, imagines strongly dominating (92.7%). Both specific wasp mimics (e.g., Chrysotoxum verrali) and bee mimics (e.g., Eristalis spp.) were not avoided by foraging reed warblers. The presence of a parasitic cuckoo chick did not affect host foraging behaviour with respect to overall dominance of hoverflies in the diet (warbler 3.3%, cuckoo 3.8%). Key words: Batesian mimicry, faunistics, hoverfly, Syrphidae, predation, diet, foraging.

Introduction Hoverflies (Syrphidae) are considered a good example of visual Batesian mimicry (Stubbs & Falk, 1983; Ruxton et al., 2004) and their similarity to various hymenopteran models also extends to behavioural mimicry (Howart et al., 2004). In theory, Batesian mimicry should serve as an adaptation for avoiding the risk of predation (Krebs & Davies, 1993). However, hoverflies sometimes form an important part of the diet of various insectivorous avian species (Krištín, 1986, 1988, 1991, 1994). During the study of interactions between the reed warbler Acrocephalus scirpaceus (Hermann, 1804) host and parasitic common cuckoo Cuculus canorus L., 1758 chicks a large sample of food delivered to both kinds of nestlings by warbler parents and fosterers respectively was obtained (Grim & Honza, 1996, 1997, 2001). Hoverflies were found frequently in the samples indicating they may form an important part of the warbler’s diet. The aim of the present study was to analyse hoverfly diversity in the diet of the model common species of insectivorous passerine – the reed warbler – in detail. Material analysed in the above mentioned papers

together with an additional 6,000 prey items of food predated by reed warblers during three breeding seasons at two sites were included. To my knowledge this is by far the largest sample of reed warbler and cuckoo nestlings diet collected so far. This should enable the importance of hoverflies in the food of this common insectivorous songbird to be determined. Material and methods The field work was carried out from May to mid-July in 1994, 1996 and 1997 field seasons on two fish pond systems near the villages of Lednice and Lužice in the SE part of the Czech Republic (47◦ 40 N, 16◦ 48 E), about 60 km SE of the city of Brno. Both pond systems are situated in a flat agricultural lowland landscape and are surrounded by deciduous woods and parkland. The distance between the areas is approximately 20 km. Reed warblers are known to build nests in various species of plants, however all the nests used in this study were placed in reed Phragmites australis vegetation. Both the Lednice and Lužice study plots have a relatively high parasitism rate of cuckoos in the nests of reed warblers (KLEVEN et al., 2004). Among various methods for obtaining food samples from nestlings the neck–collar method enables the most ac-

236 curate analysis of the quantity of food allocated to nestlings and precise prey identification (JOHNSON et al., 1980). A plastic coated wire ligature placed around the nestling neck hinders the swallowing of food but is loose enough not to strangle the chick. Neck-collars were applied for one hour and food delivered by parents was removed every 20 min to prevent chicks from regurgitating food accumulated by the ligature. For more details on field procedures see GRIM & HONZA (2001). For ecological analyses of hoverfly taxocenes found in food samples a simple ecological classification of hoverflies published by LÁSKA & MAZÁNEK (1998) was adopted. In total, analyses included 2,131 prey items delivered to reed warbler nestlings (n = 49 chicks, 189 samples) and 6,414 food items found in the samples from cuckoo nestlings (n = 32 chicks, 84 samples). Sample sizes in particular analyses may differ due to the fact that some imagines and all larvae and pupae were not identified to species level.

Results The overall number dominance of hoverflies was 3.7% (n = 8,545). Twenty-seven species (318 specimens) of hoverflies were found in food samples delivered to nestlings by reed warblers (Appendix 1). Most specimens were imagines (92.8%), the rest being both larvae (6.0%) and pupae (1.2%). The most frequent species was Episyrphus balteatus (n = 177), which formed 2.1% of all diet items and strongly dominated hoverfly taxocene in food samples (D = 55.7%). Other hoverflies were much less common, e.g., Eupeodes corollae (D = 7.9%), Melanostoma mellinum (D = 5.3%), Chalcosyrphus nemorum (D = 3.1%) and several species of the genus Platycheirus (D = 5.0%). Large-sized species, such as Eristalis tenax or Helophilus pendulus were also taken. Both perfect (Chrysotoxum verrali) and putative (Episyrphus balteatus, Sphaerophoria scripta) wasp mimics and bee mimics (Eristalis arbustorum, E. intricarius, E. tenax) were preyed upon by reed warblers. One specimen of Mallota cimbiciformis was found in the food brought to a cuckoo chick on 4 July, 1996 at Hlohovecký pond, Lednice. This is the first published record of the species in the Czech Republic (see Holinka et al., 1997). Reed warblers showed a weak but non-significant tendency to prey more on hoverfly species in proportion to their abundance in the ecosystem (ranked according to Láska & Mazánek, 1998) (rs = 0.24, n = 26, P = 0.25). There were 14 euryoecious, 7 mesophilous and 5 hygrophilous species of hoverflies in the food of reed warblers with dominances of 81.8%, 12.0% and 6.2%, respectively (n = 291 specimens). Samples contained hoverfly species that prefer woodland and parkland ecosystems (e.g., Dasysyrphus albostriatus), open habitats like agricultural fields (Sphaerophoria scripta, S. taeniata) or wet meadows (e.g., Mallota cimbiciformis, Melanostoma mellinum, Platycheirus clypeatus) and several species showing strong association with endangered marsh habitats (Eristalis intricarius, Chal-

T. Grim cosyrphus nemorum, Mesembrius peregrinus, Neoascia interrupta). However, most hoverfly species found in the food brought to warbler nests are not known to show any strong preferences for specific habitats. In all three years of study the dominance of hoverflies was slightly higher in cuckoo samples, however, pooled data from the three years showed that the difference was not significant (warbler 3.3%, cuckoo 3.8%; χ2 = 1.40, P = 0.24). Reed warbler nestling diet also showed lower species diversity of Syrphidae (11 species, 70 specimens) than that of cuckoo chicks (27 species, 248 specimens). This most probably reflects a higher sample size for the latter (2,131 vs. 6,414 prey items). Overall dominance of hoverflies (warbler and cuckoo samples pooled) decreased from 11.9% in 1994 to 3.3% in 1996 and 1.1% in 1997. The decline is highly significant (χ2 = 141.55, P < 0.0001). The Lednice area showed a significantly higher dominance of hoverflies than the Lužice area (4.5 vs. 1.8%; χ2 = 42.32, P < 0.0001). This is supported by a more robust comparison of samples from the same year (1996) and kind of nestling (cuckoo) (Lednice: 3.6%, Lužice: 2.3%; χ2 = 5.74, P = 0.016). Samples from Lednice contained 25 hoverfly species while samples from Lužice only 10 (eight species were found at both sites). This difference is probably wholly explained by a higher sample size from Lednice in comparison to Lužice (6,101 vs 2,444 prey items). Discussion Hoverflies (Syrphidae) are a classical example of Batesian mimicry. However, despite their similarity to stinging insects like bees, wasps and bumblebees, the hoverfly fauna in my samples was relatively very rich comprising 27 species with overall dominance 3.7%. In contrast, no bees, wasps or bumblebees were found in the food samples. The Order Hymenoptera was represented almost entirely by ants in the food brought by warblers (see also Grim & Honza, 1997). With respect to the relatively high dominance of hoverflies in the diet of reed warblers it is important to realize that the similarity of hoverflies and stinging insects to the human eyes does not necessarily imply that bird predators perceive the two categories of insects as similar (e.g., Dittrich et al., 1993). Birds show very different visual acuity in comparison to mammals (including humans) and may perceive hoverflies very differently. Moreover, birds may pay more attention to other cues (prey traits) than colour patterns (e.g., qualities of flight motion). Thus, a human observer may mistakenly judge a hoverfly as a mimic of a hymenopteran insect (e.g., Golding et al., 2005) when in fact there is no mimicry in the eye of the relevant beholder – the avian predator (for detailed discussion of other explanations for the maintenance of poor Batesian mimicry see Edmunds, 2000; Ruxton et al., 2004; Grim, 2005).

Hoverflies in the diet of a small passerine The diet included two species very rare in the Czech Republic: Mesembrius peregrinus and Mallota cimbiciformis. These are considered as “critically endangered” and “endangered” respectively (Láska & Mazánek, 1998). Also data from other authors indicate that research on bird foraging behaviour may provide important information on extremely rare and/or by traditional entomological methods hardly detectable species (Lauterer & Bureš, 1984) or even lead to discovery of new species for a particular country (Bureš & Pecina, 1993; this study). This shows that without detailed identification to species level of material from foraging behaviour studies important scientific information may be lost. Data from avian foraging studies provide a rich, but so far overlooked, source of important distribution data for macroecological, biogeographical and conservation studies (Grim, 2006). In a large scale study Krištín (1991) found 15 species of hoverflies in the food of 13 songbird species (n = 17,335 food items). The dominances of hoverflies was only infrequently higher than 10% in the avian species studied. The most dominant species of Syrphidae in his study were Syrphus ribesii, S. vitripennis and Episyrphus balteatus. All three species were also present in samples obtained from reed warbler and cuckoo nestlings in the present study. However, Episyrphus balteatus was the most numerous species by far. Interestingly, larvae were much more frequent (76.3%) than imagines in the study by Krištín (1991) while larvae were extremely rare in the food collected by reed warblers (present study). In general, dominance and frequency of aphidophagous hoverflies increased in several study species during rainy weather compared to sunny periods (Krištín, 1988). Data from the present study cannot test a potential effect of weather on foraging by warblers on hoverflies as food samples were collected only during sunny weather. Krištín (1986) reported even higher dominance of hoverflies in the food of the magpie Pica pica nestlings (exclusively the species Eristalis tenax; 11.4% of all food items, n = 2,537). The highest dominance of hoverflies among songbird species studied by Krištín (1991) was 23.0% in the food of the nuthatch Sitta europaea during a mass outbreak of aphids, which were preyed upon by hoverflies. However, one and two years after an outbreak the dominance of both aphids and hoverflies fell dramatically (hoverfly dominances in nuthatch diet were 0.4 and 2.6 in two years respectively). Grim & Honza (1997) reported even higher dominance (29.1%) in the food fed to cuckoo nestlings by reed warbler hosts, but this is most probably an artefact of small sample size (n = 172 prey items from one field season). After adding data from another two field seasons to material from that study the hoverfly dominance decreased to 3.7%. The diet of the closely related great reed warbler Acrocephalus arundinaceus at the same site (Lednice) showed a low dominance of hoverflies (1.1%, n

237 = 440 prey items) (Grim, 1999). Interestingly, despite the marginal occurrence of Syrphidae in the great reed warbler food their diversity was relatively high – four species (all the species were also present in the reed warbler food). Non-significant correlation between the abundance of particular hoverfly species in food samples and their relative abundance in the ecosystem probably reflects the opportunistic foraging tactics of reed warblers. This species forages in all available substrates from reeds and shrubs to herbaceous vegetation, dry mud and even agricultural fields or water surface (Grim & Honza, 1996). Moreover, it is able to use very short-term food sources. This is indicated by a large variation in diet composition among particular nests (Grim & Honza, 1996). However, a correlation between the rough ordinal scale of abundance of particular hoverfly species (Láska & Mazánek, 1998) and their dominance in the diet is a poor test of warblers’ foraging selectivity. Such a test is confounded by the microhabitat characteristics of particular nests sampled and also time during the breeding season. A quantitative comparison of actual supply of hoverflies in warbler foraging areas with composition of the diet delivered to nestlings would provide a much stronger test. Overall number dominance of hoverflies was 3.7% (n = 8,545). However, this underestimates their importance in the diet of nestlings: one ca. 10 mm long hoverfly weighs 14.0 mg (dry weight) whereas the same sized chironomid weighs only 1.3 mg (own unpublished data). Importantly, Chironomidae are by far the most dominant part of the warbler diet (42.0%; Grim & Honza, 2001; own unpublished data). Thus, the difference in weights (and consequently nutritional and energetic importance) between Chironomidae and Syrphidae indicates that the latter could contribute to the nutrition of nestlings by an order of magnitude more than would be judged from their number dominance alone. This is also supported by the high number (but clearly not weight) dominance of tiny aphids in the diet of reed warbler nestlings (13.8%; Grim & Honza, 2001; own unpublished data). The dominance of small diet items and especially aphids was consistently higher in cuckoo diet across the three field seasons (ranges: 2.4–10.3% in warbler and 6.9–36.4% in cuckoo food) indicating that hoverflies may be even more important for the growth of young cuckoos than could be expected from their numerical dominance alone. There was no effect of cuckoo parasitism on host foraging behaviour with respect to hoverflies. Interestingly, Grim & Honza (2001) reported that the presence of a cuckoo chick increases the dominance of aphids and other small food items in the diet brought by reed warblers. This is most likely explained by supernormal food demands of a cuckoo chick (Grim & Honza, 2001). Parasitic chicks require a longer nestling period than warbler chicks to develop properly and when the parasite is 8 days old or older it requires more food than

238 an average-sized host brood (Grim et al., 2003). This supernormal food consumption by the parasite leads to increased intensity of host foraging behavior (Grim & Honza, 2001). Higher foraging effort is generally known to be accompanied by decreased selectivity of foraging behaviour (see references and discussion in Grim & Honza, 2001). This mechanism may explain the increase in the dominance of aphids and other small food items in the cuckoo diet (Grim & Honza, 2001). However, Syrphidae were equally common in food brought to the hosts own and parasitic chicks which is also in line with the relationship between foraging intensity and selectivity (see above). Hoverfly diversity was higher in the cuckoo than warbler diet (27 vs 11 species) but this probably reflects a higher sample size in the former species. In summary, the high percentage of hoverflies in the food delivered by reed warbler adults indicates that putative mimicry does not provide sufficient protection against predation by songbirds at least in some hoverfly species. Hoverflies may be important for the diet of small insectivorous passerines in terms of weight dominance rather than number dominance. The finding of one new species for the Czech Republic indicates that diet analyses in insectivorous songbirds may also make an important contribution to faunistic entomological research (see Grim, 2006). Acknowledgements Author is grateful to M. HONZA, K. JANKO, O. KLEVEN, O. MIKULICA, A. MOKSNES, I. ØIEN, E. RØSKAFT and G. RUDOLFSEN for their help in the field. L. MAZÁNEK assisted with precise identification of some specimens. Special thanks go to V. BIČÍK for his help with identification of some specimens and many discussions about hoverflies. Comments by two anonymous referees improved the paper. The study was supported by a grant from the Czech Ministry of Education (Grant No. 153100012) and the Grant Agency of the Czech Republic (Grant No. 206/03/D234). References BUREŠ, S. & PECINA, P. 1993. Nález muchnice Bibio benesi (Diptera, Nematocera, Bibionidae) v potravě lindušky horské (Anthus spinoletta) z Pradědu [Occurrence of March-fly Bibio benesi (Diptera, Nematocera, Bibionidae) in water pipit diet in Praděd Mt. area]. Severní Morava 66: 49–50. DITTRICH, W., GILBERT, F., GREEN, P., MCGREGOR, P. & GREWCOCK, D. 1993. Imperfect mimicry: a pigeon’s perspective. Proc. R. Soc. Lond. B Biol. Sci. 251:195–200. EDMUNDS, M. 2000. Why are there good and poor mimics? Biol. J. Linn. Soc. 70: 459–466. GOLDING, Y., ENNOS, R., SULLIVAN, M. & EDMUNDS, M. 2005. Hoverfly mimicry deceives humans. J. Zool. 266: 395–399. GRIM, T. 1999. Potrava mláďat rákosníka velkého (Acrocephalus arundinaceus) [Food of great reed warbler (Acrocephalus arundinaceus) young]. Sylvia 35: 93–99. GRIM, T. 2005. Mimicry vs. similarity: which resemblances between brood parasites and their hosts are mimetic and which are not? Biol. J. Linn. Soc. 84: 69–78.

T. Grim GRIM, T. 2006. Avian foraging studies: an overlooked source of distribution data from macroecological and conservation studies. Diversity Distrib. (in press). GRIM, T. & HONZA, M. 1996. Effect of habitat on the diet of reed warbler (Acrocephalus scirpaceus) nestlings. Folia Zool. 45: 31–34. GRIM, T. & HONZA, M. 1997. Differences in parental care of reed warbler (Acrocephalus scirpaceus) to its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zool. 46: 135–142. GRIM, T. & HONZA, M. 2001. Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behav. Ecol. Sociobiol. 49: 322–329. GRIM, T., KLEVEN, O. & MIKULICA, O. 2003. Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Proc. R. Soc. Lond. B Biol. Sci. 270, Suppl. 1: S73–S75. HOLINKA, J., MAZÁNEK, L., VYSOKÝ, V., LÁSKA, P., MARTINOVSKÝ, J. & KULA, E. 1997. Syrphidae, pp. 224–226. In: VAŇHARA J. & ROZKOŠNÝ R. (eds) Dipterologica Bohemoslovaca 8, Folia Fac. Sci. Nat. Univ. Masaryk. Brun., Biol. 95. HOWARTH, B., EDMUNDS, M. & GILBERT, F. 2004. Does the abundance of hoverfly (Syrphidae) mimics depend on the numbers of their hymenopteran models? Evolution 58: 367– 375. JOHNSON, E.J., BEST, L.B. & HEAGY, P.A. 1980. Food sampling biases associated with the “ligature method”. Condor 82: 186–192. KLEVEN, O., MOKSNES, A., RØSKAFT, E., RUDOLFSEN, G., STOKKE, R.G. & HONZA, M. 2004. Breeding success of common cuckoos Cuculus canorus parasitising four sympatric species of Acrocephalus warblers. J. Avian Biol. 35: 394–398. KREBS, J.R. & DAVIES, N.B. 1993. An introduction to behavioural ecology. Oxford University Press, Oxford, 420 pp. KRIŠTÍN, A. 1986. Heteroptera, Coccinea, Coccinellidae and Syrphidae in the food of Passer montanus L. and Pica pica L. Biológia, Bratislava 41: 143–150. KRIŠTÍN, A. 1988. Coccinellidae and Syrphidae in the food of some birds, pp. 321–324. In: NIEMCZYK, E. & DIXON, A.F.G. (eds) Ecology and effectiveness of aphidophaga, Academic Publishing, Hague. KRIŠTÍN, A. 1991. Feeding of some polyphagous songbirds on Syrphidae, Coccinellidae and aphids in beech–oak forests, pp. 183–186. In: POLGÁR, L., CHAMBERS, R.J., DIXON, A.F.G. & HODEK, I. (eds) Behaviour and impact of Aphidophaga, Academic Publishing, Hague. KRIŠTÍN, A. 1994. Breeding biology and diet of the bee-eater (Merops apiaster) in Slovakia. Biologia, Bratislava 49: 273– 279. LÁSKA, P. & MAZÁNEK, L. 1998. Syrphidae, pp. 203–213. In: ROZKOŠNÝ, R. & VAŇHARA, J. (eds) Diptera of the Pálava Biosphere Reserve of UNESCO I., Folia Fac. Sci. Nat. Univ. Masaryk. Brun., Biol. 99. LAUTERER, P. & BUREŠ, S. 1984. Poznámky o bionomii křísa Errhomenus brachypterus (Homoptera: Cicadellidae) a jeho výskytu v potravě lejska bělokrkého (Ficedula albicollis Aves: Muscicapidae) [Notes on the bionomy of leafhopper Errhomenus brachypterus (Homoptera: Cicadellidae) and its occurrence in the food of white-collared flycatcher (Ficedula albicollis albicollis: Muscicapidae)]. Zpr. Kraj. Vlastivěd. Úst. v Olomouci 227: 18–20. RUXTON, G.D., SHERRATT, T.N. & SPEED, M.P. 2004. Avoiding attack. The evolutionary ecology of crypsis, warning signals & mimicry. Oxford University Press, Oxford, 250 pp. STUBBS, A.E. & FALK, S.J. 1983. British hoverflies – an illustrated identification guide. British Entomological & Natural History Society, 253 pp. Received January 24, 2005 Accepted May 26, 2005

Hoverflies in the diet of a small passerine

239

Appendix 1. List of hoverfly species found in the food of reed warbler/cuckoo nestlings. Eristalis arbustorum (L., 1758) – 1/0, Eristalis intricarius (L., 1758) – 0/1, Eristalis tenax (L., 1758) – 1/4, Helophilus pendulus (L., 1758) – 0/2, Mesembrius peregrinus (Loew, 1846) – 0/2, Mallota cimbiciformis (Fallén, 1817) – 0/1, Chalcosyrphus nemorum (F., 1805) – 0/10, Melanostoma mellinum (L., 1758) – 0/17, Platycheirus clypeatus (Meigen, 1822) – 3/8, Platycheirus peltatus (Meigen, 1822) – 0/1, Platycheirus scutatus (Meigen, 1822) – 0/1, Platycheirus sp. – 1/2, Scaeva pyrastri (L., 1758) 0/2, Meliscaeva auricollis (Meigen, 1822) – 0/1, Eupeodes corollae (F., 1794)

– 16/9, Eupeodes latifasciatus (Macquart, 1829) – 1/1, Eupeodes luniger (Meigen, 1822) – 2/3, Syrphus ribesii (L., 1758) – 0/2, Syrphus torvus Osten-Sacken, 1875 – 0/1, Syrphus vitripennis Meigen, 1822 – 1/4, Meligramma triangulifera (Zetterstedt, 1843) – 0/1, Episyrphus balteatus (De Geer, 1776) – 34/143, Sphaerophoria scripta (L., 1758) – 4/6, Sphaerophoria taeniata (Meigen, 1822) – 0/1, Neoascia interrupta (Meigen, 1822) – 0/2, Dasysyrphus albostriatus (Fallén, 1817) – 0/2, Chrysotoxum verrali Collin, 1940 – 0/3, Eumerus sp. – 0/1, larvae indet. – 6/13, pupae indet. – 0/4.

Biologia, Bratislava, 61/2: 239–240, 2006

FAUNISTICAL NOTES

First records of Resseliella theobaldi (Diptera, Cecidomyiidae), an important pest of raspberry from Slovakia Peter Tóth1, Monika Tóthová2 & Martina Váňová1 1

Department of Plant Protection, Slovak Agricultural University, A. Hlinku 2, SK-94976 Nitra, Slovakia; e-mail: [email protected] 2 Department of Sustainable Agriculture, Slovak Agricultural University, Mariánska 10, SK-94901 Nitra, Slovakia; e-mail: [email protected]

The raspberry cane midge, Resseliella theobaldi (Barnes, 1927) (Diptera, Cecidomyiidae), was first observed from SE England in 1920 and later became a serious pest of commercial raspberry plantations throughout Europe (WOODFORD & GORDON, 1978). Its occurrence is known from Germany (NOLTE, 1952), Sweden, Denmark (SYLVÉN, 1952), Belgium (NIJVELDT, 1954), Italy (GRASSI, 1993), Greece (VASILAKIS, 1995), Bulgaria (STOYANOV, 1963), Slovenia (MASTEN, 1958), Poland (REBANDEL, 1968), Russia (VERESHCHAGINA & KRITSKAIA, 1975), Hungary (AMBRUS, 1973) and Czech Republic (SKUHRAVÁ, 1997). No information has been available about R. theobaldi in Slovakia, but its occurrence was expected. Adults of raspberry cane midges are small (1.4–2.1 mm long), reddish brown, and because of their similarity to other midges, they are difficult (even impossible) to identify in the field (GORDON & WILLIAMSON, 1991). BARNES (1927) described adults, PITCHER (1952) presented a full description of single developmental stages and biology of the species. Larvae feed on the cortex of raspberry cane. First they are translucent, but they soon change to yellow or orange-pink. Fully grown larvae, measuring about 3.5 × 1.0 mm, fall to the soil surface to spin cocoons and pupate in the upper 1.0–4.0 cm layer (GORDON & WILLIAMSON, 1991). The direct damage caused by midge larval feeding to raspberry is superficial, but the feeding sites soon become infected by a range of fungi [e.g. Leptosphaeria coniothyrium (Fuckel)

Sacc., Didymella applanata (Niessl) Sacc.; WILLIAMSON & HARGREAVES, 1979], resulting in a disease called “midge blight”. Although females of R. theobaldi lay eggs on raspberry (Rubus ideaus L.), blackberry (Rubus fruticosus L.), loganberry (Rubus loganobaccus L.), rose (Rosa sp.), apple (Malus sp.), Haworth (Crataegus sp.), plum (Prunus domestica L.) and quince (Cydonia oblonga Mill.) (BARNES, 1944), larvae develop only on raspberry, blackberry, loganberry and rose (PITCHER, 1952). Only raspberry (R. ideaus) was recorded as a host plant in Slovakia. Material examined: Slovakia; larvae of R. theobaldi were found during July and August at 16 sites/host plant for each site was R. ideaus (col. = date of infested canes collection/larval collection; em. = date of adults emergence): Bardoňovo (48◦ 07 N, 18◦ 27 E, 205 m a.s.l.), 15.VIII.2005, 15 larvae; Bukovec (48◦ 42 N, 17◦ 29 E, 362 m a.s.l.), 1.VIII.2005, em. 24.VIII.2005, 1 , 3

; Demandice (48◦ 07 N, 18◦ 47 E, 143 m a.s.l.), 3.VIII.2005, 15 larvae; 27.VIII.2005, 57 larvae; Fabianka (48◦ 19 N, 19◦ 42 E, 190 m a.s.l.), 16.VIII.2005, 5 larvae; Imeľ (47◦ 54 N, 18◦ 08 E, , 8

; 108 m a.s.l.), 14.VIII.2005, em. 27.VIII.2005, 11 , 8

; em. 3.IX.2005, 1 , 2

; Kaem. 1.IX.2005, 4 menica nad Hronom (47◦ 50 N, 18◦ 44 E, 153 m a.s.l.), 24.VIII.2005, 3 larvae; Lok (48◦ 12 N, 18◦ 26 E, 200 m a.s.l.), 21.VIII.2005, 2 larvae; Martovce (47◦ 52 N, 18◦ 08

15. Grim T. 2006: Avian foraging studies: an overlooked source of distribution data for 15.conservation studies. macroecological and Grim T. 2006: Avian foraging studies: overlooked 12: source Diversity andan Distributions 1–3.of distribution data for macroecological and conservation studies. Diversity and Distributions 12: 1–3.

Diversity and Distributions, (Diversity Distrib.) (2006) Blackwell Publishing Ltd

BIODIVERSITY LETTER

Avian foraging studies: an overlooked source of distribution data for macroecological and conservation studies Tomas Grim

Department of Zoology, Palacky University, Tr. Svobody 26, 771 46 Olomouc, Czech Republic

Corresponding author, Tomas Grim, Department of Zoology, Palacky University, Tr. Svobody 26, 771 46 Olomouc, Czech Republic. E-mail: [email protected]

ABSTRACT

Macroecological and biogeographical studies and their applicability for biological conservation vitally depend on distribution data. These are usually taken from distribution atlases and databases or maps found in identification guides. A previous study pointed to another little explored source of data — local faunistic studies. Here, I would like to draw attention to another potential source. Papers that analysed the food composition of some taxa (e.g. insectivorous birds) are an overlooked source of rich information on taxa distributions. These studies frequently include an ‘Appendix’ with a list of food items determined to the species level. These studies also contain data on abundances, number of samples, sampling time, and geographical location as a rule. Foraging birds naturally provide data on invertebrate distributions with good spatio-temporal coverage and reasonably large samples. Importantly, birds frequently collect rare and by entomological methods hardly detectable species (i.e. those living high, in tree canopies, in very dense vegetation, or with secretive lifestyle). Data from bird dietary studies may help to ameliorate one of the most serious problems in distribution studies — zero inflation. I briefly discuss pros and cons of this so far neglected source of biogeographical data. Keywords Data sources, diversity, foraging, macroecology, methods, species richness.

There is a large body of literature on avian diet (see for example references in any standard handbook on avian biology, e.g. Cramp, 1992; del Hoyo et al., 2005). The majority of bird species is insectivorous or invertebratophagous (at least those breeding in the temperate zone; Stutchbury & Morton, 2001) and there are

thousands of studies providing data on composition of avian insect diet. Among various methods for the study of diet composition in birds the neck-collar (ligature) method yields reliable estimates of diet of birds. Neck collar is a plastic coated wire placed around the nestling neck that hinders the swallowing of food but is loose enough not to strangle the chick (see, e.g. Grim & Honza, 2001). Taxonomic composition of diet obtained by a neck collar is unaffected in comparison to some other methods (e.g. use of emetics or faecal samples) that, in contrast, do not provide unbiased data on quantity and quality of avian diet (Rosenberg & Cooper, 1990). The great advantage of the neck-collar method is that arthropod prey is kept relatively intact, thus enabling precise taxonomic determination of specimens including their sexing. Some avian dietary studies provide only rough data on taxonomic composition (i.e. food items are determined only to order or family levels, e.g. Grim & Honza, 1996, 2001). However, other dietary studies are frequently accompanied by an ‘Appendix’ with a complete list of diet items determined down to the species level (e.g. Torok, 1981; Bures, 1986, 1987, 1993, 1994; Kristin, 1986, 1994, 1995; Trnka, 1995; Grim & Honza, 1997; Exnerova et al., 2003; Grim, 2006; and many references therein). These appendices also contain data on abundances (and sometimes frequencies,

© 2006 The Author Journal compilation © 2006 Blackwell Publishing Ltd

DOI: 10.1111/j.1366-9516.2006.00251.x www.blackwellpublishing.com/ddi

INTRODUCTION Any macroecological and biogeographical study depends on sources of species-level distribution data. Recording databases, distribution atlases, and maps in identification guides are usually used as data sources for these studies. However, there are other sources of data that are utilized rarely or not at all. Keil and Konvicka (2005) drew attention to a huge and unused data source — published quantitative descriptions of local species assemblages. In this note I would like to point to still another source of distribution data for invertebrates that has, to my best knowledge, never been used for any distribution analyses and is not included in descriptive distribution atlases and local checklists as a rule. APPENDICES OF AVIAN DIETARY STUDIES: OVERLOOKED DATA CORNUCOPIA FOR MACROECOLOGISTS?

1

T. Grim i.e. proportions of samples containing particular food items). As both number of samples and the duration of sampling of an individual nestling/brood are as a standard provided in this sort of papers in Methods, it is easy to calculate the total sampling time and total abundance of food items of the particular study (which is necessary for rarefaction analysis to standardize data sets gathered with unequal sampling efforts). These studies also provide geographical data (location and coordinates) of particular study locations. The same holds not only for neck-collar studies but also for those based on pellets analyses (e.g. in owls, Trejo et al., 2005) or stomach content analyses (e.g. in passerines, Hromada & Kristin, 1996). Thus, many avian foraging studies (including master and doctoral theses and grey literature) contain all information necessary to perform macroecological distribution studies. QUALITATIVE DIETARY DATA: DECREASING INHERENT BIASES OF DISTRIBUTION STUDIES Various macroecological analyses can be done — and most in fact are done — on qualitative distribution data, i.e. presence or absence of particular taxa in a particular study location or grid cell (e.g. Jetz & Rahbek, 2002; Storch et al., 2003; Orme et al., 2005). Here, avian foraging studies can act as an important additional source of distribution data that complement standard entomological censuses. Most importantly, as avian foraging study can only add new species (or localities) to qualitative macroecological analyses, it can only decrease bias inherently present in any distribution study. On the other hand, it cannot in principle increase the sampling bias. Birds — to great surprise of experienced entomologists — frequently collect species of invertebrates that are extremely rare, i.e. are localized geographically, show unusually narrow ecological niches, and/or have very low population densities (i.e. are hardly detectable by humans; see below). Thus, birds as an effective tool able to sample rare species of insects and inaccessible habitats can help to correct one of the most serious problems in ecological studies — zero inflation. A high incidence of zero values is especially common problem in distribution data sets (Martin et al., 2005). At first sight, selectivity of avian foraging behaviour (Wolda, 1990) seems to lower benefits of dietary studies as sources of distribution data. For example, some insects defend themselves by distasteful compounds or mimic other defended insect species to avoid predation (Ruxton et al., 2004). However, passerines routinely forage even on aposematic unpalatable insects (e.g. Kristin, 1986; Exnerova et al., 2003) and their Batesian mimics (e.g. Kristin, 1986; Grim, 2006). Moreover, a brief review of previously mentioned avian foraging studies from Central Europe has shown that birds included in their diet species from all insect orders distributed in that area (the best represented group was Diptera). Although there can be a relationship between bird body size and prey size, even very small passerines (e.g. reed warbler, Acrocephalus scirpaceus, c. 12 g) are ready to take relatively large insects (up to 35 mm; Cramp, 1992; Grim & Honza, 2001). From 2

my own field experience and discussions with many experienced entomologists, I expect that the magnitude of selectivity of avian foraging is comparable to that of traditional entomological methods, but the two ‘selectivities’ may be just different. This highlights the potential of dietary studies to lower sampling biases of traditional entomological methods. Moreover, dietary selectivity cannot lead to biases if dietary data are used only as complementary sources as already stated. Additionally, avian foraging studies could in principle be used as sources of distribution data on their own. Many bird species were sampled for their food across large continental areas, i.e. at similar scales as macroecological studies are performed (e.g. Keil & Konvicka, 2005). This provides opportunities to study qualitative composition of ‘dietary taxocenes’ (i.e. taxonomically related sets of species within a ‘dietary community’) across large geographical areas. Because foraging selectivity varies between species, researchers could control for this potential confounding factor by using data only from a particular study species across the various study sites. Moreover, the study of geographical variability of birds’ foraging selectivity may help to clarify some of the trophic mechanisms behind the intensively studied geographical patterns of bird diversity (see, e.g. Jetz & Rahbek, 2002; Orme et al., 2005). To my knowledge, the avian dietary studies are usually ignored when descriptive distribution atlases and local checklists are compiled. These often represent a basic reference for a macroecological analysis. Hence, bird dietary studies could improve quality of such references as well. Importantly, birds’ foraging as a method of study of taxocene composition has some advantages in comparison to traditional entomological methods (e.g. entomological net or sets of traps). Birds forage — and consequently ‘sample’ — all vegetation strata (and air high above the ground; Cheng & Zhou, 1987) while a human entomologist with a net is limited to lowest strata (only herbaceous and shrub strata in fact). In addition to these spatial aspects there is also temporal advantage: birds forage (and neckcollar method is applicable) even in the rain when, e.g. the entomological net performs poorly. To sum up, avian foraging studies may provide biased data on composition of invertebrate taxocenes, but we should remember that various standard entomological methods also suffer from differing effectiveness and selectivity as mentioned by Keil and Konvicka (2005). CONCLUSIONS Foraging birds naturally provide data on invertebrate distributions with good spatio-temporal coverage, reasonably large samples and well intact specimens. Bird ‘entomologist’ can even work better than a human one, e.g. two rare species of hoverflies included in local taxon list for The Palava UNESCO Protected Landscape and Biosphere Reserve (Czech Republic) were found only by reed warbler ‘collectors’ despite the efforts of dozens of entomologists active in that area for several decades (Grim, 2006 and references therein). Similarly, a meta-analysis of true bugs (Heteroptera) in bird diets revealed several very rare epigeic and

© 2006 The Author Diversity and Distributions, Journal compilation © 2006 Blackwell Publishing Ltd

Overlooked data sources for macroecology canopy taxa (Exnerova et al., 2003). Thus, the study of nestling diet in birds may provide important information on extremely rare species and/or those which may be difficult to collect by traditional entomological methods (Lauterer & Bures, 1984; Exnerova et al., 2003; Grim, 2006) or even lead to the discovery of new species for a particular country (Bures & Pecina, 1993; Kristin & Patocka, 1997; Grim, 2006). Both these aspects of avian foraging studies are clearly of importance for biogeography, macroecology, and conservation biology. ACKNOWLEDGEMENTS I am grateful to Alice Exnerova, Mark E. Hauber, Anton Kristin, and two anonymous referees for their comments. REFERENCES Bures, S. (1986) Composition of the diet and trophic ecology of the collared flycatcher (Ficedula albicollis albicollis) in three segments of groups of types of forest geobiocenoses in central Moravia (Czechoslovakia). Folia Zoologica, 35, 143–155. Bures, S. (1987) Diet analysis and trophic ecology of the grey wagtail (Motacilla cinerea Tunst.) in Nizky Jesenik. Folia Zoologica, 36, 257–264. Bures, S. (1993) Food of water pipit nestlings, Anthus spinoletta, in changing environment. Folia Zoologica, 42, 213–219. Bures, S. (1994) Segregation of diet in water pipit (Anthus spinoletta) and meadow pipit (Anthus pratensis) nestlings in an area damaged by air pollution. Folia Zoologica, 43, 43–48. Bures, S. & Pecina, P. (1993) Occurrence of March-fly Bibio benesi (Diptera, Nematocera, Bibionidae) in water pipit diet in Praded Mt. area. (in Czech with summary in English). Severni Morava, 66, 49–50. Cheng, Z.Q. & Zhou, B.X. (1987) Diet analyses of the large white-rumped swift, Apus pacificus, at Chenlushan Island in the Yellow Sea and examination of their pattern of activities by radar. Acta Zoologica Sinica, 33, 180–186. Cramp, S., ed. (1992) The birds of the western Palearctic warblers, Vol. VI. Oxford University Press, Oxford. Exnerova, A., Stys, P., Kristin, A., Volf, O. & Pudil, M. (2003) Birds as predators of true bugs (Heteroptera) in different habitats. Biologia, 58, 253–264. Grim, T. (2006) An exceptionally high diversity of hoverflies (Syrphidae) in the food of the reed warbler (Acrocephalus scirpaceus). Biologia, 61, 235–239. Grim, T. & Honza, M. (1996) Effect of habitat on the diet of reed warbler (Acrocephalus scirpaceus) nestlings. Folia Zoologica, 45, 31–34. Grim, T. & Honza, M. (1997) Differences in parental care of reed warbler (Acrocephalus scirpaceus) in its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zoologica, 46, 135–142. Grim, T. & Honza, M. (2001) Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behavioral Ecology and Sociobiology, 49, 322–329. del Hoyo, J., Elliott, A. & Christie, D. (2005) Handbook of the birds of the world, Vol. 10. Lynx Editions, Barcelona.

Hromada, M. & Kristin, A. (1996) Changes in the food of the great grey shrike (Lanius excubitor) during the year. Biologia, 51, 227–233. Jetz, W. & Rahbek, C. (2002) Geographic range size and patterns of avian species richness. Science, 297, 1548–1551. Keil, P. & Konvicka, M. (2005) Local species richness of Central European hoverflies (Diptera: Syrphidae): a lesson taught by local faunal lists. Diversity and Distributions, 11, 417– 426. Kristin, A. (1986) Heteroptera, Coccinea, Coccinellidae and Syrphidae in the food of Passer montanus L. and Pica pica L. Biologia, 41, 143–150. Kristin, A. (1994) Breeding biology and diet of the bee-eater (Merops apiaster) in Slovakia. Biologia, 49, 273–279. Kristin, A. (1995) The diet and foraging ecology of the penduline tit (Remiz pendulinus). Folia Zoologica, 44, 23–29. Kristin, A. & Patocka, J. (1997) Birds as predators of Lepidoptera: selected examples. Biologia, 52, 319–326. Lauterer, P. & Bures, S. (1984) Notes on the bionomy of leafhopper Errhomenus brachypterus (Homoptera: Cicadellidae) and its occurrence in the food of white-collared flycatcher (Ficedula albicollis albicollis: Muscicapidae). (in Czech with summary in English). Zpravodaj Krajskeho Vlastivedneho Ustavu v Olomouci, 227, 18–20. Martin, T.G., Wintle, B.A., Rhodes, J.R., Kuhnert, P.M., Field, S.A., Low-Choy, S.J., Tyre, A.J. & Possingham, H.P. (2005) Zero tolerance ecology: improving ecological inference by modelling the source of zero observations. Ecology Letters, 8, 1235– 1246. Orme, C.D.L., Davies, R.G., Burgess, M., Eigenbrod, F., Pickup, N., Olson, V.A., Webster, A.J., Ding, T.S., Rasmussen, P.C., Ridgely, R.S., Stattersfield, A.J., Bennett, P.M., Blackburn, T.M., Gaston, K.J. & Owens, I.P.F. (2005) Global hotspots of species richness are not congruent with endemism or threat. Nature, 436, 1016–1019. Rosenberg, K.V. & Cooper, R.J. (1990) Approaches to avian diet analysis. Studies in Avian Biology, 13, 80–90. Ruxton, G.D., Sherratt, T.N. & Speed, M.P. (2004) Avoiding attack: the evolutionary ecology of crypsis, warning signals and mimicry. Oxford University Press, Oxford. Storch, D., Konvicka, M., Benes, J., Martinkova, J. & Gaston, K.J. (2003) Distribution patterns in butterflies and birds of the Czech Republic: separating effects of habitat and geographical position. Journal of Biogeography, 30, 1195–1205. Stutchbury, B.J.M. & Morton, E.S. (2001) Behavioral ecology of tropical birds. Academic Press, London. Torok, J. (1981) Food composition of nestling blackbirds in an oak forest bordering on an orchard. Opuscula Zoologica (Budapest), 17–18, 145–156. Trejo, A., Kun, M., Sahores, M. & Seijas, S. (2005) Diet overlap and prey size of two owls in the forest-steppe ecotone of southern Argentina. Ornitologia Neotropical, 16, 539–546. Trnka, A. (1995) Dietary habits of the great reed warbler (Acrocephalus arundinaceus) young. Biologia, 50, 507–512. Wolda, H. (1990) Food availability for an insectivore and how to measure it. Studies in Avian Biology, 13, 38–43.

© 2006 The Author Diversity and Distributions, Journal compilation © 2006 Blackwell Publishing Ltd

3

16. Grim T. 2005: Host recognition of brood parasites: implications for methodology in 16. recognition. studies of enemy Grim T. 2005: Host recognition of brood parasites: implications for methodology in Auk 122(2): 530–543. studies of enemy recognition. Auk 122(2): 530–543.

The Auk 122(2):530–543, 2005 © The American Ornithologists’ Union, 2005. Printed in USA.

HOST RECOGNITION OF BROOD PARASITES: IMPLICATIONS FOR METHODOLOGY IN STUDIES OF ENEMY RECOGNITION T
 G1 Department of Zoology, Palacký University, tř. Svobody 26, CZ–771 46 Olomouc, Czech Republic

A   .—Various studies have shown that experiments on nest defense and enemy recognition (e.g. recognition of adult brood parasites) can be confounded by many factors. However, no study has described a confounding effect of control dummy type. Here, I show experimentally that the choice of control dummy may influence the results of an experiment and lead to erroneous conclusions. I tested recognition abilities of the Blackcap (Sylvia atricapilla), currently a host rarely used by the Common Cuckoo (Cuculus canorus). Blackcaps responded very differently to two kinds of control dummies: they ignored the Eurasian Blackbird (Turdus merula) dummy, but a acked the Rock Pigeon (Columba livia) dummy as frequently as they a acked the Common Cuckoo. The differing results may be explained by the fact that the Rock Pigeon is more similar to the Common Cuckoo than the Eurasian Blackbird is, and consequently elicited more aggressive behavior than the la er. Thus, absence of discrimination in enemy-recognition studies may reflect a methodological artifact resulting from varying abilities of particular hosts to discriminate along a continuum of recognition cues. This result has serious methodological implications for further research on enemy recognition and aggression in general: a control dummy should not be too similar to the dummy brood parasite; otherwise, the chance of detecting existing recognition abilities is low. Further, I argue that coevolution only increases pre-existing aggression in the particular host species. Therefore, increment analysis (assessing changes in host antiparasitic responses during the nesting cycle while controlling for background aggression to control dummies) provides a more accurate picture of hosts’ recognition abilities than the traditional approach (when the total level of antiparasitic response is analyzed). Received 30 January 2004, accepted 3 November 2004. Key words: brood parasitism, coevolution, discrimination, methodology, nest defense, recognition.

Reconocimiento del Hospedero de los Parásitos de Nidada: Consecuencias para las Metodologías de Estudios sobre Reconocimiento del Enemigo R.—Varios estudios han mostrado que los experimentos sobre defensa de nidos y reconocimiento del enemigo (e.g. reconocimiento de parásitos de nidadas adultos) pueden ser equívocos por muchos factores. Sin embargo, ningún estudio ha descrito un efecto equívoco dado por el tipo de modelo usado como control. En este estudio, demuestro experimentalmente que la elección del modelo puede influenciar los resultados de un experimento, conduciendo a una conclusión errónea. Probé las habilidades de reconocimiento de Sylvia atricapilla, que es actualmente un hospedero raramente usado por Cuculus canorus. S. atricapilla respondió de manera muy diferente ante dos tipos de modelos control: ignoraron a modelos de Turdus merula, pero atacaron a modelos de Columba livia tan frecuentemente como atacaron a C.

1

E-mail: [email protected]

530

April 2005]

Methodology of Enemy Recognition Studies

531

canorus. Las respuestas diferenciales pueden ser explicadas por el hecho que C. livia es más similar a C. canorus que a T. merula, por lo que provocó un comportamiento más agresivo que contra T. merula. Por lo tanto, la falta de discriminación en estudios sobre el reconocimiento de enemigos puede estar reflejando un artefacto metodológico que resulta de la variación en la habilidad de un hospedero en particular de discriminar a lo largo de un continuo de señales de reconocimiento. Este resultado presenta serias consecuencias metodológicas para estudios futuros sobre reconocimiento de enemigos y sobre agresividad en general: el modelo utilizado como control no debe ser muy parecido al modelo del parásito de nidada; de lo contrario, la probabilidad de detectar la habilidad de reconocimiento existente es muy baja. Además, argumento que la coevolución sólo aumenta la agresión preexistente en una especie hospedera en particular. Por lo tanto, los análisis de incremento (la determinación de los cambios en las respuestas antiparasitarias de los hospederos durante el ciclo de nidificación mientras se controla por la agresión de fondo hacia modelos control) brindan una mejor idea sobre las habilidades de reconocimiento de los hospederos que el enfoque tradicional (cuando el nivel total de la respuesta antiparasitaria es analizado). A    subject to strong selection pressure from various environmental factors, including predators (Montgomerie and Weatherhead 1988) and brood parasites (Rothstein and Robinson 1998). Response or absence of response to those stimuli may have an important effect on an individual’s fitness. Given that brood parasites may reduce host reproductive success (Rothstein 1990), the best defense against parasitism should be to deter a parasitic female from laying her egg in a host nest in the first place (Sealy et al. 1998). However, nest defense, like any other activity around the nest, can be costly—it can a ract predators (Martin et al. 2000) or brood parasites (Banks and Martin 2001), there may be a trade-off with parental care (Ueta 1999), and the defending parent risks injury (McLean et al. 1986, McLean 1987, Montgomerie and Weatherhead 1988). Selection, then, should favor recognition of specific intruders. Importantly, the difference between the generalized nest defense and a specific response to the parasite is not relevant, per se, to host avoidance by the parasite—both responses could reduce the probability of parasitism. However, specific enemy recognition can be important evidence of host–parasite coevolution (Sealy et al. 1998). Host ability to recognize the parasite as a unique enemy and to respond aggressively has been studied in various hosts of the Brownheaded Cowbird (Molothrus ater; e.g. Robertson and Norman 1976, 1977; Briskie and Sealy 1989; Burgham and Picman 1989; Hobson and Sealy

1989; Neudorf and Sealy 1992; Gill and Sealy 1996, 2004; Gill et al. 1997a, b; Sealy et al. 1998). By contrast, relatively less is known about the enemy-recognition capabilities of host species parasitized by the Common Cuckoo (Cuculus canorus; hereaer “cuckoo”). The cuckoo hosts that have been properly tested for specific enemy recognition include only the Reed Warbler (Acrocephalus scirpaceus; Duckworth 1991, Lindholm and Thomas 2000, Honza et al. 2004), Great Reed Warbler (A. arundinaceus; Bartol et al. 2002), Meadow Pipit (Anthus pratensis; Moksnes and Røska 1989), Eurasian Blackbird (Turdus merula), and Song Thrush (T. philomelos) (Grim and Honza 2001). Other authors have studied responses of cuckoo hosts toward the parasite (e.g. Smith and Hosking 1955, Moksnes and Røska 1988, Moksnes et al. 1990, Røska et al. 2002). They found that species that are appropriate hosts and have probably been involved in a coevolutionary “arms race” with the cuckoo were significantly more aggressive toward the parasite than species that were not appropriate hosts (i.e. hole nesters, seed eaters). However, even unsuitable hosts showed some aggression toward cuckoos (e.g. Røska et al. 2002). Thus, it is possible that some species are aggressive against any intruders near the nest, including innocuous ones (Bazin and Sealy 1993). Although the results of the above studies are suggestive, their conclusions would clearly be stronger if experiments where host responses to both parasite and nonthreatening

532

T
 G

controls were compared (for detailed discussion, see Sealy et al. 1998). In general, it is important to differentiate between a generalized nest defense (i.e. host responds to various intruders—parasite, predator, food competitor, nest-site competitor, even innocuous species—with similarly aggressive responses) and a specific response to the parasite (i.e. host ignores nonthreatening intruders or shows significantly lower response to them than to the parasite at an early stage in the nesting cycle, when parasitism is a greater threat). Only the la er could be accepted as evidence of coevolution between the parasite and a particular host (Neudorf and Sealy 1992, Gill and Sealy 1996, Sealy et al. 1998). Moreover, all European species tested for specific recognition so far are either acceptors or current common hosts that reject only at intermediate frequencies (Moksnes and Røska 1989, Duckworth 1991, Lindholm and Thomas 2000, Grim and Honza 2001, Bártol et al. 2002). Virtually nothing is known about enemy-recognition abilities of strong egg-rejecters parasitized by the cuckoo. Further, from the point of view of studies of nest defense and recognition in general, it is crucial to understand factors that could confound the results of such studies. A wide array of confounding factors has received a ention (e.g. number of previous visits to a tested nest, posture of the experimental intruder, live vs. mounted predator, etc.; Knight and Temple 1986a, b; for reviews, see Montgomerie and Weatherhead 1988, Sealy et al. 1998). However, the possible confounding effect of the type of control dummy used has not been properly examined before. Although Robertson and Norman (1976, 1977) used various control dummies, they reported that responses to different control dummies were the same and lumped the results without providing details. Given that no recognition system can be perfect, every individual will commit recognition errors if presented with two stimuli that are sufficiently similar (Sherman et al. 1997). Hosts of parasitic birds vary in their discrimination capabilities—toward both eggs (e.g. Davies and Brooke 1989) and adult parasites (e.g. Sealy et al. 1998). Therefore, we can reasonably expect that tested individuals of lessdiscriminating host species will respond aggressively even to innocuous intruders if they are too similar to really threatening intruders. That could lead to confusion in the interpretation of results of enemy-recognition studies.

[Auk, Vol. 122

I studied enemy recognition and nestdefense behavior of the Blackcap (Sylvia atricapilla), a suitable host for the cuckoo (Moksnes and Røska 1995). Analysis of museum-held clutches of European passerines indicated that the Blackcap was the 16th most frequent cuckoo host in Europe (Moksnes and Røska 1995). Furthermore, Blackcaps are strong rejecters of both nonmimetic (76.9%; Moksnes et al. 1990) and mimetic (100.0%; Moksnes and Røska 1992) eggs introduced into their nests, which indicates that Blackcaps have been frequently parasitized by the cuckoo in their evolutionary history. However, there are no recent reports of cuckoo parasitism on Blackcaps. Moksnes et al. (1990) and Røska et al. (2002) recorded high levels of aggression by Blackcaps toward the cuckoo dummy; however, they did not use a control. Therefore, I investigated specific enemy recognition in the Blackcap, testing two hypotheses: (1) the generalized nest-defense hypothesis, which suggests that hosts do not recognize a brood parasite as a unique enemy and predicts similar response to parasites and innocuous controls; and (2) the specialized nest-defense hypothesis, which assumes that hosts recognize the parasite as a unique threat and predicts that hosts show aggression only (or more so) to parasites but ignore (or show significantly lower response) to controls. Additionally, I used two control dummies that differed in the degree to which they resembled the cuckoo to examine whether the choice of control dummy type may confound the ability of researchers to distinguish between those two hypotheses. M 
 Study site and species.—I conducted the study in a deciduous forest near the village of Dolní Bojanovice (48°52’N, 17°00’E), in the southeastern Czech Republic, ~60 km southeast of Brno. Data were collected from 25 April to 30 June in 2000 and 2001. Because the data were not significantly different between years (and data analyzed separately for the each field season gave qualitatively the same results), I pooled the data. Nest defense was tested with stuffed (taxidermic) dummies. I primarily followed the experimental procedure suggested by Sealy et al. (1998); however, I did not use a predator

April 2005]

Methodology of Enemy Recognition Studies

dummy, for reasons elaborated below. I tested host responses toward the cuckoo (the brood parasite), the Rock Pigeon (Columba livia domestica; hereaer “pigeon”; control 1) and the Eurasian Blackbird (adult male; hereaer “blackbird”; control 2). I used two stuffed cuckoo dummies in experiments (responses to each were identical). The pigeon dummy was pale gray (a shade similar to the cuckoo’s) overall, with two dark wing bars and a dirty white rump. The pigeon and cuckoo dummies were almost the same size (measured from the base of the bill to the wing tip: pigeon 25.0 cm, cuckoo 25.5 cm), though the cuckoo had a longer tail (17 cm) than the pigeon (12 cm). The blackbird dummy was a bit smaller (23 cm; tail 10 cm), overall black with yellow bill and eye ring. The cuckoo also had a yellow bill and eye ring. The pigeon and blackbird were chosen as controls because both are completely innocuous for the Blackcap: they are neither brood parasites nor predators, and there is no confounding effect of competition for food and nest sites. Some researchers prefer to use control species that are familiar to tested hosts (e.g. McLean 1987, Moksnes and Røska 1989), whereas others argue it is be er to use controls that do not occur on the study area (e.g. Hobson and Sealy 1989). I believe that a control species that is familiar to a tested host species provides a stronger test for enemy-recognition abilities because hosts have had prior opportunities to compare the threatening and nonthreatening sympatric species and adapt their behavior correspondingly (see also Mark and Stutchbury 1994). Either way, Sealy et al. (1998) suggested that prior (in)experience with a control species should have no effect. Blackcaps, blackbirds, pigeons, and cuckoos occur in sympatry in my study area. Experimental procedure.—I performed two series of paired experiments during the laying stage, when the cuckoo represents the greatest danger to the host (Davies and Brooke 1989). In the first series (n = 20 nests), I tested responses to the cuckoo and pigeon; in the second (n = 15 nests), to the cuckoo and blackbird. These sample sizes are higher than sample sizes in several studies that observed significant differences in behavior to various intruders (e.g. Robertson and Norman 1976, 1977; Smith et al. 1984; McLean 1987; Hobson and Sealy 1989;

533

Duckworth 1991), and thus should be sufficient to test for specific recognition (for more details, see below). Because cuckoos are a threat to hosts also during incubation and nestling stages (when they depredate both eggs and nestlings at nests too advanced for successful parasitism; e.g. Jourdain 1925, Gärtner 1981), I also tested host responses to the cuckoo and pigeon dummies during incubation (n = 20) and nestling (n = 20) stages. Blackcaps start to incubate with the penultimate egg (Cramp 1992), and some nests were tested on a day when the penultimate or last egg was laid. Those nests are included in the “laying stage” group. Models were presented at random to eliminate order effects (Kamil 1988). First, I a ached one of the dummies in a life-like position to a branch ~0.5 m from each nest, level with it and facing the nest rim. Timing started aer I retreated to the blind, set ≥15 m from the nest. Aer one parent appeared near the nest and became aware of the dummy, I observed reactions of nest owners for 5 min (from the moment the first parent arrived, even when it did not respond aggressively to the dummy). Presentation of the second mount at the same place was separated by 30 min to avoid habituation or carry-over aggression (Sealy et al. 1998). Each nest was tested only once to avoid pseudoreplication. In 2000, almost all Blackcaps breeding in the forest were ringed (M. Honza, V. Mrlík, M. Čapek, P. Procházka unpubl. data). In 2001, I observed only one ringed bird among tested individuals; therefore, probably no particular bird was tested in both years. Intensity of Blackcap responses varied from quiet watching at a short distance to vigorous a acks (i.e. flying at and hi ing the dummy; close passes are not included in a acks; see below). Because the frequency of alarm calling (“Tak-calls”; Cramp 1992), a acks, and so forth was too high to be recorded exactly on the datasheet, but discrete categories of host behavior (see below) were clearly definable, I categorized behaviors according to relative scales (see also McLean et al. 1986, Pavel and Bureš 2001). Both male and female responses were combined for each of the defense categories if both parents were present. I scored the frequency of alarm calling (0 to 3; 0 = no vocalizations, 1 = overall time spent calling <1 min, 2 = calling <3 min, 3 = calling >3 min) and a acks (i.e. bird contacted dummy; 0 to 2; 0 = no a acks, 1 = <5 a acks,

T
 G

534

2 = >5 a acks). Total level of nest defense was ranked on a scale depending on the risk taken by tested bird(s) (0 to 4; 0 = no response, i.e. silent watching of dummy; 1 = few vocalizations, bird[s] >5 m from dummy; 2 = more vocalizations, bird[s] <5 m from dummy; 3 = frequent vocalizations and close passes; 4 = frequent vocalizations and a acks). Total level of nest defense is the overall effort of nest owners (both male and female) defending the nest. I also recorded delay in arrival of nest owner(s) (in minutes), number of individuals that responded, and time spent <1 m from the dummy (in minutes). Data analyses.—I analyzed data in three ways (see also Maloney and McLean 1995). First, I compared differences in responses to dummies using five variables probably related to quality of nest defense (latency to response, frequency of alarm calling, time spent <1 m from a dummy, frequency of a acks, and number of individuals performing nest defense) to make different aspects of host behavior comparable with other studies (see Sealy et al. 1998). The five variables describe all behaviors Blackcaps performed when confronted with dummies (no Blackcaps flu ered or dived above the dummy, as other species do, e.g. Collared Flycatchers [Ficedula albicollis]; T. Grim unpubl. data). Second, I tested for differences between responses to dummies in Blackcaps’ total level of nest defense (see above). Third, because the different measures of nest defense were intercorrelated, I performed principal components analysis (PCA) on alarm calling, a acks, and number of individuals responding (I included the la er factor because it theoretically could have an important effect on probability of deterring the cuckoo from the nest; I did not include latency and time spent <1 m from a dummy, for reasons given below). Because the responses were measured on an ordinal scale (see above), I analyzed data by nonparametric Wilcoxon signed-ranks tests and Mann-Whitney U-tests. To keep the experimentwise error rate at α = 0.05, I used Bonferroni correction. R  C

 V   Avian nest defense can be influenced by various factors (Montgomerie and Weatherhead

[Auk, Vol. 122

1988). Thus, before statistically testing for recognition abilities in Blackcaps, I checked whether the level of nest defense was influenced by the reproductive value of the nest (calculated as a product of clutch or brood size and age of the clutch or brood), time of season, time of day, number of previous visits to the nest, and nest concealment. I found no significant effect of those confounding variables (T. Grim unpubl. data). Moreover, all those variables were held constant because of the within-subject design (Kamil 1988) of the study. Results of PCA on alarm calling, a acks, and number of individuals showed that the first principal component (PC1) explained 59.7% of variance in the data. The first principal component was positively correlated with the remaining three nest-defense variables (alarm calling: rs = 0.77, P = 0.001; a acks: rs = 0.73, P = 0.001; number of individuals: rs = 0.79, P = 0.001). Intensity of nest defense (PC1) did not change during the nesting cycle, from laying to nestling stages (responses to cuckoo: Kruskal-Wallis ANOVA, χ2 = 0.34, df = 2, P = 0.84; responses to pigeon: χ2 = 2.09, df = 2, P = 0.35; the same result was obtained for total level of nest defense on the ordinal scale). Given that coevolution between a parasite and host may only increase pre-existing general aggression (see below), I controlled for possible variation in general (“background”) nest defense during the nesting cycle by subtracting intensity of aggression toward the control (pigeon) from intensity of aggression to the parasite (cuckoo). That stronger test confirmed that there is no relationship between anti-cuckoo aggression and nesting cycle (Kruskal-Wallis ANOVA, χ2 = 0.13, df = 2, P = 0.94). Recognition abilities can be age-dependent (Smith et al. 1984). Older birds breed earlier (Sæther 1990) and can show be er enemy recognition (Smith et al. 1984) or stronger nest defense (Hobson and Sealy 1989), which could have confounded the results. However, logistic regression showed no effect of time of season (date of the first egg laid) on distribution of birds responding “correctly” (more response to cuckoo) and “incorrectly” (same response to the two dummies, stronger response to pigeon, or no response to either dummy; r2 = 0.02, χ2 = 1.23, df = 1, P = 0.27, n = 60). Order of presentation of dummies had no effect on any variables included in comparisons when data from first and second presentations

April 2005]

535

Methodology of Enemy Recognition Studies

of the cuckoo dummy were compared (MannWhitney U-tests: all nonsignificant). The same result was obtained from comparison of first and second presentations of pigeon and blackbird dummies and when data for incubation or incubation and nestling periods were included. There were no differences in response to first and second dummy (Wilcoxon matched-pairs tests: all nonsignificant). Thus, there was no confounding effect of habituation. C

–P
 E  Responses toward the cuckoo and pigeon dummies were very similar in all three nesting stages (Table 1). Blackcaps approaching both types of dummies typically u ered alarm calls (“Tak-calls”) at a very high frequency (~30 per 10 s; T. Grim pers. obs.; see also sonogram in Cramp 1992). Both the cuckoo and pigeon dummies were frequently a acked (proportion of experiments when a dummy was a acked at least once; laying stage: cuckoo: 42.9%, pigeon: 35.0%: χ2 = 0.33, df = 1, P = 0.57; all stages pooled:

cuckoo 38.3%, pigeon 28.3%: χ2 = 1.35, df = 1, P = 0.24). There were no significant differences between dummies in the frequency with which alarm calling and a acks were elicited from the Blackcaps (Table 1). The result was consistent among three types of analyses of data (Table 1). C

–B  E  Blackcap responses to the cuckoo and innocuous pigeon were not significantly different, despite the large sample size (n = 60) when data for all nesting stages were pooled. Therefore, I tested for responses to the cuckoo and another control dummy—the blackbird—in the second series of experiments. I found clear differences in host behavior (Table 2). Responses to the cuckoo were again aggressive (40% of experiments with an a ack) and did not differ significantly from responses to cuckoos during laying stage in the first series of experiments (MannWhitney tests, all nonsignificant). In striking contrast, when confronted with the blackbird dummy, Blackcaps never u ered alarm calls,

T  1. Summary of Blackcap responses to dummy cuckoo (C) and pigeon (P) during three nesting stages. In the far right column, data from all stages are pooled. Response variable a

Dummy type

Egg laying (n = 20)

Incubation (n = 20)

Latency (minutes)

C P C P C P C P C P C P C P

10.1 ± 2.5 11.8 ± 2.6 2 2 3.0 ± 0.4 3.0 ± 0.5 0 0 1.3 ± 0.1 1.4 ± 0.1 3 3 0.06 ± 0.34 –0.14 ± 0.26

4.1 ± 0.7 4.6 ± 0.7 2 2 2.5 ± 0.5 2.3 ± 0.5 0 0 1.6 ± 0.1 1.7 ± 0.1 2 2 0.2 ± 0.3 0.0 ± 0.3

Alarm b Less than 1 m (minutes) A acks c Number of individuals Total level d PC1 e

Nestling (n = 20) 5.6 ± 1.3 5.1 ± 1.3 2 2 3.3 ± 0.5* f 2.9 ± 0.5 0 0 1.7 ± 0.1 1.8 ± 0.1 3 3 0.3 ± 0.3 0.4 ± 0.3

All stages (n = 60) 6.6 ± 1.0 7.2 ± 1.0 2 2 2.9 ± 0.3* f 2.7 ± 0.3 0 0 1.5 ± 0.1 1.6 ± 0.1 3 3 0.2 ± 0.2 0.1 ± 0.2

a Values for latency, time spent <1 m from a dummy, number of individuals, and PC1 are means ± SE; values for alarm, a acks, and total level are medians. b Ordinal scale: 0 = no vocalizations, 1 = overall time spent calling <1 min, 2 = calling <3 min, 3 = calling >3 min. c Ordinal scale: 0 = no a acks (i.e. no bird contacted dummy), 1 = <5 times, 2 = >5 times. d Ordinal scale: 0 = no response (i.e. silent watching of dummy); 1 = few vocalizations, bird(s) >5 m from the dummy, 2 = more vocalizations, bird(s) <5 m from dummy; 3 = frequent vocalizations and close passes; 4 = frequent vocalizations and a acks. e PCA performed on alarm calling, a acks, and number of individuals. For details, see text. f Asterisks indicate results of Wilcoxon signed-ranks test between models; P < 0.05. Difference in “Less than 1 m“ is not significant aer Bonferroni correction.

T
 G

536

[Auk, Vol. 122

T  2. Differences in responses of Blackcaps to stuffed dummies. Comparisons of host behaviors in the paired experiment (cuckoo–blackbird; n = 15) and unpaired experiment (pigeon–blackbird; (n = 20, 15) are shown. For brief explanations of response variables, see Table 1; for details, see text. Cuckoo–blackbird a Response

Parasite

Control

t

Latency Alarm Less than 1 m A acks Number of individuals Total level PC1

11.1 ± 1.1 2 3.2 ± 0.5 0 1.7 ± 0.1 3 0.41 ± 0.0

11.2 ± 1.5 0 4.0 ± 0.3 0 1.2 ± 0.1 0 –1.50 ± 0.3

0.68 –3.97 2.50 –2.65 –3.22 –3.97 –4.45 ± 0.11

a b c

Pigeon–blackbird b P 0.50 0.0001* c 0.013 0.008* 0.001* 0.0001 0.0001

U 1.15 –4.24 1.32 –2.49 0.94 –4.82 –3.71

P 0.25 0.0001* 0.19 0.01* 0.35 0.0001 0.0002

Results of Wilcoxon matched-pairs tests; t-values lower than zero indicate higher response toward the cuckoo. Results of Mann-Whitney tests. U-values lower than zero indicate higher response toward the pigeon. Asterisks indicate differences significant at P = 0.05 aer sequential Bonferroni test (for the first five variables).

never a acked the dummy, and frequently (46.7% of experiments; n = 15) resumed incubation (latency to incubation = 1.4 ± 0.4 min; n = 7). Incubation behavior was never observed during the first series of experiments, even when tested birds were not aggressive toward either the cuckoo or pigeon dummies. When faced with the blackbird near their nests, three Blackcaps u ered weak squeaky sounds, but no individual u ered alarm calls. Latency to response did not differ between the cuckoo and blackbird trials (Table 2). I hypothesize that this variable probably reflects general incubation and nest-a entiveness patterns of the host and has no relationship with nest defense. This is supported by (1) nonsignificant correlations between latency and all measures of host nest-defense behavior (Spearman rank correlations, n = 35; all nonsignificant) and (2) the fact that the significant decline of latency with age of nest completely disappears when data from egg-laying period (before incubation starts) are excluded (linear regressions; whole nesting cycle: r2 = 0.19, F = 15.6, df = 1 and 68, P = 0.0002; without egg-laying period: r2 = 0.01, F = 0.43, df = 1 and 33, P = 0.52). In the cuckoo–pigeon experiments, the time spent near a dummy positively correlated with aggression. By contrast, in the cuckoo–blackbird trials, time spent near the dummy was correlated with absence of aggression (an inevitable effect of incubation behavior—when incubating, Blackcaps were <1 m from the dummy). Because time spent near the dummy shows opposing relationships in two respective series of experiments, I did not include that variable

in the PCA, because it would confound results. The possible confounding effect was confirmed when I recalculated PCA for frequency of alarm calling, time spent <1 m from the dummy, frequency of a acks, and number of individuals responding; the difference in responses to cuckoo and blackbird dummies was not significant (Wilcoxon matched-pairs test: t = –1.64, P = 0.10, n = 15). That result is clearly spurious, because the difference in responses could hardly be higher than that shown by Blackcaps toward the cuckoo as compared with the blackbird. In summary, Blackcaps showed significantly lower response toward the blackbird dummy than toward the cuckoo dummy in all measured variables (except latency to response; see above). Three types of analyses of the data gave the same results (Table 2). Responses to the blackbird were also significantly different from reactions to the pigeon during the laying stage (Table 2). D
 S 
 G  R
 The two series of experiments lead to different interpretations. The cuckoo–pigeon experiment indicated that Blackcaps do not recognize the cuckoo as a special enemy (they a ack the innocuous pigeon at the same level of aggression as they do the cuckoo). That result supports the hypothesis that host responses can be described as a generalized nest defense (hypothesis 2; see above; Sealy et al. 1998). Coevolution with the cuckoo (together with other forces, like predation) could contribute to the host’s aggressive

April 2005]

Methodology of Enemy Recognition Studies

behavior, but it did not lead to the evolution of a nest defense based on specific recognition of various intruders. The cuckoo–blackbird experiment supported the alternative hypothesis (1) that the Blackcap is capable of recognizing the cuckoo. The Blackcaps behaved adaptively—they vigorously a acked the parasite, whereas they completely ignored the innocuous blackbird. Absence of conspicuous response toward nonthreatening intruders is adaptive, because loud alarm calls could increase the conspicuousness of the nest to predators or other brood parasites and, in turn, negatively influence reproductive success of aggressive individual(s) (e.g. McLean et al. 1986, Martin et al. 2000). This second series of experiments shows that interpretation of a coevolutionary relationship between Blackcaps and cuckoos based on cuckoo–pigeon experiments was erroneous and is explainable as a methodological artifact (see below). The nonsignificant difference in responses toward the cuckoo and pigeon in the laying stage is hardly explainable by low sample size (n = 20), because (1) many studies revealed significant differences with similar or lower sample sizes (e.g. Robertson and Norman 1976, 1977; Knight and Temple 1986b; McLean 1987; Duckworth 1991; McLean and Maloney 1998); (2) cuckoo–blackbird experiments in the present study clearly show that even smaller sample size (n = 15) is sufficient to reveal significant differences in Blackcaps behavior; and (3) inclusion of data from incubation and nestling periods gave qualitatively the same results in all analyses, despite the big sample size (n = 60) for a nonparametric paired test (see above). This analysis (where data from all three nesting stages are pooled) makes good sense, because nest stage does not influence intensity of nest defense in the Blackcap (see above). The cuckoo is not only a parasite but also a predator of eggs and nestlings of small passerines (e.g. Jourdain 1925, Gärtner 1981). Thus, it would be adaptive for a host to a ack and recognize the cuckoo in all stages of nesting. E R
 
 Nest defense coupled with enemy recognition is an important strategy for hosts to avoid brood parasitism, because other strategies (egg ejection, nest desertion) may be more costly (even successful ejectors lose one or more of

537

their own eggs because laying cuckoos remove them; Davies and Brooke 1989). Results of the present study support the hypothesis that this antiparasitic adaptation evolved in the Blackcap. Thus, Blackcaps recognize both parasitic eggs (Moksnes et al. 1990) and parasitic adults as special threats. However, that ability is limited, given that Blackcaps regularly a acked the pigeon dummy. The similar responses to the pigeon and cuckoo dummies might be understood in the light of the observation of Smith and Hosking (1955) that Willow Warblers (Phylloscopus trochilus) a acked a cuckoo dummy and a cuckoo head without a body. Both the pigeon and cuckoo dummies have plain gray heads; if the gray head is a stimulus for antiparasitic aggression in Blackcaps, those hosts could easily a ack not only cuckoos but also any nonthreatening intruder with a gray head. On the other hand, the yellow eye ring and bill are probably not recognition cues for Blackcaps, given that blackbirds share those traits with cuckoos and are not a acked at all. These hypotheses require further investigation by manipulating potential recognition cues (see e.g. Gill et al. 1997b). Comparable data from Great Reed Warblers indicate that they show be er recognition of cuckoos, given that they commit much fewer recognition errors during nest defense than Blackcaps. In 10% of experiments, Great Reed Warblers mobbed or a acked an innocuous Eurasian CollaredDove (Streptopelia decaocto) control dummy (Bártol et al. 2002), whereas Blackcaps mobbed or a acked a nonthreatening pigeon mount in 90% of experiments. However, this comparison is only tentative, because the Eurasian CollaredDove and pigeon resemble the cuckoo to different degrees (at least to human eyes). To conclude that Blackcaps recognize cuckoos as a special enemy, it was necessary to use nonthreatening dummies as controls. However, almost all hosts of the cuckoo have been tested without appropriate controls (e.g. Moksnes et al. 1990, Moksnes and Røska 1988, Røska et al. 2002). Those studies provide important insights into host behavior; however, their results would have been strengthened by controlled experiments, given that some tested species may be aggressive toward brood parasites not because they co-evolved with them, but because they a ack any intruder near their nests (that possibility is supported by the fact that even species

538

T
 G

that could not have coevolved with cuckoos, because of inaccessible nests or inappropriate diet, sometimes a ack them; Moksnes et al. 1990, Røska et al. 2002). Brood parasitism is not the only force selecting for aggression against intruders; the same pressure is provided by predators (Curio et al. 1985) and competitors for food (e.g. Robinson 1992) and nest sites (e.g. Garcia and Arroyo 2002). Thus, host aggression against the parasite dummy (without a control experiment) provides only weak support for the hypothesis that brood parasitism is a force selecting for evolution of nest defense. Furthermore, some controlled studies found no differences in responses to parasite and control dummies for some species tested (e.g. Robertson and Norman 1976, 1977; Bazin and Sealy 1993; Grim and Honza 2001; Honza et al. 2004). However, the results of those studies, too, may not be conclusive—experiments with more-dissimilar control dummies (than those used in the above-mentioned papers) may show that even those hosts recognize brood parasites as special enemies, but only poorly. C
!
 
 "  P    O I  P-  A
: I 
 
 I  

 R  It is generally believed that because brood parasites pose the greatest threat to their hosts during the egg-laying period, responses to them should decrease in later stages of the nesting cycle if a host recognizes the parasite as a special enemy (e.g. Briskie and Sealy 1989, Hobson and Sealy 1989, Sealy et al. 1998). In his study of enemy recognition in Field Sparrows (Spizella pusilla), Burhans (2001) concluded that Field Sparrows probably do not recognize Brownheaded Cowbirds as a special enemy, because responses to them increased from incubation to nestling stage (though insignificantly). However, responses to a nonthreatening Fox Sparrow (Passerella iliaca) control increased between the two stages as well. More importantly, responses (frequency of alarm calling) to the control increased much more (+120%) than responses to Brown-headed Cowbirds (+20%). Thus, this reanalysis of Burhans’ (2001) data shows that the overall responsiveness of Field Sparrows clearly increased between incubation and nesting stages, regardless of the type of dummy (parasite, predator, control; see table 1 in Burhans 2001).

[Auk, Vol. 122

Importantly, if a host a acks both a parasite and innocuous intruders at a similar rate, it cannot be claimed that a acks on the parasite are the result of coevolution. Coevolution can only increase pre-existing aggressiveness; thus, a host‘s response to a parasite is not equivalent to the overall level of aggression. Rather, it is only the difference in aggression toward the parasite as compared with the host’s response to an innocuous enemy (this is analogous to the fact that the predation cost of begging is the increase in the rate of predation caused by begging, not the overall rate of predation; Haskell 1999). Therefore, the measured response to Brown-headed Cowbirds should be adjusted to the increase in general host responsiveness by subtracting the response to the Fox Sparrow control from the response to the Brown-headed Cowbird. Aer that adjustment, the results are the opposite of those reported by Burhans (2001): response to Brown-headed Cowbirds at incubation stage (70.3 alarms per 5 min) is actually higher than that at nestling stage (38.3 alarms per 5 min)—which suggests that Field Sparrows recognize the Brown-headed Cowbird as a special enemy (as also indicated by significant difference between responses to parasite and control dummies at incubation stage). In conclusion, if there was coevolution between Field Sparrows and Brown-headed Cowbirds, the result is not an overall response of the former to the la er, but only an “aggression increment” (i.e. the difference between aggressiveness to parasite and to nonthreatening intruder). A possible exception, in which the incremental increase in aggressiveness cannot explain the differential behavior of hosts to parasites and predators, is found in the Yellow Warbler (Dendroica petechia). That species preferentially uses specific alarm calls and nest-protection behavior toward Brown-headed Cowbirds (Gill and Sealy 1996, Gill et al. 1997b), and populations allopatric with Brown-headed Cowbirds do not express those behaviors (Gill and Sealy 2004). However, in other studies, no specific antiparasitic behavior (different from antipredator behavior) has been reported. W A T D  R

T"
T
 C
 
 D The discrimination threshold of any recognition system is set by a trade-off between

April 2005]

Methodology of Enemy Recognition Studies

acceptance errors (e.g. a acks on pigeon) and rejection errors (e.g. no response to cuckoo; see e.g. Sherman et al. 1997). By recognition errors, I mean a nonadaptive response to any stimulus (e.g. a ack on innocuous intruder or absence of a ack on threatening intruder). Increasing similarity of tested and control stimuli inevitably leads to a higher rate of occurrence of recognition errors (e.g. mimetic parasitic eggs elicit higher frequency of acceptance errors than nonmimetic eggs; e.g. Davies and Brooke 1989). Similarly, the pigeon (which is more similar to the cuckoo than the blackbird is, with regard to overall coloration, size, and shape) elicited a high frequency of recognition errors, whereas the blackbird (which exhibits more cues for recognition) elicited no recognition errors. In general, animals discriminate according to degree of resemblance between stimuli (i.e. there is a continuum of discrimination abilities along the continuum of resemblances; e.g. Rothstein 1982, Di rich et al. 1993, Caley and Schluter 2003). Anecdotal observations (e.g. cuckoo a acking a wild pigeon, probably mistaking it for a territorial conspecific intruder; Radford 1991) indicate that birds commit similar kinds of recognition errors under natural conditions. It is important to stress that the blackbird test is a stronger test of recognition abilities than the pigeon test, because the blackbird is less similar to the cuckoo than the pigeon is. Similarly, absence of rejection of conspecific eggs is not evidence of absence of egg recognition in a particular species (Moksnes and Røska 1992). A test with nonmimetic eggs provides much more reliable results; if a host does not reject even highly nonmimetic eggs, we can safely conclude that it has no recognition ability; if a host does not reject mimetic eggs, no firm conclusions can be drawn. On the other hand, tests with mimetic eggs can provide additional information on the quality (degree) of recognition. The same holds true for adult-parasite dummy experiments (see also Kamil 1988). To explain the existence of any behavior, the costs and benefits associated with it need to be understood. For example, acceptance of parasitic eggs could be explained by low parasitism rate, or high costs of rejection, or both (Davies et al. 1996). To understand why Blackcaps recognized some intruders (e.g. blackbird) but not others (e.g. pigeon), we would have to obtain information on costs and benefits associated with their

539

responses, the probabilities of encounters with different intruders, and the effectiveness of deterring dangerous enemies from host nests. A complicating factor is that even unsuccessful aggression against a parasite could have benefits—if birds “know” they were parasitized, they tend to reject parasite eggs more frequently (Davies and Brooke 1989, Moksnes and Røska 1989). Therefore, the effectiveness of nest defense behavior is generally hard to establish (Sealy et al. 1998). However, both mathematical models and direct measurements of breeding success of hosts showing various levels of nest defense and enemy recognition would shed more light on the issue. P
  C
 
 D: D U
 C
" ! C

S  Some authors have compared host responses to a brood parasite and a predator (e.g. Burgham and Picman 1989, Duckworth 1991, Neudorf and Sealy 1992, Soler et al. 1999). Theory predicts that responses to predators should increase during the nesting cycle (as the value of host progeny increases), whereas responses to parasites should decrease, because they pose the most threat during the early laying stage (Sealy et al. 1998). However, the observation that host responses to a parasite and predator are the same (e.g. Burhans 2001) may not necessarily mean that a host does not recognize a parasite as a specific threat. On one hand, one host species— Yellow Warbler—was reported to show a unique response (specific alarm calls and nest-protection behavior) that apparently evolved in response to brood parasitism and not nest predation (Gill and Sealy 1996, 2004). On the other hand, there is no reason to expect that hosts have to evolve some novel antiparasitic behavior—why not use old and well-established antipredator behaviors (see discussion of the evolution of novel antiparasitic responses vs. strengthening of pre-existing anti-intruder adaptations in Hosoi and Rothstein [2000])? Recognition of a parasite may merely be manifested in different pa erns of nest defense against parasites than against predators during a nesting cycle (Neudorf and Sealy 1992). However, brood parasites also prey on host nestlings (Rothstein and Robinson 1998), which makes antiparasitic aggression adaptive throughout the nesting period. That effect is

540

T
 G

probably more important in cuckoos, which are clearly predators of both eggs and nestlings (e.g. Jourdain 1925, Gärtner 1981), than in Brownheaded Cowbirds, which only sometimes prey on nestlings (see McLaren and Sealy 2000 and references therein). Thus, lack of difference in responses to parasite and predator would not provide unambiguous positive evidence for absence of recognition in a cuckoo host (e.g. Blackcaps a ack cuckoos throughout the nesting period at the same level; see above; see also Briskie and Sealy 1989, Neudorf and Sealy 1992). However, it may provide the key test in Brown-headed Cowbird hosts (Sealy et al. 1998). In other words, use of a predator mount may be informative in experiments with hosts of some parasites (e.g. Brown-headed Cowbirds), but less so in experiments with others (e.g. cuckoos). Therefore, use of innocuous species as a control should generally provide a stronger test of host enemy-recognition abilities than use of a predator mount. However, it should be noted that the actual magnitude of host-nest predation by cuckoos as compared with Brown-headed Cowbirds is not well known at present. C

 The results of the present study indicate that (1) absence of discrimination in enemyrecognition studies may reflect a methodological artifact, (2) the narrow similarity of tested and control stimuli can lead to erroneous inferences about coevolution, (3) biased results might not be avoided even by using a control, (4) inclusion of some behavioral variables into a composite measure of nest defense (PCA) can confound results, (5) the Blackcap as an effective egg rejecter is highly aggressive toward the brood parasite, and (6) the Blackcap recognizes the parasitic cuckoo as a special enemy. Further, I suggest that the use of the predator dummy is important in experiments with Brown-headed Cowbird hosts (see Sealy et al. 1998) but may be less informative with cuckoo hosts because of possible differences in pa erns of predation by the two brood parasites on hosts nests during the nesting cycle. Finally, coevolution only increases pre-existing aggression of the particular host species. Therefore, the increment analysis (testing for changes in host responses to parasites during the nesting cycle while controlling for background aggression toward control dummy) provides a be er test

[Auk, Vol. 122

of host recognition abilities than the traditional approach (when the total level of antiparasitic response is analyzed and the confounding effect of background aggression is not controlled). This sort of approach is well established in other areas of research—for example, in controlling for background nest desertion in egg rejection experiments (e.g. Grim and Honza 2001, Lahti and Lahti 2002) or controlling for background predation in studies of cost of begging (Haskell 1999). Study of background aggression requires a standard experimental approach across study species. Background aggression should also be taken into account in comparative studies of host aggression (e.g. Røska et al. 2002), because validity of their results is based on an unproved assumption that background aggression does not differ among species. The finding that an absence of discrimination may be a methodological artifact has important implications to future studies of enemy recognition. In cases where researchers find no significant differences in responses to tested and control stimuli, it would be useful to employ another control that is less similar to a brood parasite. That suggestion accords with a general rule that researchers should spread out the levels of the independent variable (e.g. a type of enemy dummy), so that effects are detected if they exist (Kamil 1988). Additional experiments can later focus more finely on a quality of recognition abilities, if they exist. Such an approach leads to more reliable results as shown by the present study. These results also point to an important yet usually overlooked problem that pervades the brood-parasite literature in general and hampers our understanding of coevolutionary interactions: the need to divide continuous variables into discrete categories (Grim 2005). In theory, parasitic eggs are “mimetic” or “non-mimetic,” hosts are “acceptors” or “rejecters,” and adult parasites are “recognized” or “not recognized” by their hosts. In reality, there is a continuum of discrimination abilities along the continuum of resemblances. The present study documents how that phenomenon may confuse experimental results and their interpretations. A
"  I thank M. Čapek, M. Honza, V. Mrlík, and P. Procházka for their help in the field during

April 2005]

Methodology of Enemy Recognition Studies

the first field season. V. Bičík and Z. Vermouzek kindly provided stuffed dummies. Earlier versions of the manuscript benefited from detailed critical suggestions by F. Alvarez, A. Moksnes, V. Pavel, V. Remeš, E. Røska, B. G. Stokke, E. Tkadlec, and K. Weidinger. I am especially indebted to S. G. Sealy and S. A. Gill for their detailed comments, which enormously improved the manuscript. I am also grateful to three anonymous referees for their constructive comments. When working on this paper, I was supported by grants from the Research Council of Norway, the Czech Ministry of Education (grant no. 153100012), and the Grant Agency of the Czech Republic (grant no. 206/03/D234). No nest was abandoned aer experiments were performed. The experiments were done under license from The Central Commission for Animal Welfare of the Czech Republic (no. 065/ 2002–V2) and in accordance with the laws and ethical guidelines of the Czech Republic. L   C  B , A. J., and T. E. M  . 2001. Host activity and the risk of nest parasitism by Brownheaded Cowbirds. Behavioral Ecology 12: 31–40. B
, I., Z. K  , C. M
 , E. R$  ,

 T. K. 2002. Responses of Great Reed Warblers to experimental brood parasitism: The effects of a cuckoo Cuculus canorus dummy and egg mimicry. Journal of Avian Biology 33:420–425. B , R. C.,  S. G. S . 1993. Experiments on the responses of a rejector species to threats of predation and cowbird parasitism. Ethology 94:326–338. B, J. V.,  S. G. S . 1989. Changes in nest defense against a brood parasite over the breeding cycle. Ethology 82:61–67. B , M. C. J.,  J. P . 1989. Effect of Brown-headed Cowbirds on the evolution of Yellow Warbler anti-parasite strategies. Animal Behaviour 38:298–308. B , D. E. 2001. Enemy recognition by Field Sparrows. Wilson Bulletin 113:189–193. C , M. J.,  D. S . 2003. Predators favour mimicry in a tropical reef fish. Proceedings of the Royal Society of London, Series B 270:667–672. C , S. 1992. The Birds of the Western Palearctic, vol. VI. Oxford University Press, Oxford.

541

C
, E., K. R , and U. Z . 1985. Brood defence in the Great Tit (Parus major): The influence of life-history and habitat. Behavioral Ecology and Sociobiology 16:273–283. D !, N. B., and M.  L. B

. 1989. An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. 1. Host egg discrimination. Journal of Animal Ecology 58:207–224. D !, N. B., M.  L. B

,  A. K . 1996. Recognition errors and probability of parasitism determine whether Reed Warblers should accept or reject mimetic cuckoo eggs. Proceedings of the Royal Society of London, Series B 263: 925–931. D , W., F. G , P. G, P. MG
,

 D. G"
. 1993. Imperfect mimicry: A pigeon’s perspective. Proceedings of the Royal Society of London, Series B 251: 195–200. D"
 , J. W. 1991. Responses of breeding Reed Warblers Acrocephalus scirpaceus to mounts of Sparrowhawk Accipiter nisus, cuckoo Cuculus canorus and jay Garrulus glandarius. Ibis 133:68–74. G  , J. T.,  B. E. A

. 2002. Intra- and interspecific agonistic behaviour in sympatric harriers during the breeding season. Animal Behaviour 64:77–84. G& , K. 1981. Das Wegnehmen von Wirtsvogeleiern durch den Kuckuck Cuculus canorus. Ornithologische Mi eilungen 33: 115–131. G, S. A., P. M. G, L. M. S ,  S. G. S . 1997a. Does nest defence deter or facilitate cowbird parasitism? A test of the nesting-cue hypothesis. Ethology 103:56–71. G, S. A., D. L. N
,  S. G. S . 1997b. Host responses to cowbirds near the nest: Cues for recognition. Animal Behaviour 53:1287–1293. G, S. A.,  S. G. S . 1996. Nest defence by Yellow Warblers: Recognition of a brood parasite and an avian nest predator. Behaviour 133:263–282. G, S. A.,  S. G. S . 2004. Functional reference in an alarm signal given during nest defence: Seet calls of Yellow Warblers denote brood-parasitic Brown-headed Cowbirds. Behavioral Ecology and Sociobiology 56: 71–80.

542

T
 G

G, T. 2005. Mimicry vs. similarity: Which resemblances between brood parasites and their hosts are mimetic and which are not? Biological Journal of the Linnean Society 84: 69–78. G, T.,  M. H
 . 2001. Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to experimental parasitism by the Common Cuckoo Cuculus canorus. Biologia, Bratislava 56:549–556. H , D. G. 1999. The effect of predation on begging-call evolution in nestling Wood Warblers. Animal Behaviour 57:893–901. H

, K. A.,  S. G. S . 1989. Responses of Yellow Warblers to the threat of cowbird parasitism. Animal Behaviour 38:510–519. H
 , M., T. G, M. Č , J., A. M
,

 E. R$  . 2004. Nest defence, enemy recognition and nest inspection behaviour of experimentally parasitised Reed Warblers Acrocephalus scirpaceus. Bird Study 51:256–263. H

, S. A.,  S. I. R
 . 2000. Nest desertion and cowbird parasitism: Evidence for evolved responses and evolutionary lag. Animal Behaviour 59:823–840. J
 , F. C. R. 1925. A study on parasitism in the cuckoos. Proceedings of the Zoological Society of London 1925:639–667. K , A. C. 1988. Experimental design in ornithology. Pages 313–346 in Current Ornithology, vol. 5 (R. F. Johnston, Ed.). Plenum Press, New York. K , R. L.,  S. A. T. 1986a. Why does intensity of avian nest defense increase during the nesting cycle? Auk 103:318–327. K , R. L.,  S. A. T. 1986b. Methodological problems in studies of avian nest defense. Animal Behaviour 34:561–566. L  , D. C.,  A. R. L  . 2002. How precise is egg discrimination in weaverbirds? Animal Behaviour 63:1135–1142. L
, A. K.,  R. J. T
 . 2000. Differences between populations of Reed Warblers in defences against brood parasitism. Behaviour 137:25–42. M 
, R. F.,  I. G. ML . 1995. Historical and experimental learned predator recognition in free-living New Zealand Robins. Animal Behaviour 50:1193–1201. M , D.,  B. J. M. S  . 1994. Response of a forest-interior songbird to

[Auk, Vol. 122

the threat of cowbird parasitism. Animal Behaviour 47:275–280. M  , T. E., J. S
,  C. M. 2000. Nest predation increases with parental activity: Separating nest site and parental activity effects. Proceedings of the Royal Society of London, Series B 267:2287–2293. ML , C. M.,  S. G. S . 2000. Are nest predation and brood parasitism correlated in Yellow Warblers? A test of the cowbird predation hypothesis. Auk 117:1056–1060. ML , I. G. 1987. Response to a dangerous enemy: Should a brood parasite be mobbed? Ethology 75:235–245. ML , I. G.,  R. F. M 
. 1998. Brood parasitism, recognition, and response: The options. Pages 255–269 in Parasitic Birds and Their Hosts (S. I. Rothstein and S. K. Robinson, Eds.). Oxford University Press, New York. ML , I. G., J. N. M. S ,  K. G. S "  . 1986. Mobbing behaviour, nest exposure, and breeding success in the American Robin. Behaviour 96:171–186. M
, A.,  E. R$  . 1988. Responses of Fieldfares Turdus pilaris and Bramblings Fringilla montifringilla to experimental parasitism by the Common Cuckoo Cuculus canorus. Ibis 130:535–539. M
, A.,  E. R$  . 1989. Adaptations of Meadow Pipits to parasitism by the Common Cuckoo. Behavioral Ecology and Sociobiology 24:25–30. M
, A.,  E. R$  . 1992. Responses of some rare cuckoo hosts to mimetic model cuckoo eggs and to foreign conspecific eggs. Ornis Scandinavica 23:17–23. M
, A.,  E. R$  . 1995. Egg-morphs and host preference in the Common Cuckoo Cuculus canorus: An analysis of cuckoo and host eggs from European museum collections. Journal of Zoology (London) 236:625–648. M
, A., E. R$  , A. T. B

, L. K
, H. M. L ,  H. C. P. 1990. Behavioural responses of potential hosts towards artificial cuckoo eggs and dummies. Behaviour 116:64–89. M
 
, R. D.,  P. J. W  . 1988. Risks and rewards of nest defense by parent birds. Quarterly Review of Biology 63:167–187. N
, D. L.,  S. G. S . 1992. Reactions of four passerine species to threats of predation and cowbird parasitism: Enemy

April 2005]

Methodology of Enemy Recognition Studies

recognition or generalized responses? Behaviour 123:84–105. P !, V.,  S. B. 2001. Offspring age and nest defence: Test of the feedback hypothesis in the Meadow Pipit. Animal Behaviour 61:297–303. R 
, A. P. 1991. Cuckoo a acking woodpigeon. British Birds 84:195. R
 
, R. J.,  R. F. N
 . 1976. Behavioral defenses to brood parasitism by potential hosts of the Brown-headed Cowbird. Condor 78:166–173. R
 
, R. J.,  R. F. N
 . 1977. The function and evolution of aggressive host behavior towards the Brown-headed Cowbird (Molothrus ater). Canadian Journal of Zoology 55:508–518. R

, D. 1992. Habitat use and foraging behavior of the Scarlet Robin and the Flame Robin at a site of breeding-season sympatry. Wildlife Research 19:377–395. R$  , E., A. M
, B. G. S
, V. B',

 C. M
 . 2002. Aggression to dummy cuckoos by potential European Cuckoo hosts. Behaviour 139:613–628. R
 , S. I. 1982. Mechanisms of avian egg recognition: Which egg parameters elicit responses by rejecter species? Behavioral Ecology and Sociobiology 11:229–239. R
 , S. I. 1990. A model system for coevolution: Avian brood parasitism. Annual Review of Ecology and Systematics 21:481–508. R
 , S. I.,  S. K. R

. 1998. Parasitic Birds and Their Hosts. Oxford University Press, New York.

543

SX , B.-E. 1990. Age-specific variation in reproductive performance of birds. Pages 251–283 in Current Ornithology, vol. 7 (D. M. Power, Ed.). Plenum Press, New York. S , S. G., D. L. N
, K. A. H

,  S. A. G. 1998. Nest defense by potential hosts of the Brown-headed Cowbird: Methodological approaches, benefits of defense, and coevolution. Pages 194–211 in Parasitic Birds and Their Hosts (S. I. Rothstein and S. K. Robinson, Eds.). Oxford University Press, New York. S , P. W., H. K. R!,  D. W. P. 1997. Recognition systems. Pages 69–96 in Behavioural Ecology (J. R. Krebs and N. B. Davies, Eds.). Blackwell Scientific, Oxford. S , J. N. M., P. A,  I. G. ML . 1984. Age, experience, and enemy recognition by wild Song Sparrows. Behavioral Ecology and Sociobiology 14:101–106. S , S.,  E. H
. 1955. Birds Fighting. Faber and Faber, London. S
, J. J., M. S
, T. P\-C
  , S. A ^,  A. P. M$. 1999. Antagonistic antiparasite defenses: Nest defense and egg rejection in the magpie host of the Great Spo ed Cuckoo. Behavioral Ecology 10:707–713. U , M. 1999. Cost of nest defense in Azurewinged Magpies. Journal of Avian Biology 30:326–328.

Associate Editor: M. Briingham

17. Grim T. & Honza M. 2001: Differences in behaviour of closely related thrushes (Turdus 17. philomelos and T. merula) to experimental parasitism by the common cuckoo Cuculus canorus. Grim T. & Honza M. 2001: Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to experimental parasitism by the common cuckoo Cuculus Biologia 56(5): 549–556. canorus. Biologia 56(5): 549–556.

Biologia, Bratislava, 56/5: 549—556, 2001

Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to experimental parasitism by the common cuckoo Cuculus canorus Tomáš Grim1 & Marcel Honza2 1

Laboratory of Ornithology, Faculty of Sciences, Palacký University, Tř. Svobody 26,CZ–77146 Olomouc, Czech Republic; e-mail: [email protected] 2 Institute of Vertebrate Biology, AS CR, Květná 8, CZ–60365 Brno, Czech Republic; e-mail: [email protected]

Grim,

T. & Honza, M., Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) towards experimental parasitism by the common cuckoo Cuculus canorus. Biologia, Bratislava, 56: 549—556, 2001; ISSN 0006-3088. The common cuckoo Cuculus canorus parasitizes many passerines, but some common species sympatric with the brood parasite are rarely used as hosts. Potential host species may escape brood parasitism using methods such as high rejection of cuckoo eggs or high aggressiveness towards female parasite. We tested the responses of two common species, the song thrush Turdus philomelos and blackbird T. merula, not regularly parasitised by the cuckoo, to artificial cuckoo eggs and dummies. Both species rejected model parasitic eggs (song thrush 58.3%, blackbird 66.7%). Song thrushes showed very low levels of aggression toward a stuffed dummy, while blackbirds were very aggressive. Neither species discriminated between the cuckoo and control pigeon dummies. We observed one case of intraspecific nest parasitism in the song thrush. This is probably the first documentation of intraspecific nest parasitism in this species. Both our and previously published data indicate that the rejection behaviour of Turdus species evolved as a defence against intraspecific nest parasitism. This behaviour contributes to cuckoos avoiding these potential host species. However, other nonexclusive factors (e.g. diet composition) could explain more fully why thrushes are not victimized by the cuckoo. Key words: brood parasitism, nest defence, mimicry, aggression, co-evolution, egg rejection.

Introduction The common cuckoo Cuculus canorus (Linnaeus, 1758) is a brood parasite whose eggs have been found in nests of more than one hundred species of small passerines (Moksnes & Rskaft, 1995). However, only five to ten host species are para-

sitised regularly (Rothstein & Robinson, 1998). Other common species of open-nesting passerines sympatric with the cuckoo (e.g. Turdus, Emberiza, Carduelis) are parasitised rarely or not at all (Moksnes & Rskaft, 1995). Although brood parasitism by the cuckoo and its North-American counterpart, the brown-

549

headed cowbird Molothrus ater (Boddaert, 1783) is currently subject to intensive research (for reviews see Ortega, 1998; Rothstein & Robinson, 1998), few authors have tried to explain low levels of parasitism in particular host species. Davies & Brooke (1989) and Moksnes & Rskaft (1992) found that several rare cuckoo hosts (e.g. the reed bunting Emberiza schoeniclus Linnaeus, 1758 and willow warbler Phylloscopus trochilus Linnaeus, 1758) show fine egg discrimination. Peer & Bollinger (1997) reported that low synchronization between breeding cycles of the host (common grackles Quiscalus quiscula Linnaeus, 1758) and parasite (brown-headed cowbirds) is responsible for low parasitism rates in common grackles. Brown-headed cowbirds avoid parasitising eastern kingbirds Tyrannus tyrannus (Linnaeus, 1758) because their eggs would be wasted – kingbirds reject almost 100% of parasitic eggs (Sealy & Bazin, 1995). Mermoz & Fernandéz (1999) explained the low frequency of parasitism in scarlet-headed blackbirds Amblyramphus holosericeus (Scopoli, 1786) by shiny cowbirds Molothrus bonariensis (Gmelin, 1789) with the fact that the scarlet-headed blackbird shows a high level of nest attentiveness and therefore the parasitic female cannot lay without being noticed and attacked by the nest owners. Several hypotheses were proposed to explain why a particular bird species is not victimized by a brood parasite including host breeding success, egg acceptance/rejection status, intensity of nest defence, host care and diet, nest type and habitat (Ortega, 1998). In this paper we test two of these hypotheses (egg rejection and nest defence) in two common passerines: the song thrush Turdus philomelos (C.L. Brehm, 1831), and blackbird T. merula (Linnaeus, 1758). Observations of parasitism in the two species are extremely rare. Lack (1963) reported only 3 parasitized nests among 22,656 blackbird nests in England. Moksnes & Rskaft (1995) found only 11 and 21 cuckoo eggs laid in the nests of the two respective species (n = 11, 870 clutches of European passerines held in museum egg-collections). In addition, none of these cuckoo eggs matched the eggs of the song thrush and blackbird in appearance. These data indicate that neither species are regularly used as fosterers by the common cuckoo. Kleven et al. (1999) found that the size of the host had a significant effect on the growth of cuckoo nestlings – cuckoo chicks cared for by larger host species, such as the great reed warbler Acrocephalus arundinaceus (Linnaeus, 1758), were significantly heavier at fledging than nestlings raised

550

by a smaller host, the reed warbler Acrocephalus scirpaceus (Hermann, 1804). In the light of this finding it is interesting that no Turdus species is regularly used as a host by the cuckoo despite the body size of these potential hosts. Therefore we made an attempt to explain why the song thrush and blackbird are not parasitised by the common cuckoo. Responses of the two species to parasitic eggs were tested only in Great Britain (Davies & Brooke, 1989) and Norway (Moksnes et al., 1990). We focused on nest defence behaviour and parasite recognition abilities of the song thrush and blackbird, which have not been studied, to date. Methods The study was conducted in a deciduous forest nearby Dolní Bojanovice village (48◦ 520 N, 17◦ 000 E), about 60 km SE of Brno, South Moravia, Czech Republic. Data were collected from 25 April to 30 June 2000 and 2001. For the egg experiments we used natural Chinese quail Coturnix chinensis (Linnaeus, 1766) eggs painted blue to mimic eggs of the cuckoo gent parasitising the redstart Phoenicurus phoenicurus (Linnaeus, 1758) (Moksnes & Rskaft, 1995). We introduced experimental eggs to “host” nests at the egg-laying or early incubation stages (from day 2 till day 6; day 0 is the day when the first egg was laid; the clutch size in both species tested is 3–5; Cramp, 1988). To minimize disturbance to the hosts, we did not remove any host eggs (as female cuckoos do) because the experimental removal of one host egg has no effect on the rejection rates of model eggs (Davies & Brooke, 1989). We experimentally parasitised 17 song thrush and 8 blackbird nests. The egg was considered accepted if it remained in the nest for six days. This criterion is used because almost all observed rejections in different host species appeared before the sixth day after the egg was introduced, see e.g. Moksnes et al. (1990); Davies & Brooke (1989) reported that most rejections took place within 3 days after clutch completion. Nest defence by hosts was tested with stuffed dummies of the common cuckoo and the feral pigeon Columba livia f. domestica (Gmelin, 1789) as a control. We chose the pigeon as a control species because it is about the same size and shape as a cuckoo but provides no threat to the either species tested. The experimental design followed the standard procedure suggested by Sealy et al. (1998). First, one of dummies (in a life-like position) was attached to a branch about 0.5 m from a focal nest. The head of the dummy was directed to the nest. After the first parent appeared near the nest and became aware of the dummy, reactions of nest owners were observed for 5 min from a hide set up 20 m from the focal nest. Presentation of the second mount at the same place was separated by a 30 min interval to avoid habituation or carry-over aggression (Sealy et al., 1998). The order in which models

Table 1. The outcome of the observed nesting attempts and results of model egg experiments at song thrush and blackbird nests. There were no significant interspecific differences in all measured outcomes (chi-square and Fisher’s exact probabilities tests, all n.s.). Song thrush (n = 55)

Blackbird (n = 25)

Outcome

Predated Experimental egg accepted Experimental egg ejected Nest deserted

Control

Experiment

Control

Experiment

27/38 – – 0/38

5/17 5/12 4/12 3/12

12/17 – – 0/17

2/8 2/6 3/6 1/6

Note: some of successful experimental nests were predated after the egg experiments finished, therefore the breeding success is actually lower than is shown in the table as noted in the results.

were presented was randomized. All experiments with stuffed dummies were performed by one author (T.G.) to avoid possible observer bias. The number of nests tested was 15 for the song thrush and 6 for the blackbird. In the song thrush 6 nests were tested at the egg stage and 9 at the nestling stage, while all blackbird nests were tested at the nestling stage. The intensity of nest defence varied from quiet watching of the nest from a distance to vigorous mobbing of the dummy. We adjusted our categorization of host responses to natural variation in responses observed during our experiments. Blackbirds usually quickly attacked the mount directly (the mount was immediately removed to avoid its destruction), so we did not quantify host reactions as a number of particular behaviours per 5 min of observation. Instead, we quantified the behaviour of both blackbirds and song thrushes on relative subjective scales (see e.g. Mermoz & Fernández, 1999). We recorded the behavioural variables suggested by Sealy et al. (1998) to enable interspecific comparisons (see e.g. Gill et al., 1997). We measured the delay in arrival of nest owners from the moment the dummy was attached near the focal nest (latency of response) in minutes, which should reflect the general level of nest attentiveness. Latency in response can be taken as a rough measure of time the parents spent at the nest which is crucial with respect to interaction with the brood parasitic cuckoo which lays extremely quickly to avoid host attacks. In addition we quantified several parameters of host reactions: vocalizations (0 to 3, i.e. from no vocalizations to very strong and permanent vocalizations), contact attacks (0 = none, 1 = one or more), overall level of response (0 = no response, i.e. silent watching of a dummy, 1 = few vocalizations, bird(s) usually more than 10 m from the dummy, 2 = more vocalizations, bird usually less than 10 m from the dummy, 3 = strong vocalizations and close passes, 4 = strong vocalizations and contact attacks). These parameters should reflect the risks taken by birds when defending their nests. Each nest was tested only once to avoid pseudoreplication.

Results Egg discrimination Altogether we observed 66 song thrush and 31 blackbird nesting attempts (11 song thrush and 6 blackbird nests were not followed to fledging and their final fate is unknown). The proportions of successful breeding attempts (nests with a known fate) of song thrushes (41.8%, n = 55) and blackbirds (44.0%, n = 25) were not statistically significantly different (χ2 = 0.0003, d.f. = 1, P = 0.85). The main reason for nesting failure was the predation of clutch (Tab. 1). We experimentally parasitised 17 song thrush and 8 blackbird nests and only 12 and 6 artificially parasitised nests survived the 6 days acceptance/rejection criterion, for each species respectively. Song thrushes rejected 58.3% and blackbirds 66.7% of parasitic eggs (Tab. 1). Rejection methods used by both species were ejection of the parasite egg and desertion of the parasitised nest (Tab. 1). Blackbirds rejected parasitic eggs more quickly than song thrushes (medians: 1 and 4 days; Mann-Whitney test, Z = 2.038, P = 0.040, n = 7). At three predated song thrush nests the parasitic eggs were accepted for 1, 2 and 4 days before predation of the clutch (the remaining 2 nests were predated before the first check). At predated blackbird nests the parasitic eggs remained in the nests for 1 and 3 days before being predated. One song thrush and one blackbird deserted their experimentally parasitised nests. Both desertions occurred less than three days after the parasitic egg was added. This behaviour is probably a response to parasitism as no control nests (which were regularly checked but not used in experiments) were deserted. We observed one case of intraspecific nest parasitism in the song thrush. The parasitic egg

551

Table 2. Intraspecific comparison of intensity of song thrush and blackbird responses to cuckoo and pigeon dummies. Values are medians. There were no differences in responses to the two types of dummies in both host species in any of the measured behavioural parameters (Wilcoxon matched pairs tests, all n.s.). Latency of response was measured in minutes, other parameters measured on ordinal scale (see Methods). Song thrush (n = 15)

Blackbird (n = 6)

Parameter

Latency of response Vocalizations Contacts Overall level of nest defence

Cuckoo

Pigeon

Cuckoo

Pigeon

7 1 0 1

8 1 0 1

12 2 1 4

6 1 1 4

Table 3. Interspecific comparison of intensity of song thrush and blackbird responses to stuffed dummies. Values are medians. The results of Mann-Whitney tests are shown. Latency of response was measured in minutes, other parameters measured on ordinal scale (see Methods). Differences which remained significant after sequential Bonferroni test (Rice, 1989) are indicated with an asterisk (P < 0.05). Parameter

Latency of response Vocalizations Contacts Overall level of nest defence

Song thrush (n = 15)

Blackbird (n = 6)

Z

P

7 1 0 1

9 2 1 4

1.686 2.352 2.335 2.182

n.s. * * *

was laid seven days after the host clutch completion (the appearance of eggs both before and after the host’s laying period is considered as an indication of brood parasitism, see e.g. Ringsby et al., 1993). Five days later the parasitic egg was destroyed due to partial predation of the clutch. As far as we know, this is the first observation of intraspecific brood parasitism in the song thrush. Nest defence We found no differences in song thrush responses to mounts between the egg (n = 6) and nestling (n = 9) stage in latency of response, vocalizations, contacts and overall level of nest defence (MannWhitney tests, P > 0.05 in all cases). Therefore the data were pooled. Neither song thrushes or blackbirds distinguished between the cuckoo and the control species – there were no differences between the responses of either species to either type of dummy in any of the measured behavioural parameters (Wilcoxon matched pairs tests, all n.s.; Tab. 2). There was no relationship between the latency of response and overall level of nest defence (song thrush: rs = −0.355, P = 0.194, n = 15; blackbird: rs = 0.655, P = 0.158, n = 6). The number of attacking individuals had no effect on the overall level of aggression in the song thrush

552

(Mann-Whitney test, Z = 1.687, P = 0.092, n = 15). Interestingly, during five experiments on blackbirds the overall level of aggression was highest (level 4) and both parents attacked a dummy together in all five cases. In one experiment only the female was present showing no aggression at all. The male was never observed at this nest. This female also accepted the experimental blue egg, which remained in the nest for 15 days and was not ejected even after nestlings hatched. There were no significant differences in the latency of response between species (Tab. 3). However, the number of individuals performing nest defence behaviour differed between species – in the song thrush usually only one parent responded, but in the blackbird both male and female attacked dummies (Fisher’s exact probabilities test, P = 0.046). Moreover, blackbirds vocalized more strongly, attacked the mount directly with higher a probability and showed higher overall level of nest defence than song thrushes (Tab. 3). Song thrushes responded very weakly to dummies. They usually stayed more than 10 meters from the nest uttering very few vocalizations about 20 times per 5 min observation period. Comparable data for blackbirds are lacking because blackbirds uttered alarm calls very quickly and almost continu-

ously, moreover, the nest owners immediately attacked the mount directly and therefore the experiments were stopped earlier thus making sensible interspecific comparison of call rates impossible (see Methods). Only two studied song thrushes responded extremely aggressively and one even showed redirected aggression towards its own nest (this unusual observation is described in detail by Grim, 2000). On the other hand, blackbirds behaved very aggressively – they continually and intensively vocalized and mobbed dummies vigorously. In conclusion, the low level of song thrush nest defence would present almost no risk to a female cuckoo. On the other hand, the extremely vigorous nest defence behaviour shown by blackbirds would threaten a female cuckoo’s life. Three song thrush and three blackbird nests were tested with both parasitic eggs and dummies. Dummy experiments were performed after the egg experiments finished. Parents in four nests accepted parasitic eggs. It might be expected that acceptors are naive breeders (Lotem et al., 1992) that would also defend their nests poorly. However, intensity of response to mounts was at levels 1 and 2 for the two song thrush nests and 0 and 4 for the two blackbird nests. Parasitic eggs were rejected at two other nests. Intensity of response to dummies was at level 1 for one song thrush nest and 4 for the blackbird nest. Our data sets are too small to be analysed statistically, however, they show some inconsistency in the host responses to eggs versus dummies.

Discussion Egg discrimination The breeding success of the study species found during our study seasons (song thrush: 41.8%, blackbird: 44.0%) is similar to that found in previous studies (e.g. Osborne & Osborne, 1980; Hatchwell et al., 1996). It is within the range of breeding success of bird species commonly used as hosts by the cuckoo. Therefore the level of breeding success in the song thrush and blackbird probably has no effect on the avoidance of these species by the cuckoo. The results of the egg experiments are similar to those reported by Davies & Brooke (1989). They found that song thrushes rejected 27.3% and blackbirds 59.1% of non-mimetic redstart type eggs (i.e. the same type as we used in our study). Overall levels of rejection of several types of alien eggs by song thrushes and blackbirds were 58.5% and 61.8% respectively (Davies

& Brooke,

1989). These values are higher than the rejection rates of commonly parasitised hosts (Davies & Brooke, 1989). In a study by Moksnes et al. (1990) 20% of song thrush pairs rejected a model cuckoo egg and two blackbirds also rejected them. Four experimentally parasitised nests in our study were deserted. We consider this as a method of rejection because Davies & Brooke (1989) showed that desertion can serve as a method of rejection – hosts tested by them deserted nests parasitised with non-mimetic eggs more frequently then nests parasitised with mimetic eggs. The cuckoo probably does not use some hosts (e.g. the reed bunting and spotted flycatcher Muscicapa striata Pallas 1764) because they show high rejection rates of parasitic eggs (Davies & Brooke, 1989). However, great reed warblers Acrocephalus arundinaceus in Hungary reject 39% of cuckoo eggs and are still heavily parasitised (Moskát & Honza, 2000). Other commonly used hosts are also rejecters (Davies & Brooke, 1989). Our data and the results of Davies & Brooke (1989) and Moksnes et al. (1990) show that both the song thrush and blackbird can recognize and reject parasitic eggs. However, these studies also indicate that the egg rejection behaviour of the song thrush and blackbird cannot alone explain the low level of parasitism in these species. We documented one case of intraspecific nest parasitism in the song thrush. We found no other report on intraspecific nest parasitism by the song thrush in the literature. However, Grendstad et al. (1999) reported that one of the studied redwing Turdus iliacus (Linnaeus, 1766) nests was parasitized by the closely related fieldfare Turdus pilaris (Linnaeus, 1758). Interestingly, Ringsby et al. (1993) detected a relatively high intraspecific parasitism rate in fieldfares (11.5%). Redwings discriminate against intraspecific eggs introduced to their nests and show stronger aggression toward conspecific compared to the female cuckoo. This strongly suggests that redwings evolved this behaviour as a defence against intraspecific and not interspecific nest parasitism ( Grendstad et al., 1999). Nest defence Blackbirds displayed aggressive nest defence behaviour and readily attacked dummies with contact. The blackbird is generally very aggressive and intraspecific fights can lead to death (Cramp, 1988). Its parental anti-predatory behaviour is also reported to be very intense (Cramp, 1988). The song thrush is similarly described as aggressive by

553

Cramp

(1988), however, in our study only two of 15 tested pairs attacked the dummy directly and the overall level of aggressiveness was very low compared to the blackbird. We found that both studied species did not discriminate between the cuckoo and control pigeon mounts. This result indicates that the behaviour of song thrushes and blackbirds was a result of generalized nest defence and not a response to a specific threat provided by the parasite. However, Duckworth (1991) found that reed warblers Acrocephalus scirpaceus (commonly parasitised by the cuckoo throughout their range) can differentiate cuckoos from similar sparrowhawks Accipiter nisus (Linnaeus, 1758). The reed warbler’s response was more aggressive to a cuckoo than to a sparrowhawk during incubation, but the reaction to the cuckoo disappeared after fledging although the parents still responded strongly to a sparrowhawk at that stage. Therefore our results indicate that there probably was no coevolutionary arms-race between the cuckoo and Turdus species. This hypothesis is supported by the fact that none of the cuckoo eggs found in song thrush and blackbird nests were mimicking host species eggs (Moksnes & Rskaft, 1995). Fieldfares show aggression towards a hooded crow Corvus corone (Linnaeus, 1758) dummy (Meilvang et al., 1997) but they do not consider the cuckoo to be a potential threat (Moksnes & Rskaft, 1988). Grendstad et al. (1999) also concluded that redwings evolved alien egg rejection as a defence against intraspecific brood parasitism. We observed one case of intraspecific nest parasitism in the song thrush which also gives support to the hypothesis that Turdus species have not been in the co-evolutionary arms race with the cuckoo. The overall level of nest defence by song thrushes was very low. Thus, the song thrush would not prevent a female cuckoo trying to lay an egg in a song thrush nest. However, the very high intensity of blackbird aggression suggests that the cuckoo would risk serious injury if attacked at a blackbird nest during the parasitism act. Molnár (1944) reported several observations of dead female cuckoos under great reed warbler nests. Females were evidently killed by the nest owners. The blackbird is a much larger bird than the great reed warbler (100 g vs 30 g), thus, a female cuckoo would probably risk much more at a blackbird than warbler nest. The poor level of nest defence shown by the song thrush and strong aggression shown by blackbirds in our study are consistent with the results of dummy experiments performed

554

by V. Bièík in Czech Republic (unpublished data; in verb). Mermoz & Fernández (1999) described a low rate of parasitism in scarlet-headed blackbirds Amblyramphus holosericeus with a non-specific life-history trait, i.e. high levels of nest attentiveness combined with aggressive host behaviour against all intruders. Scarlet-headed blackbirds do not recognize the shiny cowbird Molothrus bonariensis as a specific threat, but 98% of time the nest is guarded by at least one parent. The high level of nest attentiveness cannot play an important role in the cuckoo’s avoidance of blackbirds and song thrushes as these species arrive at their nest with long delays (Tabs 1, 2), i.e. nest attentiveness is very poor and the probability of encountering a brood parasite at the host nest is consequently minimal. We tested whether the cuckoo avoids parasitising song thrush and blackbird nests because these species have low breeding success, strong egg rejection behaviour or intensive nest defence. We found that both species discriminated parasitic non-mimetic eggs. Moreover, blackbirds, although they do not recognize the female cuckoo as a specific threat (they attacked the pigeon with the same vigour), have a high level of generalized nest defence that could threaten the female cuckoo. However, both these behaviours are shared with commonly parasitised hosts (e.g. Phylloscopus warblers or the great reed warbler are also strongly aggressive to the cuckoo). Thus, these factors cannot alone explain why cuckoos avoid parasitising them. Ortega (1998) presents 12 hypotheses trying to explain the absence of parasitism in some passerine species. It is unlikely that most of these could explain the avoidance of Turdus species specifically (e.g. short host incubation period, insufficient amount of parental care, unsuitable habitat, well concealed hosts nests). However, other hypotheses are more likely to explain the absence of parasitism in thrushes. Moksnes & Rskaft (1988) hypothesized that Turdus nests could be too deep for a cuckoo nestling to evict host eggs or nestlings. However, Moksnes et al. (1990) reported an observation of a four-day old cuckoo chick ejecting a fieldfare nestling weighing 10.0 g. Turdus nestlings grow very quickly (Cramp, 1988), thus cuckoo nestling could have big problems competing with host chicks. In addition, cuckoos and cowbirds can live only on an insectivorous diet. Their breeding success in nests of granivorous birds is very low – diet composition sufficiently explains the absence of parasitism in

e.g. greenfinch Carduelis chloris (Linnaeus, 1758), linnet C. cannabina (Linnaeus, 1758) and bullfinch Pyrrhula pyrrhula (Linnaeus, 1758) (Davies & Brooke, 1989). The song thrush and blackbird feed their nestlings mainly on molluscs and earthworms respectively (Cramp, 1988). These food items are probably indigestible for the cuckoo nestling that must be brought up on insects. However, this hypothesis could be rigorously tested only by cross-fostering experiments, i.e. transferring nestling cuckoos to Turdus nests. In a preliminary experiment it was found that a cuckoo nestling did not survive in a blackbird nest. In conclusion, our results suggest that the intensity of antiparasitic behaviour shown by song thrushes probably has a low effect on brood parasite choice. However, strong generalized nest defence by blackbirds probably constrains the use of this species by the cuckoo, although the low level of nest attentiveness in this species reduces the benefits of strong nest defence. Nevertheless, other hypotheses (food, nest type) need testing before decisive conclusions can be drawn. Our observations, coupled with previous findings, support the hypothesis that Turdus species have evolved rejection behaviour against intraspecific nest parasitism. Acknowledgements We are grateful to M. Èapek, V. Mrlík and P. Procházka for their help during the field work. Suggestions by two anonymous referees substantially improved the MS. The research was supported by Palacký University internal grant (no. 3210–3013) operated to T. Grim and grant 206/00/P046 operated to M. Honza. References

Cramp,

S. (ed.) 1988. The Birds of the Western Palearctic. Vol. V. Oxford University Press, Oxford & New York, 1063 pp. Davies, N. B. & Brooke, M. L. 1989. An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. I. Hosts egg discrimination. J. Anim. Ecol. 58: 207–224. Duckworth, J. W. 1991. Responses of breeding reed warblers Acrocephalus scirpaceus to mounts of sparrowhawk Accipiter nisus, cuckoo Cuculus canorus and jay Garrulus glandarius. Ibis 133: 68– 74.

Gill, S. A., Grieef, P. M., Staib, L. M. & Sealy, S. G. 1997. Does nest defence deter or facilitate cowbird parasitism? A test of the nesting-cue hypothesis. Ethology 103: 56–71.

Grendstad, L. C., Moksnes, A. & Rskaft, E. 1999. Do strategies against conspecific brood parasitism occur in redwings Turdus iliacus? Ardea 87: 101–111. Grim, T. 2000. An interesting observation of redirected activity in the song thrush (Turdus philomelos). Sylvia 36(2): 161–163.

Hatchwell, B. J., Chamberlain, D. E. & Perrins, C. M. 1996. The reproductive success of blackbirds Turdus merula in relation to habitat structure and choice of nest site. Ibis 138: 256– 262.

Kleven, O., Moksnes, A., Rskaft, E. & Honza, M. 1999. Host species affects the growth rate of cuckoo (Cuculus canorus) chicks. Behav. Ecol. Sociobiol. 47: 41–46. Lack, D. 1963. Cuckoo hosts in England. Bird Study 10: 185–201. Lotem, A., Nakamura, H. & Zahavi, A. 1992. Rejection of cuckoo eggs in relation to host age: a possible evolutionary equilibrium. Behav. Ecol. 3: 128–132. Meilvang, D., Moksnes, A. & Rskaft, E. 1997. Nest predation, nesting characteristics and nest defence behaviour of fieldfares and redwings. J. Avian Biol. 28: 331–337. Mermoz, M. E. & Fernández, G. J. 1999. Low frequency of shiny cowbird parasitism on scarletheaded blackbirds: anti-parasite adaptations or nonspecific host life-history traits? J. Avian Biol. 30: 15–22. Moksnes, A. & Rskaft, E. 1988. Response of fieldfares Turdus pilaris and bramblings Fringilla montifringilla to experimental parasitism by the cuckoo Cuculus canorus. Ibis 130: 535–539. Moksnes, A. & Rskaft, E. 1992. Responses of some rare cuckoo hosts to mimetic cuckoo eggs and to foreign conspecific eggs. Ornis. Scand. 23: 17– 23. Moksnes, A. & Rskaft, E., 1995. Egg-morphs and host preference in the common cuckoo (Cuculus canorus): an analysis of cuckoo and host eggs from European museum collections. J. Zool. 236: 625– 648.

Moksnes, A., Rskaft, E., Braa, A. T., Korsnes, L., Lampe, H. M. & Pedersen, H. C. 1990. Be-

havioural responses of potential hosts towards artificial cuckoo eggs and dummies. Behaviour 116: 64–89. Molnár, B. 1944. Cuckoo in the Hungarian Plain. Aquila 51: 100–112. Moskát, C. & Honza, M. 2000. Effect of nest and nest site characteristics on the risk of cuckoo Cuculus canorus parasitism in the great reed warbler Acrocephalus arundinaceus. Ecography 23: 335– 341. Ortega, C. 1998. Cowbirds and other brood parasites. The University of Arizona Press, Tucson, 374 pp. Osborne, P. & Osborne, L. 1980. The contribution of nest site characteristics to breeding-success

555

among blackbirds Turdus merula. Ibis 122: 512– 517. Peer, B. D. & Bollinger, E. K. 1997. Explanations for the infrequent cowbird parasitism on common grackles. Condor 99: 151–161. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43: 223–225.

Ringsby, T. H., Moksnes, A., Rskaft, E. & Lerkelund, H. E. 1993. Do conspecific brood parasitism and antiparasite strategies occur in fieldfares Turdus pilaris? Fauna norv., Ser. C, Cinclus 16: 45–53. Rothstein, S. I. & Robinson, S. K. (eds) 1998. Parasitic birds and their hosts. Studies in coevolution. Oxford University Press, New York & Oxford, 444 pp.

556

Sealy, S. G. & Bazin, R. C. 1995. Low-frequency of observed cowbird parasitism on eastern kingbirds – host rejection, effective nest defense, or parasite avoidance? Behav. Ecol. 6: 140–145.

Sealy, S. G., Neudorf, D. L., Hobson, K. A. & Gill, S. A. 1998. Nest defence by potentional

hosts of the brown-headed cowbird: methodological approaches, benefits of nest defense, and coevolution. pp. 194–211. In: Rothstein, S. I. & Robinson, S. K. (eds) Parasitic birds and their hosts, Studies in coevolution, Oxford University Press, New York & Oxford. Received November 14, 2000 Accepted June 6, 2001

18. Honza M., Grim T., Čapek M., Moksnes A. & Røskaft E. 2004: Nest defence, enemy 18. recognition and nest inspection behaviour of experimentally parasitised reed warblers Acrocephalus scirpaceus. Honza M., Grim T., Čapek M., Moksnes A. & Røskaft E. 2004: Nest defence, enemy recognition and nest inspection behaviour of experimentally parasitised reed warblers Bird Study 51(3): 256–263. Acrocephalus scirpaceus. Bird Study 51(3): 256–263.

Bird Study (2004) 51, 256–263

Nest defence, enemy recognition and nest inspection behaviour of experimentally parasitized Reed Warblers Acrocephalus scirpaceus MARCEL HONZA1*, TOMÁS GRIM2, MIROSLAV CAPEK JR1, ARNE MOKSNES3 and EIVIN RØSKAFT3 1Department

of Avian Ecology, Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic, Kvetná 8, 603 65 Brno, Czech Republic, 2Laboratory of Ornithology, Palacky University, Tr. Svobody 26, 771 46 Olomouc, Czech Republic and 3Department of Biology, Norwegian University of Science and Technology, NTNU, N-7491 Trondheim, Norway

Capsule Reed Warblers in a regularly parasitized population do not recognize Cuckoo Cuculus canorus as a special enemy and do not change their behaviour at nest immediately after being parasitized. Aims To assess if an intruder near the nest influences the behaviour of the Cuckoo host. Methods Host responses to Cuckoo, control Pigeon dummies and human intruder were observed. Host behaviour at 71 nests was video-recorded for 30 minutes at four experimental groups of nests: Cuckoo dummy, Cuckoo dummy + Cuckoo egg, Pigeon dummy, human intruder. Results Reed Warblers did not respond differently to the Cuckoo and the control species. The experimental procedure had no significant effect on the behaviour of hosts during the study period. We were unable to find any differences in the time spent at the nest, clutch inspection behaviour and nest defence behaviour between morning and afternoon experimental groups. Our results do not support the hypothesis that afternoon laying by the Cuckoo is maintained by a selection pressure from the host. We observed no ejection or egg-pecking during the 30-min period after the experimental parasitism. Conclusions Low aggression and non-specificity of host responses in our study area are in line with the fact that the Reed Warbler is an intermediate rejecter of Cuckoo eggs as expected from the spatial habitat structure hypothesis.

One important factor affecting reproductive success and consequently nest defence behaviour in many passerine birds is brood parasitism (Rothstein 1990), because the successful act of parasitism often reduces host fitness dramatically. Clearly, the best protection for a host against parasitism is to avoid being parasitized. Hosts can avoid parasitism by breeding in safe sites (Alvarez 1993, Øien et al. 1996, Honza et al. 1998, Moskát & Honza 2000, Clarke et al. 2001) or by a vigorous nest defence (Moksnes et al. 1990, Sealy et al. 1998, Grim & Honza 2001, Røskaft et al. 2002b). There is a huge interspecific variation in defensive behaviour of potential host species against territory intruders. Some host species discriminate between the parasite and control species that pose no threat to them (Burgham & Picman 1989, Moksnes et al. 1993a) and some hosts also respond differently to brood *Correspondence author. Email: [email protected]

© 2004 British Trust for Ornithology

parasites and predators (Duckworth 1991, Gill & Sealy 1996). Cuckoo Cuculus canorus hosts that reject at very high frequencies show little variation in their defence behaviour (Øien et al. 1999). Variation in the aggressive response of a host to the parasite is influenced by various factors, e.g. by an occurrence of Cuckoos in the particular locality (hosts breeding in sympatry with the parasite are more aggressive than those breeding in allopatry; Røskaft et al. 2002b). Habitat selection by the host also plays an important role – host species breeding near trees only (Cuckoo vantage points) are more aggressive against a Cuckoo dummy and reject parasitic eggs at higher rates than host species breeding both near and further away from trees (Røskaft et al. 2002b, 2002c) probably because of gene flow between parasitized (near trees) and unparasitized (farther away from trees) populations. In the case where a host is not successful in deterring

Behaviour at nest of parasitized Reed Warbler

a brood parasite, it is adaptive to recognize and reject the parasitic egg (Rothstein 1990). There are great differences in reactions of various hosts towards parasitic eggs. Cuckoo hosts normally exhibit some delay in their response towards the parasitic egg (Davies & Brooke 1988, Moksnes et al. 1990, Amundsen et al. 2002), but Moksnes et al. (1994) and Soler et al. (2003) documented relatively short times to ejection in Chaffinches Fringilla coelebs, Blackcaps Sylvia atricapilla, Sub-Alpine Warblers S. cantillans and Blackbirds Turdus merula. Only a few studies have paid any attention to host behaviour immediately after the act of experimental Cuckoo parasitism: Moksnes et al. (1994) observed rejection behaviour of Chaffinches and Blackcaps and Martín-Vivaldi et al. (2002) video-recorded behaviour of three potential Cuckoo hosts to determine the effort needed to puncture experimentally added parasitic eggs (see also Soler et al. 2003). To our knowledge only one study has focused on host clutch inspection behaviour after the presentation of a stuffed Cuckoo dummy (Moksnes et al. 1993a). On the other hand several studies have been conducted on hosts of Brown-headed Cowbird Molothrus ater immediately after the act of parasitism (Rothstein 1977, Briskie & Sealy 1987, Sealy & Neudorf 1995, Sealy 1996, Sealy & Lorenzana 1998). Most of these studies did not, however, examine host behaviour in detail. More importantly, in all these studies (except Moksnes et al. 1993a) only experimentally parasitized nests were observed and there were no observations of unparasitized control nests. This makes a general interpretation of the results difficult because of problems of separating specific responses to parasitism and general nest defence behaviour. The Reed Warbler Acrocephalus scirpaceus is one of the most frequently used Cuckoo hosts in Europe (Moksnes & Røskaft 1995). However, the Reed Warbler is not very aggressive towards the Cuckoo (Duckworth 1991, Lindholm & Thomas 2000, Røskaft et al. 2002a, 2002b). We investigated whether host behaviour at the nest is influenced by the type of intruder (parasitic Cuckoo versus non-threatening intruder) during the presence of the intruder and also after it leaves the vicinity of the nest. As a more detailed knowledge of changes in all aspects of host behaviour during the course of a day is essential to understand parasitic adaptations of the Cuckoo (see also Moksnes et al. 2000) we performed experiments both in the morning and in the afternoon. We report behaviour of Reed Warblers in detail with respect to

257

the time of day and the kind of intruder, immediately after nests have been tested experimentally. We made the following predictions. (1) If hosts can distinguish between their enemies and non-enemies, they should respond with higher intensity of nest defence to a brood parasite than to control dummies of a non-threatening species. Nest defence could be performed either as aggression (see Moksnes et al. 1990) or sitting in the nest to prevent parasites access to the nest (see Hobson & Sealy 1989). We predicted that Reed Warblers would show a higher level of nest defence towards a Cuckoo dummy than to a nonthreatening species. (2) The presence of a stuffed Cuckoo dummy in close proximity to the nest is reported to facilitate host egg discrimination behaviour (Davies & Brooke 1988, Moksnes & Røskaft 1989, Moksnes et al. 2000), i.e. a Cuckoo dummy increases the likelihood that a host will reject a Cuckoo egg. We therefore predicted that hosts should show a higher intensity of clutch inspection behaviour after being faced with a Cuckoo dummy compared with a nonthreatening control species. (3) The Cuckoo, unlike its hosts, lays its eggs in the afternoon (Wyllie 1981; but see Honza et al. 2002). Davies & Brooke (1988) suggested that this behaviour has evolved because the Cuckoo has a lower probability of encountering hosts during the parasitism act in the afternoon. Such behaviour should be adaptive because a physical presence of the host at its nest can serve as an effective defence against successful parasitism (Hobson & Sealy 1989; however, Moksnes et al. 2000 found no support for this hypothesis). Alternative explanations for the unique laying pattern of the Cuckoo are that hosts spend more time inspecting their clutches in the morning than in the afternoon (Davies & Brooke 1988) or they are more aggressive in the morning. MATERIALS AND METHODS Study area and fieldwork

The study sites were two pond systems in the southeastern part of the Czech Republic near Lednice (48°48′N, 16°48′E) and Luzice (48°51′N, 17°04′E). The two systems are 25 km apart. We searched systematically for Reed Warbler nests in the littoral vegetation during the breeding periods between 15 May and 30 June in 1997 and 1998. We located nests in vegetation consisting mostly of Common Reed Phragmites australis and to a lesser extent of Reedmace Typha angustifolia surrounding the ponds. The fish © 2004 British Trust for Ornithology, Bird Study,

51, 256–263

258

M. Honza et al.

ponds are situated in flat agricultural lowland landscape and are mostly surrounded by deciduous woods (see Hudec 1975 and Moksnes et al. 1993b, for more detailed descriptions of the study area). The frequency of Cuckoo parasitism in Reed Warbler nests in our study areas is 15.0% (Øien et al. 1998) and Reed Warblers reject 37.5% of parasitic eggs (29.9% by desertion, 7.6% by ejection; Øien et al. 1998). Experimental procedure and video recordings

We conducted experiments on nests during the egglaying period (after the host had laid three or four eggs) to mimic a natural Cuckoo laying. A total of 71 nests was tested out of which 35 were tested during the morning (05:00–08:00 hours Central European Time, CET) and 36 during the afternoon (15:00–20:00 hours CET). To standardize our procedure, experiments were not conducted during rainy or windy days. We established four experimental groups. (1) Cuckoo dummy group (n = 30). The Cuckoo dummy was presented 1 m from the nest. (2) Cuckoo dummy + Cuckoo egg group (n = 11). After the presentation of the Cuckoo dummy, the nest was experimentally parasitized with a real Cuckoo egg from the study area resembling the eggs of the Great Reed Warbler Acrocephalus arundinaceus (the egg was removed after the video recording was finished, see later). (3) Pigeon Columba livia f. domestica group (n = 16). The nest owner’s reactions were tested with a stuffed Pigeon as a control non-threatening species. (4) Human intruder group (n = 14). One researcher (M.H.) as a control only visited the nest. Each nest was tested only once (i.e. one type of experiment) to avoid pseudoreplication. To reduce the number of confounding variables (size, plumage colour, shape of a bird), we used Pigeon as the control species because its size and overall colour (grey) is similar to Cuckoo and it poses no threat to Reed Warblers (see also Sealy et al. 1998). Some researchers have used control species that are familiar to tested hosts (Moksnes & Røskaft 1989, Grim & Honza 2001), while others performed experiments with model species not occurring in the study area (Bazin & Sealy 1993). The Pigeon lives in sympatry with the Cuckoo in the nearest vicinity of reedbeds in our study area, therefore Reed Warblers probably have had a chance to gain prior experience with this species. Nevertheless, Sealy et al. (1998) suggested that prior experience (or its absence) with a control species should have no effect. The human intruder experiment was used to ascertain if responses to dummies are specific reactions to bird intruders or a © 2004 British Trust for Ornithology, Bird Study,

51, 256–263

general response to any form of nest disturbance. A Sony video camera (CCD-TR 660E Hi 8) was mounted on a stand and placed at a distance of 3–4 m and levelled 0.4 m above the nests. The camera was powered with a 12-volt car battery and provided with a shelter painted dark green to be inconspicuous. Since all the nests were situated above water, the battery was placed on the shore and connected to the camera with a cable. Reed Warblers are known to be tolerant to human presence (Lindholm & Thomas 2000). They start to feed their broods almost immediately after cameras or hides are placed near their nests (latency to start of feeding for cameras = 5.00 ± 0.99 min (mean ± sd), n = 5; latency to start of feeding for hides = 3.08 ± 2.10 min, n = 13) and the age of brood does not influence this latency (r = –0.04, P = 0.90, n = 13; T. Grim, M. Honza, B. Matysioková & K. Voslajerová unpubl. data). Furthermore, feeding frequencies observed at the nest immediately after placing cameras or hides are the same as those reported by Kilner et al. (1999) who left cameras near the nest for several hours for birds to habituate. Thus, the effect of cameras on Reed Warbler behaviour is undetectable and it is highly unlikely that it could confound our results (moreover, we were interested in among-group differences and all groups were treated identically with cameras). Also the behaviour of the closely related Great Reed Warbler is not significantly affected by a video camera near the nest (M. Honza & C. Moskát, unpubl. data). Finally, responses of Reed Warblers towards a stuffed Cuckoo in the current study were almost the same as those in a previous study (Røskaft et al. 2002a) where host behaviour was directly observed without cameras. Nevertheless, we left the place for 1 hour allowing the birds to habituate to the set-up. After 1 hour, we presented the Cuckoo or stuffed Pigeon dummy mounted on a wooden pole 1 m from the nest. The dummies were in perching position and at the same height as the focal nest. The behaviour of the nest owners was observed during a 5-min period from a distance of 10 m. We recorded latency time to arrival, latency time to alarm calling from the first arrival and time spent within 1 m of the Cuckoo/Pigeon dummy. The overall level of nest defence was rated on an ordinal scale: 0 = no response (no bird arrived during the 5-min period); 1 = silent watching; 2 = mobbing (i.e. flights around the dummy and alarm calls); 3 = contact attack(s); see Moksnes et al. (1990) for the description of these behaviours. All observations were made by one researcher (M.H.) to avoid possible observer bias.

Behaviour at nest of parasitized Reed Warbler

After presentation of the dummy and clutch manipulation (see above), the dummy was removed and the camera was switched on. Nests were videotaped for 30 minutes. In the case of the human intruder group, we followed a similar pattern. The camera was set up 1 hour before an observer approached the nest. The observer stood near the nest for 2 min (almost all the pairs responded up to this time). After this period the camera was switched on and the observer left. The videotapes were analysed in the laboratory. From the tapes we recorded the arrival time (time to first sight of the nest owner/s), brooding time (time from when the camera was switched on until the host started brooding), look-1 (time spent with nest inspection behaviour, from the moment when the bird first looked into the nest until the moment when it first sat on the clutch) and look-2 (total time spent inspecting the nest after the bird sat on it. Look-1 is not included in look-2. Nest inspection behaviour is defined as the bird looking at the clutch. The number of arrivals at the nest (this variable should reflect the amount of host activity at the nest) and nest time (total amount of time spent at the nest) were also recorded. RESULTS Nest defence and enemy recognition (prediction 1)

There were no statistically significant differences in any of the behavioural variables when Reed Warblers were facing Cuckoo or Pigeon dummies (Table 1). Absence of significant discrimination was not caused by absence of nest defence – some Reed Warblers responded with alarm calling; however, only two individuals attacked the Cuckoo dummy. Pigeons were never attacked. The difference in the number of experiments with attacks on Cuckoo and Pigeon dummies was, however, not significant (Fisher’s exact probabilities test, ns).

259

Table 1. Responses of Reed Warblers to a Cuckoo dummy and a stuffed Pigeon. Values are medians (results of Mann–Whitney tests are shown). Nest defence: 0 = no response, 1 = silent watching, 2 = mobbing, 3 = contact attack (see Methods for further explanation).

Variable Latency to first response (min) Latency to alarm calling after first response (min) Time spent within 1m from the nest (s) Median level of nest defence

Cuckoo (n = 41)

Pigeon (n = 16)

U

P

3.0

2.0

1.35

0.17

1.2

0.2

1.57

0.12

8.5 1

42.0 1

1.69 1.56

0.09 0.12

Clutch inspection behaviour (prediction 2)

The experimental procedure had no significant effect on Reed Warbler’s behaviour during the 30-min period immediately after the experiment (Kruskal–Wallis ANOVA tests, P > 0.10 in all cases; Table 2). After seeing the Cuckoo dummy at the nest, Reed Warblers did not tend to arrive at their nests significantly earlier, they did not spend significantly more time with nest inspection behaviour and their activity was not significantly different from those who saw the Pigeon dummy near their nests. The addition of a Cuckoo egg to the host nest had no effect on Reed Warbler behaviour. There was no significant difference in the way a human intruder affected host behaviour compared to dummies (Table 2). Only six Reed Warbler pairs did not visit their nests during the 30-min period (two in Cuckoo – morning, two in Cuckoo – afternoon, one in Cuckoo + egg replacement – morning, one in Pigeon – afternoon). We observed no ejection or egg pecking of the Cuckoo egg during the 30-min observation period. Effect of time of day on host behaviour (prediction 3)

Reed Warblers generally tended to spend slightly more

Table 2. Behaviour of Reed Warblers during the first 30-min period following the dummy experiments. Results are means ± sd. See Methods for explanations of the measured variables.

Variable Arrival time (s) Brooding time (s) Look-1 (s) Look-2 (s) No. of arrivals Nest time (min)

Cuckoo (n = 30) 341 353 4.7 9.9 2.5 18.5

± ± ± ± ± ±

369 377 5.5 11.2 1.4 8.4

Cuckoo + egg (n = 11) 379 389 10.4 9.1 3.0 17.8

± ± ± ± ± ±

510 508 14.1 3.8 1.9 10.4

Pigeon (n = 16) 424 388 6.9 5.5 2.1 18.5

± ± ± ± ± ±

507 502 8.0 4.2 1.0 10.1

Human (n = 14) 170 189 4.4 5.7 2.0 22.4

± ± ± ± ± ±

152 156 6.6 6.0 0.7 6.5

Kruskal–Wallis test (H)

P

2.47 1.98 2.98 5.45 1.59 2.96

0.48 0.57 0.39 0.14 0.66 0.41

© 2004 British Trust for Ornithology, Bird Study,

51, 256–263

260

M. Honza et al.

time at the nest in the morning (70.4% of the observation time) than in the afternoon (56.9%), although the differences were not statistically significant (Table 3). Furthermore, we found no significant differences in clutch inspection behaviour by Reed Warblers between the morning and afternoon experimental groups (Table 3). Finally, the intensity of nest defence was almost identical between the morning and afternoon (Table 3). Interestingly, host responses were highly variable among individuals. Individuals that quickly responded to the dummies also quickly arrived at their nests during the following 30-min period (rs = 0.404, n = 41, P = 0.009). Furthermore, Reed Warblers who arrived at their nests quickly reacted more immediately with alarm calling, while those individuals arriving later in response to the dummy tended to delay their alarm calling response (rs = 0.955, n = 41, P < 0.001). DISCUSSION Nest defence and enemy recognition (prediction 1)

Our prediction 1 of a more intense nest defence behaviour against the Cuckoo dummy than against the Pigeon dummy was not supported. In general, the level of nest defence by Reed Warblers in our study area was relatively low, only 4.9% of pairs attacked the Cuckoo dummy which is in accordance with previous findings of Røskaft et al. (2002a, 2002b). A low level of aggression provides potential for errors in enemy recognition which could explain why responses to the Cuckoo and Pigeon were not significantly different.

Absence of specific recognition of the Cuckoo could be explained by the fact that the Reed Warbler is an intermediate rejecter (rejection rate of natural Cuckoo eggs is 37.5% in our study areas; Øien et al. 1998) – egg rejection and aggression against adult parasites evolve in concert, i.e. acceptors are less aggressive than rejecters (Røskaft et al. 2002b). Thus, intermediate rejecters could be expected to show low levels of aggression and also poorer abilities to recognize adult parasites. Our results are in line with this general relationship between parasitic egg and adult parasite related adaptations. Low level of aggression and poor enemy recognition in the Reed Warbler are also expected from the spatial habitat structure hypothesis (Røskaft et al. 2002c) – species with some populations breeding in the vicinity of trees (Cuckoo observation points) and some breeding away from trees generally have lower egg rejection rates (Røskaft et al. 2002c) and less aggressive response to the parasite (Røskaft et al. 2002c) than species always breeding close to Cuckoo perches in trees. In the former (including Reed Warbler) there is lower selection pressure for antiparasitic adaptations (including specific enemy recognition) than in the latter. Thus, our results (low aggression, poor enemy recognition) accord with the spatial habitat structure hypothesis. Despite being regularly parasitized (Moksnes et al. 1993b, Øien et al. 1998) Reed Warblers in our study area behaved similarly to individuals from unparasitized populations in Britain (Lindholm & Thomas 2000; see also Røskaft et al. 2002). Duckworth (1991) found that Reed Warblers in his study area in England recognized

Table 3. Effect of time of day on Reed Warbler behaviour at their nests during the first 30-min period following the dummy experiment. The data presented are medians (results of Mann–Whitney tests are shown). Nest defence was rated on an ordinal scale (see Table 1). Sample sizes for morning/afternoon experiments are: Cuckoo 14/16, Cuckoo + egg 7/4, Pigeon 9/7, human 5/9. Variable Time spent at the nest (% of total observation)

Nest inspection = Look-1 (s)

Nest defence (0–3)

Experimental procedure

Morning

Afternoon

U

P

Cuckoo Cuckoo + egg Pigeon Human Mean Cuckoo Cuckoo + egg Pigeon Human Mean Cuckoo Cuckoo + egg Pigeon Human Total

70.8 72.3 66.7 71.7 70.4 2.5 3.0 4.5 1.0 3.0 1.5 2 1 – 1

54.5 40.2 56.5 76.4 56.9 2.0 7.0 3.0 3.0 3.0 1 1 1 – 1

1.15 0.96 0.89 0.33 1.29 0.21 0.54 0.93 0.82 0.24 0.55 0.93 0.00 – 0.58

0.25 0.34 0.37 0.74 0.20 0.83 0.59 0.35 0.41 0.81 0.58 0.35 1.00 – 0.56

© 2004 British Trust for Ornithology, Bird Study,

51, 256–263

Behaviour at nest of parasitized Reed Warbler

the Cuckoo as a special enemy while our results did not support the hypothesis. However, caution is needed in this comparison because the authors of the two studies did not follow the same experimental procedure – we presented dummies 1 m from the focal nest while Duckworth (1991) placed dummies at two distances, either directly at the nest or 3 m from the nest. He found that Reed Warblers responded much more aggressively to the dummy on the nest compared to one 3 m from the nest (see also Røskaft et al. 2002). More importantly, Duckworth (1991) observed significant differences in Reed Warbler responses towards different mounts only when dummies were placed directly at the nest – there were no significant differences when dummies were placed 3 m from focal nests. Nest inspection behaviour (prediction 2)

We predicted that the host would show higher nest inspection activity after being exposed to a Cuckoo dummy than to a Pigeon dummy. We obtained no support for this prediction because Reed Warblers did not modify their behaviour in respect to the kind of intruder (Cuckoo, Pigeon or human). Including a Cuckoo egg in the nest in addition to the Cuckoo dummy did not influence the host behaviour. This could indicate that the sight of the Cuckoo at the nest was not a strong enough cue to indicate parasitism and thus release defence behaviour immediately after the act of parasitism. Moksnes et al. (1993a) have reported similar results for Meadow Pipit Anthus pratensis where the presence or absence of Cuckoo female and/or egg dummies had no significant effect on incubation and nest checking behaviours. An explanation may be that the Reed Warbler needs time to discover parasitism. Like many other species it usually rejects parasitic eggs several days after being parasitized (Davies & Brooke 1988, Moksnes et al. 1990, Grim & Honza 2001, Amundsen et al. 2002). On the other hand the Meadow Pipit, another frequently used Cuckoo host which also is an intermediate rejecter, deserts its nest very quickly (sometimes within a few minutes after the act of parasitism) and always within 24 h after experimental manipulation (Moksnes et al. 1993a). However, even in this quickly rejecting species there were no differences in the behaviour of individuals who could see a Cuckoo dummy at their nests and those who could not (Moksnes et al. 1993a). Davies & Brooke (1988) showed that a Cuckoo mount increases the probability that the Reed Warbler will reject the parasitic egg. Our observations indicate

261

that this effect is not detectable immediately after the parasitism act. Why the Cuckoo lays in the afternoon (prediction 3)

Reed Warblers did not spend more time at their nests in the morning compared with the afternoon, which gives no support for prediction 3. Davies & Brooke (1988) measured the temperature of Reed Warbler eggs in the morning and in the evening and found morning broods to be significantly warmer and concluded that Reed Warblers spent more time at their nests in the morning. However, our observations are in accordance with those obtained by Moksnes et al. (2000) who found no difference between the length of time that Reed Warblers spent at their nests in the forenoon and afternoon. Therefore, the absence of changes in nest attentiveness during the day does not support the view that Cuckoo afternoon laying has evolved because the host is away from its nest more during this part of the day. Cuckoo afternoon laying could also result from a higher intensity of clutch inspection behaviour and nest defence by hosts in the morning. Our study does not support either of these behaviours because Reed Warblers showed no differences in clutch inspection and nest defence behaviours during the day. Absence of immediate ejections

We did not observe any pecking or ejections during the 30-min period after we added a parasitic egg to a focal nest. We used natural Cuckoo eggs which in our study area do not correspond very well to Reed Warbler eggs as judged by the human eye (Edvardsen et al. 2001). However, Reed Warblers in our study area accept 62.5% of these natural parasitic eggs (Øien et al. 1998) and even highly non-mimetic eggs are frequently accepted (43.7%; Stokke et al. 1999). Low aggression and absence of specific responses to the Cuckoo both during dummy experiments and in the 30-min period thereafter could result from the fact that the arms-race between the Cuckoo and Acrocephalus warblers in our study area is at a relatively early stage as indicated by host acceptance of badly matching Cuckoo eggs and a low match between parasitic and host eggs (Edvardsen et al. 2001). Honza et al. (2001) hypothesized that the fact that 86.3% of Cuckoo eggs laid in Reed Warbler nests belong to the Sylvia egg morph could be indicative of host switching due to construction of fish ponds in Moravia in the © 2004 British Trust for Ornithology, Bird Study,

51, 256–263

262

M. Honza et al.

16th century. The low level of host defences in our study area could represent an example of evolutionary lag. Alternatively, this finding could be explained by gene flow between parasitized and unparasitized populations, see Røskaft et al. (2002). To explain poor Reed Warbler antiparasitic defences in our study area it is helpful to consider other sympatric hosts. The Great Reed Warbler is currently parasitized more frequently than the Reed Warbler (Kleven et al. 1999). Some 30 years ago Reed Warbler and Great Reed Warbler were equally common (Hudec 1975). At present the Great Reed Warbler is almost absent from the study area (but is still preferred as a host). If in the past Cuckoos also laid their eggs preferentially in the nests of the Great Reed Warbler then it is possible that the selection for specific host adaptations in the Reed Warbler was weak before the recent decline of the Great Reed Warbler in the study area. In conclusion, low level and specificity of host responses after presentation of stuffed dummies in our study area could be expected from the fact that the Reed Warbler is an intermediate rejecter of Cuckoo eggs (Røskaft et al. 2002b). This could result from a relatively short co-evolution between host and parasite (see also Edvardsen et al. 2001, Honza et al. 2001) and is in line with the spatial habitat structure hypothesis (Røskaft et al. 2002c). ACKNOWLEDGEMENTS We are grateful to E. Tkadlec for helpful comments on the manuscript. We thank O. Kleven, O. Mikulica, I. Øien and G. Rudolfsen for their help in the field. Financial support was received from the Grant Agency of the Czech Republic (grant 206/00/P046 and A6093203) and Ministry of Education grants VS-96019 and MSM 153100012. This study has been carried out under permission given to M.H. and in accordance with the laws and ethical guidelines of the Czech Republic.

REFERENCES Alvarez, F. 1993. Proximity of trees facilitates parasitism by Cuckoos Cuculus canorus on Rufous Warblers Cercotrichas galactotes. Ibis 135: 331. Amundsen, T., Brobakken, P.T., Moksnes, A. & Røskaft, E. 2002. Rejection of Cuckoo eggs in relation to female age in the bluethroat. J. Avian Biol. 33: 366–370. Bazin, R.C. & Sealy, S.G. 1993. Experiments on the responses of a rejecter species to threats of predation and Cowbird parasitism. Ethology 94: 326–338. Briskie, J.V. & Sealy, S.G. 1987. Responses of Least Flycatchers to

© 2004 British Trust for Ornithology, Bird Study,

51, 256–263

experimental inter- and intraspecific brood parasitism. Condor 89: 899–901. Burgham, M.C.J. & Picman, J. 1989. Effect of Brown-headed Cowbirds on the evolution of Yellow Warbler anti-parasite strategies. Anim. Behav. 38: 298–308. Clarke, A.L., Øien, I.J., Honza, M., Moksnes, A. & Røskaft, E. 2001. Factors affecting Reed Warbler risk of brood parasitism by the Common Cuckoo. Auk 118: 534–538. Davies, N.B. & Brooke, M.L. 1988. Cuckoos versus Reed Warblers: adaptations and counter-adaptations. Anim. Behav. 36: 262–284. Duckworth, J.W. 1991. Responses of breeding Reed Warblers Acrocephalus scirpaceus to mounts of Sparrowhawk Accipiter nisus, Cuckoo Cuculus canorus and Jay Garrulus glandarius. Ibis 133: 68–74. Edvardsen, E., Moksnes, A., Røskaft, E, Øien, I.J. & Honza, M. 2001. Egg mimicry in Cuckoos parasitizing four sympatric species of Acrocephalus warblers. Condor 103: 829–837. Gill, S.A. & Sealy, S.G. 1996. Yellow Warbler nest defence: recognition of a brood parasite and an avian nest predator. Behaviour 133: 263–282. Grim, T. & Honza, M. 2001. Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to experimental parasitism by the Common Cuckoo Cuculus canorus. Biologia 56: 549–556. Hobson, K.A. & Sealy, S.G. 1989. Responses of yellow warblers to the threat of Cowbird parasitism. Anim. Behav. 38: 510–519. Honza, M., Øien, I.J., Moksnes, A. & Røskaft, E. 1998. Survival of Reed Warbler Acrocephalus scirpaceus clutches in relation to nest position. Bird Study 45: 104–108. Honza, M., Moksnes A., Røskaft E. & Stokke, B.G. 2001. How are different Common Cuckoo Cuculus canorus egg morphs maintained? An evaluation of different hypotheses. Ardea 89: 341–352. Honza, M., Taborsky, B., Taborsky, M., Teuschl, Y., Vogl, W., Moksnes, E. & Røskaft, E. 2002. Behaviour of female common Cuckoos, Cuculus canorus, in the vicinity of host nests before and during egg laying: a radiotelemetry study. Anim. Behav. 64: 861–868. Hudec, K. 1975. Density and breeding of birds in the Reed swamps of Southern Moravian ponds. Acta Sci. Nat. Brno 9(6): 1–40. Kilner, R.M., Noble, D.G. & Davies, N.B. 1999. Signals of need in parent-offspring communication and their exploitation by the Common Cuckoo. Nature 397: 667–672. Kleven, O., Moksnes, A., Røskaft, E. & Honza, M. 1999. Host species affects the growth rate of Cuckoo (Cuculus canorus) chicks. Behav. Ecol. Sociobiol. 47: 41–46. Lindholm, A.K. & Thomas, R.J. 2000. Differences between populations of Reed Warblers in defences against brood parasitism. Behaviour 137: 25–42. Martín-Vivaldi, M., Soler, M. & Møller, A.P. 2002. Unrealistically high costs of rejecting artificial model eggs in Cuckoo Cuculus canorus hosts. J. Avian Biol. 33: 295–301. Moksnes, A. & Røskaft, E. 1989. Adaptations of Meadow Pipits to parasitism by the Common Cuckoo. Behav. Ecol. Sociobiol. 24: 25–30. Moksnes, A. & Røskaft, E. 1995. Egg-morphs and host preference in the Common Cuckoo (Cuculus canorus): an analysis of Cuckoo and host eggs from European museum collections. J. Zool. Lond. 236: 625–648. Moksnes, A., Røskaft, E., Braa, A.T., Korsnes, L., Lampe, H.T. & Pedersen, H.C. 1990. Behavioural responses of potential hosts towards artificial Cuckoo eggs and dummies. Behaviour 116: 64–89.

Behaviour at nest of parasitized Reed Warbler

Moksnes, A., Røskaft, E. & Korsnes, L. 1993a. Rejection of Cuckoo (Cuculus canorus) eggs by Meadow Pipits (Anthus pratensis). Behav. Ecol. 4: 120–127. Moksnes, A., Røskaft, E., Bicík, V., Honza, M. & Øien, I.J. 1993b. Cuckoo Cuculus canorus parasitism on Acrocephalus warblers in Southern Moravia in the Czech Republic. J. Ornithol. 134: 425–434. Moksnes, A., Røskaft, E. & Solli, M.M. 1994. Documenting puncture ejection of parasitic eggs by Chaffinches Fringilla coelebs and Blackcaps Sylvia atricapilla. Fauna Norv. Ser. C Cinclus 17: 115–118. Moksnes, A., Røskaft, E., Hagen, G.L., Honza, M., Mørk, C. & Olsen, P.H. 2000. Common Cuckoo Cuculus canorus and host behaviour at Reed Warbler Acrocephalus scirpaceus nests. Ibis 142: 247–258. Moskát, C. & Honza, M. 2000. Effect of nest and nest site characteristics on the risk of Cuckoo Cuculus canorus parasitism in the Great Reed Warbler Acrocephalus arundinaceus. Ecography 23: 335–341. Øien, I.J., Honza, M., Moksnes, A. & Røskaft, E. 1996. The risk of parasitism in relation to distance from Reed Warbler nests to Cuckoo perches. J. Anim. Ecol. 65: 147–153. Øien, I.J., Moksnes, A., Røskaft, E. & Honza, M. 1998. Costs of Cuckoo Cuculus canorus parasitism to Reed Warblers Acrocephalus scirpaceus. J. Avian Biol. 29: 209–215. Øien, I.J., Moksnes, A., Røskaft, E., Edvardsen, E., Honza, M., Kleven, O. & Rudolfsen, G. 1999. Conditional host responses to Cuckoo Cuculus canorus parasitism. In Adams, N.J. & Slotow, R.H. (eds) Proc. 22nd Int. Ornithol. Congr. University of Natal, Durban, South Africa: 3125–3145. Røskaft, E., Moksnes, A., Meilvang, D., Bicík, V., Jemelíková, J. & Honza, M. 2002a. No evidence for recognition errors in Acrocephalus warblers. J. Avian Biol. 33: 31–38. Røskaft, E., Moksnes, A., Stokke, B.G., Bicík, V. & Moskát,

263

C. 2002b. Aggression to dummy Cuckoos by potential European Cuckoo hosts. Behaviour 139: 613–628. Røskaft, E., Moksnes, A., Stokke, B.G., Moskát, C. & Honza, M. 2002c. The spatial habitat structure hypothesis of host populations explains the pattern of rejection behavior in hosts and parasitic adaptations in Cuckoos. Behav. Ecol. 13: 163–168. Rothstein, S.I. 1977. Cowbird parasitism and egg recognition in the northern oriole. Wilson Bull. 89: 21–32. Rothstein, S.I. 1990. A model system for coevolution: avian brood parasitism. Annu. Rev. Ecol. Syst. 21: 481–508. Sealy, S.G. 1996. Evolution of host defences against brood parasitism: implications of puncture-ejection by a small passerine. Auk 113: 346–355. Sealy, S.G. & Lorenzana, J.C. 1998. Yellow Warblers (Dendroica petechia) do not recognize their own eggs. Bird Behav. 12: 57–66. Sealy, S.G. & Neudorf, D.L. 1995. Male Northern Orioles eject Cowbird eggs: implications for the evolution of rejection behavior. Condor 97: 369–375. Sealy, S.G., Neudorf, D.L., Hobson, K.A. & Gill, S.A. 1998. Nest defense by potential hosts of the Brown-headed Cowbird: methodological approaches, benefits of defense, and coevolution. In Rothstein, S.I. & Robinson, S. Parasitic Birds and Their Hosts. Studies in Coevolution: 194–211. Oxford University Press, New York & Oxford. Soler, M., Martin-Vivaldi, M. & Perez-Contreras, T. 2003. Identification of the sex responsible for recognition and the method of ejection of parasitic eggs in some potential common Cuckoo hosts. Ethology 108: 1–10. Stokke, B.G., Moksnes, A., Røskaft, E., Rudolfsen, G. & Honza, M. 1999. Rejection of artificial cuckoo (Cuculus canorus) eggs in relation to variation in egg appearance among reed warblers (Acrocephalus scirpaceus). Proc. R. Soc. Lond. B 266: 1483–1488. Wyllie, I. 1981. The Cuckoo. London, Batsford.

(MS received 27 February 2003; revised MS accepted 20 November 2003)

© 2004 British Trust for Ornithology, Bird Study,

51, 256–263

19. Grim T.: Does conspicuous nest defence by the blackcap attract nest defence “helpers”? 19. (subm.). Grim T.: Does conspicuous nest defence by the blackcap attract nest defence “helpers”? (subm.).

Does conspicuous nest defence attract nest defence “helpers”? Tomáš Grim1 Department of Zoology Palacký University tr. Svobody 26 CZ-771 46 Olomouc Czech Republic e-mail: [email protected] 1

Abstract One hypothesized function of conspicuous mobbing of predators by bird nest owners is to attract third-party predators (predator attraction hypothesis) or neighbouring birds (calling for help hypothesis). These may help the nest owners by distracting and/or attacking the mobbed intruder near the nest. I experimentally tested these hypotheses in the blackcap Sylvia atricapilla, a small passerine with a highly aggressive and conspicuous nest defence behaviour. I elicited blackcaps’ aggressive responses by presenting stuffed dummies of the brood parasitic common cuckoo Cuculus canorus and controls near their nests (n = 75 nests). At 32% nests blackcap’s responses to dummies attracted birds from 21 passerine species (at 68% nests no neighbours appeared during trials). Up to 15 individuals were attracted per experiment. No potential predators of the cuckoo were attracted despite living in the study area, thus rejecting the predator attraction hypothesis. Most of the attracted birds were heterospecifics and rarely participated in mobbing, thus the calling for help hypothesis was not supported as well. The number of attracted birds was a positive function of the owner’s intensity of nest defence, the most likely cue for attraction being vocal signals (rates of alarm calling) but not visual cues (rates of attacks). Suitable and unsuitable cuckoo hosts did not differ in their behaviour in the vicinity of defended nests. Based on discussion of possible costs and benefits for both blackcaps and attracted neighbours I conclude that the observed pattern of the positive correlation between the intensity of nest defence and the number of attracted birds is most likely a proximate by-product of the conspicuous nest defence by blackcaps and is selectively neutral for the study species.

1

INTRODUCTION Nest defence is an important component of bird behavioural strategies for successful reproduction. Montgomery & Weatherhead (1988) reviewed multitude of life-history factors and other variables that may elucidate the large observed intra-specific variance in the intensity of nest defence among birds, e.g. the reproductive value of the nest, parent sex, the type of intruder etc. Generally, across this variance the parents should try to deter a predator or a brood parasite from destroying their reproductive output. However, a conspicuous mobbing (e.g. loud alarm calls) may be a signal not only to the mobbed predator but also to third-parties – neighbouring birds (Rohwer et al. 1976) as well as additional predators (Högstedt 1983). Under the former scenario (Rohwer et al. 1976), alarm calls may serve as a request for help from kin or other conspecifics and heterospecifics. The “calling for help” hypothesis predicts that attracted birds should actively help to mob the predator by alarm calling (to confuse it) or even attacking it (to drive it away). Under this hypothesis it may also be expected that conspecifics will numerically dominate among attracted birds in semi-colonial breeders (Krams et al. 2006) and in species with high breeding densities (Pavel & Bures 2001) and will be more common than expected by chance from their abundance. Alternatively or additionally, one of the possible functions of mobbing is the attraction of a second predator which is more dangerous to a mobbed primary intruder than to an individual defending its nest (Bourne 1977). This may be especially important when there is a high discrepancy between a size, armament and the physical capability of the defending parent and the dangerous primary intruder (Krams & Krama 2002). Under the “predator attraction” hypothesis there should be a positive correlation between the intensity of the nest defence (mainly its vocal component) and the number of attracted predators. Both conspecifics (Pavel & Bures 2001) and heterospecifics (Hurd 1996) are known to be attracted by alarm calling birds. Although the attraction of neighbours may benefit nest owners (see above), it also has associated costs (e.g. an increased risk of nest predation, Krama and Krams 2005). Both the predator attraction and calling for help hypotheses were originally formulated in the context of predation on distress calling birds (Rohwer et al. 1976, Bourne 1977) they equally apply in the context of avian nest defence (Pavel & Bures 2001). The majority of avian nest defence studies focused on nest owners (Montgomerie & Weatherhead 1988), whereas other birds attracted by them and their behaviour received marginal attention (Winkler 1995, Krams & Krama 2002, Krams et al. 2006). Therefore I tested the predictions of the predator attraction and calling for help hypotheses experimentally in a small European passerine, the blackcap Sylvia atricapilla, which shows highly aggressive and conspicuous nest defence (Grim 2005). I provoked blackcap’s aggressive responses by presenting stuffed dummies of the common cuckoo Cuculus canorus near their nests in an area where the parasite is found commonly in sympatry with the study species. The cuckoo is known to have parasitized the blackcap previously (Moksnes & Røskaft 1995). The current absence of the cuckoo parasitism in the blackcap may be explained by strong evolved antiparasitic defences by this host against both parasitic eggs (Honza et al. 2004) and adults (Grim 2005). The blackcap is substantially smaller than the cuckoo (~20 vs. ~120 g) and nests at high densities in Central European forests (personal observations). Therefore, this species seems to be a good model to test both the predator attraction (see the body size discrepancy) and calling for help hypotheses (see the high breeding densities). I tested the host’s responses throughout laying, incubation and nestling periods. The cuckoo is a threat to blackcaps not only during laying and early incubation stages when successful parasitism is possible but also at the stage of late incubation and nestlings as the cuckoo is known to predate both eggs and nestlings at unparasitized nests (e.g. Milburn 1915, Headley & Jourdain 1919, Jourdain 1925, Marchant 1972, Wyllie 1975, Gärtner 1981). In line with this the intensity of nest defence against the cuckoo by blackcaps does not change during the nesting cycle despite the blackcaps’ ability to recognize the cuckoo as a special threat (see discussion in Grim 2005). I predicted that more aggressive and conspicuous nest defence should attract more neighbours than less aggressive response to dummies. Further, attracted birds should be predators potentially dangerous to the cuckoo (predator attraction hypothesis) and/or neighbouring conspecifics or heterospecifics (calling for help hypothesis). Under both hypotheses these attracted birds should actively mob the dummy near the nest.

2

METHODS I studied blackcaps’ behaviour in the south-eastern part of the Czech Republic in a deciduous forest in the vicinity of Dolní Bojanovice village (48° 52' N, 17° 00' E), about 60 km south-east of Brno. Fieldwork was done from late April to late June 2000 and 2001. I generally followed the experimental procedure suggested by Sealy et al. (1998) except for the use of the predator dummy (for explanation and discussion see Grim 2005). During experiments the dummy stuffed in a life-like position was attached to a branch about 0.5m from each nest, level with it and facing the nest rim. Responses of nest owners were observed for 5 minutes (from the moment when the first parent arrived) from a cryptic blind set up at least 15 m from the nest. I tested each host nest (n = 75) with both the cuckoo and a control dummy (pigeon Columba livia or blackbird Turdus merula; see Grim 2005) and randomised the order of dummies. There was no significant effect of the order of presentation on any of the studied behavioural variables (Grim 2005) including the number of attracted birds (U36,39 = 1.20, P = 0.23). Therefore I pooled the data from the first and second presentations of the cuckoo dummy. Moreover, when I used only data from the first experiment at each nest I obtained qualitatively identical results. For detailed description of dummies, experimental procedures and the rationale for the selection of controls see Grim (2005). To assess the intensity of nest defence I used both ordinal scales (see e.g. Grim & Honza 2001, Pavel & Bures 2001) and composite measures of parental behaviour (e.g. Olendorf & Robinson 2000, Krist 2004). First, I categorised behaviours according to relative scales. I scored the intensity of vocalizations on an ordinal scale from 0 to 2: 0 = no vocalizations, 1 = overall time spent calling < 3 min., 2 = calling > 3 min. Attacks were scored from 0 to 2: 0 = no attacks, 1 = < 5 attacks, 2 = > 5 attacks). I also noted delay in arrival of nest owner(s) in minutes and the number of individuals that responded. The total level of nest defence was measured in two ways. First, I ranked the total level of nest defence on a scale depending on the risk taken by tested bird(s) (see Montgomerie & Weatherhead 1988) from 0 to 2: 0 = silent watching of a dummy or only few vocalizations, bird(s) > 5 m from the dummy, 1 = more vocalizations, bird(s) < 5m m from the dummy, and close passes (i.e. mobbing), 2 = frequent vocalizations and attacks. Second, I created a composite measure of nest defence by means of principal component analysis (PCA) on alarm calls, attacks and the number of nest owners responding (see Grim 2005). PC1 had Eigenvalue 1.70 and was strongly positively correlated with all included variables (P < 0.0001; for details see Grim 2005). I recorded all other birds attracted by blackcap responses to dummies and also their behaviour as a dichotomous variable: silent watching of the dummy and aggressive blackcaps vs. participation in the nest defence by means of mobbing. Mobbing included alarm vocalizations (passive defence) and close passes and dives (active defence, sensu Winkler 1995). A substantial part of attracted birds appeared only in control experiments (see difference between the cuckoo and all trials data sets in Table 1). Therefore, in the descriptive part of the study I pooled data from the cuckoo and control experiments within each nest to increase sample sizes. Thus, there was no pseudoreplication as each nest was treated as a single data point. When testing for the possible effects of host behaviour on the number of attracted birds (log-transformed data) I used both ordinal (the index of nest defence intensity, alarms, and contacts) and continual (PC1) measures of host responses. To increase sample size and thus the power of the test I also fitted general linear mixed models (PROC MIXED in SAS; normal error distribution, parameters estimated by REML, degrees of freedom calculated using Kenward-Roger method) with the PC1 as predictor, the log-transformed number of attracted birds as a response and the brood identity as a random effect. Blackcaps responded to the pigeon dummy similarly to the cuckoo dummy but totally ignored the blackbird dummy (Grim 2005). Therefore, the inclusion of the pigeon and blackbird dummy data in the last analysis provides a stronger test of the hypotheses under study as it spreads out the level of the independent variable (i.e. increases variation in blackcaps’ nest defence behaviour; Kamil 1988). There are various factors that may influence nest defence intensity and thus confound results of nest defence studies (reviewed in Montgomerie & Weatherhead 1988).

3

However, in the current study I was not interested in the effects of e.g. season, reproductive value of the nest or number of previous visits on the intensity of nest defence but on the effect of the nest owners’ responses to intruders on other birds in the vicinity of the defended nest. Thus, my aim was to study the second link in the chain “independent variables → the intensity of nest defence → the number and behaviour of the attracted birds” but not the first one (this link was studied separately, see Grim 2005). In the previous study, I established that the behaviour of blackcaps was independent of the reproductive value of the nest (product of the age of clutch/brood and the size of clutch/brood), the time of season, the time of day, the number of previous visits to the nest and nest concealment (Grim 2005). In addition I found that the main pattern observed in this study, an increase in the number of attracted birds with the intensity of reaction (PC1), remained significant in a general linear mixed model (GLMM) with explanatory variables including the dummy type (cuckoo, pigeon, blackbird), the rank of presentation (first vs. second), dummy*rank interaction, breeding stage (laying, incubation, nestlings), dummy*stage interaction, date in season, reproductive value and nest concealment (poor, average, excellent) as fixed effects and nest identity as a random effect, even when all these non-significant terms were kept conservatively in the model (the effect of PC1 in the full model: F1,121 = 4.61, P = 0.034). The final model, which was selected according to corrected Akaike’s information criterion (AICc), contained only PC1 (the effect of PC1 in the final reduced model: F1,144 = 10.33, P = 0.0016). This analysis substantiates the exclusion of these non-significant factors from further analyses (Results). I also tested for relationships between abundance and frequency (proportion of experiments where birds appeared) of attracted birds during experiments and their respective availability, i.e. breeding densities (pairs/ha). Breeding densities for 9 most common species breeding in the study area were kindly provided by M. Capek. Finally, I tested whether there are any differences in abundance, frequency and participation in mobbing of attracted birds and their suitability as cuckoo hosts. Suitability of observed species as cuckoo hosts was assigned according to Moksnes & Røskaft (1995), Røskaft et al. (2002) and Grim (2006). RESULTS Characteristics of attracted birds “community” At 32.0% nests (n = 75, cuckoo and control experiments pooled at each nest) the vigorous nest defence activity of blackcaps attracted in total 108 individuals of 21 passerine species (Table 1). At 15 of these 24 nests both the cuckoo and control experiments attracted neighbours. At some nests (n = 11), particular species appeared both during the first and second experiment, thus, it is possible that these were the same individuals (attracted individuals were not individually marked). After correcting for this possibility, the total number of attracted birds was 83 (i.e. 25 individuals were perhaps attracted to both experiments within the particular nest; Table 1). However, at 9 out of 15 nests where both experiments attracted neighbours these attracted birds’ “communities” differed as for species composition and abundance of particular species between the first and second experiment within a nest. Thus, there was a substantial turnover of attracted species and individuals between the experiments within a nest. Therefore, the total and corrected abundance (Table 1) should be taken as an upper and lower limit of the actual range of abundance respectively. Taking breeding densities as a surrogate measure of bird availability the blackcaps formed 25.7% of breeding pairs in the study area while only 8.4% of attracted birds were conspecifics (Table 1). Attracted blackcaps only alarm called and never attacked the dummy. The most individuals attracted during one experiment were 15 birds of 6 different species during a cuckoo presentation. In the total sample, the smallest species was the goldcrest Regulus regulus, the biggest was the jay Garrulus glandarius. Attracted birds usually stayed about 3 to 5 m from the focal nest and silently watched it. However, individuals of several species consistently alarm called (the chaffinch Fringilla coelebs, great tit Parus major, blackbird Turdus merula, robin Erithacus rubecula, and blackcap; Table 1). Such passive defence was observed only at 24.0% nests (n = 75) and at 10 of these nests (n = 18) the passive defence was elicited in both experiments at the particular nest. I observed solely one case of active defence: one nuthatch Sitta europaea vigorously dived above the dummy in one experiment.

4

During one experiment a female blackcap from the tested nest tried to drive away a “watching” nuthatch. Breeding density of the particular attracted species predicted neither their abundance nor frequency (both total and corrected; Table 1) during the experiments (Spearman correlations, n = 9, all P > 0.60). The abundance and frequency of attracted birds were strongly positively correlated (total: rs = 0.970, n = 21, P < 0.0001; corrected: rs = 0.938, n = 21, P < 0.0001). Out of 21 species of attracted birds 9 can be considered suitable while 12 as unsuitable cuckoo hosts due to their hole nestling habits or diet composition (Table 1). The abundance and frequency (both total and corrected) of suitable and unsuitable hosts did not differ (Mann-Whitney tests: all P > 0.50). Suitable (3 out of 9) and unsuitable (3 out of 12) hosts participated in mobbing with the same probability (Fisher exact test: P = 1.00). Determinants of attraction The number of attracted birds significantly increased with the ordinal measure of the intensity of nest defence in the cuckoo dummy experiments (Fig. 1; Kruskal-Wallis ANOVA: χ2 = 8.29, df = 2, P = 0.0159). The trend for attracted birds that helped with mobbing was similar but insignificant (Fig. 1; χ2 = 4.13, df = 2, P = 0.127). As the index of nest defence combines potential vocal and visual cues triggering the attraction of other birds I analysed separately these hypothesized effects of alarm calls and attacks. Although the rates of calling and frequencies of contacts were positively correlated (rs = 0.35, n = 75, P = 0.002) it was still possible to disentangle their respective effects. Higher rates of alarm calls attracted more birds (Fig. 2; χ2 = 8.26, df = 2, P = 0.0161). There was a similar trend for the number of attracted birds that participated in mobbing (Fig. 2; χ2 = 5.01, df = 2, P = 0.0815). However, there was no effect of contact attacks on the total number of attracted birds (Fig. 3; χ2 = 0.19, df = 2, P = 0.908). In contrast, the number of attracted mobbers tended to increase with the intensity of attacks by nest owners (χ2 = 4.90, df = 2, P = 0.0864). When no alarm calls were given by blackcaps other birds were never attracted (Fig. 2), whereas birds were attracted even when no contacts when performed (Fig. 3). These contrasting results indicate that the cue for attraction of alien birds was vocal (alarm calls) and not visual stimulus (attacks). However, when attracted birds were confronted with blackcaps showing higher intensity of nests defence (attacks) they too increased the intensity of their reaction (Fig. 3). Further I compared numbers of nests with low (alarm = 1) and high (alarm = 2, see Methods and Fig. 2) frequency of alarm calling by nest owners. The proportion of nests which attracted at least some neighbours did not differ between trials with low (44.1%, n = 34) and high (30.8%, n = 26) vocalization output by nest owners (χ2 = 1.11, df = 2, P = 0.292). Similarly the frequency of alarm calling did not affect the proportion of nests where neighbours participated in mobbing (26.5 vs. 26.9%; χ2 = 0.002, df = 2, P = 0.969). These results suggest that prolonged calling at the nest does not benefit the nest owners by attracting additional neighbours to join a mobbing group. Additional analyses also consistently showed a positive relationship between the number of alien birds attracted to the focal nest and the intensity of that nest owners’ responses to the cuckoo dummy (PC1). There was a significant increase in the number of attracted birds with the intensity of reaction to the cuckoo dummy (F1,73 = 5.72, P = 0.0193). Correlation on the same data was also significant (rs = 0.25, n = 75, P = 0.0291). When I restricted the analyses to attracted birds that mobbed the relationship was again consistent (F1,73 = 6.41, P = 0.0135; rs = 0.27, n = 75, P = 0.0208). However, the number of attracted birds that participated in mobbing only weakly increased with the group size (rs = 0.30, n = 23, P = 0.168). When I analyzed all data (both the cuckoo and control experiment from each nest) and controlled for the effect of the particular nest there was again a significant positive relationship between the intensity of reaction (PC1) and the total number of attracted birds (GLMM: F1,144 = 10.33, P = 0.0016). Excluding non-mobbing attracted birds showed identical results (GLMM: F1,144 = 9.04, P = 0.0031). The type of dummy, rank of presentation and their interaction were insignificant in both total and aggressive attracted birds analyses (GLMM, all P > 0.10). The positive relationship between the intensity of nest defence and the number of attracted birds cannot be explained by an increasing availability of potential mobbers during the nesting season (see Pavel & Bures 2001) as there was no effect of date in season on the number of attracted birds per cuckoo dummy experiments (rs = –0.031, n = 75, P = 0.794) or

5

per nest (cuckoo and control experiments pooled at each nest; rs = 0.0098, n = 75, P = 0.933). Sex of the defending parent may influence the number of attracted birds (Winkler 1995). However, there were no differences in the number of attracted birds or attracted aggressive birds both per cuckoo trials and per nest between nests divided into categories 1) female more aggressive, 2) male more aggressive and 3) both nest owners similarly aggressive (Kruskal-Wallis ANOVAs, all P > 0.05). DISCUSSION As predicted, the more intense nest defence by blackcaps attracted more neighbouring birds to the defended nest. The cue for this attraction was most likely the rate of alarm calling which concurs with results of previous studies in birds (Winkler 1995, Hurd 1996). The rate of contact attacks seemed to be unimportant for the attraction process itself but might affected the behaviour of foreign birds once attracted to the focal nest by increasing the probability of their alarm calling (which was the primary behavioural response of attracted birds). No raptors were attracted despite living in the study area forest (e.g. sparrowhawk Accipiter nisus, buzzard Buteo buteo; own observations), which falsifies the predator attraction hypothesis. Alien blackcaps appeared near defended nests at lower frequencies than expected from their very high breeding density which seems not to support the calling for help hypothesis. Although the kin relationships between blackcap nest owners and attracted blackcaps are unknown, the level and frequency of their “helping” with nest defence seem to be extremely low to be favoured by kin selection even in the potential case of close relatedness between nest owners and attracted blackcaps (moreover, extra-pair matings were never reported for the blackcap, thus helping to own extra-pair chicks in cuckolded nests is unlikely). This argument is even stronger as at their own nests blackcaps are highly aggressive: 42.7% of the tested pairs (n = 75) physically attacked (contacted and pecked) the dummy near their nests but not a single blackcap attracted to an alien nest attacked the dummy there. Mobbing based on reciprocal altruism between nest owners and attracted neighbours was reported recently in pied flycatchers Ficedula hypoleuca (Krams et al. 2006). In this species conspecific neighbours arrived at tested nests in 100% cases (under natural conditions when they were not experimentally constrained from arriving), whereas conspecifics arrived in only 9.3% of cases at blackcap nests in my study area (Table 1). In another study (Krams & Krama 2002) heterospecific neighbours attended 100% of mobbing trials with chaffinches whereas blackcaps attracted any neighbours only at 32% nests in my study. Thus, there seems to be only very little opportunity and motivation for blackcaps to reciprocate while mobbing with both conspecifics and heterospecifics. Nonetheless, a test of this reciprocal altruism hypothesis would require a specifically designed study with individually marked nest owners. The probability of being attracted did not correlate with the breeding density of attracted bird species in the study area (see also Winkler 1995). Breeding densities may be a too rough measure of attracted birds “availability”. More likely, the attracted birds were just those whose territories overlapped with territories of tested blackcaps or birds which happened to be near tested nests, e.g. when foraging there. From the methodological point of view the results of the current study should be conservative. Unsurprisingly, it was harder to note all behaviours of the more aggressive and active individuals in comparison to passive ones that just silently watched the dummy. Thus, there was a higher risk of overlooking some attracted birds in the former in comparison to the latter experiments. This logistic trade-off may have caused an underestimation of the true relationship between the nest defence intensity and the number of attracted birds. Coupled with the consistently significant results of the analyses of various data sets (cuckoo trials, all data) and the various measures of the nest defence intensity (ordinal scales, PCA) this suggests that the results of the current study are robust. Costs and benefits of attractions for blackcaps and attracted birds The rejection of the calling for help hypothesis is further supported by observations that attracted birds only rarely “helped” with nest defence – most of them just silently watched blackcaps defending their nests. The generally passive behaviour of attracted birds and their arrival after blackcaps started to respond aggressively clearly rejects a hypothesis that the

6

intensity of nest defence by blackcaps was influenced by potential mobbers (see Halupka & Halupka 1997). Interestingly, even in species with obligatory helpers these helpers may have no effect on the nest defence intensity of the breeding pair (Veen et al. 2000). Through such a passive behaviour the attracted birds may gain information on the presence of predators in, or in the vicinity of, their own territories. Additional benefits for attracted birds that alarm called may be distracting the predator (or brood parasite) from the general area, i.e. away from their own nests. In other words, attracted birds participating in the “passive defence” (sensu Winkler 1995) may force the mobbed intruder to move on by informing it that its presence is revealed and the probability of its foraging success is consequently decreased not only at the particular nest but also at those around. Attracted birds may also directly benefit from alerting their mates (Yasukawa 1989) and/or informing their own chicks at nearby nests that there is a risk of predation; thus chicks may be silenced by parental vocal signals and the probability of their death decreased (Knight & Temple 1986, Gill & Sealy 2003). Attracted mobbers might in principle benefit in the future if parents whom they helped will reciprocate; however, this is unlikely in short-lived small passerines (Winkler 1995) showing low philopatry (Grim 2005). Finally, attracted birds may learn alarm calls of their neighbours, preferentially of those more aggressive (Morton 1976), and later use them in mobbing contexts themselves (Goodale & Kotagama 2005) and/or to attract aggressive “nest defence helpers” to their own nests when necessary. All these benefits may compensate for time, energy and risk of injury costs resulting from taking part in mobbing of a predator at alien nests (Winkler 1995). Loud alarm calling may result in additional costs of revealing the nest location to eavesdropping predators (Krama & Krams 2005). Blackcap alarm calling attracted very few potential predators of their nests – only one jay (Table 1) and no great-spotted woodpeckers Dendrocopos major. Both these nest predators are common in the study area (personal observations). Even the attracted jay did not find the defended blackcap’s nest. Thus, there seems to be very low immediate costs of conspicuous mobbing to blackcaps apart from time and energy expenditures. However, Krama & Krams (2005) showed experimentally that loud mobbing may inform eavesdropping predators about the location of the nest with the predation event taking place at a later time. The attraction of neighbouring birds may also benefit nest owners as some of attracted birds alarm called and even attacked the dummy and thus increased the intensity of nest defence in favour of the nest owners. The limitation of the current study to test this is the use of stuffed dummies which cannot, in contrast to live intruders, be forced to leave the vicinity of the tested nest. It remains to be determined whether and how the presence of many birds affects the behaviour of living primary intruders (Bildstein 1982). This suggestion points to the need to study this question with real cuckoos and not just stuffed dummies. Nevertheless, potential benefits of predator distraction would be probably very low for blackcaps under natural conditions as the attracted birds participated infrequently both in active and passive nest defence (at 1 and 24% of nests respectively). More importantly, laying female cuckoos may succeed in parasitising a host nest even when the nest owners attack the laying cuckoo female physically by pecking (Wyllie 1981; see also Moksnes et al. 2000). Thus, the low frequency of neighbours’ responses coupled with low documented success of even very aggressive behaviours to brood parasites adults suggest that benefits to nest owners of attracting neighbours are negligible if any. Therefore, the observed pattern of positive correlation between the nest defence intensity and the number of attracted birds is, from nest owners’ point of view, best explained as a proximate response of attracted birds to the noise around defended nests. Universal structure of alarm and distress calls is generally thought to explain such interspecific attractions (Högstedt 1983, Hurd 1996). To sum up, the attraction of neighbouring birds by nest defending blackcaps seems to be neither costly nor beneficial for the tested species. Thus, from the point of view of “calling for help” the nest defence by the blackcap seems to be selectively neutral. Attracted birds and their suitability as cuckoo hosts Four of the attracted species that alarm called (chaffinch, blackbird, robin, and blackcap) are all known to be sometimes parasitized by the cuckoo (Moksnes & Røskaft 1995). However, the nuthatch which was the only species attacking the dummy and the great tit that alarm called were never reported as cuckoo hosts and may safely be considered

7

unsuitable for cuckoo parasitism due to their hole nesting habits (Moksnes & Røskaft 1995). In contrast, no chiffchaff Phylloscopus collybita, icterine warbler Hippolais icterina or whitethroat Sylvia communis ever participated in mobbing despite being suitable cuckoo hosts. Even cuckoo hosts known to show highly aggressive responses to cuckoo dummies at their own nests (e.g. Phylloscopus warblers, yellowhammer Emberiza citrinella, blackbird, chaffinch, blackcap; Grim 2005, Grim & Honza 2001, Røskaft et al. 2002, own unpublished data) did not physically attack the dummy near alien nest. This provides additional evidence against calling for help hypothesis. In general, there were no differences between suitable and unsuitable cuckoo hosts in abundance, frequency or participation in mobbing (Table 1). This may result for several reasons: (i) attracted birds did not recognize the cuckoo as a special threat and ignored it (acceptance errors sensu Sherman et al. 1997), or (ii) attracted birds simply observed blackcaps nest defence to gain information about possible predators in their territories overlapping with blackcap territories and did not participate in nest defence at alien nests which would provide them with no benefits but would be costly in terms of energy and potential injuries. As the intensity of alarm calling by blackcaps strongly positively correlated with some of their risk taking behaviours (attacks, time spent < 1m from the dummy, both P < 0.0001; see Grim 2005) the alarm calling in the blackcap is obviously directed at the intruder near the nest and not at any secondary predators, kin or reciprocal altruists. To sum up, although blackcaps’ nest defence attracted substantial number of birds from a wide variety of genera and families this attraction seems to be, from nest owners’ point of view, a by-product of proximate factors (e.g. conspicuousness of nest defence behaviour) and did not provide blackcaps with significant immediate benefits from attracted birds. This conclusion is further supported by previous observations that showed that parasitic cuckoos may not refrain from laying into host’s nest even when being both under heavy physical attacks from nest owners and in the presence of other mobbing birds (Wyllie 1981) not to speak about silent watching by nest owners or neighbours (Moksnes et al. 2000). Thus, results of the present study support neither the predator attraction (Högstedt 1983) nor calling for help (Rohwer et al. 1976) hypotheses. However, this does not preclude the adaptiveness of nest defence helpers attraction in other species (Pavel & Bures 2001, Krams & Krama 2002) or accruance of benefits (e.g. information on presence of predators) for attracted birds themselves. I thank S. A. Gill, M. E. Hauber, V. Pavel, and V. Remes for comments on the draft. I am grateful to M. Capek for providing breeding density data. I was supported by grants MSM6198959212 and GACR 206/03/D234. REFERENCES Bildstein, K.L. 1982. Responses of northern harriers to mobbing passerines. J. Field Ornithol. 53: 7–14. Bourne, W.R.P. 1977. The function of mobbing. British Birds 70: 266–267. Gärtner, K. 1981. Das Wegnehmen von Wirtsvogeleiern durch den Kuckuck Cuculus canorus. Ornith. Mitt. 33:115–131 Gill, S.A. & Sealy, S.G. 2003. Tests of two functions of alarm calls given by yellow warblers during nest defence. Can. J. Zool. 81: 1685–1690. Goodale, E. & Kotagama, S.W. 2005. Context-dependent vocal mimicry in a passerine bird. Proc. R. Soc. Lond. B. 273: 875–880. Grim, T. 2005. Host recognition of brood parasites: Implications for methodology in studies of enemy recognition. Auk 122: 530–543. Grim, T. 2006. Cuckoo growth performance in parasitized and unused hosts: not only host size matters. Behav. Ecol. Sociobiol. (in press). Grim, T. & Honza, M. 2001. Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to experimental parasitism by the common cuckoo Cuculus canorus. Biologia 56: 549–556. Halupka, K. & Halupka, L. 1997. The influence of reproductive season stage on nest defence by meadow pipits (Anthus pratensis). Ethol. Ecol. Evol. 9: 89–98.

8

Headley, F.W. & Jourdain, F.C.R. 1919. Adult cuckoo killing nestling birds. Brit. Birds 13: 57. Högstedt, G. 1983. Adaptation unto death: function of fear screams. Am. Nat. 121: 562– 570. Honza, M., Prochazka, P., Stokke, B.G., Moksnes, A., Røskaft, E., Capek, M. & Mrlik, V. 2004. Are blackcaps current winners in the evolutionary struggle against the common cuckoo? J. Ethol. 22: 175–180. Hurd, C.R. 1996. Interspecific attraction to the mobbing calls of black-capped chickadees (Parus atricapillus). Behav. Ecol. Sociobiol. 38: 287–292. Jourdain, F.C.R. 1925. A study on parasitism in the cuckoos. Proc. Zool. Soc. Lond. 44: 639–667. Kamil, A.C. 1988. Experimental design in ornithology. In Johnston, R.F. (ed) Current Ornithology Vol. 5: 313–346. New York: Plenum Press. Knight, R.L. & Temple, S.A. 1986. Nest defense in the American goldfinch. Anim. Behav. 34: 887–897. Krama, T. & Krams, I. 2005. Cost of mobbing call to breeding pied flycatcher, Ficedula hypoleuca. Behav. Ecol. 16: 37–40. Krams, I. & Krama, T. 2002. Interspecific reciprocity explains mobbing behaviour of the breeding chaffinches, Fringilla coelebs. Proc. R. Soc. Lond. B 269: 2345–2350. Krams, I., Krama, T. & Igaune, K. 2006. Mobbing behaviour: reciprocity-based cooperation in breeding pied flycatchers Ficedula hypoleuca. Ibis 148: 50–54. Krist, M. 2004. Importance of competition for food and nest sites in aggressive behaviour of collared flycatcher Ficedula albicollis. Bird Study 51: 41–47. Maloney, R.F., & Mclean, I.G. 1995. Historical and experimental learned predator recognition in free-living New Zealand Robins. Anim. Behav. 50:1193–1201. Marchant, S. 1972. Destruction of nest contents by cuckoos. Emu 72: 29–31. McLean, I.G., Smith, J.N.M. & Stewart, K.G. 1986. Mobbing behavior, nest exposure, and breeding success in the American robin. Behaviour 96: 171–185. Milburn, C.E. 1915. Adult cuckoo killing nestling meadow pipits. Brit. Birds 9: 95–96. Moksnes, A., Røskaft, E., Hagen, L.G., Honza, M., Mork, C. & Olsen, P.H. 2000. Common cuckoo Cuculus canorus and host behaviour at reed warbler Acrocephalus scirpaceus nests. Ibis 142: 247–258. Moksnes, A, & Røskaft, E. 1995. Egg-morphs and host preference in the common cuckoo Cuculus canorus: an analysis of cuckoo and host eggs from European museum collections. J. Zool. 236: 625–648. Morton, E.S. 1976. Vocal mimicry in the thick-billed euphonia. Wilson Bull. 88: 485–487. Pavel, V. & Bures, S. 2001. Offspring age and nest defence: test of the feedback hypothesis in the meadow pipit. Anim. Behav. 61: 297–303. Olendorf, R. & Robinson, S.K. 2000. Effectiveness of nest defence in the Acadian flycatcher Empidonax virescens. Ibis 142: 365–371. Rohwer, S., Frettwell, S.D. & Tuckfield, R.C. 1976. Distress screams as a measure of kinship in birds. Am. Midl. Nat. 96: 418–430. Røskaft, E., Moksnes, A., Stokke, B.G., Bicik, V. & Moskat, C. 2002. Aggression to dummy cuckoos by potential European cuckoo hosts. Behaviour 139: 613–628. Sealy, S.G., Neudorf, D.L., Hobson, K.A. & Gill, S.A. 1998. Nest defense by potential hosts of the brown-headed cowbird: methodological approaches, benefits of defense, and coevolution. In Rothstein, S.I. & Robinson, S.K. (eds) Parasitic Birds and Their Hosts: 194–211. New York: Oxford University Press. Sherman, P.W., Reeve, H.K. & Pfenning, D.W. 1997. Recognition systems. In Krebs, J.R. & Davies, N.B. (eds) Behavioural Ecology: 69–96. Oxford: Blackwell Scientific. Veen, T., Richardson, D.S., Blaakmeer, K. & Komdeur, J. 2000. Experimental evidence for innate predator recognition in the Seychelles warbler. Proc. R. Soc. Lond. B 267: 2253–2258. Winkler, D.W. 1994. Antipredator defense by neighbors as a responsive amplifier of parental defense in tree swallows. Anim. Behav. 47: 595–605. Wyllie, I. 1975. Study of cuckoos and reed warblers. Brit. Birds 68: 369–378. Yasukawa, K. 1989. The costs and benefits of a vocal signal: the nest-associated “chit” of the female red-winged blackbirds, Agelaius phoeniceus. Anim. Behav. 38: 866–874.

9

Table 1. The overview of species attracted by nest defence activities of blackcaps. Both upper (Total) and lower (Corrected) estimates of the numbers of attracted birds are shown (see text for details). n = abundance (i.e., number of individuals), F = frequency (i.e., number of nests). Total abundance = the sum of all attracted individuals observed during both cuckoo and control experiments. Total frequency = the total number of experiments where the species was observed. Corrected abundance = the number of attracted individuals discounting those that probably appeared both in the first and second experiment at the particular nest. Corrected frequency = the number of nests where the species was observed (i.e. experiments pooled within nests). The species indicated with an asterisk mobbed the dummy as a rule. Species Fringilla coelebs * Parus major * Parus caeruleus Sylvia atricapilla * Aegithalos caudatus Certhia familiaris Parus palustris Sitta europaea * Phylloscopus collybita Erithacus rubecula * Muscicapa striata Turdus merula * Carduelis chloris Emberiza citrinella Hippolais icterina Phylloscopus sibilatrix Regulus regulus Serinus serinus Sylvia communis Carduelis carduelis Garrulus glandarius Total = 21 species

Total n/F (n = 150 experiments) 18 / 14 12 / 8 9/7 9/9 8/4 7/5 7/4 7/4 4/4 3 3 3 3 2 2 2

/ / / / / / /

3 3 3 2 2 2 2

2/2 2/2 2/2 2/1 1/1 108 / 39

Corrected n/F (n = 75 nests) 12 / 9 11 / 7 8/6 7/7 4/2 6/4 7/4 6/3 4/4 2 2 2 2 1 1 2

/ / / / / / /

2 2 2 1 1 1 2

1 1 1 2 1 83

/ / / / / /

1 1 1 1 1 24

Cuckoo experiments n/F (n = 75 experiments) 7/6 9/5 7/6 5/5 4/2 5/3 5/3 5/2 0/0

Suitable cuckoo host? Yes No No Yes No No No No Yes

2 1 2 1 1 1 2

/ / / / / / /

2 1 2 1 1 1 2

Yes Yes No No Yes Yes Yes

1 1 1 0 1 61

/ / / / / /

1 1 1 0 1 23

No No Yes No No –

Figure 1. The effect of the intensity of nest defence (ordinal scale, see Methods) on the total number of birds attracted (open bars) and number of attracted birds that participated in mobbing (solid bars) of the cuckoo dummy (mean+SE). Sample sizes are shown above columns. Figure 2. The effect of alarm calling intensity (ordinal scale, see Methods) on the total number of birds attracted (open bars) and number of attracted birds that participated in mobbing (solid bars) of the cuckoo dummy (mean+SE). Sample sizes are shown above columns. Figure 3. The effect of frequency of contact attacks (ordinal scale, see Methods) on the total number of birds attracted (open bars) and number of attracted birds that participated in mobbing (solid bars) of the cuckoo dummy (mean+SE). Sample sizes are shown above columns.

10

Attracted birds

3.0 2.5 2.0 1.5 1.0 0.5 0.0

18

28

29

0

1

2

The index of nest defence (rank)

Attracted birds

Figure 1.

3.0 2.5 2.0 1.5 1.0 0.5 0.0

15

34

26

0

1

2

Alarm calls (rank)

Attracted birds

Figure 2.

3.0 2.5 2.0 1.5 1.0 0.5 0.0

46

12

17

0

1

2

Contact attacks (rank) Figure 3.

11

20. Grim T. 2002: Why is mimicry in cuckoo eggs sometimes so poor? 20. Journal of Avian Biology 33(3): 302–305. Grim T. 2002: Why is mimicry in cuckoo eggs sometimes so poor? Journal of Avian Biology 33(3): 302–305.

Forum JOURNAL OF AVIAN BIOLOGY 33: 302–305, 2002

Why is mimicry in cuckoo eggs sometimes so poor? Toma´sˇ Grim, Laboratory of Ornithology, Palacky´ Uni7ersity, Trˇ. S7obody 26, 771 46 Olomouc, Czech Republic. E-mail: [email protected]

I propose that the existence of imperfect adaptations (e.g. egg mimicry) in brood parasites and their hosts (e.g. discrimination abilities) could reflect age-dependent territory and nest-site selection patterns of the host. Studies of various passerines indicate that (1) older breeders tend to occupy nest sites of higher quality than do young birds (ideal despotic distribution resulting from interference competition), (2) nest-site selection affects the risk of parasitism in various habitats, (3) egg recognition in passerines has a strong learning component (therefore naive breeders tend to accept whereas older birds tend to reject parasitic eggs). Because young naive birds, who tend to accept parasitic eggs, usually breed in low-quality areas where they are frequently parasitised, while old experienced birds, who tend to reject parasitic eggs, breed in high-quality areas where they are rarely parasitised, the distribution of acceptors and rejecters with respect to the risk of parasitism is non-random, i.e. populations of some host species may consist of heavily parasitised acceptors and weakly parasitised rejecters. Therefore, the selection pressure exerted by the host on the parasite should be weaker than if brood parasitism was randomly distributed among naive and experienced breeders and affect adaptations such as egg mimicry. This could explain the existence of imperfect adaptations in some brood parasite-host systems.

Brood parasite-host associations are generally accepted as an excellent example of coevolution because we can be reasonably certain about the reciprocal evolution of adaptations and counter-adaptations (Rothstein 1990). However, despite the fact that brood parasites exert a strong selection pressure on their hosts (Rothstein 1990), defence mechanisms of many hosts against parasites are frequently poor. This is clearly evident from the mere existence of brood parasitism. The existence of imperfect adaptations has been generally explained in terms of genetic, economical and other types of constraints (Dawkins 1982). Earlier published ideas on constraints affecting hostbrood parasite coevolution include several mechanisms (see Table 1). Here I propose a new type of constraint that could play an important role in retarding parasitehost coevolution. The ‘‘nest-site constraint’’ suggests that age-dependent patterns of nest-site selection by hosts can result in higher parasitism rates in acceptors (young birds) than rejecters (older birds); this would 302

slow down an arms race and lead to the existence of imperfect adaptations in both parasites and their hosts.

The nest-site constraint: parasitised acceptors and non-parasitised rejecters Coevolution between brood parasites and hosts could be constrained by age-related patterns of nest-site selection in some host species. For example, in the reed warbler Acrocephalus scirpaceus individuals that start laying early lay larger eggs, produce larger clutches and generally occupy territories located far away from trees (Øien et al. 1996). In contrast, individuals that initiate laying later, lay smaller eggs and clutches and occupy territories closer to trees (Øien et al. 1996). Because early laying, larger eggs and larger clutches are characteristics of older, more experienced breeders in passerines (e.g. Saether 1990), including the closely related great reed warbler A. arundinaceus (Lotem et al. 1992), reed warblers nesting farther away from trees are presumably older, experienced individuals. Furthermore, the available evidence suggests that the increasing distance from trees (potential cuckoo Cuculus canorus perches) lowers the incidence of cuckoo parasitism of reed warbler nests (Øien et al. 1996). This suggests that young, inexperienced reed warblers tend to breed near trees, i.e. in areas highly susceptible to cuckoo parasitism (hence low-quality areas), whereas older, experienced individuals tend to breed in areas farther away from trees (hence high-quality areas), where they are less likely to be parasitised by cuckoos. Moreover, experienced breeders can discriminate between their own and cuckoo eggs (this ability is acquired during their ontogeny; see Lotem et al. 1992, 1995) whereas first-time breeders cannot. Thus the age/experience-related learning constraint makes it impossible for naive breeders to respond in an appropriate way to parasitic laying by the cuckoo. Because naive breeders (accepJOURNAL OF AVIAN BIOLOGY 33:3 (2002)

tors) are parasitised more often than older breeders (rejecters), the selection pressure driving the evolution of adaptations in the brood parasite, such as egg mimicry, should be weakened. Furthermore, the fact that naive breeders avoid parasitism later in life by avoiding areas with high rates of parasitic laying (i.e. through their territory and nest-site selection rather than through egg discrimination and rejection), suggests that selection for egg discrimination behaviour should be weaker in this situation than in situations when the probability of being parasitised is the same for naive and experienced breeders. To conclude, I propose that the weak selection pressure exerted on cuckoos by inexperienced host individuals should slow down the evolution of mimicry in parasitic eggs. This should, in turn, weaken selection for discrimination ability in hosts. The fact that both nest-site selection and discrimination behaviour in the host are age-dependent offers an explanation for the existence of poor egg mimicry in some parasites and low recognition capabilities in some of their hosts. The essence of the nest-site constraint is the age-dependent susceptibility to parasitism mediated through the age-dependent pattern of nest-site selection. This

age-related susceptibility to parasitism in hosts can theoretically reach two extreme states. First, acceptors are parasitised almost exclusively (thus, parasitic egg mimicry is very poor). Second, almost only rejecters are parasitised (thus, mimicry is near perfect). Every parasitised host population must lie somewhere in between these two extremes. If the population is biased in the direction of ‘‘only acceptors parasitised’’, then selection for mimicry is inevitably lower than if there is an equal probability that a parasitic egg will be laid in the nest of an acceptor or rejecter. The nest-site constraint should apply to any brood parasites that can successfully parasitise naive breeders lacking the egg discrimination ability (parasitising individuals that do not reject the parasite’s eggs is highly adaptive for a parasite; however, this pattern of parasitism is a by-product of the pattern of nestsite selection by the host). By weakening the intensity of selection for specific adaptations, the naive part of the host population could facilitate successful parasitism of the host, thereby also maintaining the brood parasite’s population. Therefore, the age of breeders is an important confounding variable which should be incorporated into models of brood parasitism dynamics and should be controlled for in field experiments.

Table 1. Overview of constraints on brood parasite-host coevolution. Genetic constraints

Weak-selection constraints

Ecological constraints Behavioural constraints

Morphological constraints

(1) gene flow between parasitised and non-parasitised populations (host defence in an isolated population is an example of local adaptation and its necessary prerequisite is philopatry) (2) low mutation rates resulting in the absence of beneficial mutations (3) low rates of parasitism (acceptance of parasitic eggs could be adaptive if parasites are rare and risk of recognition errors is high – the evolutionary equilibrium hypothesis) (4) short exposure of hosts to parasitism (acceptance of parasitic eggs could reflect short evolutionary interaction between parasite and host – the evolutionary lag hypothesis) (5) life/dinner principle and rare enemy effect (parasites are always ahead in the arms race because they have more to lose compared to hosts) (6) lack of suitable nest sites for re-nesting after nest desertion (7) nestling eviction behaviour exhibited by the common cuckoo, which reduces benefits of host’s discrimination (8) mafia effect – hosts ejecting parasitic eggs are penalised by clutch predation by a parasite (9) cognitive constraints such as poor recognition and discrimination abilities of the host (10) parasite egg mimicry making egg recognition by the host difficult (11) within-clutch and between-clutch variation in appearance of host eggs – lower intra-clutch and higher inter-clutch variation of host eggs facilitates recognition of parasitic eggs (12) small host size relative to that of the brood parasite – hosts with smaller bills tend to reject by desertion which is more costly than egg ejection; due to lower fitness returns for deserters, the desertion behaviour should spread more slowly than other defences

JOURNAL OF AVIAN BIOLOGY 33:3 (2002)

Soler et al. 1999

Rothstein 1990 Rohwer and Spaw 1988, Røskaft et al. 1990 Rothstein 1990

Dawkins and Krebs 1979 Petit 1991 Lotem 1993 Soler et al. 1995 McLean and Maloney 1998 Rothstein 1990 Øien et al. 1995, Stokke et al. 1999

Rohwer and Spaw 1988, Davies and Brooke 1989, Moksnes et al. 1991

303

Natural selection should favour individuals with adaptive nest-site choice, and young naive individuals should also be selected for breeding at safe sites. Nevertheless, the nest-site constraint cannot be avoided by naive breeders because the constraint is maintained by intra-specific competition; older, experienced breeders in general monopolise high-quality breeding sites, forcing naive breeders to occupy low-quality territories (ideal despotic distribution, see e.g. Hill 1988, Møller 1991, Petit and Petit 1996). Therefore, it is unlikely that young breeders could avoid the negative effects of the nest-site constraint. The effect of a nest-site constraint is almost inevitable if three conditions are met: (1) breeders’ age affects nest-site selection, (2) nest-site selection affects the risk of parasitism, (3) breeders’ age affects probability of rejection of parasitic eggs. Are these assumptions generally plausible? First, the effect of age on nest-site selection of birds is well established for many passerines; in general older breeders tend to monopolise higher-quality nest sites than do young birds (e.g. Petit and Petit 1996). This phenomenon has been established for species breeding in forests (Møller 1991), in more open heterogeneous mixed habitats (Hill 1988), and in open country (Brooke 1979). Second, nest-site selection is known to affect the probability of parasitism in several species breeding not only in reeds (Øien et al. 1996, Moska´ t and Honza 2000), but also in vineyards (Alvarez 1993), fields (Burhans 1997, Hauber and Russo 2000), prairies (Clotfelter 1998), and forests (Larison et al. 1998). Third, various studies indicate that egg recognition in passerines has a strong learning component (Victoria 1972, Rothstein 1990, Lotem et al. 1992, 1995), providing support for the idea that in general naive breeders are acceptors while older birds are rejecters. As a result, it is plausible that the nest-site constraint could slow down the arms race not only in the reed warbler-common cuckoo association but also in other brood parasite-host systems. Although the effect of nest-site constraint is almost inevitable if the above-mentioned conditions are met, the hypothesis generates a testable prediction: host species showing the age-dependent pattern of nest-site selection and hence age-dependent susceptibility to parasitism should be parasitised by cuckoos laying less mimetic eggs than species in which the nest-site constraint is absent. The paradox of parasitised acceptors and non-parasitised rejecters should operate also at a higher general level. Current favourite hosts usually accept parasitic eggs while most hosts that are strong rejecters are not parasitised (Davies and Brooke 1989). Therefore, the evolution of egg mimicry driven by host species with a weak discrimination ability should be slower than that driven by host species with high discrimination ability. Thus, current mimicry in cuckoo eggs should in general be poorer than if cuckoos did also parasitise strong rejecters. 304

Acknowledgements – I am grateful to N. B. Davies, J. Picman, V. Remesˇ, E. Røskaft, M. Soler, E. Tkadlec and two anonymous referees for helpful comments on earlier versions of the manuscript.

References Alvarez, F. 1993. Proximity of trees facilitates parasitism by cuckoos Cuculus canorus on rufous warblers Cercotrichas galactotes. – Ibis 135: 331. Brooke, M. de L. 1979. Differences in the quality of territories held by wheatears (Oenanthe oenanthe). – J. Anim. Ecol. 48: 21 – 32. Burhans, D. E. 1997. Habitat and microhabitat features associated with cowbird parasitism in two forest edge cowbird hosts. – Condor 99: 866 – 872. Clotfelter, E. D. 1998. What cues do brown-headed cowbirds use to locate red-winged blackbird host nests? – Anim. Behav. 55: 1181 – 1189. Davies, N. B and Brooke, M. de L. 1989. An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. I. Host egg discrimination. – J. Anim. Ecol. 58: 207 – 224. Dawkins, R. 1982. The Extended Phenotype. – Oxford University Press, Oxford. Dawkins, R. and Krebs, J. R. 1979. Arms races between and within species. – Proc. R. Soc. Lond. B 205: 489 – 511. Hauber, M. E. and Russo, S. A. 2000. Perch proximity correlates with higher rates of cowbird parasitism of ground nesting song sparrows. – Wilson Bull. 112: 150 – 153. Hill, G. E. 1988. Age, plumage brightness, territory quality, and reproductive success in the black-headed grosbeak. – Condor 90: 379 – 388. Larison, B., Laymon, S. A., Williams, P. L. and Smith, T. B. 1998. Song sparrows vs. cowbird brood parasites: impacts of forest structure and nest-site selection. – Condor 100: 93 – 101. Lotem, A. 1993. Learning to recognize nestlings is maladaptive for cuckoo Cuculus canorus hosts. – Nature 362: 743 – 745. Lotem, A., Nakamura, H. and Zahavi, A. 1992. Rejection of cuckoo eggs in relation to host age: a possible evolutionary equilibrium. – Behav. Ecol. 3: 128 – 132. Lotem, A., Nakamura, H. and Zahavi, A. 1995. Constraints on egg discrimination and cuckoo-host co-evolution. – Anim. Behav. 49: 1185 – 1209. McLean, I. G. and Maloney, R. F. 1998. Brood parasitism, recognition and response: the options. – In: Rothstein, S. I. and Robinson, S. K. (eds). Parasitic birds and their hosts. Oxford University Press, New York, pp. 255 – 272. Moksnes, A., Røskaft, E. and Braa, A. T. 1991. Rejection behavior by common cuckoo hosts towards artificial brood parasite eggs. – Auk 108: 348 – 354. Møller, A. P. 1991. Clutch size, nest predation, and distribution of avian unequal competitors in a patchy environment. – Ecology 72: 1336 – 1349. Moska´ t, C. and Honza, M. 2000. Effect of nest and nest site characteristics on the risk of cuckoo Cuculus canorus parasitism in the great reed warbler Acrocephalus arundinaceus. – Ecography 23: 335 – 341. Øien, I. J., Honza, M., Moksnes, A. and Røskaft, E. 1996. The risk of parasitism in relation to distance from reed warbler nests to cuckoo perches. – J. Anim. Ecol. 65: 147 – 153. Øien, I. J., Moksnes, A. and Røskaft, E. 1995. Evolution of variation in egg colour and marking pattern in European passerines: adaptations in a coevolutionary arms race with the cuckoo, Cuculus canorus. – Behav. Ecol. 6: 166 – 174. Petit, L. J. 1991. Adaptive tolerance of cowbird parasitism by prothonotary warblers: a consequence of nest-site limitation? – Anim. Behav. 41: 425 – 432. JOURNAL OF AVIAN BIOLOGY 33:3 (2002)

Petit, L. J. and Petit, D. R. 1996. Factors governing habitat selection by prothonotary warblers: Field tests of the Fretwell-Lucas models. – Ecol. Monogr. 66: 367 – 387. Rohwer, S. and Spaw, C. D. 1988. Evolutionary lag versus bill-size constraints: a comparative study of the acceptance of cowbird eggs by old hosts. – Evol. Ecol. 2: 27 – 36. Røskaft, E., Orians, G. H. and Beletsky, L. D. 1990. Why do red-winged blackbirds accept eggs of brown-headed cowbirds? – Evol. Ecol. 4: 35 –42. Rothstein, S. I. 1990. A model system for coevolution: avian brood parasitism. – Annu. Rev. Ecol. Syst. 21: 481 – 508. Saether, B-E. 1990. Age-specific variation in reproductive performance of birds. – In: Power, D. M. (ed.). Current Ornithology. Plenum Press, New York, pp. 251 – 283.

JOURNAL OF AVIAN BIOLOGY 33:3 (2002)

Soler, J. J., Martinez, J. G., Soler, M. and Møller, A. P. 1999. Genetic and geographic variation in rejection behavior of cuckoo eggs by European magpie populations: an experimental test of rejecter-gene flow. – Evolution 53: 947 – 956. Soler, M., Soler, J. J., Martinez, J. G. and Møller, A. P. 1995. Magpie host manipulation by great spotted cuckoos: evidence for an avian mafia? – Evolution 49: 770 – 775. Stokke, B. G., Moksnes, A., Røskaft, E., Rudolfsen, G. and Honza, M. 1999. Rejection of artificial cuckoo Cuculus canorus eggs in relation to variation in egg appearance among reed warblers Acrocephalus scirpaceus. – Proc. R. Soc. Lond. B. 266: 1483 – 1488.

305

21. Grim T. & Šumbera 2006: A new record of the endangered white-winged nightjar 21. from Beni, Bolivia. (Eleothreptus candicans) Grim T. & Šumbera 2006: A new record of the endangered white-winged nightjar The Wilson Journal of Ornithology 118(1): 109–112. (Eleothreptus candicans) from Beni, Bolivia. The Wilson Journal of Ornithology 118(1): 109–112.

WILSON BULLETIN Thursday Jan 05 2006 04:43 PM Allen Press • DTPro System

GALLEY

SHORT COMMUNICATIONS

wils 118_115 Mp_109 File # 19TQ

109

The Wilson Journal of Ornithology 118(1):109–112, 2006

A New Record of the Endangered White-winged Nightjar (Eleothreptus candicans) from Beni, Bolivia Toma´sˇ Grim,1,3 and Radim Sˇumbera2 ABSTRACT.—The ecology of the White-winged Nightjar (Eleothreptus candicans) is poorly known. Only three breeding populations (one from Brazil and two from Paraguay) are known, and populations are decreasing due to continuing destruction of cerrado habitat. On 14 September 2003, we took several photos of an unidentified nightjar in Beni Biosphere Reserve, Departmento Beni, Bolivia. The bird was later determined to be an adult male White-winged Nightjar. Interestingly, the only previous record for Bolivia was a male collected in 1987 at the same locality and time of year. Because the White-winged Nightjar is nonmigratory and secretive, we hypothesize that there may be a sustainable population of White-winged Nightjars in Bolivia, and the paucity of sightings may be due to the species’ low detectability. Received 16 December 2004, accepted 11 October 2005.

The White-winged Nightjar (Eleothreptus candicans), a member of the Caprimulgidae (Cleere 1999, Pople 2004), was recently reclassified from the genus Caprimulgus to the genus Eleothreptus (Cleere 2002). Its known range and population size are very small, and its ecology has received attention only recently (Pople 2003). Parker et al. (1996) assigned the species High Conservation Priority and the IUCN lists the species as Endangered (IUCN Red List; Pople 2004). E. candicans is threatened by ongoing loss of its cerrado habitat (heavy grazing, trampling, invasive grasses, habitat conversion to plantations, and large-scale, uncontrolled grass fires; Cleere 1999, Pople 2004). Until the 1980s, White-winged Nightjars were known only from two museum specimens collected at the beginning of the 19th century in Oric¸anga, Sa˜o Paulo state, and Cuiaba´, Mato Grosso state, Brazil (Sclater 1866). Only three populations have been found, all 1 Dept. of Zoology, Palacky ´ Univ., Tr. Svobody 26, 771 46 Olomouc, Czech Republic. 2 Dept. of Zoology, Univ. of South Bohemia, Branisˇovska´ 31, 37005 Ceske´ Budejovice, Czech Republic. 3 Corresponding author; e-mail: [email protected]

in southern Brazil and eastern Paraguay: Emas National Park, Brazil (Rodrigues et al. 1999); ˜ u, Mbaracayu´ Forest Nature ReAguara´ N serve, Paraguay (Lowen et al. 1996, Clay et al. 1998); and a recently discovered population at Laguna Blanca, Departmento San Pedro, central Paraguay (Anonymous 2002). Additionally, in 1987 a single male was captured and collected at the Beni Biological Station, Departmento Beni, Bolivia (Davis and Flores 1994). Despite specific searches for the species in subsequent years, however, it has not been relocated at Beni (Brace et al. 1997, Brace 2000, Pople 2004; R. Brace and J. Hornbuckle in litt.). ˜ u have resulted in a Surveys in Aguara´ N population estimate of 40–150 individuals (Clay et al. 1998, Pople 2003) at that location. The number of birds observed in Emas National Park was 12 in September 1985 and only 1 in October 1990 and in November 1997 (Rodrigues et al. 1999). Although there are no other recently published records from Emas, the national park probably supports a sizeable population of E. candicans (Pople 2004) because Emas encompasses a large extent of apparently suitable habitat. The recently discovered population at Laguna Blanca in Paraguay is estimated to include a minimum of 30 birds (R. P. Clay in litt.). On 14 September 2003 at 22:00 EDT, we photographed an unidentified nightjar on a termite mound between the Beni Biological Station (Estacio´n Biolo´gica del Beni; 148 509 S, 668 179 W) and Laguna Normandia (;1.5 km northwest of the station; see Fig. 3 in Brace et al. 1997), Departmento Beni in northern Bolivia. Later the bird was unambiguously identified as a male E. candicans (Fig. 1). Because it lacked visible wear on the remiges and pale flecking in the contour plumage, it is probable that the individual had recently completed a molt. If the species undergoes the same pattern of molt in both Beni Biosphere

WILSON BULLETIN Thursday Jan 05 2006 04:43 PM Allen Press • DTPro System

110

GALLEY

wils 118_115 Mp_110 File # 19TQ

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 118, No. 1, March 2006

FIG. 1. Adult male White-winged Nightjar (Eleothreptus candicans) photographed on 14 September 2003 in Beni Biosphere Reserve, Departmento Beni, Bolivia. Photo by R. Sˇumbera.

Reserve and Paraguay (i.e., replacement of flight feathers in a single post-nuptial molt), it suggests that the species may breed considerably earlier in Bolivia than in Paraguay (where it breeds mainly between September and December). Beni Biological Station is 180 km west of Trinidad and 50 km east of San Borja on El ´ stancia. El Porvenir E ´ stancia lies in Porvenir E the Llanos de Mojos, which is a lowland plain (;200 m elevation) characterized as savanna with forest islands. The habitat where we observed the White-winged Nightjar is a seasonally inundated savanna with a high density of termite mounds (Fig. 2). Ours is only the second record of Whitewinged Nightjar in Bolivia, the first having been made in September 1987 (Davis and Flores 1994). Interestingly, both observations were made near Beni Biological Station at the same time of year (11 September 1987 and 14 September 2003). Despite a number of research programs that have been conducted at the station (A. B. Hennessey in litt.), there had been no additional records of White-winged Nightjar after 1987. R. C. Brace and J. Horn-

buckle (in litt.), for example, searched for White-winged Nightjars and conducted mistnetting from mid-July through the end of August every year from 1992 to 1999, but recorded no White-winged Nightjars. Although the White-winged Nightjar is considerably less conspicuous than many other sympatric nightjar species common in Bolivia (R. G. Pople in litt.), it seems unlikely that there would be so few observations of the species if the area supported a small resident population. Rather, the two individuals recorded during the last 2 decades may have come from an undiscovered population elsewhere in the northern Bolivian lowlands. However, E. candicans is presumed to be a resident species. Indeed, radio-tracking work in Paraguay (Pople 2003) revealed that White-winged Nightjars are year-round residents, and a study of captive birds revealed a post-nuptial molt pattern typical of a nonmigratory species. Therefore, the occurrence of the two individuals at Beni Biological Station during the same time of year may indicate that some birds make local movements, possibly in response to fires (Pople 2004).

WILSON BULLETIN Thursday Jan 05 2006 04:43 PM Allen Press • DTPro System

wils 118_115 Mp_111 File # 19TQ

GALLEY

SHORT COMMUNICATIONS

111

FIG. 2. Typical habitat of the White-winged Nightjar—wet savanna with termite mounds providing perches above the surrounding young vegetation. The forest in the background is Florida Fragment south of Laguna Normandia, 1.5 km northwest of Beni Biological Station, Departmento Beni, Bolivia. The photo in Figure 1 was taken within this area. Photo by T. Grim.

Neotropical savannas are under increasing human pressure due to large-scale conversion of grassland habitats to pastures (Marris 2005). Although the White-winged Nightjar is a typical savanna dweller and is adapted to irregular and small-scale fires, it likely has been negatively affected by regular and largescale burning in recent years (Brace et al. 1997, Pople 2004). Conservation of savanna habitats—including cerrado, the primary habitat for E. candicans—has been neglected thus far. Because savanna habitats are facing greater threats than Amazonian rainforests, the conservation of cerrado habitat should become a top priority in the Neotropics (Marris 2005). Our observation highlights the importance of Beni Biosphere Reserve for threatened (n 5 4) and near-threatened (n 5 15) bird species in Bolivia (Brace et al. 1997). Among these 19 species are 11 that rely wholly or partially on savanna habitat. So far, 500 bird species have been reported from Beni Biosphere Re-

serve (Brace et al. 1997, Brace 2000). We add to this list one more species: on the same day (14 September 2003) that we observed the White-winged Nightjar, we also recorded one Black-throated Saltator (Saltator atricollis). We hypothesize that Departmento Beni in northern Bolivia holds a resident population of E. candicans, and that the paucity of records from Bolivia reflects the lack of intensive searches during the correct season and the limited detectability of this species. We concur with Brace et al. (1997) that more information on the White-winged Nightjar’s status is required, and we hope that our observation provides an impetus for further research on this elusive species. ACKNOWLEDGMENTS We are grateful for the detailed and helpful comments and suggestions by R. P. Clay and two anonymous referees. We thank R. C. Brace and A. B. Hennessey for their comments on the manuscript and G. Dryden for reviewing our translation to English.

WILSON BULLETIN Thursday Jan 05 2006 04:43 PM Allen Press • DTPro System

112

GALLEY

wils 118_115 Mp_112 File # 20TQ

THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 118, No. 1, March 2006

LITERATURE CITED ANONYMOUS. 2002. A new population of the Whitewinged Nightjar. World Birdwatch 24:5. BRACE, R. C. 2000. The avifauna of the Beni Biological Station: records to 1999. Estacio´n Biologica del Beni, Bolivia. BRACE, R. C., J. HORNBUCKLE, AND J. W. PEARCE-HIGGINS. 1997. The avifauna of the Beni Biological Station, Bolivia. Bird Conservation International 7:117–159. CLAY, R. P., D. R. CAPPER, J. MAZAR BARNETT, I. J. BURFIELD, E. Z. ESQUIVEL, R. FARIN˜A, C. P. KENNEDY, M. PERRENS, AND R. G. POPLE. 1998. Whitewinged Nightjars Caprimulgus candicans and cerrado conservation: the key findings of Project ˜ u 1997. Cotinga 9:52–56. Aguara´ N CLEERE, N. 1999. Family Caprimulgidae (nightjars). Pages 302–386 in Handbook of the birds of the world, vol. 5: Barn-owls to hummingbirds (J. del Hoyo, A. Elliott, and J. Sargatal, Eds.). Lynx Edicions, Barcelona, Spain. CLEERE, N. 2002. A review of the taxonomy and systematics of the Sickle-winged and White-winged nightjars (Caprimulgidae). The Bulletin of the British Ornithologists’ Club 122:168–179. DAVIS, S. E. AND E. FLORES. 1994. First record of White-winged Nightjar Caprimulgus candicans for Bolivia. The Bulletin of the British Ornithologists’ Club 114:127–128.

LOWEN, J. C., L. BARTRINA, T. M. BROOKS, R. P. CLAY, AND J. TOBIAS. 1996. Project YACUTINGA ’95: bird surveys and conservation priorities in eastern Paraguay. Cotinga 5:14–19. MARRIS, E. 2005. The forgotten ecosystem. Nature 437:943–944. PARKER, T. A., III, D. F. STOTZ, AND J. W. FITZPATRICK. 1996. Ecological and distributional databases for Neotropical birds. Pages 118–407 in Neotropical birds: ecology and conservation (D. F. Stotz, J. W. Fitzpatrick, T. A. Parker, III, and D. K. Moskovits). University of Chicago Press, Chicago, Illinois. POPLE, R. G. 2003. The ecology and conservation of the White-winged Nightjar Caprimulgus candicans. Ph.D. dissertation, University of Cambridge, United Kingdom. POPLE, R. G. 2004. White-winged Nightjar Eleothreptus candicans. In Threatened birds of the world 2004. CD-ROM. BirdLife International, Cambridge, United Kingdom. RODRIGUES, F. H. G., A. HASS, O. J. MARINI-FILHO, M. M. GUIMARA˜ES, AND M. A. BAGNO. 1999. A new record of White-winged Nightjar Caprimulgus candicans in Emas National Park, Goia´s, Brazil. Cotinga 11:83–85. SCLATER, P. L. 1866. Additional notes on the Caprimulgidae. Proceedings of the Zoological Society of London 1866:581–590.

22. Honza M., Picman J., Grim T., Novák V., Čapek M. & Mrlík V. 2001: How to hatch from 22. A study of the common cuckoo. an egg of great structural strength. Honza M., Picman J., Grim T., Novák V., Biology Čapek M. & Mrlík V. 2001: How to hatch from Journal of Avian 32(3): 249–255. an egg of great structural strength. A study of the common cuckoo. Journal of Avian Biology 32(3): 249–255.

JOURNAL OF AVIAN BIOLOGY 32: 249–255. Copenhagen 2001

How to hatch from an egg of great structural strength. A study of the Common Cuckoo Marcel Honza, Jaroslav Picman, Toma´sˇ Grim, Vı´t Nova´k, Miroslav C& apek, Jr. and Vojteˇch Mrlı´k

Honza, M., Picman, J., Grim, T., Nova´k, V., C& apek, M., Jr. and Mrlı´k, V. 2001. How to hatch from an egg of great structural strength. A study of the Common Cuckoo. – J. Avian Biol. 32: 249 – 255. Brood parasitism represents a unique mode of avian reproduction that requires a number of adaptations. For example, to reduce chances of puncture ejection of their eggs by small hosts, brood parasites may have been selected for laying eggs of unusually great structural strength. However, great structural strength of eggshells should hinder hatching. The goals of our study were to establish if chicks of the Common Cuckoo Cuculus canorus have more difficulty with hatching out of their strong eggs than chicks of species with eggs of similar size, and whether they possess any mechanisms facilitating hatching. To achieve these goals, we compared hatching pattern and selected body characteristics of chicks of the Common Cuckoo with those of another altricial species with eggs of a similar size, the Great Reed Warbler Acrocephalus arundinaceus. Although the rate of pecking was similar in the two species, the Common Cuckoo chicks started pecking earlier in relation to their emergence and consequently required more time and a greater cumulative number of pecks for breaking open their eggs than did young Great Reed Warblers. The two species also differed with respect to the pattern of opening their shells; in contrast to the warbler chicks, which enlarged the original pip circularly, the cuckoo chicks opened the egg by systematically creating a long narrow slit until they emerged. Finally, our study of hatched young revealed several differences; the Cuckoo hatchlings were significantly heavier, had a longer forearm, and their egg tooth was located significantly farther from the tip of the beak. The edge used for cutting through the shell was also significantly longer than that of hatchling Great Reed Warblers. To conclude, our data suggest that hatching is more difficult for a Cuckoo than for a Great Reed Warbler and that Cuckoos possess several mechanisms to overcome the problems of hatching from a structurally strong egg. M. Honza, M. C& apek, Jr., and V. Mrlı´k, Department of A6ian Ecology, Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic, K6eˇtna´ 8, 603 65 Brno, Czech Republic. J. Picman, Department of Biology, Uni6ersity of Ottawa, 30 Marie Curie, Ottawa, Ontario K1N 6N5, Canada. T. Grim, Laboratory of Ornithology, Palacky´ Uni6ersity, Trˇ. S6obody 26, 771 46 Olomouc, Czech Republic. V. No6a´k, Department of Zoology and Ecology, Masaryk Uni6ersity, Kotla´rˇska´ 2, 611 37 Brno, Czech Republic. E-mail: [email protected]

Avian eggs characteristically have hard calcareous eggshells whose main function is to provide mechanical protection for the developing embryo from outside pressures. Egg strength is highly species-specific because it represents a compromise between different selective forces such as the need for mechanical protection of an embryo, which favours great strength, and the necessity for the young to hatch without assistance from its parents presumably favouring low strength of an egg. There exists a highly significant relationship between ©

egg strength and egg size (Ar et al. 1979, Picman et al. 1996). However, available data suggest that brood parasitic species of birds such as cuckoos and cowbirds lay eggs that are stronger than expected for their size (Spaw and Rohwer 1987, Picman 1989a, Brooker and Brooker 1991, Picman and Pribil 1997). The Common Cuckoo Cuculus canorus is an obligate brood parasite that breeds across the Palearctic from western Europe to Japan (Cramp and Simmons 1985). This cuckoo species is known to have parasitized more

JOURNAL OF AVIAN BIOLOGY

JOURNAL OF AVIAN BIOLOGY 32:3 (2001)

249

than 125 passerine species, but only 11 of these are considered to be major hosts (Moksnes and Røskaft 1995, Wyllie 1981). Since Common Cuckoo parasitism characteristically reduces hosts’ reproductive success to a very low level (Øien et al. 1998), there should be strong selection for host defences against parasitism. During the evolution of this brood parasitism system a number of adaptations and counter adaptations have evolved, described by Davies and Brooke (1989a, b) and Øien et al. (1995) as a co-evolutionary arms race between the Common Cuckoo and its hosts. The rejection of the brood parasite’s egg(s) by the host is an important trait that may have selected for increased eggshell strength in the cuckoo, thus precluding rejection of cuckoo eggs by small hosts through puncture ejection. Alternatively or in addition, the extra strength could simply reflect a residual characteristic of the much larger cuckoo egg before it evolved its small size (Lack 1968, Spaw and Rohwer 1987, Picman 1989a, J. Picman and M. Honza unpubl. data). In addition, increased strength of the Common Cuckoo eggs should also prevent their breakage during laying (Lack 1968). The great strength of cuckoo eggs has recently been demonstrated in a study that showed that Common Cuckoo eggs are 2.2 times stronger against outside pressure than eggs of the same size laid by their passerine hosts (J. Picman and M. Honza unpubl. data). Similar evidence also exists for inside strength of eggs of this cuckoo species (J. Picman and M. Honza, unpubl. data). This suggests that the great structural strength of cuckoo eggshells should make hatching difficult for young Common Cuckoos. Therefore, there should be intense selection on Common Cuckoos for various adaptations facilitating hatching. In this study we examine hatching-related adaptations in the Common Cuckoo and one of its major hosts, the Great Reed Warbler Acrocephalus arundinaceus. Our earlier studies have established that eggs of these two species are similar in size (t-test, P\ 0.05; mean egg volume in ml 9SD: 3.0719 0.026 vs 3.1289 0.43 for Common Cuckoo and Great Reed Warbler, respectively, J. Picman and M. Honza unpubl. data). To measure the strength of eggs, we used a mechanical puncture-resistance tester (see Picman 1989b). Both the outside and inside strengths of eggs are substantially greater in the Common Cuckoo than in the Great Reed Warbler (J. Picman and M. Honza unpubl. data). Based on these findings we suggest that the Common Cuckoo chicks should: (1) experience greater difficulty during hatching and (2) possess specific mechanisms facilitating their hatching. More specifically, we propose the following predictions. First, the great inside eggshell strength should increase the effort required for both egg pipping and subsequent shell breakage during emergence. Therefore, we expect that Common Cuckoo chicks should start pecking earlier relative to their emergence and/or that they should exhibit a higher 250

pecking rate than their hosts’ chicks, which hatch from eggs of normal strength. Second, the necessity to break through the unusually strong eggshell could also result in several changes in the egg tooth used for puncturing and then breaking the shell open. For example, a strong shell may require an egg tooth that is larger and possibly also located closer to the forehead on the upper mandible (thereby presumably increasing its leverage). Third, we predict that natural selection has favoured long forearms and tarsi in cuckoo chicks, because both are presumably used in the final stages of hatching (stretching legs and spreading forearms should contribute to the opening of eggshell by forcing the shell apart; see e.g. Pettingill 1970). If this is true, then hatchling Common Cuckoos should have longer forearms than hatchling Great Reed Warblers.

Methods Egg collection We collected Common Cuckoo and Great Reed Warbler eggs between 15 May and 30 June 1999, in the south-eastern part of the Czech Republic, in pond areas near the villages of Lednice and Luzˇ ice (47° 40% N, 16° 48% E), about 40 and 60 km south and south-east of Brno, respectively. We searched the littoral vegetation surrounding fish ponds for nests. The vegetation consisted of common reed Phragmites australis and narrow-leaf cattail Typha angustifolia stands which provide suitable nesting sites for the two main Common Cuckoo hosts, the Great Reed Warbler and Reed Warbler Acrocephalus scirpaceus. We included only one egg from each female cuckoo. This was made possible because individual female Common Cuckoos lay eggs of an individual-specific appearance (Wyllie 1981). If a nest contained more than one cuckoo egg, we selected one that was different in appearance from those collected earlier. In this study we used only those eggs that were freshly laid or that had been incubated for a maximum of three days. If the stage of embryonic development was not precisely known, we estimated it by the floatation method (Hays and Lecroy 1971). Egg volume (ml) was calculated using Hoyt’s (1979) formula: volume =0.507× length× breadth2.

Incubation We artificially incubated a total of 25 Common Cuckoo and 25 Great Reed Warbler eggs of which 18 (72.0%) and 17 (68%) eggs hatched. Incubation took place in an incubator (Octagon 20, Brinsea, UK), with temperature ranging from 37.7° to 37.8°C and relative humidity (RH) from 55 to 65%. Eggs were turned automatically 24 times a day (once every hour). Two to three days JOURNAL OF AVIAN BIOLOGY 32:3 (2001)

before expected hatching, we transferred the eggs to another incubator (Octagon 10, Brinsea, UK), where we maintained the same temperature, but where we increased RH to 75–80%. The eggs were not turned in the second incubator. Both incubators were transparent and double-skinned. During the last two days prior to the expected hatching day, we listened for bill-pecking every two hours during the day and every four hours during the night. When we recorded the first bill-pecking, we placed a small, sensitive tie-clip microphone, SONY ECM T 140 (connected to a Marantz PMD 222 audio tape-recorder), in the incubator in such a way that it almost touched the egg. The bill-pecking was then recorded automatically on audio tape during 5min intervals, with 30 min between consecutive recording sessions until chick emergence (emergence was defined as the time when the chick pushed the shell cap off and emerged from the egg; Hamburger and Oppenheim 1967). Just before the eggs began to hatch, we started video-recording the eggs using Sony CCD-TR 660 E Hi 8 video-cameras. Immediately after emergence, hatched chicks were euthanased by cold and then stored in 60% alcohol for later analyses of selected meristic traits. During the hatching process, we measured the time from the first pip on the shell to chick emergence. Only those eggs where the interval between the last examination and the first crack in the shell was shorter than 30 min were included in the following analyses. The half-time between the two events (last examination and the first crack) was then considered as the starting time for all calculations. It is known that certain stimuli to which developing avian embryos are exposed (e.g. click communication) may affect the timing of emergence and duration of the hatching process (e.g. Vince 1969, McMaster and Sealy 1998). Therefore, we incubated eggs that were at the same stage of embryonic development together. Since it is known that avian embryos are behaviourally responsive to light stimulation (Oppenheim 1968), we maintained a constant light regime throughout the incubation period.

beak (see Fig. 1). The selected traits of the emerged chicks were measured using the stereomicroscope system OLYMPUS and MicroImage software (accuracy 0.1 mm).

Results Hatching pattern; egg-pecking The first Common Cuckoo started to peck 16 h before it emerged, whereas the first Great Reed Warbler chick started to peck 7 h before its emergence. On average, the Common Cuckoo chicks started pecking earlier (x¯ = 7.109 4.70 (SD) h, n=18) before their emergence than the Great Reed Warbler chicks (x¯ = 4.159 4.01 (SD) h, n =17); see also Figs 2 and 3). The difference between the two species was statistically significant (t-test; t =0.55, d.f. =43, PB0.05). The difference in the timing of egg pecking between the two species is further evident from the fact that 50% of Cuckoo chicks were pecking 8 h before their emergence versus 2.5 h for Great Reed Warbler chicks. However, chicks of the two species exhibited similar rates of pecking (36 and 37 pecks/min for Common Cuckoo and Great Reed Warbler, respectively; t=1.802, d.f.= 24, P\ 0.05) and in general also similar temporal patterns of pecking throughout the hatching period (see Fig. 2a and b). However, the examination of the temporal pattern of egg pecking revealed that the mean pecking rates of both species increased gradually to their peak

Egg measurements To determine eggshell thickness, we obtained a small shell fragment (about 2 mm2) with pincers from three randomly selected areas along the widest area of an egg and measured its thickness with a micrometer (accuracy 0.001 mm). Hatchling weight was obtained by removing the chick from alcohol, drying it for 5 min at 20°C, and then weighing it (accuracy 1 mg). For a comparative analysis of meristic traits between the Common Cuckoo and the Great Reed Warbler, we selected traits that are likely to play a role in hatching: length of culmen, forearm length, body length (measured from shoulder to tail), tarsus length, length and height of the egg tooth, and distance of the egg tooth from the tip of the JOURNAL OF AVIAN BIOLOGY 32:3 (2001)

Fig. 1. Measurements of bird embryos. (A) shoulder to tail length; (B) forearm length; (C) culmen length; (D) tarsus length. The measurements on the egg tooth included distance from the front edge of the tooth to the tip of the beak (L1), length of the front edge of the egg tooth (L2), tooth height (L3) and length of the cutting (rear) edge of the tooth (L4).

251

chicks than by the Great Reed Warbler chicks (Fig. 3; Wilcoxon signed ranks test: Z= − 9.43, PB 0.001). Finally, also as a result of the different timing of initiation of egg pecking, the hatching period (time between the first egg pip and chick emergence) was longer in the Common Cuckoo (416.669 219.25 min) than in the Great Reed Warbler (119.669 149.8 min; t-test; t = 3.36, d.f.= 16, PB0.001).

Hatching pattern; opening of the eggshell The Common Cuckoo and Great Reed Warbler chicks exhibited species-specific patterns of opening the eggshell. The Common Cuckoo chicks first made a crack in the shell with the egg-tooth and then, by back-stretching the neck and using the long edge of the egg tooth (designated as L4 in Fig. 1) as a cutting device made a narrow slit approximately 1 mm wide. The average length of this slit immediately before emergence was 10.8492.79 mm (n = 13). In five (38.5%) chicks, we recorded a second pip located up to 2 mm from the original one. These chicks then continued breaking the egg from these holes as described above. In contrast to the cuckoo chicks, after having made the first crack in the shell, the Great Reed Warbler chicks were enlarging the original pip circularly (i.e. by increasing the diameter of the initial hole) until emerging when the mean (9 SD) length and width of the opening

Fig. 2. Pecking frequency (number of pecks/5 min) of the embryos. Shown are means and S.D. Only intervals in which at least three embryos pecked are considered. The time between successive recording intervals was 30 min. (a) Common Cuckoo, (b) Great Reed Warbler.

levels, which were reached about 1.5 h and 1 h before emergence in the Common Cuckoo and Great Reed Warbler, respectively (Fig. 2a and b). This increase in pecking rate was statistically significant for both species (Generalized Linear Mixed Model, slope for Common Cuckoo =4.086, d.f.=13, t=3.70, P= 0.0027; slope for Great Reed Warbler =4.797, d.f.= 11, t = 4.55, P =0.0008). Just before emergence, the mean pecking rates declined slightly for chicks of both species (Fig. 2a and b). The total number of pecks made by the chicks of the two species reflects their different timing of initiation of pecking. Overall, hatching required a significantly higher total number of pecks by the Common Cuckoo 252

Fig. 3. Cumulative number of pecks by hatching embryos of the Common Cuckoo ( ) and Great Reed Warbler ( ) as recorded at 30-min intervals. JOURNAL OF AVIAN BIOLOGY 32:3 (2001)

Table 1. Comparison of selected meristic traits (shown are means9 SD; sample sizes in brackets) between hatchling Common Cuckoos and Great Reed Warblers. The two species were compared by a two-tailed t-test. Characteristic

Cuckoo

Warbler

P

Nestling mass (g) Length (mm) Forearm (mm) Tarsus (mm) Bill length (mm) Egg tooth*: L1 (mm) L2 (mm) L3 (mm) L4 (mm)

2.5579 0.354 19.8279 0.810 9.4209 0.298 8.0579 0.621 7.8209 0.408

(16) (14) (14) (14) (14)

2.15990.314 19.844 90.639 7.74690.766 8.52690.697 7.714 90.465

(16) (15) (15) (15) (18)

0.002 0.951 0.000 0.068 0.505

1.03549 0.101 0.3179 0.051 0.2499 0.044 0.6229 0.082

(14) (14) (14) (14)

0.773 90.181 0.33890.215 0.222 90.093 0.51590.148

(18) (18) (18) (18)

0.000 0.736 0.327 0.018

* For measurements see Fig. 1.

was 4.06 91.27 mm and 3.76 91.05 mm (n = 6), respectively.

Comparison of hatching-related mechanisms Although the Common Cuckoo eggs had slightly smaller volume (3.0699 0.306 ml; n=22) than the Great Reed Warbler eggs (3.10590.172 ml; n= 25), the difference between the two species was not statistically significant (t-test: t= −0.505, d.f.=45, P \ 0.05). In both species, egg volume was highly significantly positively correlated with hatchling mass (Great Reed Warbler: rs =0.678, n=17, P= 0.008; Common Cuckoo: rs =0.635, n= 18, P=0.007; Table 1, Fig. 4). A comparison of eggshell thickness showed that Common Cuckoo eggs had thicker shells (0.11890.009 mm; n=22) than those of Great Reed Warblers (0.0959 0.007 mm, n =19; t-test: t= 8.44, d.f.=39, PB 0.001). A comparison of hatchling characteristics showed that Common Cuckoos and Great Reed Warblers differed statistically significantly in four out of nine characteristics that we examined (Table 1). More specifically, the hatchling cuckoos were heavier, had longer forearms and egg tooth (designated as L4, Fig. 1), and their egg tooth was placed farther from the tip of the beak (or closer to forehead; see L1 on Fig. 1). Hatchlings of the two species did not differ in the remaining two egg tooth parameters and there was no significant difference in tarsus length between the two species (Table 1).

Great Reed Warbler chicks (resulting in a significantly longer hatching period in the Common Cuckoo chicks), and (2) the total number of pecks was greater for Common Cuckoos than for Great Reed Warblers. However, the rates of pecking (the number of pecks per unit of time) and the temporal pecking patterns (i.e. changes in pecking frequency during the course of the hatching period) were similar. A possible explanation for the similar temporal hatching patterns is that similar pecking behaviour among species may reflect common ancestry. This view is supported by the fact that the pecking rates of the two species we studied (36 pecks/min and 37 pecks/min for Cuckoo and Reed warbler, respectively) are relatively similar to the rates of bill-clapping (i.e. rapid opening of the upper and lower mandibles) by duck embryos (27 pecks/min) reported by Oppenheim (1972). These observations also

Discussion Pecking effort We predicted that, in comparison with Great Reed Warblers which lay structurally weaker eggs, Common Cuckoos should experience greater difficulty in hatching. This prediction was supported by the following observations: (1) relative to emergence, the Common Cuckoo chicks started pecking much earlier than the JOURNAL OF AVIAN BIOLOGY 32:3 (2001)

Fig. 4. Relationship between egg volume (ml) and nestling mass of the Common Cuckoo ( ) and Great Reed Warbler ( ).

253

suggest that there are physiological constraints on the frequency of egg pecking by avian embryos. Therefore, the greater hatching difficulty is more likely to be reflected in changes in the duration of hatching and thus in the total number of pecks rather than in the pecking rate.

Eggshell opening After an avian embryo has made the first pip-crack in the eggshell, there is a relatively long interval until it begins the final stage of hatching (Abbott and Craig 1960, Oppenheim 1970). Oppenheim (1972) found that the time interval between pipping and emergence varied among species from 15 h in the House Wren Troglodytes aedon to almost 41 h in the Northern Bobwhite Colinus 6irginianus. However, our results suggest that the time between pipping and emergence is shorter in the two species we studied. The observed variation among species could at least partly be caused by size-related differences in egg size and incubation period, variation in eggshell strength, and different conditions during hatching. For example, the discrepancy between our and Oppenheim’s (1972) results could be explained by different humidity conditions in the two studies of hatching. While we kept relative humidity during the final stage of hatching between 75 and 80%, Oppenheim (1972) hatched chicks at 55–65%. In any case, our results support our prediction that Common Cuckoos must use more effort to hatch than do Great Reed Warblers. Furthermore, the prediction that Common Cuckoo chicks should experience greater difficulty in hatching was supported by the observation that five (38.5%) chicks made an additional crack to the first one. In contrast, additional cracks were not made by Great Reed Warbler embryos. According to Oppenheim (1972), after avian embryos have made the initial crack in the shell, they usually do not make any additional cracks until emergence. Despite Oppenheim’s finding (1972) that pre-hatching behaviour and hatching behaviour are similar in species representing all forms of birds (precocial, semi-altricial, altricial), some species appear to have a characteristic pattern of hatching as shown by the condition of their eggshells after hatching. For example, American Woodcock Scolopax minor and Willet Catoptrophorus semipalmatus split eggshell longitudinally, ripping open the seam rather than breaking the shell into pieces (Wetherbee and Barlett 1962). Furthermore, the Australian Brushturkey Alectura lathami exhibits an unusual hatching pattern; for example, artificially incubated eggs fail to hatch if they are exposed to light. Further, instead of simply removing a shell cap, the chick breaks the entire shell into small pieces (Baltin 1969). Finally, in the Ostrich Struthio camelus, chicks essentially explode out of their strong-shelled eggs shattering the shell into many pieces (Sauer and Sauer 1966). 254

The hatching pattern of the Great Reed Warbler chicks conforms to the general hatching pattern described by Gill (1994) for most birds. However, the Common Cuckoo chicks exhibit a completely different pattern of breaking the shell, and we suggest that this could be the result of selection favouring adaptations facilitating hatching from structurally unusually strong eggs.

Do Common Cuckoos possess any adaptations facilitating hatching? The second goal of our study was to examine selected structural characteristics of Common Cuckoo and Great Reed Warbler hatchlings and establish if any of those in the Cuckoo could facilitate hatching from the species’ strong eggs. Our analyses demonstrated that, immediately after hatching, the Common Cuckoo chicks are heavier, have longer forearms and somewhat longer tarsi than the Great Reed Warbler chicks, despite the fact that Common Cuckoo egg is somewhat (although not significantly) smaller. This, along with the fact that the incubation period of brood parasites is unusually short (O’Connor 1984), suggests that embryonic development is faster in this species than in the Great Reed Warbler. The importance of the egg tooth for hatching has already been discussed by Clark (1961), and in our study we concentrated on the size and placement of this structure on the upper mandible. We found that the egg tooth had a longer cutting edge in Common Cuckoo chicks than in Great Reed Warblers, but the two species did not differ in the other two egg tooth parameters (height and length of the front edge). We hypothesize that (1) the placement of the egg tooth closer to the forehead in the cuckoo increases the amount of pressure that the chick can exert on the shell through pecking (due to increased leverage) and (2) that the longer upper edge of the egg tooth presents a longer cutting blade, which should also facilitate hatching from structurally strong eggs. It is known that inorganic constituents of avian eggshells (mainly calcium carbonate) have higher specific mass than the organic constituents (matrix proteins and polysaccharides) and that they play a dominant role in determining eggshells strength (Romanoff and Romanoff 1949, Burley and Vadehra 1989). Since eggs of brood parasites are structurally much stronger than those of their hosts (Spaw and Rohwer 1987, Picman 1989a, J. Picman and M. Honza unpubl. data), their shells presumably contain greater amounts of calcium carbonate than would be expected for their size. This is supported by data showing a significantly greater thickness of eggshells of the Brown-headed Cowbird Molothrus ater (Picman 1989a) and a somewhat, although not significantly, greater thickness of shells of JOURNAL OF AVIAN BIOLOGY 32:3 (2001)

the Common Cuckoo (J. Picman and M. Honza unpubl. data). Furthermore, Common Cuckoo eggs have shells of unusually high density (Picman and Pribil 1997, J. Picman and M. Honza unpubl. data). These findings suggest that brood parasites must ingest large quantities of calcium to be able to form eggshells. Because much of the calcium that is needed for embryonic development (especially for the formation of skeleton) is derived from the eggshell, the rapid development may play an important role in a specialized pattern of decalcification that may, in turn, facilitate hatching. Acknowledgements – We are grateful to Anne Bekoff and Michael Brooker and their valuable comments to an earlier version of this manuscripts. This study was supported by a Grant no. A 6087801 awarded by the Czech Academy of Sciences to M. Honza, and by a NSERC Research Grant awarded to J. Picman.

References Abbott, U. K. and Craig, R. M. 1960. Observations on hatching time in three avian species. – Poultry Sci. 39: 827 – 830. Ar, A., Rahn, H. and Paganelli, C. V. 1979. The avian egg: mass and strength. – Condor 81: 331– 337. Baltin, S. 1969. Zur Biologie und Ethologie des TalegallaHuhns (Alectura lathami Gray) unter besonderer Beru¨ cksichtigung des Verhaltens wa¨ hrend der Brutperiode. – Z. Tierpsychol. 26: 524–572. Brooker, M. G. and Brooker, L. C. 1991. Egg-shell strength in Cuckoos and Cowbirds. – Ibis 133: 406–413. Burley, R. W. and Vadehra, D. V. 1989. The Avian Egg. Chemistry and Biology. – John Wiley and Sons, New York. Clark, G. A. Jr. 1961. Occurrence and timing of egg teeth in birds. – Wilson Bull. 73: 268–278. Cramp, S. and Simmons, K. E. L. (eds). 1985. The Birds of the Western Palearctic. Vol. 4. – Oxford University Press, Oxford. Davies, N. B. and Brooke, M. de L. 1989a. An experimental study of co-evolution between the Cuckoo, Cuculus canorus, and its hosts. I. Host egg discrimination. – J. Anim. Ecol. 58: 207–224. Davies, N. B. and Brooke, M. de L. 1989b. An experimental study of co-evolution between the Cuckoo Cuculus canorus, and its hosts. II. Host egg markings, chick discrimination and general discussion. – J. Anim. Ecol. 58: 225– 236. Gill, F. B. 1994. Ornithology. 2nd ed. – W. H. Freeman and Company, New York. Hamburger, V. and Oppenheim, R. 1967. Pre-hatching motility and hatching behaviour in the chick. – J. Exp. Zool. 162: 133– 160. Hays, H. and Lecroy, M. 1971. Field criteria for determining incubation stage in eggs of the Common Tern. – Wilson Bull. 83: 425–429.

JOURNAL OF AVIAN BIOLOGY 32:3 (2001)

Hoyt, O. F. 1979. Practical methods of estimating volume and fresh weight of bird eggs. – Auk 96: 73 – 79. Lack, D. 1968. Ecological Adaptations for Breeding in Birds. – Methuen & Co, London. McMaster, D. G. and Sealy, S. G. 1998. Short incubation periods of Brown-headed Cowbirds: How do Cowbird eggs hatch before Yellow Warbler eggs? – Condor 100: 102– 111. Moksnes, A. and Røskaft, E. 1995. Egg morphs and host preference in the common Cuckoo (Cuculus canorus): an analysis of Cuckoo and host eggs from European museum collections. – J. Zool. Lond. 236: 625– 648. O’Connor, R. J. 1984. The Growth and Development of Birds. – John Wiley and Sons, New York. Øien, I. J. E., Moksnes, A. and Røskaft, E. 1995. Evolution of variation in egg colour and marking pattern in European passerines: adaptations in a co-evolutionary arms race with the Cuckoo, Cuculus canorus. – Behav. Ecol. 6: 166– 174. Øien, I. J. E., Moksnes, A., Røskaft, E. and Honza, M. 1998. Costs of Cuckoo Cuculus canorus parasitism to Reed Warblers Acrocephalus scirpaceus. – J. Avian Biol. 29: 209– 215. Oppenheim, R. W. 1968. Light responsivity in chick and duck embryos just prior to hatching. – Anim. Behav. 16: 276– 280. Oppenheim, R. W. 1970. Some aspects of embryonic behaviour in the Duck (Anas platyrhynchos). – Anim. Behav. 18: 335 – 352. Oppenheim, R. W. 1972. Prehatching and hatching behaviour in birds: a comparative study of altricial and precocial species. – Anim. Behav. 20: 644– 655. Pettingill, O. S. Jr. 1970. Ornithology in Laboratory and Field. – Burgess Publishing Company, Minneapolis. Picman, J. 1989a. Mechanism of increased puncture resistance of eggs of Brown-headed Cowbirds. – Auk 106: 577 – 583. Picman, J. 1989b. An inexpensive device for measuring puncture resistance of eggs. – J. Field Ornithol. 60: 190– 196. Picman, J. and Pribil, S. 1997. Is greater eggshell density an alternative mechanism by which parasitic Cuckoos increase the strength of their eggs? – J. Ornithol. 138: 531– 541. Picman, J., Pribil, S. and Picman, K. 1996. The effect of intraspecific egg destruction on the strength of Marsh Wren eggs. – Auk 113: 599– 607. Romanoff, A. L. and Romanoff, A. J. 1949. The Avian Egg. – John Wiley and Sons, New York. Sauer, E. G. F. and Sauer, E. M. 1966. The behaviour and ecology of the South African Ostrich. – Living Bird 5: 45 – 75. Spaw, C. D. and Rohwer, S. 1987. A comparative study of eggshell thickness in cowbirds and other passerines. – Condor 89: 307– 318. Vince, M. 1969. Embryonic communication, respiration, and the synchronization of hatching. – In: Hinde, R. A. (ed.). Bird Vocalizations. Cambridge Univ. Press, Cambridge, pp. 233– 260. Wetherbee, D. K. and Barlett, L. M. 1962. Egg teeth and shell rupture of the American Woodcock. – Auk 79: 117. Wyllie, I. 1981. The Cuckoo. – Batsford, London. (Recei6ed 24 July 2000, re6ised 21 December 2000, accepted 12 January 2001.)

255

23. Grim T.: Publication and citation biases: a possible role of social activities? 23. (subm.) Grim T.: Publication and citation biases: a possible role of social activities? (subm.)

Explaining publication output among scientists: a possible role of social activities? Tomáš Grim1 1 Department of Zoology Palacký University tř. Svobody 26 771 46 Olomouc Czech Republic e-mail: [email protected]

Abstract Publication success is a crucial aspect in both the practice and evaluation of scientific work. Unsurprisingly, publication and citation biases received increasing attention recently. Despite a plethora of papers on the issue no study so far considered a possible effect of social activities on publication output. Here I apply the well known “trade-off” principle on scientific productivity. In theory, any investment (time, energy) to leisure time social activities should be traded-off against investment to scientific productivity. One of the most frequent leisure time activities in the world and especially in Europe is beer drinking. Using data from the Czech Republic, that has the highest per capita beer consumption rate in the world, I show that with increasing per capita beer consumption come lower numbers of papers, total citations, and citations per paper (a surrogate measure of paper quality). All trends were consistent across analyses where either the length of publication activity or the researchers’ age was controlled for. In addition I found all these predicted trends in the comparison of two separate geographic areas within the Czech Republic that are also known to differ in beer consumption rates. The results indicate that leisure time social activities could influence the quality of scientific work and may be potential sources of publication and citation biases. Keywords: publication bias; trade-off; social activities

1

Publication success (i.e. number and citation rate of papers) is widely used in assessments of the academic performance at all levels from individual scientists to nations (Cassey and Blackburn 2004, Leimu and Koricheva 2005). Clearly, both publication and citation success are influenced by variety of factors, e.g. statistical (non)significance of results, number of coauthors, nationality or scientific field (Møller and Jennions 2002, Leimu and Koricheva 2005, Wong and Kokko 2005). However, to my knowledge no study so far paid attention to other factors less tightly connected to publication process itself. Trade-offs are a crucial concept on which behavioral ecology in general (Krebs and Davies 1993) and life-history theory in particular (Stearns 1992) is based. Here I apply the idea of trade-offs to scientific productivity. Obviously, the limiting resources – time and energy – invested into any activity, such as leisure activities, must be traded off against time and energy invested into any other activity, such as scientific work. One of the most common leisure time activities in the world, especially in Europe, is beer drinking, so it seems logical to expect a trade-off between beer drinking and publication output (A. P. Møller, personal communication). The Czech Republic, with the highest consumption of beer per capita in the world (160 liters per year; Guinness book of records, data from 1999, web version), seems to be a good candidate to test this hypothesis. I surveyed all researchers studying avian evolutionary biology and behavioral ecology, my own discipline, in the Czech Republic who published at least one paper in a peer-reviewed journal listed by the Web of Science (http://portal17.isiknowledge.com/portal.cgi) and published outside the Czech Republic during the last two decades. I asked how many glasses or bottles of beer they drank per week and recalculated this value to estimate annual consumption in liters. In addition, I collated data on their year of birth to control for age effects. I collated the data in May 2002 and repeated the study four years later in March 2006 with the same subjects when available. Beer consumption appeared to be highly consistent with advancing years (linear regression of 2006 against 2002 consumption: R2 = 0.90, F1,9 = 61.52, P < 0.0001). I conducted three separate analyses. First, I analyzed data from the first census in 2002 (n = 18). Second, I reanalyzed an updated publication data set for this original group of researchers in 2006 using the 2002 beer consumption data (n = 18) to test for the temporal consistency of observed relationships. Third, I collated data on new researchers who started to publish after the first census in 2002 and analyzed the total data set (n = 34). I found that the amount of beer consumed per year significantly and negatively correlated with the total number of papers published, with the total number of citations received and with the average number of citations received per paper (a surrogate measure of paper quality) (Table 1, Figure 1). The results were remarkably consistent when I controlled for either (i) the length of publication activity, i.e. the period from publishing the first paper until the date of my research, or (ii) the researchers’ age (Table 1). Moreover, majority of trends was consistently significant in all three separate analyses (Table 1). The non-significance of age-corrected measures in the final 2006 data set (Table 1c) is most likely caused by the inclusion of new (i.e. post-2002) researchers whose data are less representative due to the short period of their publication activity. Generally, inhabitants of Bohemia (western part of the Czech Republic) are known to drink more beer than people from Moravia (eastern part of the country). This difference was confirmed for my sample of researchers: researchers from Bohemia drank significantly more beer per capita per year (median 200.0 liters) than those from Moravia (median 37.5 liters; Mann-Whitney test: U17,17 = –2.84, P = 0.005). Therefore I predicted lower measures of publication output for the former in comparison to latter group of researchers. Indeed, researchers from Bohemia tended to publish fewer papers per year (U17,17 = 1.87, P = 0.06), were less cited per year (U17,17 = 2.99, P = 0.003), and showed lower citation rate per paper per year (U17,17 = 2.30, P = 0.02). When controlling for researchers’ age (instead of the length of publication activity), the results were qualitatively similar for the number of papers (U17,17 = 3.29, P = 0.001), citations (U17,17 = 2.95, P = 0.003) and citations per paper (U17,17 = 2.61, P = 0.009). Moreover, the results remained qualitatively similar for the original subset of researchers both in 2002 and 2006 (Bohemia n = 6, Moravia n = 12; results not shown; all 12 tests with P values: 0.007 < P < 0.09). Bohemian and Moravian researchers did not differ in the average age when they started to publish (median: 28 vs. 27 years; U17,17 = –0.95, P = 0.33). Moreover, there is no evidence for discrimination in funding support against researchers from Bohemia (personal inquiry at the Grant Agency of the Czech Republic).

2

Thus, beer drinking appears to negatively influence publication output not only at individual but also at regional level. Although the current study is based on a limited sample (i.e. researchers from a single country focusing on a single scientific discipline) it is important to stress that the majority of evolutionary biology and behavioral ecology studies are also based on data from a single population of a single study species. In fact, my single country approach is advantageous in comparison to some studies that pooled data from various states (e.g. Wong and Kokko 2005) as it cannot in principle be confounded by differences among countries (cultural, nutritional, or funding-related). These may be especially important when studying social behaviour of humans (Buss 2004). In addition, I was able to get valid responses for my analyses from most Czech avian ecologists publishing in foreign journals during the study period (34 out of 38 researchers). This 89% respondent success is noticeably higher than in other studies of publication success (e.g. 40% in Cassey and Blackburn 2004). These results indicate that considering leisure time social activities (e.g. beer drinking) could be useful for understanding a quality of scientific work and potential sources of publication and citation biases. Importantly, the principle of trade-offs seems to be applicable not only to the realm of life-history traits of organisms but also to the social life of humans. Acknowledgements I am grateful to A. P. Møller for our discussion which resulted in the formulation of “beerpublications trade-off” hypothesis. I thank M. E. Hauber, M. Krist, R. Leimu, A. Moksnes, V. Remeš, E. Røskaft and many others for discussing the exciting results with me and encouraging me to publish them. References Buss DM 2004. Evolutionary Psychology: The New Science of the Mind. Boston: Allyn & Bacon. Cassey P, Blackburn TM. 2004. Publication and rejection among successful ecologists. BioScience 54: 234–239. Krebs JR, Davies NB. 1993. An introduction to behavioural ecology. Oxford: Blackwell. Leimu R, Koricheva J. 2005. What determines the citation frequency of ecological papers? Trends in Ecology and Evolution 20: 28–32. Møller AP, Jennions MD. 2002. Testing and adjusting for publication bias. Trends in Ecology and Evolution 16: 580–586. Stearns SC. 1992. The evolution of life-histories. Oxford: Oxford University Press. Wong BBM, Kokko H. 2005. Is science as global as we think? Trends in Ecology and Evolution 20: 745–476.

3

Table 1 Correlations between beer consumption per capita per year in liters and several measures of publication output of Czech researchers studying avian evolutionary biology and behavioral ecology: a) correlations based on data collected in 2002, b) correlations for the same group of subjects as in a) re-analyzed four years later, c) correlations for the whole data set obtained in 2006 (“old” and “new” researchers pooled). In each of the three data sets the “per year” measures are corrected by the length of publication activity (i.e. the number of years elapsed since the first publication up to the time of collection data for the analyses) and “age” measures are corrected by the age of researchers at the time of the two respective censuses. a) 2002 dataset (n = 18) Papers per year Citations per year Citations per paper per year Papers / age Citations / age Citations per paper / age

rs –0.70 –0.63 –0.47 –0.54 –0.55 –0.40

P 0.001 0.006 0.049 0.02 0.019 0.10

b) 2002 subset in 2006 (n = 18) Papers per year Citations per year Citations per paper per year Papers / age Citations / age Citations per paper / age

–0.68 –0.70 –0.44 –0.61 –0.66 –0.48

0.002 0.001 0.07 0.007 0.003 0.04

c) 2006 dataset (n = 34) Papers per year Citations per year Citations per paper per year Papers / age Citations / age Citations per paper / age

–0.55 –0.40 –0.35 –0.34 –0.22 –0.17

0.0008 0.02 0.04 0.048 0.21 0.35

Figure 1 Number of publications per capita per year published by Czech avian ecologists up to 2006 plotted against their beer consumption per capita per year in liters. Both data sets shown are Box-Cox transformed (thus neither the output score nor the consumption score values enable the identification of particular persons included in this research). The negative relationship between beer consumption and publication success is significant not only for the whole data set (rs = –0.55, n = 34, P = 0.0008) but also for “old” (included in the first survey in 2002; z) and “new” researchers (included in 2006; {) analyzed separately (“old”: rs = –0.68, n = 18, P = 0.002; “new”: rs = –0.52, n = 16, P = 0.04).

Publication output score

2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 1

2

3

4

5

Beer consumption score

4

6

24. Grim T., Røskaft E., Moksnes A., Stokke B. G., Honza M., Moskat C., Kleven O. & 24. in parasitic birds: why are thrushes not Rudolfsen G.: Constraints on host choice parasitized by cuckoos? Grim T., Røskaft E., Moksnes A., Stokke B. G., Honza M., Moskat C., Kleven O. & Rudolfsen G.: Constraints on host choice in parasitic birds: why are thrushes not (subm.) parasitized by cuckoos? (subm.)

Constraints on host choice in parasitic birds: why are thrushes not parasitized by cuckoos? Tomáš Grim1, Eivin Røskaft2, Arne Moksnes2, Bård G. Stokke2, Csaba Moskat3, Oddmund Kleven4, Marcel Honza5 1 Department of Zoology, Palacký University, Tr. Svobody 26, CZ–771 46 Olomouc, Czech Republic 2 Department of Biology, Norwegian University of Science and Technology, NTNU, N–7491 Trondheim, Norway 3 Animal Ecology Research Group of the Hungarian Academy of Sciences, c/o Hungarian Natural History Museum, H–1083 Budapest, Ludovika ter 2, Hungary 4 Zoological Museum, Natural History Museums and Botanical Garden, University of Oslo, P.O. Box 1172 Blindern, N-0318 Oslo, Norway 5 Institute of vertebrate Biology, Academy of Sciences of the Czech Republic, Kvetna 8, CZ– 603 65 Brno, Czech Republic

Abstract Brood parasites generally lay their eggs into nests of a wide range of host species. However, some potential hosts commonly occurring in sympatry with parasites are apparently avoided by them. Few attempts have been made to explain the absence of parasitism in potentially suitable hosts of brood parasites in general and of the cuckoo (Cuculus canorus) in particular. Therefore, we tested potential explanations for why thrushes Turdus spp. are not parasitized by the cuckoo despite breeding commonly in sympatry with the parasite and building conspicuous nests. We studied four European thrush species and compared their responses to parasitism and life-history traits with those of acceptors and rejecters of cuckoo parasitism. No single factor could alone explain the extremely low parasitism rates in thrushes. However, our data showed the importance of integrating various factors when explaining absence of parasitism in these potential host species. We found that 1) the antiparasitic behaviour of the host does not vary significantly between sympatric and allopatric populations, 2) thrushes do not recognise the cuckoo as a special threat, 3) cuckoo nestlings experience very low survival rates in some thrush nests, and 4) some thrushes show low tolerance towards alien eggs or nestlings and high non-specific aggression near their nests. In generally, life-history traits of thrushes are similar to those in unsuitable hosts (seed-eaters) and acceptors (which did not experience coevolutionary arms-race with the cuckoo) and differ from these traits in rejecters. These results coupled with previous findings (e.g. there are no specialized cuckoo gens parasitizing particular Turdus spp.) indicate that thrushes are poor quality hosts, were not regularly parasitized by cuckoos in the past and have not evolved specific adaptations against interspecific brood parasitism. Presumably, their egg discrimination behaviour evolved in response to conspecific brood parasitism, which was documented in all the four studied thrushes and other related species. We conclude that there has been no long-term co-evolutionary armsrace between the thrushes and the cuckoo. However, adaptations selected through conspecific brood parasitism may provide a preadapted protection against cuckoos. Key words: brood parasitism, coevolution, recognition, nest defence, host choice

1

Introduction Both generalist (e.g. brown-headed cowbird Molothrus ater) and specialist (e.g. European cuckoo Cuculus canorus) brood parasites lay their eggs in nests of a wide range of passerine species (Friedmann and Kiff 1985, Moksnes and Røskaft 1995). The eggs of a specialist cuckoo female are not laid randomly – some bird species are parasitized frequently, others rarely and some are not utilized at all (Moksnes and Røskaft 1995). Selection of a proper host is critical for the reproductive success of all brood parasites. It can be viewed as a two-level process whereby a selection by a parasite among potential host species is followed by a selection of particular individuals within the chosen host species. There is some evidence that brood parasites avoid low quality host species (Mason 1986b, Wiley 1988, Rothstein & Robinson 1998, Sackmann & Reboreda 2003) and that high quality individuals are parasitized more frequently than at random (Brown and Brown 1991, Soler et al. 1995, Poysa 2003; but see Grim 2002). At the initial stages in the evolution of parasitic habits in birds we expect individuals to start to lay their eggs primarily into the nests of other species according to their availability. This implies that species with the highest breeding densities or the most conspicuous nests will be utilized. In the more advanced stages of brood parasite-host coevolution we expect increasing specialization, i.e. avoidance of some hosts, by the brood parasite (e.g. parasitism by larger brood parasite on smaller hosts is more successful than vice versa; Slagsvold 1998; but see Sackmann and Reboreda 2003). The fact that particular cuckoo gentes are specialists, i.e. each cuckoo strain parasitizes only one or very few host species (Moksnes and Røskaft 1995, Gibbs et al. 2000), suggests a strong selection pressure on the host choice by the parasite. A low reproductive success in the nests of particular host species can result from various factors, e.g. strong rejection of alien eggs (Sealy and Bazin 1995), unsuitable food delivered to nestlings (Middleton 1977) or intense competition with host nestlings (e.g., Scott and Lemon; for review see Ortega 1998). Host body size may positively influence parasite nestling growth rate in evicting parasites (Kleven et al. 1999, but see Grim subm.) or parasitism of median-sized hosts may yield highest reproductive success for a non-evicting parasite (Kilner 2003, Kilner et al. 2004). Brood parasites may benefit by utilizing larger host species as fledgling size strongly positively influences survival rate (e.g. Magrath 1991). In fact, parasitism rate in a local area in the Czech Republic is higher in the great reed warbler Acrocephalus arundinaceus, which is a large host, compared to the smaller reed warbler Acrocephalus scirpaceus host (Moksnes et al. 1993, Kleven et al. 1999). Parasitism rates generally increase with host nest densities both within (Alvarez 2003) and across species (Soler et al. 1999a). However, the higher frequency of parasitism in great reed warblers found by Kleven et al. (1999) was not a byproduct of local host breeding densities, since the great reed warbler was more frequently parasitized despite being much less common than the reed warbler in the study area. Blackbird Turdus merula, song thrush T. philomelos, fieldfare T. pilaris and redwing T. iliacus breed in high densities in their respective habitats over the most of Europe (Cramp 1988, Hagemeijer and Blair 1997). All four species build large and open conspicuous nests that can easily be found by the cuckoo. Thrushes are a striking exception to the rule found by Soler et al. (1999a) where host population size was the best predictor of parasitism rate. For example, the blackbird is roughly as common as the six most frequently used cuckoo host species in Europe taken together (population size data from Hagemeijer and Blair 1997). However, the parasitism rate of the blackbird is more than 300 times lower than that among these common hosts as indicated by data from museum egg collections (Glue and Morgan 1972, Moksnes and Røskaft 1995). Available data from other species of thrushes lead to the same conclusion (Moksnes and Røskaft 1987, 1988, 1995). These negligible rates of parasitism are apparently not a by-product of a rapid ejection of foreign eggs, because experimentally parasitized thrushes do not reject eggs significantly faster than other species (Moksnes and Røskaft 1988, Moksnes et al. 1990, Grim and Honza 2001b). As there are observations of thrushes rearing a young cuckoo up till fledging (e.g. Glue and Morgan 1972) the big puzzle is why are thrushes almost never parasitized by the cuckoo despite their abundance, a relatively large body size and very poor nest concealment. Because of these pronounced differences thrushes provide an ideal model for the study of host choice by parasitic birds in general. Previous studies have indicated the importance of several factors for host choice in the cuckoo: host population size, the length of the nestling period, the length of the breeding

2

period, nest type, host antiparasitic behaviour and host nest-site selection (Moksnes and Røskaft 1992, Soler et al. 1999a, Røskaft et al. 2002b,c). Further, host breeding strategy (presence or absence of brood reduction strategy) was suggested as another important factor moulding host choice by brood parasites (Soler 2002). However, so far very little effort has been made in attempts to explain the absence (Peer and Bollinger 1997) or low rates of parasitism in potential host species of parasitic birds. Most of the species studied in this respect so far are parasitized infrequently but still orders of magnitude more than thrushes (e.g. Briskie et al. 1990: Empidonax minimus: 2.8%, Sealy and Bazin 1995: Tyrannus tyrannus: 0.2%, Mermoz and Fernandez 1999: Amblyramphus holosericeus: 14.2%). Hypotheses on host avoidance have been tested almost entirely among North-American cowbird hosts, whereas European cuckoo hosts have received marginal attention (Moksnes and Røskaft 1992). Most importantly, none of previous studies reached strong conclusive results, as usually most of the factors involved remained untested. However, it was shown that the common whitethroat (Procházka & Honza, 2003), blackcap (Honza et al. 2004), redbacked shrike (Lovászi & Moskát 2004) and chaffinch (Stokke et al. 2004) used to be parasitised by the cuckoo in the past. Cuckoos avoid these winners of the arms race because they retained highly developed egg discrimination ability. Here we review 16 factors that may play important roles in host choice by parasitic birds (see also Ortega 1998). Some of these factors can be outright rejected for thrushes and will not be discussed further (factors 14–16). We tested all the other 13 potential factors for the cuckoo avoidance of potential thrush hosts. 1. Breeding synchrony – Some hosts are avoided simply because their breeding cycles do not overlap or do so only weakly with those of the parasite. We compared laying dates of thrushes, currently parasitized passerines and the cuckoo. 2. Breeding density – Rare species can hardly be utilized as hosts, because the cuckoo female would have problems to find sufficient numbers of nests to parasitize. We compared European population sizes and breeding densities of thrushes and current common hosts. 3. Predation rates – High predation rates on host nests render parasitism less beneficial for the cuckoo. We tested whether predation rates of thrushes nests are higher than those of nests of current common hosts. 4. Nest defence – Intense nest defence by hosts may keep parasites away from the nest. This hypothesis predicts that the intensity of nest defence among thrushes should be higher than that among currently used hosts. Further, for any birds it is not adaptive to attack intruders near the nest that do not pose a threat to them (Grim 2005). Therefore, if thrushes have not been in a co-evolutionary arms race with the cuckoo they should show no aggression against the parasite. This second hypothesis predicts that thrushes should not respond to a stuffed cuckoo dummy near their nests, but should attack a crow (Corvus cornix), which over large parts of Europe acts as an effective nest predator. 5. Nest attentiveness – Low parasitism rates may result from the fact that parents guard their nests (Møller 1987). Combined effects of high attentiveness and non-specific aggression may therefore result in a very low frequency of parasitism (Mermoz and Fernandez 1999). “The latency to response” in nest defence experiments (see Methods) can be used as a rough measure of nest attentiveness. We tested whether thrushes show shorter latency (i.e. higher nest attentiveness) than some commonly used cuckoo hosts. Due to the lack of parasitism in our model species the direct effect of host nest attentiveness on parasitism frequencies cannot be tested. 6. Egg rejection – Cuckoo females should avoid parasitism of strong egg rejecters. This hypothesis predicts that egg rejection rates in thrushes are higher than rejection rates of current common hosts. 7. Short incubation period – Hatching earlier than host eggs is critical for a parasite, because late hatching can impair the eviction of large host nestlings. Thus, hatching success of parasitic eggs in the nests of some passerines could be low because the host’s incubation period is shorter than that of the parasite. Hatching later than host nestlings can be especially detrimental for cuckoos in thrush nests, because thrush nestlings develop very quickly (Lübcke and Furrer 1985, Carlson and Moreno 1986, Cramp 1988). We compared the lengths of incubation periods between thrushes and currently used cuckoo fosterers. 8. Host egg size – incubation and egg eviction constraints – Smaller eggs in the presence of larger eggs suffer from inefficient incubation (Wood and Bollinger 1997) and clutch size has an effect on egg hatchability (Lerkelund et al. 1993). Therefore, large host eggs can cause

3

insufficient incubation of parasitic eggs. This may lead to two outcomes: 1) a cuckoo egg does not hatch at all or 2) the parasitic egg hatches later than those of the host (see Hypothesis 11). This hypothesis predicts that 1) thrush eggs should be significantly larger than eggs of current common hosts and 2) cuckoo eggs should show low hatchability in thrush nests. In addition, it is predicted that the cuckoo chick would not be able to evict host eggs or nestlings due to their big size and would have to compete with the host chicks for food. 9. Nest size/shape – Host eggs cannot be evicted from cavities (Rutila et al. 2002) and only with increased effort from too deep or big open nests (Nakamura 1990). We tested the ability of cuckoo nestlings to evict host eggs from relatively deep thrush nests. We also compared the size of thrush nests with that of the great reed warbler that presently is one of the largest hosts in Europe. To disentangle between the effects of nest sizes and egg sizes on the eviction success we tested for eviction of large thrush and small reed warblers eggs from both large thrush and small reed warbler nests. 10. Food type/amount – Some host species, as e.g. the greenfinch Carduelis chloris and linnet C. cannabina are avoided by the cuckoo because they feed their nestlings with seeds, which is an unsuitable diet for the insectivorous cuckoo nestling (Glue and Morgan 1972, Davies and Brooke 1989a, Alvarez 1994b). Thrushes feed their offspring with a variety of non-insect diet items, including molluscs (song thrush) and earthworms (blackbird, fieldfare) at relatively high percentages (Simms 1978, Lübcke and Furrer 1985, Cramp 1988, Norman 1994). We reviewed published data on host and parasite nestling diet and tested if there are significant differences in nestling diet composition between thrushes and current common hosts. Further, we tested the hypothesis experimentally. We predicted that cuckoo nestlings should show 1) significantly lower growth rates and/or 2) lower survival rates (excluding the effect of predation) in thrush nests compared to these measures in the nests of two common hosts (the reed warbler and great reed warbler). 11. Competition with host chicks – Competition with large host nestlings may severely constrain the survival of brown-headed cowbird nestlings (Manson 1986, Røskaft et al. 1990, Kilner 2003). The presence of host nestlings may also decrease the cuckoo breeding success if the parasite is not successful in evicting all host progeny (Rutila et al 2002). Competition with quickly growing thrush nestlings (Lübcke and Furrer 1985, Cramp 1988) could therefore reduce the survival probability of the parasitic nestling. We therefore tested for differences in cuckoo survival rates between two experimental groups: nestling cuckoo raised alone in a thrush nest and nestling cuckoo raised together with host nestlings in the same nest. 12. Chick rejection – Some hosts are able to recognise and successfully discriminate against alien chicks (Fraga 1998, Lichtenstein 2001, Langmore et al. 2003). However, this behaviour is believed to be generally absent among hosts of parasitic birds in Europe (Davies and Brooke 1988, 1989a; but see Grim et al. 2003, Grim subm.). We observed host behaviour at nests with cross-fostered cuckoo nestlings to find whether hosts refuse to feed parasitic nestlings (which could explain cuckoo nestlings low survival apart from indigestible food brought by fosterers). 13. Host breeding strategy – Soler (2002) predicted that a host decision to adjust its reproductive effort at the stage of eggs vs. nestlings may have important consequences on parasite chick’s success. Some hosts adjust their breeding effort at the stage of eggs, preferentially feed smaller chicks and all hatchlings fledge as a rule (clutch adjusters). Other potential hosts adjust their breeding effort post-hatch, feed preferentially larger nestlings and not all hatchlings fledge (brood reducers). If the parasitic chick is smaller than host chicks (as would be a case in thrushes) it should thrive in the nests of clutch adjusters but not in the nests of brood reducers. We compared host breeding strategy between current hosts and thrushes. 14. Breeding habitat – Cuckoo females can successfully breed only in areas where there are vantage points in trees (Øien et al. 1996, Røskaft et al. 2002b). Therefore many passerines breeding in open areas escape parasitism through their habitat selection. This explanation cannot hold for any species of thrushes, as all thrushes breed in wooded areas. 15. Nest type – cuckoos are unable to parasitize hole nesters successfully (but see Rutila et al. 2002). This explanation can be rejected because all thrushes are open-nesters. 16. Nest concealment – Nest-site selection could work as an effective antiparasitic defence (e.g. Briskie et al. 1990, Moskát & Honza 2000, Grim 2002) and some passerines with well-

4

hidden nests may, at least partly, avoid being parasitized. This hypothesis can be rejected because thrushes nests are by far the most conspicuous and easily found in forest habitats. Previous research have shown that thrushes show some kind of antiparasitic behaviour (Davies and Brooke 1989a, Moksnes et al. 1990, Meilvang et al. 1997, Grendstad et al 1999, Grim and Honza 2001b, Moskát et al. 2003a). Although a well-developed nest cleaning behaviour might contribute to antiparasite defence (Moskát et al. 2003b), antiparasite behaviour in general can be explained by two different hypotheses – interspecific vs. conspecific brood parasitism hypothesis. Their predictions are reviewed in Table 1. Theoretically, in the arms-race between brood parasites and their hosts the latter can evolve a lowered intraclutch-variation and a higher interclutch variation in response to the mimetic eggs laid by the brood parasite (Davies & Brooke 1989, Øien et al. 1995, Soler & Møller 1996, Stokke et al. 2002). Because there are no indications that thrushes have been utilized as hosts by a brood parasite laying mimetic eggs, we predict that the intraclutch variation should be higher and interclutch variation should be lower in thrushes than in species in which a cuckoo egg-morph exist or in species that show a high rejection rate of parasitic eggs and a high level of aggression towards the cuckoo. A failure to find such a pattern could either indicate that thrushes have been utilized by a brood parasite laying mimetic eggs in the past and retained this behaviour (Rothstein 2001) or alternatively that conspecific brood parasitism has led to the evolution of these traits. Material and methods We studied responses of thrushes in 20 localities in Northern and Central Europe (Table 2; see Moksnes and Røskaft 1987, 1989, 1992, Moksnes et al. 1990, 1993, Grim and Honza 2001b, Moskát et al. 2003a for descriptions of these areas). Thrushes breed in sympatry with the cuckoo in thirteen of these areas, while seven studied populations are not in contact with the parasite during the breeding period. Because of sample sizes limitations we pooled data from some geographically close localities with the same sympatry/allopatry status. Thus, we compared host responses in 6 sympatric and 6 allopatric populations. The host population was considered sympatric when the cuckoo was breeding in that particular area. Other, mainly town, populations were considered allopatric. Blackbird populations in towns are generally non-migratory and show very high philopatry (e.g. Formánek 1958, Havlín 1963, Lack 1966, Karlsson and Källander 1977, Desrochers and Magrath 1993, Móra et al. 1998, Streif and Rasa 2001). Because the cuckoo avoids towns, these populations can safely be regarded as allopatric. Moreover, our own data from Hungary show very high site fidelity in both town and countryside blackbird populations (96.0% of recoveries in the city for birds ringed in Hungarian towns, T. Csörgö unpublished; 87.9% of recoveries in countryside for birds ringed there Z. Karcza unpublished). Fieldfares can also show high fidelity to their natal areas (84.9% recoveries within 6 km from original nest; Norman 1994). The start of egg laying in nests found during incubation was determined by assuming that the incubation period lasts 13 days in the redwing and song thrush and 14 days in the fieldfare and blackbird (Cramp 1988). The length and width of all host eggs were measured to the nearest 0.1 mm with a calliper and the egg volume was calculated according to the formula: 0.51*length*breadth2 (Hoyt 1979). After clutch completion the host clutch was photographed on a Kodak grey card. Intraclutch and interclutch variation was determined on a scale from 1 (low) to 5 (high) by four experienced persons (for details see Øien et al. 1995). The mean value of the scores was used in the further analyses. Assessments of intra- and interclutch variability showed high repeatability among observers (Øien et al. 1995). The human and avian colour visions show some differences (e.g. sensitivity to UV wavelengths; Cherry and Bennett 2001). Thus, measurements of UV reflectance would perhaps improve the present study. Unfortunately, intraclutch variation evaluation and painting of mimetic and nonmimetic model eggs based on human standards were done before the study of Cherry and Bennett (2001) was published. As birds respond differently to mimetic and nonmimetic eggs defined by human standards (e.g. Davies and Brooke 1988, 1989a, Moksnes 1992, Moksnes and Røskaft 1992, Stokke et al 1999, Bartol et al. 2002) we believe that the method we used is still satisfactory. Further, a recent study of another

5

European host passerine found no effect of UV-light on host decision to reject or accept the parasitic egg (Aviles et al. 2006). We measured the following nest variables: inner and outer diameters (both measured twice, crosswise), inside depth, outside height (all in mm) and height of the nest above ground (in cm). Nest volume was calculated according to formula: 4/3*π*a*b2 (a=larger radius, b=smaller radius of the nest). The index of nest cup steepness was calculated as the nest cup inside depth divided by the nest cup inner width. We predicted that cuckoo chicks will show lower eviction success in deeper and/or steeper nests. Information on breeding season, breeding density, breeding success, clutch size and egg size in thrushes and 12 common hosts were obtained from our own unpublished data from studied populations. We also analysed data from the literature (Arheimer 1978, Hudec 1983, Lübcke and Furrer 1985, Carlson and Moreno 1986, Hatchwell et al. 1996, Cramp 1988) to include information from more geographical areas. Data about nestling diet were extracted from the literature (Arheimer 1978, Török 1981, Lübcke and Furrer 1985, Török 1985, Carlson and Moreno 1986, Cramp 1988, Török and Ludvig 1988, Kristin 1992, Chamberlain et al. 1999, Grim and Honza 1997, 2001a). We included only data obtained by neck-collar method (Grim and Honza 2001a), which is the most precise nestling food sampling method (e.g. analysis of stomachs did not reveal earthworms in the diets of blackbirds and song thrushes despite the fact that earthworms are dominant part of the food for nestlings of both thrushes, see review Table 6 in Török 1985). Data from different studies were averaged. Egg experiments We tested rejection abilities of thrushes with model eggs made of plaster of Paris or hard plastic and painted 1) blue to mimic eggs of the cuckoo gens parasitizing the redstart Phoenicurus phoenicurus (Moksnes and Røskaft 1995) and 2) brown resembling those of the meadow pipit Anthus pratensis (Moksnes et al. 1990). Blue redstart type models were used as mimetic for the song thrush and as nonmimetic for the other three species. Brown meadow pipit type models were used as nonmimetic for all species. However, these meadow pipit type eggs were less contrasting to the fieldfare, redwing and blackbird eggs than to the song thrush eggs than was the blue model egg. The parasitic egg was introduced to the host nest during the egg-laying or incubation stages. We tested for the possible effect of breeding stage by comparing rejection rates between four time categories: egg laying, 1–3 days of incubation, 4–9 days of incubation, and 10 days of incubation to hatching. We fitted regression models with “breeding stage”, “sympatry/allopatry” and their interaction as factors and the host reaction to the parasitic egg (accepted/rejected) as a response separately for each combination of host species and egg types. When model was insignificant we removed the interaction and fitted a reduced model. Only in fieldfares the breeding stage significantly influenced their responses towards parasitic eggs (55.7% rejection in laying and early incubation, 0% rejection later, χ2=11.28, df=1, P=0.01), therefore we excluded the data from the nests parasitised after early incubation. As we found no effects of the breeding stage on rejection rates of model eggs both in sympatry and allopatry we pooled the data in subsequent analyses in the other three species. One host egg was removed in some of the experiments while in most of the experiments no host egg was removed. However, this did not influence our results (Moksnes et al. 1993; see also Davies and Brooke 1989a). An effort was made to monitor experimental nests daily during a 6-day-period following the experimental parasitism. This period was regarded sufficient to determine if the host rejects or accepts parasitic eggs (Davies and Brooke 1989, Moksnes et al. 1990). Nests predated before the 6-day-period finished were excluded from the further analyses. We scored three kinds of responses: acceptance, ejection and desertion. Desertion can be considered as a rejection because 1) parasitized nests are deserted more frequently than unparasitized nests (Grim and Honza 2001b) and 2) hosts tested with nonmimetic eggs desert more often than those parasitized with mimetic models (Davies and Brooke 1989a). Dummy experiments Responses of thrushes and control species to adult brood parasites were tested using a stuffed cuckoo dummy and a hooded crow dummy as a control. The dummy was placed

6

between 0.5 and 1.0 m from the focal nest. The responses of nest owners were observed for 10 min after the first parent appeared near the nest and became aware of the dummy. We scored responses on the following scale: 1) no reaction=host(s) saw the dummy but ignored it, 2) distress calls=host(s) uttered distress calls, alarm calls or song, 3) mobbing=host(s) performed dives or flights around the dummy and 4) attacks=host(s) aggressively attacked the dummy with contact attacks (see also Moksnes et al. 1990, Røskaft et al. 2002b). In cases of attacks the dummy was immediately removed to avoid its destruction. As some responses were rare in some data sets we pooled scores 1 and 2 as “no aggression” and scores 3 and 4 as “aggression”. If no birds arrived at the focal nest during a 30-min period after the dummy was placed near the nest, the response was scored as “no reaction”. Excluding data from such experiments had no effect on results from fieldfares (n=1), redwings (n=4), song thrushes (n=10) or blackbirds (n=5). Each nest was tested only once to avoid pseudoreplication and only one kind of dummy (cuckoo or crow) was presented near each nest. Nest defence experiments were performed during egg-laying, incubation and young nestling stages when the adult cuckoo is a threat to hosts (the cuckoo female preys upon young host nestlings, Wyllie 1981). At some nests both egg and nest defence experiments were done. In such cases, an aggression experiment was performed after the egg discrimination experiment was finished. We tested for the possible effect of breeding stage on host responses to dummies. We fitted regression models with “breeding stage”, “sympatry/allopatry” and their interaction as factors and host reaction to dummies (no aggression/aggression) as a response separately for each combination of host species and dummy type. When model was insignificant we removed the interaction and fitted a reduced model. We found no effects of the breeding stage both in sympatry and allopatry on host aggression to both the cuckoo and the crow, therefore we pooled the data in subsequent analyses. Cross-fostering experiments No studied thrush nests were naturally parasitized by the cuckoo. Therefore, we crossfostered cuckoo chicks after hatching to thrush nests to test hypotheses 8–12. We did not introduce cuckoo eggs as these could be rejected by hosts. This risk was confirmed at two song thrush nests later used for cross-fostering where model eggs (meadow pipit type) were rejected after 1 and 2 days after the acts of artificial parasitism respectively. It is noteworthy that the two song thrush pairs rejected meadow pipit type eggs but neither refused to feed parasitic cuckoo chicks introduced to their nest later. This suggests that egg recognition and nestling recognition are two different cognitive processes (Redondo 1993). We weighed chicks to the nearest 0.1 g with a Pesola spring balance. As a control we measured the growth of cuckoo nestlings in nests of reed warblers and great reed warblers (some of these data are analysed in Kleven et al. 1999). Sample sizes in these experiment are relatively small due to ethical reasons (cuckoo nestlings showed very low survival in thrush nests and we felt that even these small sample sizes are sufficient to test the hypotheses 8–12 adequately). Comparisons of life-history traits We established two groups of cuckoo hosts for comparison of life-history variables and responses to parasitic eggs and dummies (Tables 2, 3): 1) “Rejecters” – Hosts that show relatively high egg rejection (Davies and Brooke 1989a, Moksnes 1992, Moksnes et al. 1990, 1994) and aggression to adult cuckoos (Moksnes et al. 1990, Røskaft et al. 2002b). We included four species of Acrocephalus warblers, the reed warbler, marsh warbler (A. palustris), great reed warbler and sedge warbler (A. schoenobaenus). All species are commonly parasitized (reed warbler is the most common host in Europe, Moksnes and Røskaft 1995), and can be regarded as suitable hosts in all respects of the interaction with the parasite. The most detailed data were available for this genus. Further we included strong rejecters (the following species did not significantly differ from Acrocephalus warblers in the traits listed in Tab. 3): the willow warbler (Phylloscopus trochilus), chaffinch (Fringilla coelebs), blackcap (Sylvia atricapilla), reed bunting (Emberiza schoeniclus). These hosts are suitable but currently not used. 2) “Acceptors” – Meadow pipit, redstart, dunnock (Prunella modularis) and robin (Erithacus rubecula). Frequently used hosts showing low rejection (< 40%) of nonmimetic cuckoo eggs

7

and less aggression against the laying cuckoo (Davies and Brooke 1989a, Moksnes et al. 1990, Røskaft et al. 2002b). Scores for intra- and interclutch variation for the species included were obtained from Øien et al. (1995), Stokke et al. (2002), and used for each of the following four groups: 1. Turdus species in Europe (n=4), 2. Acrocephalus species in Europe: Aquatic warbler (A. paludicola), Moustached warbler (A. melanopogon), plus four species of the genus from Table 2 (n=6), 3. Rejecters (i.e. suitable cuckoo hosts showing mean rejection rate ≥ 80% in Europe): Redbacked shrike (Lanius collurio), Icterine warbler (Hippolais icterina), Willow warbler (Phylloscopus trochilus), Chiffchaff (P. collybita), Subalpine warbler (Sylvia cantillans), Whitethroat (S. communis), Blackcap (S. atricapilla), Yellow wagtail (Motacilla flava), Chaffinch (Fringilla coelebs), Brambling (F. montifringilla), Reed bunting (Emberiza schoeniclus), Yellowhammer (E. citrinella) (n=12), 4. Seed-eaters and other acceptors in Europe: Robin, Wren (Troglodytes troglodytes), Dunnock, Hawfinch (Coccothraustes coccothraustes), Bullfinch (Pyrrhula pyrrhula), Pine grosbeak (Pinicola enucleator), Common crossbill (Loxia curvirostra), Scarlet rosefinch (Carpodacus erythrinus), Common redpoll (Carduelis flammea), Siskin (C. spinus), Greenfinch (C. chloris), Goldfinch (C. carduelis), Linnet (C. cannabina), Serin (Serinus serinus) (n=14). A potentially weak point in our analysis is the problem with phylogenetic ancestry. However, in 2 out of these groups only members of one genus were included. Regarding the remaining two groups (seed-eaters and rejecters) we have dealt with the problem by taking the mean of all members of the same genus within each of the two groups and then used this mean in the calculation of the total mean for the groups. This means that the group seed-eaters was reduced from 11 to 7 data points (+3 other acceptor-species), while the group rejecters was reduced from 12 to 7 data points. However, the traits we investigated should be related to the length of evolutionary time of parasite-host interaction which is most likely unrelated to phylogenetic distances between hosts (Aviles & Møller 2004) as cuckoo gentes colonized their hosts independently of phylogenetic relationships of hosts (Gibbs et al. 2000). All statistical tests are two-tailed and values are give as mean±SE. Results and discussion We observed no natural parasitism by the cuckoo in studied Turdus spp. nests in our study areas (N=281 experimental nests in sympatry with the cuckoo; Table 2). Overall parasitism rate in control nests of 12 passerines was 12.0% (N=3152). The parasitism was especially high in the Acrocephalus spp. group (17.7%), with the marsh warbler as the most frequently parasitized species (44.8%). The acceptor group suffered on average higher parasitism rate (4.0%) than the rejecter group (0.3%) (Table 3). 1. Breeding synchrony – Within the thrush group blackbirds started to breed ~3 weeks earlier that the other three species which in turn did not differ from each other (ANOVA: R2=0.30, F3,641=91.01, P<0.0001, Tukey-Kramer P<0.01). Blackbirds and song thrushes normally start breeding several weeks before the first cuckoo arrive from their wintering grounds and significantly earlier (about one month) than current cuckoo hosts (Table 3). The distribution of laying dates significantly differs between cuckoos and both blackbirds (χ2=721.96, df=1, P<0.0001) and song thrushes (χ2=442.82, df=1, P<0.0001) respectively (Fig. 1; data from Hudec 1983). Nevertheless, thrush nests are still available for the cuckoo when it parasitizes other hosts in late May, throughout June and early July (Fig. 1). Weak overlap of breeding seasons can contribute to low selection pressure exerted by parasites on their hosts (King 1973, Carey 1982, Ortega & Cruz 1991). Thus, the selection pressure from cuckoos that would parasitize blackbirds would be lower (other things being equal) than selection for antiparasitic adaptations in e.g. reed warblers, which all breed in synchrony with the parasite. In the lowlands (Norway Allopatry) redwings and fieldfares usually start laying in the beginning of May, while the cuckoo arrives to this latitude in the middle of May when the majority of these two thrush species have reached the incubation stage. In the mountain areas (Norway Sympatry) there are overlap between Turdus and cuckoo egg-laying (own unpubl. data).

8

2. Breeding density – Comparison of European population sizes of thrushes and 12 passerines parasitized by the cuckoo shows that thrushes are available for parasitism in about the same quantities as other hosts (Table 3). Interspecific comparative study by Soler et al. (1999) indicated that overall population size of potential host influences positively host choice by the cuckoo. However, rufous bush chat Cercotrichas galactotes breeds only in very limited geographic areas, however, it is parasitized by a specialized cuckoo gent (Alvares 1994a). This indicates that local breeding densities may play larger role in host choice by the cuckoo than overall population sizes. However, thrushes have similar breeding densities as common current cuckoo hosts (Tab. 3). Therefore this factor probably cannot explain the absence of parasitism in thrushes. 3. Breeding success – Breeding success of thrushes is well within the range of breeding success of species frequently parasitized by the cuckoo (Table 3). Therefore, this factor is probably of a minor importance for host choice by the cuckoo. Nevertheless, thrushes tend to have lower breeding success (maybe because of the high conspicuousness of their nests) than current cuckoo hosts, which lowers the benefits of parasitism in thrushes for the cuckoo. 4. Nest defence – Out of 595 dummy experiments 65.7% were done during laying or early incubation. Fieldfares ignored the cuckoo dummy in all experiments. However, redwings, song thrushes and blackbirds sometimes attacked the dummy (Table 4). Overall levels of thrushes aggression towards cuckoo dummies are similar to acceptors and significantly lower than those in rejecters (Table 3). The host behaviour towards the cuckoo was not influenced by host body size across studied species (rs=–0.33, n=16, P=0.22). Sympatry/allopatry distinction had no effect on behaviour of song thrushes and blackbirds against the crow (Table 4). Unfortunately, no data to test this relationship were available for fieldfares and redwings. As sympatry with the parasite had no effect on host nest defence behaviour, we pooled the data and compared the responses to the cuckoo and crow dummies. Fieldfares and redwings attacked the crow much more frequently than the cuckoo (Table 4), while the responses of song thrushes and blackbirds were similar towards the two kinds of dummies (Table 4). Thrushes were less aggressive than Acrocephalus warblers and rejecters and were similar to acceptor hosts. Therefore the intensity of nest defence cannot be responsible of extremely low rates of parasitism in thrushes. On the other hand, nest defence can contribute to avoidance of some hosts by other brood parasites (e.g. Briskie et al. 1990). Some cuckoo hosts, e.g. the reed warbler, meadow pipit, red-backed shrike, great reed warbler and blackcap, recognise the cuckoo as a special enemy at least in some areas (Duckworth 1991, Moksnes et al. 1993, Bartol et al. 2002, Lovászi & Moskát 2004, Grim 2005). However, aggression towards dummy cuckoo by song thrushes and blackbirds can be an expression of generalized nest defence – they attack both the cuckoo and control innocuous pigeon (Columba livia f. domestica) dummies at about the same rate (Grim and Honza 2001b). Interestingly, redwings attack dummies of conspecifics and fieldfares more intensely than cuckoo mounts (Grendstad et al. 1999). Data from another study showed that responses of thrushes were more similar to reactions of unsuitable than suitable hosts (Røskaft et al. 2002b). Therefore, thrush responses probably did not evolve as a specific response to brood parasitic cuckoo. 5. Nest attentiveness – We obtained sufficient data for latencies to respond to a dummy for song thrushes and blackbirds. Nest stage when experiments were done did not affect latencies in any of the two thrush species (Kruskal-Wallis ANOVAs, P=0.54 and 0.47 respectively) therefore the data were pooled across breeding stages. In song thrushes, average latencies (in minutes) to response to cuckoo (11.31±1.63) and crow (9.04±2.12) dummies did not differ (U36,27=1.68, P=0.09) and were not influenced by sympatry (11.09±1.53) and allopatry (8.63±2.48) status of tested nests (U44,19=1.57, P=0.12). After excluding nests where nest owners did not appear during the experiment, the latencies were significantly longer in cuckoo (7.57±0.99) than in crow (4.27±0.99) experiments (U30,22=2.41, P=0.016) and were slightly longer in sympatry (6.86±0.85 min.) than in allopatry (4.63±1.41) trials (U36,16=1.87, P=0.060). In blackbirds, average latencies to response to cuckoo (8.54±1.71) and crow (9.43±1.81) dummies did not differ (U35,51=1.29, P=0.20) but were much longer in sympatry (15.35±2.43) than in allopatry (6.35±1.36) with the cuckoo (U26,60=4.32, P<0.0001). After

9

excluding nests where nest owners did not appear during the experiment, the latencies were significantly longer in cuckoo (5.77±1.24) than in crow (2.24±0.64) experiments (U31,38=3.35, P=0.0008) and were much longer in sympatry (7.24±1.43) than in allopatry (2.71±0.73) trials (U17,52=4.02, P<0.0001). Latencies to response to cuckoos were 7.5 min. in blackcaps (Grim 2005) and 3.0 in reed warblers (Honza et al. 2004). Although increased nest attentiveness may decrease the costs of brood parasitism in some hosts (Tewksbury et al. 2002) our observations do not support the hypothesis that thrushes evolved the increased nest attentiveness as a defence against brood parasitism. In fact, observed differences in experimental latencies in respect to dummy type and presence of the cuckoo in the particular area were consistently against predictions from interspecific parasitism hypothesis but in line with predictions from conspecific parasitism hypothesis (Table 1). 6. Egg rejection – The time during breeding cycle when a parasitic egg was added to the host nest did not significantly affect rejection rates (except in fieldfares where we therefore excluded data from more advanced breeding stages, see Methods). This is in line with previous findings in thrushes and other cuckoo hosts (Davies and Brooke 1988, 1989a, Moksnes et al. 1990). Either way 81.8% of egg experiments (n=532) were done during egglaying or the first three days of host incubation period. Song thrushes rejected significantly more meadow pipit type models than more mimetic redstart type eggs, while the other three species tended to reject more the latter type (Table 4). Contrary to what could be expected from interspecific parasitism hypothesis (Table 1), fieldfares, redwings and song thrushes tended to show stronger rejection of parasitic eggs in allopatry with the cuckoo, but the differences never reached significance (Table 4). Fieldfares, redwings and song thrushes spent non-significantly longer time with rejecting blue models while blackbirds rejected them highly significantly earlier (Table 4). Sympatry or allopatry of thrush populations had no effect on the time required to recognise and reject model eggs (Table 4). Frequency of rejection of parasitic eggs was strongly positively correlated with aggression against cuckoo in the whole sample of experimental and control species (rs=0.76, N=16, P=0.0006) but not among the four thrushes (rs=0.80, P=0.20). Thrushes responded to experimental intraspecific parasitism in some localities. In the Czech Republic, song thrushes rejected 57.10% (n=7) and blackbirds 20.80% (n=24) of conspecific eggs. In contrast, blackbirds in Hungary deserted only 5.9% (n=17) of nest parasitized with conspecific eggs which was lower than desertion rate of non-manipulated nests (10%). This suggests that desertion was not response to parasitism itself. In Norway, redwings reject 20.8 of natural conspecific eggs added to their nests (Grendstad et al. 1999). Latencies to rejection for conspecific eggs in the song thrush (2.75±0.85, n=4) were very similar to those for blue models (Table 4) and those in blackbirds (2.80±0.86, n=5) were very close to latencies to rejection of spotted model eggs. 7. Short incubation period – There was no tendency for shorter incubation periods in thrushes compared to used hosts, moreover, the clutch sizes of thrushes are within the range of clutch sizes of currently used hosts (Table 3). Thus hypothesis 7 can be rejected. This factor probably plays minimal role in host choice in general because of extremely short incubation periods in parasitic birds (Davies 2000; but see Scott and Lemon 1996). However, thrush egg can sometimes hatch extremely early within 9–10 days (Cramp 1988) which is shorter period that time needed for hatching of cuckoo eggs (11 days). This may lead to large size disadvantage for newly hatched cuckoo chick in early hatched thrush nests. 8. Host egg size – On average, thrushes lay much bigger (about three times) eggs than current used hosts (Table 3). Moskát et al. (2003a) reported that many artificial cuckoo eggs used for experimental parasitism in blackbirds’ nests rolled down into the bottom of deep nest-cups. Although they were almost totally or even totally covered by the much bigger blackbird eggs, so the parasitic eggs were hardly visible, blackbirds were able to reject them. However, blackbirds were not able to rotate these deeply positioned parasitic eggs. Although these eggs were artificial, this phenomenon (Peer and Bollinger 1997) might cause the failure of real cuckoo eggs in deep thrush nests. To test for the effect of egg size on cuckoo eviction success we introduced song thrush eggs into parasitized reed warbler nests shortly after a cuckoo chick hatched. All song thrush eggs were successfully evicted by cuckoo chicks from reed warbler nests within

10

a day (n=7). This indicates that the egg size itself does not constrain cuckoo chick eviction behaviour. Further support for this conclusion comes from the fact that 2–3 days old cuckoo chicks had no problem to evict blackbird eggs (which are even larger than song thrush eggs) from old used blackbird nests experimentally placed on active great reed warbler nests (n=8). 9. Nest size/shape – Mean nest depths (in cm) decreased in the order fieldfare (6.86±0.13), song thrush (6.70±0.11), blackbird (6.32±0.07) and redwing (5.65±0.09). The differences were significant except between the fieldfare and song thrush (ANOVA, R2=0.16, F3,279=15.76, P<0.0001; Tukey-Kramer HSD: P<0.05). The index of nest cup steepness decreased in the similar order song thrush (0.73±0.01), fieldfare (0.66±0.01), redwing (0.65±0.01) and blackbird (0.63±0.01). Song thrush nests cups were significantly steeper than those of the three other species which in turn did not differ from each other (ANOVA: R2=0.15, F3,279=16.55, P<0.0001; Tukey-Kramer HSD: P<0.005). Cuckoo chicks in fieldfare nests had a low success in evicting host clutch: at two nests no eggs were evicted and at the third nest only 2 out of 3 eggs were evicted. At none of those three nests were cuckoos able to evict subsequently hatched host chicks. Also one day old cuckoo chicks did not succeed in evicting host eggs from natural active blackbird nests (n=3) but older cuckoo chicks (n=8) were more successful (100% eviction success). We did not include the three younger cuckoo chicks in nest depth and steepness comparison (see below) as these chicks were deserted by hosts before the eviction instinct ceases, thus, these experiments are inconclusive as for chick ability to evict host offspring. Cuckoo chicks were unable to evict any song thrush eggs from 4 song thrush nests attached to reed warbler nests (each nest with 2 song thrush eggs exposed for three days). At 6 natural song thrush nests cuckoo chicks successfully evicted some host eggs only from one nest. This nest was relatively shallow (4.5 cm deep), was slightly inclined and the cuckoo chick accomplished the eviction of the whole host clutch only when 5 days old. Video-recordings confirmed that chicks tried to evict experimental eggs but were unable to push them up to the nest rim. Unsuccessful evictions were not caused by large host egg size as cuckoo chicks 1) were also unable to evict small reed warbler eggs from song thrush nests (n=4) and in contrast 2) were able to evict song thrush eggs from reed warbler nests in all cases (see above). We conclude that the song thrush nest design was the major constraint on eviction behaviour and that most likely the quality of nest cup surface has played a significant role. Song thrush nests have hard and smooth lining, which can hardly provide any toehold for a cuckoo chick trying to evict host clutch/brood. This was confirmed by video-recordings showing that cuckoo’s legs during eviction process slipped on the nest cup lining and effectively prevented the chick to push host eggs higher than half-way to the nest rim. To sum up, in line with the significant differences in nest sizes and shapes among thrushes the cuckoo chicks were significantly less successful in evicting at least one host egg from deep and steep song thrush and fieldfare nests (2 out of 13 chicks) in comparison to shallower and less steep blackbirds nests (8 out of 8 chicks; χ2=17.90, df=1, P<0.0001). Cuckoos hatched in thrush nests are sometimes able to evict host eggs (Davies 2000) but our experiments at fieldfare, song thrush and blackbird nests show that this happens only very rarely. Indirect evidence for the impairing effect of host nest size and/or shape on cuckoo eviction process comes from anecdotal observations of cuckoos raised together with host young. This is the best evidence that parasitic chicks are unable to evict host clutch/brood. Such observations are common among large hosts and especially among thrushes (Grim & Procházka in prep.). 10. Food type/amount – Thrushes feed their nestlings with significantly less Diptera and Araneida (which are almost exclusively brought to nestlings by common fosterer species) and significantly more earthworms (which are almost absent in the diet of current common hosts) (Table 3). Some common hosts (dunnock, meadow pipit, and white wagtail, see Cramp 1988), also feed their nestling on earthworms and seeds, but the percentages of these food types are negligible (much less than 1%). Cuckoo nestlings are never fed earthworms or food items of that size in the nest of their commonest host, the reed warbler (Grim & Honza 1997, 2001a). However, direct observation of feeding of cuckoo chicks in song thrush nests (n=2 chicks, 19 hours) showed that cuckoo chicks were fed mainly with large earthworms and despite this achieved higher growth rates than in reed warbler nests (Grim subm.). Obviously, diet composition does not constrain cuckoo parasitism of

11

thrushes. Recent observations show that cuckoo chicks may even digest considerable amounts of plant diet (grapes; Martin-Galvez et al. 2005). However, seed or fruit diet can protect some potential hosts from being victimized by brood parasites, which are insectivorous (Middleton 1977, Manson 1986, Wiley 1988, Davies and Brooke 1989a, Kozlovic et al. 1996). Cuckoo chicks cannot survive on purely seed diet as well (Alvarez 1994b) but this does no apply to thrushes which are invertebratophagous. Our data tentatively support Soler et al. (1999a) conclusion that diet generally plays a minor role in host choice by cuckoos. 11. Competition with host chicks – An average body size for thrushes (85 g) is five time higher than that for common hosts (17 g) – a highly significant difference (Table 3). The largest commonly used European hosts (the great reed warbler and red-backed shrike) are more than twice smaller than the smallest thrush – the redwing. Because of general positive correlation between bird body size and egg size thrushes lay large eggs. Consequently, already at the time of hatching thrush nestlings have size advantage over cuckoo hatchling, e.g. a fieldfare chick weights about 10 g shortly after hatching (Norman 1994) while the cuckoo hatchling weighs only 2.5–3.0 g (Kleven et al. 1999, Honza et al. 2001). Moreover, thrush nestlings grow extremely quickly (Cramp 1988). E.g. fieldfare chicks attain a weight of 20–30 g within three days after hatching (Lübcke and Furrer 1985, Carlson and Moreno 1986, Norman 1994). At that age the cuckoo nestling weights only 7–8 g irrespective of the size of the fosterer (Kleven et al. 1999, Grim subm.) which suggests that there are physiological constraints on growth in cuckoos. Being smaller than nestmates is a huge disadvantage for any nestling (e.g. Kilner and Johnstone 1997) and may be fatal in nests of brood reducers (Soler 2002). Therefore the weight advantage at hatching coupled with extremely fast growth rates of thrushes could have an important negative impact on cuckoo nestling growth and, in turn, also survival. Fieldfares and blackbirds clutches hatch asynchronously (Cramp 1988), which could further decrease their suitability as hosts (these hosts start to incubate before clutch completion, thus parasitic eggs not laid very early in the laying period of the hosts could have very low chance of hatching in advance to host eggs). We found that cuckoo chicks raised together with fieldfare nestlings showed very poor growth and survived only for 1, 4 and 12 days respectively while host nestlings fledged successfully. Low survival of the former most likely resulted from the very fast growth of host nestlings which resulted in higher size asymmetries between parasitic and host chicks. Data from other host species indicate that the competition with host nestlings can have very deleterious effect on cuckoo growth and survival even in much smaller hosts than in thrushes. Cuckoo chicks showed poor competitive ability in comparison to rufous bush chat chicks (Martin-Galvez et al. 2005) and even presence of redstart nestlings in the nest significantly decreases cuckoo chicks growth and survival (Rutila et al. 2002). Molnár (1939) reported two cases when the cuckoo chick hatched a few hours later than the first host great reed warbler nestlings and these cuckoo chicks were not able to evict their nest-mates. Host nestlings started to grow faster than the cuckoo chicks in these nests. Although the survival of cuckoo chicks in these mixed broods nest were not documented in details, each died within 5 days. The hypothesis that a large host body size per se is an important constraint on host choice by the cuckoo is supported by data from Japan. In this area the cuckoo parasitizes the azure-winged magpie Cyanopica cyana (Nakamura 1990), which weights more than 70g (which is similar to European song thrushes). Cuckoo nestlings in magpie nests spent long time trying to evict host eggs and chicks (Nakamura 1990). Thus, cuckoos are usually unable to evict all eggs/nestlings of large host and consequently suffer from impaired growth probably because of competition with quickly growing host nestling. For comparison, in Acrocephalus group we never observed a nest, under natural circumstances, where cuckoo would be raised together with host nestlings. Host size seem to be an important determinant of parasite eviction behaviour in general: some cuckoos (Clamator, Scythrops) parasitizing large hosts do not evict and other cuckoos (Eudynamys) evict host progeny only in areas where they parasitize small hosts (Davies 2000). However, even a competition with small passerine nestlings can have seriously deleterious effect on cuckoo survival in a host nest (Molnar 1939, Rutila et al. 2002, Soler 2002, Martin-Galvez et al. 2005).

12

12. Chick rejection – We found some inter-species variability in host responses to parasitic chicks. Cuckoo chicks cross-fostered to song thrush nests survived well (3 fledged), none chick was deserted and chicks even grew better than in the reed warbler nests (n=6, Grim subm.). In these experiments the host eggs were removed by the observer (5 cases) or due to cuckoo chick eviction behaviour (1 case) thus cuckoo chicks were the sole occupants of the song thrush nests. However, when host young were present in experiments with fieldfares no cuckoo chick survived until fledging but the reason was not outright chick rejection by hosts (see above). In 3 cases out of 4 blackbird fosterers in Hungary did not feed the cuckoo chick and abandoned the nest, so the cuckoo nestling died. In another case we found the cuckoo chick under the nest, but the female blackbird continued to brood her eggs just before hatching. Chicks survived only for 1 or 2 days (1 and 3 chicks respectively). The experiments in the Czech Republic also showed extremely low success of cuckoo chicks in nests of blackbirds even in the absence of host own chicks: cuckoo chicks grew very poorly and out of 6 transferred chicks no survived until fledging. As a control for possible effect of cross-fostering process itself we transferred cuckoo chicks among 1) reed warbler nests, 2) great reed warbler nests and 3) between reed and great reed warbler nests (Kleven et al. 1999, Grim & Honza 2001b). No nestlings suffered from low growth or survival due to cross-fostering. Both song thrushes and fieldfares were willing to feed cuckoo chicks while blackbirds seemed to feed cuckoo chicks only at low rates (Czech Republic) or refused to feed them completely (Hungary). Lowered feeding rates to parasitic chicks are one of host defences documented at chick stage in several hosts of other brood parasites (Payne et al. 2001). For example, Lichtenstein (2001) found that rufous-bellied thrushes (Turdus rufiventris) preferentially fed their own nestlings and discriminated against parasitic shiny cowbird (Molothrus bonariensis) chicks. Our data show that while some thrushes (namely fieldfares and song thrushes) do not discriminate against cuckoo chicks some other species may to do so (blackbirds). There seems to be intraspecific variance in blackbird responses to cuckoo chicks as there are documented cases of large cuckoo chicks in blackbird nests which presumably fledged successfully (L. Sevcik, pers. comm.). This is unsurprising as also we observed high intraspecific variability in blackbird responses to parasitic chicks and eggs. 13. Host breeding strategy – Our data seem to be in line with Soler (2002) “host breeding strategy” hypothesis. Cuckoo chicks did not survive in nests of brood reducers (fieldfares and blackbirds) but grew well in nests of clutch adjusters (song thrushes). The low cuckoo chick success in fieldfare nest may be explained by hosts preference for larger chicks (cuckoo chick has disadvantage of smaller hatching size and lower growth rate in comparison to host chicks). In experiments with song thrushes and blackbirds there were no host chicks present. Despite this our observation may be in line with the host breeding strategy hypothesis. Due to blackbird brood reduction strategy it is highly unlikely that cuckoo chicks in the past survived in blackbird nests frequently enough to evolve an ability to exploit blackbirds feeding rules and/or to adapt to diet delivered to blackbird chicks. In contrast, cuckoo chicks in song thrush nests would survive and this may explain their successful fledging from nests of this hosts. Intra- and inter-clutch variation The four comparative groups (Turdus, rejecters, Acrocephalus, acceptors) did not show significant differences in intraclutch variation (ANOVA, F3,23=2.24, P=0.11), however, they differed in interclutch variation (ANOVA, F3,23=3.37, P=0.036). In a comparison of single groups Turdus group was more similar to seed eaters/acceptors than to rejecters: intraclutch variation Turdus vs. acceptors (P=0.54), Turdus vs. rejecters (P=0.059), interclutch variation Turdus vs. acceptors (P=0.42), Turdus vs. rejecters (P=0.14). There were no significant differences in intra- or interclutch variation between the Turdus and Acrocephalus group. Although abandoned host populations may retain highly developed egg discrimination, their supposed adaptations in high interclutch and low intraclutch variation might be destroyed in the absence of selection pressure by cuckoo parasitism (Moskát et al. 2002, Lahti 2005).

13

Conspecific parasitism During the study we observed very low levels of conspecific brood parasitism in blackbirds (N=1), song thrushes (N=1) and fieldfares (N=26). Further we found two blackbirds nests with 8 resp. 7 eggs which may suggest intraspecific parasitism. Thrushes: suitable or unsuitable cuckoo hosts? The blackbird, song thrush, fieldfare and redwing are 43rd, 49th, 69th and 84th most commonly parasitized species in Europe (Moksnes and Røskaft 1995). In other words, they are evidently parasitized only extremely scarcely, probably by cuckoo females unable to find nests of their regular hosts. Various passerines breeding in sympatry with brood parasites may be considered unsuitable hosts due to low synchronization of breeding cycles (Carey 1982, Peer and Bollinger 1997), inaccessible nests (Davies and Brooke 1989a), nest concealment (Briskie et al. 1990), nest attentiveness (Mermoz and Fernandéz 1999; see also Tewksbury et al. 2002), vigorous nest defence (Mermoz and Fernandez 1999), high rejection of parasitic eggs (Rothstein 1975, Moksnes and Røskaft 1992), inappropriate incubation of parasitic eggs because of large size of host eggs (Peer and Bollinger 1997), too short incubation period (Scott & Lemon 1996), high rejection of parasitic eggs (Honza et al. 2004, Lovászi & Moskát 2004, Stokke et al. 2004), unsuitable chick diet (Middleton 1977, Davies and Brooke 1989a), breeding strategy (Soler 2002) or high rejection of parasitic chicks (Langmore et al. 2003). Intense predation pressure on host nests (Mason 1986b), nest design constraining the eviction of host eggs (Rutila et al. 2002) and large size of host eggs that cannot be evicted by parasitic hatchling (this study) may also play a role. So far there was an inconsistency in classification of thrushes into “suitable” or “unsuitable” categories. Davies and Brooke (1989a) considered that the blackbird and song thrush are suitable hosts for the cuckoo, while other authors (Moksnes et al. 1990, Moksnes and Røskaft 1995) classified Turdus spp. as unsuitable hosts for the parasite. Our data support the latter view. Nevertheless, there are anecdotal observations of cuckoo nestlings successfully fledging from blackbirds nest (see above). However, even non-zero reproductive success of “thrush” cuckoos cannot lead to the conclusion that the blackbird is a suitable host. The important variable is the difference between cuckoo reproductive success in the nest of a thrush species and other host species available in the same area. The low breeding success of parasitic chicks in thrush nests indicates that these species are parasitized only by mistake or in the last resort when a nest of the main host is not available for parasitism. “Suitable hosts” and “unsuitable hosts” are discrete categories. However, (un)suitability of the host is a continuous variable: almost all factors influencing the probability of successful recruitment from a host nest are continual (egg rejection rate, intensity of nest defence, incubation period length, host egg size, host nest size, food quality and quantity, etc.). This makes strict categorization of potential hosts difficult. However, selection in cuckoo females should lead to a preference for the most suitable hosts, thus lowering the parasitism in hosts of low or intermediate quality. This could be the major evolutionary force behind the fact that some potentially suitable but poor quality hosts are avoided by the cuckoo. Røskaft et al. (2002b) showed that a more detailed classification of host species (to 5 categories) can better explain the pattern of parasitism than traditional suitable-unsuitable distinction. Interestingly, a category “large nest and eggs” (i.e. thrushes) shows almost identical level of aggression against dummy cuckoos as the category “seed eaters” and “hole nesters”, i.e. as birds, which clearly cannot serve as hosts for the cuckoo indicating that thrushes experienced similar selection pressures as the two latter categories. This results is in accordance with the finding that only 12% of tree-nesting songbirds have their cuckoo egg-morph, which is more similar to that for hole-nesting birds (8%) than for those nesting in low vegetation (48%; Moksnes and Røskaft 1995). Most cuckoo hosts build nests in low vegetation or on the ground (Moksnes and Røskaft 1995) while thrushes mainly nest in trees in their original nesting habitat (whereas in an urbanized environment, e.g. in city parks of Budapest, blackbirds mainly breed in thorny bushes.) Thus also in their nest-site selection thrushes resemble nonparasitized species more than current common hosts. Furthermore, the meadow pipit cuckoos gens in Norway also parasitize other species breeding on ground (Lapland buntings Calcarius lapponicus) or in low vegetation (reed buntings) in the area but they avoid parasitism in

14

tree-nesting bramblings and fieldfares despite their abundance (Moksnes and Røskaft 1995). This indicates that nest-site selection is an important factor contributing to absence of parasitism in thrushes. When explaining interspecific differences in parasitism rates it is important to take into account both benefits and costs associated with parasitism in particular species from the parasite’s point of view. E.g. least flycatchers do not reject cowbird eggs and thus it should be more advantageous for the brown-headed cowbird to parasitize this species than laying in nests of yellow warblers (Dendroica petechia), which reject at high frequencies. However, flycatchers are less common, show stronger nest defence and breed higher above ground than yellow warblers (cowbird females searches for host nest primarily in low vegetation). Consequently warblers are parasitized about six times more than flycatchers – high costs of nest searching and overcoming host nest defence probably explain relatively lower parasitism in flycatchers (Briskie et al. 1990). The effect of sympatry vs. allopatry Several host species frequently parasitized by the common cuckoo, great spotted cuckoo (Clamator glandarius) and brown-headed cowbird are known to show stronger antiparasitic responses (egg rejection, aggression against female parasite, lower intraclutch variation) in areas of sympatry with the parasite (meadow pipit: Davies and Brooke 1989a, Aviles and Møller 2003, magpie: Soler and Møller 1990, American robin and yellow warbler: Briskie et al. 1992, Reed Warbler: Lindholm and Thomas 2000; see also Brown et al. 1990, Røskaft et al. 2002b). On the other hand, allopatric great reed warbler population in Greece showed higher (100%) egg rejection rate than sympatric Hungarian population (70%; Moskát et al. 2002). This indicates that Greek population was recently abandoned by the cuckoo because of high host antiparasitic response. However, in general host population in sympatry show stronger defenses against parasitism than those in allopatry. Therefore we expected that if the egg rejection behaviour by thrushes was result of coevolutionary arms-race with the cuckoo we should not find this behaviour in allopatric populations. However, we found the opposite: thrushes from sympatric areas generally showed lower rejection than in allopatric populations (Table 4). Similarly, Davies and Brooke (1989a) found even higher rejection of nonmimetic blue eggs (70.0%; the same type of egg we used) in allopatric redwing population in Iceland than we found in allopatric redwing population in Europe (49.3%) although a difference is not significant (χ2=1.55, P=0.21). Moreover, a gene flow between European and Icelandic populations in this species is probably minimal as Iceland is inhabited by distinct redwing subspecies (T. i. coburni; Davies and Brooke 1989a). This comparison further points to the conspecific parasitism explanation for redwing responses. Bolen et al. (2000) and Rothstein (2001) showed that host populations, which are no longer parasitized can retain rejection behaviour for very long periods (see also Cruz and Wiley 1989). This is in line with finding that various hosts of parasitic birds make almost no rejection errors (Røskaft et al. 2002a; rejection errors could be important force leading to disappearance of host defenses in the absence of parasitism, see evolutionary equilibrium hypothesis; Rothstein 1990, Lotem et al 1992, 1995) and ejection costs can be very low (e.g. in blackbirds; Moskát et al. 2003a). However, it should be noted that we should expect decline in egg rejection in the absence of previous parasitism pressure even when a host makes no recognition errors – long term accumulation of mutations in genes, which are not under selection pressure can disrupt the particular phenotype (e.g. egg rejection ability). Results of Bolen et al. (2000) and Rothstein (2001) indicate that the possibility that thrushes were formerly parasitized by the cuckoo cannot be rejected. However, the circumstantial evidence against the interspecific parasitism hypothesis is so strong (see Results) that we conclude that cuckoo parasitism must have played only a minor (if any) role in the evolution of antiparasitic defenses in thrushes. Briskie et al. (1992) and Soler et al. (1999b) explained strong responses to parasites in allopatric populations with a gene flow from sympatric to allopatric areas. Gene flow would lead to decrease of host defenses in sympatry and increase of these defenses in allopatry (Barabás et al. 2004). However, several host responses in our study species were even higher in allopatry – this finding clearly cannot be explained by gene flow. Moreover, our study populations (except the fieldfare population in Stjørdal) are most probably nonmigratory (see Methods).

15

Conspecific brood parasitism There are indications that fieldfares commonly parasitize conspecifics (Ringsby et al. 1993) and even a case of redwing nest parasitized by the fieldfare was reported (Grendstad et al. 1999). Redwings probably also parasitize conspecifics frequently, since they show good discrimination of conspecific eggs and are much more aggressive towards conspecifics and fieldfares than to the cuckoo (Grendstad et al. 1999). Further, intraspecific nest parasitism was documented also in the song thrush (Grim and Honza 2001b) and during this study we also detected at least one case on intraspecific parasitism in the blackbird in the Czech Republic (Honza and Grim, unpublished data) and another two cases in the blackbird population in Hungary (Moskát et al. 2003a). In blackbirds clutches of up to 9 eggs were recorded presumably laid always by two females (Cramp 1988). Also in other congeners the conspecific nest parasitism was documented (Yom-Tov 2001). These data indicate that parasitizing conspecifics/congeners could be an ancestral breeding strategy for thrushes. Currently low levels of conspecific brood parasitism in blackbirds and song thrushes could be explained by the evolution of host defenses (just as in the cases of interspecific parasitism). Blackbirds and mistle thrushes (T. viscivorus; Martin-Vivaldi et al. 2002) are very good rejecters, and so is the American robin (even though it is not a common cowbird host). The fact that the American robin is a good rejecter could indicate that thrushes have maintained a high rejection rate for a long time, perhaps even from a common ancestor (see also Rothstein (2001) for a discussion of retention of rejection behaviour). The lack of rejection costs in thrushes (Martin-Vivaldi et al. 2002) can validate such a scenario. Interestingly, it seems like it is the largest thrush species that are the best rejecters. It is important to note, that even if no cuckoo chick would survive after hatching in a thrush nest there would still exist a selection pressure for host defenses, i.e. both for aggression against parasite (which removes one or more host eggs or even young nestlings, Wyllie 1981) and egg rejection (as the cuckoo hatchling sometimes evicts host eggs and young). Hosts responses can still be a result of co-evolution with a parasite. More to the point, because of their large body size thrushes should experience no problems when grasping alien eggs (Rohwer and Spaw 1988, Moksnes et al. 1991). Further, thrushes lay eggs that are much larger than cuckoo eggs. Hence they are easily distinguishable. Therefore, evolution of antiparasitic defenses in thrushes could be driven even by low parasitism pressure as there would be no or very low opposing selection resulting from recognition errors or rejection costs. However, if there was such a coevolution between the cuckoo and thrushes it must have been very weak (see above). We cannot exclude the possibility that highly developed nest-cleaning behaviour might contribute to egg rejection ability of thrushes, as it was set up in the case of the blackbird (Moskát et al. 2003a), but this also could be the side effect of inter- or intraspecific parasitism (cf. Rothstein 1975, Moskát et al. 2003b). It is important to stress that interspecific and conspecific parasitism are not alternative explanations for the evolution of antiparasitic host behaviour. Both kinds of parasitism provide the selection for the very same adaptations – namely egg rejection and low intraclutch variation. However, when considering all factors involved (see above) we can conclude that the role of interspecific cuckoo parasitism in the evolution of antiparasitic behaviour in thrushes was probably low. In general, different thrush species defend against cuckoo parasitism at different stages. E.g. song thrushes show relatively low aggression and low rejection of alien eggs but depth and cup steepness of their nests effectively force the cuckoo to share the parental care with host young. Moreover, very large eggs and short incubation period of song thrushes may decrease cuckoo hatching success down to zero. In contrast, blackbirds are more aggressive, reject more alien eggs, but their nests are too shallow to prevent the cuckoo from evicting host progeny. Thus, in the case of successful eviction of their eggs/young by the cuckoo chicks egg acceptors may imply another line of defence: chick discrimination. Conclusions In conclusion, no single factor can alone explain the cuckoo avoidance of thrushes. Thrushes lack both main advantages – smaller body size and longer incubation period – that were shown to facilitate the start of parasitism in new host species (Slagsvold 1998). Absence of these traits increases eviction and nestling competition constraints. Further,

16

nest defence, egg rejection and host breeding strategy play complementary roles in avoidance of thrushes by cuckoos. These factors independently and additively decrease the benefits of cuckoo parasitism in thrushes. The contribution of these factors is species specific (e.g. nest defence plays some role in blackbirds but not in fieldfares). It is suggested that selection forced cuckoos to specialize on more profitable hosts. Further, fieldfares and redwings probably did not experience any coevolution with the cuckoo, and interspecific brood parasitism probably played only minor (if any) role in the evolution of antiparasitic behaviour of blackbirds and song thrushes. In general, thrushes’ behaviour was better predicted by conspecific than interspecific parasitism hypothesis. Conspecific nest parasitism presumably resulted in antiparasitic behaviours that can serve as a preadapted defence against interspecific parasites as well. In general, this study shows the importance of considering all potential factors involved and necessity of interspecific comparisons for understanding an absence of parasitism in potential hosts of parasitic birds. Acknowledgements V. Beran, A. Dvorská, T. Csörgö, Z. Karcza, P. Procházka P. Samaš and Z. Strachoňová helped with the fieldwork. Financial support was received from following institutions and grants: TG (grant from the Research Council of Norway, the Czech Ministry of Education grants No. 153100012 No. MSM6198959212 and post-doc grant from the Grant Agency of the Czech Republic 206/03/D234), ER and AM (grant from Nansen Foundation), BGS (grant from the Research Council of Norway, no. 151641/432). References Alvarez F (1994b) Rates of weight increase of cuckoo (Cuculus canorus) and host (Cercotrichas galactotes) chicks. Ardeola 41:63–65 Alvarez F. 2003. Parasitism rate by the common cuckoo Cuculus canorus increases with high density of host's breeding pairs. - Ornis Fennica 80: 193-196. Alvarez, F. (1994a) A gens of cuckoo Cuculus canorus parasitizing rufous bush chat Cercotrichas galactotes. Journal of Avian Biology 25: 239–243. Arheimer, O. 1978. Rödvingetrastens Turdus iliacus L. häckningsbiologi i subalpin ängsbjörkskog vid Ammarnäs i svenska Lappland. PhD thesis, Göteborg University, Sweden. Avilés J. M., Soler J. J., Pérez-Contreras T., Soler M. & Møller A. P. 2006: Ultraviolet reflectance of great spotted cuckoo eggs and egg discrimination by magpies. Behav Ecol 17: 310–314. Aviles, J. M. and Møller, A. P. 2003. Meadow pipit (Anthus pratensis) egg appearance in cuckoo (Cuculus canorus) sympatric and allopatric populations. Biol. J. Linn. Soc. 79: 543–549. Barabás, L., Gilicze, B., Takasu, F. & Moskát, C. 2004. Survival and anti-parasite defense in a host metapopulation under heavy brood parasitism: a source-sink dynamic model. - J. Ethol. 22: 143-151 DOI 10.1007/s10164-003-0114-y Bártol, I., Z. Karcza, C. Moskát, E. Røskaft, and T. Kisbenedek. 2002. Responses of great reed warblers to experimental brood parasitism: the effects of a dummy and egg mimicry. Journal of Avian Biology 33: 420–425. Bolen, G. M., Rothstein S. I., and C. H. Trost. 2000. Egg recognition in yellow-billed and black-billed magpies in the absence of interspecific parasitism: Implications for parasite-host coevolution. Condor 102: 432–438. Briskie, J. V., Sealy S. G., and K. A. Hobson. 1990. Differential parasitism of least flycatchers and yellow warblers by the brown-headed cowbird. Behavioral Ecology and Sociobiology 27: 403–410. Briskie, J. V., Sealy S. G., and K. A. Hobson. 1992. Behavioral defenses against avian brood parasitism in sympatric and allopatric host populations. Evolution 46: 334–340. Brown C.R., Brown M.B. 1991: Selection of high-quality host nests by parasitic cliff swallows. Anim. Behav. 41: 457–465. Brown, R. J., M. N. Brown, M. de L. Brooke, and N. B. Davies. 1990. Reactions of parasitized and unparasitized populations of Acrocephalus warblers to model cuckoo eggs. Ibis 132: 109–111. Carey, M. 1982. An analysis of factors governing pair-bonding period and the onset of laying in indigo buntings. Journal of Field Ornithology 53: 240–248.

17

Carlson, A., and J. Moreno. 1986. Foraging behaviour and parental care in the fieldfare. Ardea 74: 79–90. Cramp, S., editor. 1988. The birds of the Western Palearctic. Tyrant Flycatchers to Thrushes. Volume V. Oxford University Press, Oxford, UK. Cramp, S., editor. 1992. The birds of the Western Palearctic. Warblers. Volume VI. Oxford University Press, Oxford, UK. Cruz, A., and J.W. Wiley. 1989. The decline of an adaptation in the absence of a presumed selection pressure. Evolution 43: 55–62. Davies, N. B. 2000. Cuckoos, cowbirds and other cheats. T & AD Poyser, London, UK. Davies, N. B., and M. de L. Brooke. 1988. Cuckoos versus reed warblers: adaptations and counteradaptations. Animal Behaviour 36: 262–284. Davies, N. B., and M. de L. Brooke. 1989a. An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. I. Host egg discrimination. Journal of Animal Ecology 58: 207–224. Davies, N. B., and M. de L. Brooke. 1989b. An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. II. Host egg markings, chick discrimination and general discussion. Journal of Animal Ecology 58: 225–236. Desrochers, A., and R. D. Magrath. 1993. Environmental predictability and remating in European blackbirds. Behavioral Ecology 4: 271–275. Duckworth, J. W. 1991. Responses of breeding reed warblers Acrocephalus scirpaceus to mounts of sparrowhawk Accipiter nisus, cuckoo Cuculus canorus and jay Garrulus glandarius. Ibis 133: 68–74. Formánek J., 1958: Migration of blackbird (Turdus merula) and song thrush (T. ericetorum) in Czechoslovakia. Sylvia 15: 23–41. (in Czech) Friedmann, H., and L. F. Kiff. 1985. The parasitic cowbirds and their hosts. Proceedings of West. Found Vert. Zool. 2: 225–304. Gibbs HL, Sorenson MD, Marchetti K, Brooke MD, Davies NB, Nakamura H 2000: Genetic evidence for female host-specific races of the common cuckoo. Nature 407: 183-186. Glue D. and Morgan R. 1972: Cuckoo hosts in British habitats. Bird Study 19: 187–192. Grendstad, L. C., A. Moksnes, and E. Røskaft. 1999. Do strategies against conspecific brood parasitism occur in redwings Turdus iliacus? Ardea 87: 101–111. Grim, T. 2002. Why is mimicry in cuckoo eggs sometimes so poor? Journal of Avian Biology 33: 302–305. Grim, T., and M. Honza. 1997. Differences in parental care of reed warbler (Acrocephalus scirpaceus) to its own nestlings and parasitic cuckoo (Cuculus canorus) chicks. Folia Zoologica 46: 135–142. Grim, T., and M. Honza. 2001a. Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to experimental parasitism by the common cuckoo Cuculus canorus. Biologia 56: 549–556. Grim, T., and M. Honza. 2001b. Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behavioral Ecology and Sociobiology 49: 322–329. Grim, T., Kleven, O. and Mikulica, O. 2003. Nestling discrimination without recognition: a possible defence mechanism for hosts towards cuckoo parasitism? Proc. R. Soc Lond. B 270: S73–S75. Hagemeijer, W. J. M., and M. S. Blair, editors. 1997. The EBCC atlas of European breeding birds. Their distribution and abundance. T & AD Poyser, London, UK. Hatchwell, B. J., D. E. Chamberlain, and C. Perrins. 1996. The reproductive success of blackbirds Turdus merula in relation to habitat structure and choice of nest site. Ibis 138: 256–262. Honza, M., J. Picman, T. Grim, V. Novak, M. Capek, and V. Mrlik. 2001. How to hatch from an egg of great structural strength. A study of the common cuckoo. Journal of Avian Biology 32: 249–255. Honza, M., Procházka, P., Stokke, B. G., Moksnes, A., Røskaft, E., Capek, M. Jr. & Mrlík, V. 2004. Are blackcaps current winners in the evolutionary struggle against the common cuckoo? - J. Ethol. 22: 175-180. Hoyt, O. F. 1979. Practical methods of estimating volume and fresh weight of bird eggs. Auk 96: 73–77. Hudec K. (ed.) 1983: Fauna ČSSR. Ptáci 3/I. Academia, Praha. (in Czech)

18

Chamberlain, D. E., B. J. Hatchwell, and C. M. Perrins. 1999. Importance of feeding ecology to the reproductive success of blackbirds Turdus merula nesting in rural habitats. Ibis 141: 415–427. Cherry, M. I., and A. T. D. Bennett. 2001. Egg colour matching in an African cuckoo, as revealed by ultraviolet-visible reflectance spectrophotometry. Proc. R. Soc. Lond. B 268: 565–571. Karlsson, J., and H. Källander. 1977. Fluctuations and density of suburban populations of the blackbird Turdus merula. Ornis Scandinavica 8: 139–144. Kilner, R., and R. A. Johnstone. 1997. Begging the question: are offspring solicitation behaviours signals of needs? Trends in Ecology and Evolution 12: 11–15. King, J.R. 1973. Reproductive relationships of the rufous-collared sparrow and the shiny cowbird. Auk 90: 19–34. Kleven, O., A. Moksnes, E. Røskaft, and M. Honza. 1999. Host species affects the growth rate of cuckoo (Cuculus canorus) chicks. Behavioral Ecology and Sociobiology 47: 41– 46. Kozlovic, D. R., Knapton, R. W. and Barlow, J. C. 1996. Unsuitability of the house finch as a host of the brown-headed cowbird. Condor 98: 253–258. Kristin A. 1992. Trophische Beziehungen zwischen Singvögeln und Wirbellosen im EichenBuchenwald zur Brutzeit. Der Ornithologische Beobachter 89: 157–169. Lack, D. 1966. Population studies of birds. Clarendon Press, Oxford, UK. Langmore, N. E., Hunt S. and Kilner, R. M. 2003. Escalation of a coevolutionary arms race through host rejection of brood parasitic young. Nature 422: 157–160. Lerkelund H. E., Moksnes A., Røskaft E. and Ringsby T. H. 1993: An experimental test of optimal clutch size of the fieldfare – with a discussion on why brood parasites remove eggs when they parasitize a host species. Ornis Scandinavica 24: 95–102. Lichtenstein, G. 2001. Low success of shiny cowbird chicks parasitizing rufous-bellied thrushes: chick-chick competition or parental discrimination? Animal Behaviour 61: 401–413. Lindholm, A. K. 1999. Brood parasitism by the cuckoo on patchy reed warbler populations in Britain. Journal of Animal Ecology 68: 293–309. Lindholm, A., and R.J. Thomas. 2000. Differences between populations of reed warblers in defences against brood parasitism. Behaviour 137: 25–42. Lovászi, P. & Moskát, C. 2004. Break-down of arms race between the red-backed shrike (Lanius collurio) and common cuckoo (Cuculus canorus). - Behaviour 141: 245-262. Lübcke, W., and R. Furrer. 1985. Die Wacholderdrossel. A. Ziemsen Verlag, Wittenberg Lutherstadt, Germany. Magrath, R.D. 1991. Nestling weight and juvenile survival in the blackbird, Turdus merula. Journal of Animal Ecology 60: 335–351. Mason, P. 1986a. Brood parasitism in a host generalist, the shiny cowbird: I. The quality of different species as hosts. Auk 103: 52–60. Mason, P. 1986b: Brood parasitism in a host generalist, the shiny cowbird: II. Host selection. Auk 103: 61–69. Meilvang, D., A. Moksnes, and E. Røskaft. 1997. Nest predation, nesting characteristics and nest defence behaviour of fieldfares and redwings. Journal of Avian Biology 28: 331– 337. Mermoz, M. E., and G. J. Fernandez. 1999. Low frequency of shiny cowbird parasitism on scarlet headed blackbirds: anti-parasite adaptations or nonspecific life-history traits? Journal of Avian Biology 30: 15–22. Middleton A. L. A. 1977. Effect of cowbird parasitism on American goldfinch nesting. Auk 94: 304–307. Moksnes A., Røskaft E. and Braa A. T. 1991: Rejection behavior by common cuckoo hosts towards artificial brood parasite eggs. Auk 108: 348–354. Moksnes, A. 1992. Egg recognition in chaffinches and bramblings. Animal Behaviour 44: 993–995. Moksnes, A., and E. Røskaft. 1987: Cuckoo host interactions in Norwegian mountain areas. Ornis Scandinavica 18: 168–172. Moksnes, A., and E. Røskaft. 1988. Response of fieldfares Turdus pilaris and bramblings Fringilla montifringilla to experimental parasitism by the cuckoo Cuculus canorus. Ibis 130: 535–539.

19

Moksnes, A., and E. Røskaft. 1992. Responses of some rare cuckoo hosts to mimetic cuckoo eggs and foreign conspecific eggs. Ornis Scandinavica 23: 17–23. Moksnes, A., and E. Røskaft. 1995. Egg-morphs and host preference in the common cuckoo (Cuculus canorus): an analysis of cuckoo and host eggs from European museum collections. Journal of Zoology 236: 625–648. Moksnes, A., E. Røskaft, A. T. Braa, L. Korsnes, H. M. Lampe, and H. C. Pedersen. 1990. Behavioural responses of potential hosts towards artificial cuckoo eggs and dummies. Behaviour 116: 64–89. Moksnes, A., E. Røskaft, and L. Korsnes. 1993. Rejection of cuckoo (Cuculus canorus) eggs by meadow pipits (Anthus pratensis). Behavioral Ecology 4: 120–127. Moksnes, A., E. Røskaft, and M. M. Solli. 1994. Documenting puncture ejection of parasitic eggs by chaffinches Fringilla coelebs and blackcaps Sylvia atricapilla. Fauna norvegica Series C, Cinclus 17: 115–118. Moksnes, A., E. Røskaft, V. Bičík, M. Honza, and I. J. Øien. 1993. Cuckoo Cuculus canorus parasitism on Acrocephalus warblers in Southern Moravia in the Czech Republic. Journal für Ornithologie 134: 425–434. Møller, A. P. 1987: Intraspecific nest parasitism and anti-parasite behaviour in swallows, Hirundo rustica. Anim. Behav. 35: 247–254. Molnár, B. 1939. About the cuckoo: observations and experiments on the eviction instinct of the cuckoo chick. Published by the author, Szarvas, Hungary. (In Hungarian) Móra V., Csörgö T. & Karcza Z. 1998: Survival of blackbird (Turdus merula) populations in relation to its migratory status. Ornis Hungarica 8, Suppl. 1: 187–198. (In Hungarian with English summary) Moskát C. & Honza M. 2000. Effect of nest and nest site characteristics on the risk of cuckoo Cuculus canorus parasitism in the great reed warbler Acrocephalus arundinaceus. - Ecography 23: 335-341. Moskát, C., J. Szentpéteri, and Z. Barta. 2002. Adaptations by great reed warblers to brood parasitism: a comparison of populations in sympatry and allopatry with the common cuckoo. Behaviour 139: 1313–1329. Moskát, C., Karcza, Z. and T. Csörgö. 2003a. Egg rejection in European blackbirds (Turdus merula): the effect of mimicry. Ornis Fennica 80: 86–91. Moskát, C., Székely, T., Kisbenedek, T., Karcza, Z. and Bártol, I. 2003b. The importance of nest cleaning in egg rejection behaviour of great reed warblers Acrocephalus arundinaceus. J. Avian Biol. 34: 16–19. Nakamura, H. 1990. Brood parasitism of the cuckoo Cuculus canorus in Japan and the start of new parasitism on the azure-winged Magpie Cyanopica cyana. Japanese Journal of Ornithology 39: 1–18. Nicolai, J. 1974. Mimicry in parasitic birds. Scientific American 231: 92–98. Norman, D. 1994. The fieldfare. Hamlyn, London, UK. Øien I.J., Honza M., Moksnes A. and Røskaft E. 1996. The risk of parasitism in relation to the distance from reed warbler nests to cuckoo perches. Journal of Animal Ecology 65: 147–153. Øien, I.J., Moksnes, A. and Røskaft, E. 1995: Evolution of variation in egg color and marking pattern in European passerines: adaptations in a coevolutionary arms race with the cuckoo, Cuculus canorus. Behav. Ecol. 6: 166–174. Ortega, C. 1998. Cowbirds and other brood parasites. The University of Arizona Press, Tucson, USA. Ortega, C. P. and Cruz A. 1991. A comparative-study of cowbird parasitism in yellow-headed blackbirds and red-winged blackbirds. Auk 108: 16–24. Peer, B. D., and E. K. Bollinger. 1997. Explanations for the infrequent cowbird parasitism on common grackles. Condor 99: 151–161. Poysa H. 2003: Parasitic common goldeneye (Bucephala clangula) females lay preferentially in safe neighbourhoods. Behav. Ecol. Sociobiol. 54: 30–35 Procházka, P. & Honza, M. 2003. Do common whitethroats (Sylvia communis) discriminate against alien eggs? -J. Ornithol. 144: 354-363. Ringsby, T. H., A. Moksnes, E. Røskaft, and H. E. Lerkelund. 1993. Do conspecific brood parasitism and antiparasite strategies occur in fieldfares Turdus pilaris? Fauna norvegica Series C, Cinclus 16: 45–53.

20

Rohwer, S. and Spaw, C. D. 1988. Evolutionary lag versus bill-size constraints: a comparative study of the acceptance of cowbird eggs by old hosts. – Evol. Ecol. 2: 2736. Røskaft, E., A. Moksnes, B. G. Stokke, C. Moskát, and M. Honza. 2002c. The spatial habitat structure hypothesis of host populations explains the pattern of rejection behavior in hosts and parasitic adaptations in cuckoos. Behavioral Ecology 13: 163–168. Røskaft, E., A. Moksnes, D. Meilvang, V. Bičík, J. Jemelíková, and M. Honza. 2002a. No evidence for recognition errors in Acrocephalus warblers. Journal of Avian Biology 33: 31–38. Røskaft, E., G. H. Orians, and L. D. Beletsky. 1990. Why do red-winged blackbirds accept eggs of brown-headed cowbirds. Evolutionary Ecology 4: 35–42. Røskaft, E., Moksnes A., Stokke B.G., Bičík V., and Moskát C. 2002b. Aggression to dummy cuckoos by potential European cuckoo hosts. Behaviour 139: 613–628. Rothstein S. I. 1975. An experimental and teleonomic investigation of avian brood parasitism. Condor 77: 250–271. Rothstein, S. I. 1990. A model system for coevolution: avian brood parasitism. – Annu. Rev. Ecol. Syst. 21: 481–508. Rothstein, S. I. 2001. Relic behaviours, coevolution and the retention versus loss of host defences after episodes of avian brood parasitism. Animal Behaviour 61: 95–107. Rutila, J.; Latja, R.; Koskela, K. 2002: The common cuckoo Cuculus canorus and its cavity nesting host, the redstart Phoenicurus phoenicurus: a peculiar cuckoo-host system? J. Avian Biol. 33: 413–419. Sackmann P, Reboreda JC 2003: A comparative study of shiny cowbird parasitism of two large hosts, the chalk-browed mockingbird and the rufous-bellied thrush. Condor 105: 728–736. Scott D. M. and Lemon R. E. 1996. Differential reproductive success of brown-headed cowbirds with northern cardinals and three other hosts. Condor 98: 259–271. Sealy, S. G., and R. C. Bazin. 1995. Low-frequency of observed cowbird parasitism on eastern kingbirds: host rejection, effective nest defence, or parasite avoidance? Behavioral Ecology 6: 140–145. Simms, E. 1978. British thrushes. Collins, London, UK. Slagsvold T. 1998. On the origin and rarity of interspecific nest parasitism in birds. Am. Nat. 152: 264–272. Soler J.J., Soler M., Møller A.P., Martinez J.G. 1995: Does the great spotted cuckoo choose magpie hosts according to their parenting ability? Behav. Ecol. Sociobiol. 36: 201– 206. Soler, J. J., A. P. Møller, and M. Soler. 1999a. A comparative study of host selection in the European cuckoo. Oecologia 118: 265–276. Soler, J. J., J. G. Martinez, M. Soler, and A. P. Møller. 1999b. Genetic and geographic variation in rejection behavior of cuckoo eggs by European magpie populations: an experimental test of rejecter-gene flow. Evolution 53: 947–956. Soler M (2002) Breeding strategy and begging intensity: influences on food delivery by parents and host selection by parasitic cuckoos. In: Wright J & Leonard ML (eds) The evolution of Begging. pp 413–427. Kluwer, Dordrecht. Soler, M., and A. P. Møller. 1990. Duration of sympatry and coevolution between the great spotted cuckoo and its magpie host. Nature 343: 748–750. Stokke, B. G., A. Moksnes, and E. Røskaft. 2002. Obligate brood parasites as selective agents for evolution of egg appearance in passerine birds. Evolution 56: 199–205. Stokke, B. G., Moksnes, A., Røskaft, E., Rudolfsen, G. and Honza, M. 1999. Rejection of artificial cuckoo Cuculus canorus eggs in relation to variation in egg appearance among reed warblers Acrocephalus scirpaceus. – Proc. R. Soc. Lond. B 266: 1483– 1488. Stokke, B. G., Rudolfsen, G., Moksnes, A. & Røskaft, E. 2004. Rejection of conspecific eggs in chaffinches: the effect of age and clutch characteristics. - Ethology 110: 459-470. Streif M., and O. A. E. Rasa. 2001. Divorce and its consequences in the common blackbird Turdus merula. Ibis 143: 554–560. Tewksbury, J. J., T. E. Martin, S. J. Hejl., M. J. Kuehn, and J. W. Jenkins. 2002. Parental care of a cowbird host: caught between the costs of egg-removal and nest predation. Proceedings of the Royal Society of London B 269: 423–429.

21

Török, J. 1981. Food composition of nestling blackbirds in an oak forest bordering on an orchard. Opusc. Zool. Budapest 17–18: 145–156. Török, J. 1985. Comparative ecological studies on blackbird (Turdus merula) and song thrush (T. philomelos) populations. I. Nutritional ecology. Opusc. Zool. Budapest 21: 105–135. Török, J., and E. Ludvig 1988. Seasonal changes in foraging strategies of nesting blackbirds (Turdus merula L.). Behavioral Ecology and Sociobiology 22: 329–333. Wiley, J. W. 1988. Host selection by the shiny cowbird. Condor 90: 289–303. Wood, D. R., and E. K. Bollinger. 1997. Egg removal by brown-headed cowbirds: A field test of the host incubation efficiency hypothesis. Condor 99: 851–857. Wyllie, I. 1981. The cuckoo. Batsford, London, UK. Yom-Tov, Y. 2001. An updated list and some comments on the occurrence of intraspecific nest parasitism in birds. Ibis 143: 133–143.

Table 1. Interspecific and conspecific brood parasitism hypotheses predict different host behaviours in sympatry and allopatry with interspecific brood parasite (e.g. the Cuckoo) and host life-history variables. Variable • Egg rejection rate and latency to rejection • Aggression against the Cuckoo • Latency of response to dummy • Nestlings food suitable for a young Cuckoo • Host nestlings growth rate

22

Interspecific parasitism Higher and faster sympatry Higher in sympatry

Conspecific parasitism in No specific effects

No aggression or no effect of sympatry/allopatry Shorter in sympatry, shorter No specific effects in cuckoo trials Necessary Not necessary Slower or parasite

same

as

the Can be higher than in the parasite

09°14´ 17°15´

17°33´ 16°25´ 17°01´ 16°48´ 17°04´ 19°05´

55°44´

55°25´ 49°35´

49°12´ 49°32´

49°30´ 49°14´ 48°51´

48°48´ 48°51´ 47°30´

47°01´

Gundsømagle, DK Rørkær, DK Olomouc, CZ

Brno, CZ Grygov, CZ

Prerov, CZ V. Kninice, CZ D. Bojanovice, CZ Lednice, CZ Luzice, CZ Budapest, HU

Buda hills, HU

Hungary allopatry Hungary sympatry

Czech R. sympatry

Czech R. allopatry

Sweden allopatry Denmark sympatry

Total (N) Parasitised (N) Parasitised (%)

12°10´

63°04´ 62°47´ 62°43´ 62°18´ 55°42´

Tydal, NO Nerskogen, NO Innerdal, NO Kongsvold, NO Lund, SW

Norway sympatry

63°12´ 59°55´ 63°20´

Ler, NO Oslo, NO Songli, NO

19°00´

16°38´ 17°19´

11°34´ 09°37´ 08°47´ 09°37´ 13°10´

10°20´ 10°45´ 09°39´

10°57´

63°27´

Stjørdal, NO

Longitude E

Norway allopatry

Latitude N

Study area

Experimental group Turdus pilaris 593 0 0.0

-

-

1 -

2

-

-

438 -

-

152

Turdus illiacus 310 0 0.0

-

-

-

-

-

-

40 4 -

2 12

252

Turdus philomelos 153 0 0.0

-

4 -

6 8 66

6 40

5 28

2

18 1 3 2 5

3 1

3

Turdus merula 153 0 0.0

28

3 33

14 31

17 9

38

-

1 6

2 -

5

Acrocephalus scirpaceus 1612 247 15.3

-

717 895 -

-

-

-

-

-

-

-

Acrocephalus palustris 92 44 47.8

-

17 75 -

-

-

-

-

-

-

-

Acrocephalus arundinaceus 237 49 20.7

-

72 165 -

-

-

-

-

-

-

-

63 15 23.8

-

11 52 -

-

-

-

-

-

-

104 0 0.0

-

-

3 -

-

-

-

79 7 11 -

1 3

-

Acrocephalus schoenobaenu s Phylloscopus trochilus -

Fringilla coelebs 235 0 0.0

-

-

11 -

-

1

-

-

3 -

220

Sylvia atricapilla 149 0 0.0

-

15 2 -

15 75

-

-

1

-

-

41

Emberiza schoeniclus 94 2 2.1

-

1 1 -

-

-

-

-

68 8 13 -

1 -

2

Anthus pratensis 475 23 4.8

-

-

3 -

-

-

-

462 10 -

-

-

Phoenicurus phoenicurus 54 0 0.0

-

-

3 -

-

-

-

48 -

1

2

33 0 0.0

-

-

3 -

-

-

-

-

3 2

25

Prunella modularis

Table 2. Location (NO = Norway, SW = Sweden, DK = Denmark, CZ = Czech Republic, HU = Hungary) and sample sizes (overall number of nests tested with eggs and/or dummies) for allopatric and sympatric populations of Thrushes and control species studied. Numbers of nests naturally parasitised by the Cuckoo are also shown.

5 0 0.0

-

-

2 1

-

-

-

1 -

1 -

-

Erithacus rubecula

Table 3. Comparison of life-history characteristics (mean±1SE) in thrushes and two control groups of hosts (see Methods). Index of host suitability=(host breeding success in %)*(100– host rejection of parasitic non-mimetic model eggs in %)*(100–proportion of host individuals aggressive towards the cuckoo dummy in %). Differences in all parameters tested with ANOVA and Tukey-Kramer post-hoc tests (the unit of analysis was an average value for every species within a group). Groups with the same superscript are not significantly different in post-hoc comparisons. We did not use Bonferroni corrections as recommended by Nakagawa (2004).

Breeding population (*1 000 000 ex.) Breeding density (pairs/km2) Breeding success (literature; %) Average laying date (day 0=1st April) Egg size (cm3) Clutch size (literature) Clutch size (present study) Incubation period (days) Fledging period (days) Adult body size (g) Aggression to cuckoo (%) Egg rejection (%) Interclutch variation Intraclutch variation Diptera & Araneida (% dominance) Earthworms (% dominance) Eviction success (%) Index of host suitability Natural parasitism rate (%)

24

Thrushes (n=4) 17.7±8.9

Rejecters (n=8) 19.6±10.5

Acceptors (n=4) 14.6±7.5

P 0.95

P posthoc –

33.3±10.8

236.3±129.9

84.6±50.0

0.47

–

50.6±3.8

54.0±7.2

61.0±6.5

0.67

–

36.4±5.8A

60.1±2.8B

62.0±5.0B

0.002

0.009

6.1±0.5 A

2.0±0.2 B

2.1±0.1 B

<0.0001

<0.0001

4.8±0.3

4.8±0.2

5.3±0.4

0.48

–

4.7±0.3

4.9±0.2

5.6±0.2

0.05

–

13.5±0.3

13.0±0.3

13.8±0.3

0.19

–

12.4±0.8

12.2±0.5

12.6±0.4

0.88

–

84.7±9.7 A

17.0±2.6 B

17.6±1.0 B

<0.0001

<0.0001

18.1±10.6 A

69.2±11.9 B

23.2±15.2 A B

0.02

0.05

42.7±9.9 A B

70.4±7.9 A

15.8±8.6 B

0.003

0.005

3.0±0.2

3.4±0.2

2.6±0.6

0.22

–

2.2±0.2

2.0±0.1

1.6±0.2

0.08

–

29.9±17.2 A

93.6±1.5 B

95.1±2.8 B

<0.0001

0.0004

63.5±18.5 A

0.0±0.0 B

0.1±0.0 B

<0.0001

0.0007

33.0±28.0 A

100.0±0.0 B

88.5±11.5 B

0.01

0.04

22.9±6.9 A B

5.3±2.7 B

44.1±11.1 A

0.0017

0.002

0.0

0.3

4.0

0.02

Table 4. Differences in responses of the four Thrush species towards Cuckoo and Crow dummies and blue and spotted model eggs in sympatry and allopatry with the Cuckoo. Responses to dummies were analysed with chi-squares (absence of crow experiments in sympatry did not enable to built a models including all factors in fieldfare and redwing). In comparisons of responses to Cuckoo and Crow dummies data from sympatric and allopatric areas were pooled. Rejection responses to the two egg types in sympatry/allopatry were analysed with a logistic regression model that included breeding stage, egg type, sympatry/allopatry and their interactions as factors and acceptance/rejection reaction or time to rejection as a response. Effects of breeding stage and all interactions were nonsignificant in all cases and were removed. P values (significant in bold) from final reduced models that included only egg type and sympatry/allopatry status are shown. Sample sizes in parentheses. Type of experiment Cuckoo aggression)

dummy

Factor

(% Sympatry 0.0 (12)

Crow dummy (% aggression)

Dummy (% aggression)

T. pilaris

33.3 (3)

17.8 (45)

54.2 (24)

0.0. (16)

6.5. (46)

9.1 (11)

43.8 (32)

P

1.00

0.19

0.46

0.44

Sympatry –

–

16.7 (24)

21.2 (33)

Allopatry

83.2 (113)

78.4 (74)

35.0 (20)

39.6 (53)

P

–

–

0.16

0.07

Cuckoo

0.0 (28)

8.2 (49)

16.1 (56)

48.2 (56)

Crow

83.2 (113)

78.4 (74)

25.0 (44)

32.6 (86)

P

0.0001

0.0001

0.27

0.06

33.3 (21)

44.4 (18)

42.6 (47)

63.9 (122)

Spotted

24.1 (29)

31.0 (129)

75.8 (62)

58.7 (46)

P

0.14

0.36

0.0005

0.56

19.0 (21)

60.0 (75)

65.5 (55)

Egg experiments (% rejection) Sympatry 14.7 (34)

Latency to rejection (days)

T. philomelos T. merula

Allopatry

Egg experiments (% rejection) Blue

Latency to rejection (days)

T. iliacus

Allopatry

24.3 (37)

34.9 (126)

64.7 (34)

61.1 (113)

P

0.91

0.20

0.58

0.62

Blue

3.6±0.6 (5)

3.0±0.9 (7)

2.8±0.3 (18)

1.6±0. (77)

Spotted

2.6±0.5 (7)

2.8±0.2 (39)

2.2±0.2 (42)

2.9±0.3 (27)

P

0.38

0.87

0.12

0.0001

Sympatry 2.3±0.3 (3)

1.8±0.5 (4)

2.5±0.2 (42)

1.7±0.2 (36)

Allopatry

3.2±0.5 (9)

2.9±0.3 (42)

2.0±0.3 (18)

2.1±0.2 (68)

P

0.70

0.19

0.12

0.36

25

Figure 1 Distribution of laying dates (the start of particular clutch) for blackbird (filled bars; N=1708), song thrush (hatched bars; N=607) and cuckoo (open bars; N=405). Data (from Hudec 1983) are pooled for decades from the start of March (decades 1–3 = March, 4–6 = April, 7–9 = May, 10–12 = June, 13–14 = July).

30 25 20 % 15 10 5 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14

Decade (March - July)

26

Figure 2 Mean intra- (black bars) and interclutch- (gray bars) variation in egg appearance ± SD for rejecters of non-mimetic parasitic eggs (rejection rate >80%), Turdus-species, Acrocephalus-species and acceptors of non-mimetic parasitic eggs (rejection rate <20%). For more details see text. 4.5 4 3.5 3 2.5 2 1.5 1 Rejecters

Turdus

Acrocephalus

Acceptors

27

25. Krist M. & Grim T.: Are blue eggs sexually selected signal of female collared flycatchers? 25. experiment. A cross-fostering Krist M. & Grim T.: Are blue eggs sexually selected signal of female collared flycatchers? (subm.) A cross-fostering experiment. (subm.)

Are blue eggs sexually selected signal of female collared flycatchers? A cross-fostering experiment Miloš Krist1,2 and Tomáš Grim2 Museum of Natural History, nám. Republiky 5, 771 73 Olomouc, Czech Republic Department of Zoology, Palacký University, tř. Svobody 26, 771 46 Olomouc, Czech Republic 1 2

Author for correspondence: Miloš Krist Museum of Natural History Nám. Republiky 5 771 73 Olomouc Czech Republic E-mails: [email protected], [email protected] Phone: +420-585515128 Fax.: +420-58522743

ABSTRACT Impressive variation in egg colouration among birds puzzled evolutionary biologists for a long time. Most frequently studied selective forces moulding egg colouration – predation and brood parasitism – either received little empirical support or may play role only in a minority of species. A novel hypothesis suggested that egg colour may be significantly influenced by sexual selection. Females may deposit blue-green pigment biliverdin into eggshells instead of using it for themselves as a powerful antioxidant. By this handicap females may signal their quality to males which are then hypothesized to increase their paternal effort. We tested the hypothesis experimentally in the collared flycatcher (Ficedula albicollis), a species laying blue-green eggs. We cross-fostered clutches between nests to disentangle effects of female/territory quality and egg colour on paternal effort and nestling quality. Results supported two assumptions of sexual signalling through egg colour hypothesis: blue pigment seems to be a limited resource for females and female quality is positively correlated with intensity of blue-green colour. However, we did not find support for the main prediction of the hypothesis as male parental effort parameters (feeding frequencies to nestlings and intensity of nest defence) were unrelated to egg colour. We discuss possible reasons for discrepancy between our results and previous correlative analyses that supported the hypothesis that blue egg colour may be a postmating sexually selected signal in females. Key words: egg color, differential allocation, female signaling, immunity, parental investment.

1

INTRODUCTION Egg colouration has been attracting attention of biologists for decades and consequently, various hypotheses have been generated to explain variation in this trait (Underwood & Sealy, 2002). Recent research indicates that pigments may physically strengthen the eggshell (Gosler et al., 2005). However, majority of suggested explanations propose that egg colouration has mainly signalling function to either heterospecifics or conspecifics. On the heterospecific level, the most attention has been paid to nest predators and brood parasites. Eggs may have cryptic colours to preclude predators from locating eggs/nests (Tinbergen et al., 1962). Brood parasitism may select for small intraclutch variation but large interclutch variability in egg colouration in hosts (Øien et al., 1995; Soler & Møller, 1996) and hosts may select for mimetic eggs in parasites by rejecting eggs differing from their own (Davies & Brooke, 1988; Grim & Honza, 2001). On the intraspecific level egg colouration may be important for recognition of own eggs in colonial birds (Birkhead, 1978). By laying pale last egg in a clutch females may signal that they started to incubate which would diminish opportunities for successful conspecific brood parasitism (Yom-Tov, 1980; Ruxton et al., 2001). The support from observations and experiments for each of these hypotheses is equivocal (Underwood & Sealy, 2002) which also means that large amount of variation in egg colour remains unexplained. For example, open-cup nests baited with differently coloured (white, blue, and cryptic) eggs survived equally in a carefully designed study (Weidinger, 2001) despite the fact that it was conducted in a population where most nests were eventually destroyed by nest predators and where one could therefore expect strong selection for egg crypsis. One of the most striking – and most puzzling – egg colour morhps are blue eggs (Underwood & Sealy, 2002). The evolutionary significance of blue eggs was previously studied in the context of predation and brood parasitism. Götmark (1992) and Weidinger (2001) found no support for a hypothesized cryptic function of blue eggs (blending with specific nest micro-environment). On the other hand there is some evidence that the blue egg morph in the common cuckoo (Cuculus canorus) is counter-adaptation against host egg discrimination in foster species laying bluish eggs (Moksnes et al. 1995). However, this does not explain why the hosts themselves lay blue eggs in the first place. Consequently, bluegreen colouration of eggs has been considered to be mystery until recently (Underwood & Sealy, 2002). Recently, new hypothesis based on intraspecific signalling has been proposed to explain variation in egg colours (Moreno & Osorno, 2003). According to this hypothesis females colour eggs by costly pigments to signal their quality and consequently also the quality of their progeny to mates. The costs of such extended phenotype should be outweighed by benefits from increased paternal care into the current brood. Such increase of paternal care in high quality offspring is in turn predicted by the differential allocation hypothesis (Burley, 1986; Sheldon, 2000). Moreno and Osorno (2003) paid special attention to blue-green eggs when formulating their hypothesis because biliverdin, which causes the blue colour of eggs (Mikšík et al., 1996) has also strong antioxidant activity (McDonagh, 2001; Kaur et al., 2003). Thus the deposition of biliverdin into eggs may signal female capacity to control free radicals despite the handicap (Moreno & Osorno, 2003). Such signal would be in principle similar to colouration of plumage with other strong antioxidants like carotenoids, the topic that received considerable attention at present (e.g. Hill, 2002; McNett & Marchetti, 2005). So far three correlational studies tested and found support for sexual selection hypothesis of evolution of blue colouration of eggs (SSEC hypothesis). Moreno et al. (2004) found in the pied flycatcher (Ficedula hypoleuca), a species laying blue eggs, that males fed with greater frequency broods hatched from eggs with more saturated colour. Moreno et al. (2005) found in the same species a positive correlation between female immunocompetence and saturation of blue colour of their eggs. In that study, colour of eggs became also brighter and less saturated in the course of laying which suggests that the pigment was depleted as laying progressed supporting the crucial assumption that biliverdin is a limited resource for females. Finally, Soler et al. (2005) found in a comparative study positive correlations between blue colour of eggs and mating system and duration of nestling period. The two latter variables were used by Soler et al. (2005) as surrogates of intraspecific variation in paternal effort, a variable that should affect evolution of female signalling.

2

Although all three studies gave results that were consistent with SSEC hypothesis, they could not prove causality of the observed relationships because of their correlative nature. Thus authors of these studies suggested that experimental approach is needed for further progress in testing of the SSEC hypothesis (Moreno et al., 2004, 2005; Soler et al., 2005). In the two studies on the pied flycatcher (Moreno et al., 2004, 2005) the reflectance spectra of eggs were measured at wavelength above 400 and 360 nm respectively. As most passerines are able to see even shorter wavelength (Cuthill et al., 2000), further studies including UV part of spectrum in measuring of egg colour are particularly needed (Moreno et al., 2004, 2005; Soler et al., 2005). The main aim of this study was to test experimentally whether there is a causal link between egg colour and paternal effort. We performed the study on the collared flycatcher (Ficedula albicollis) which lays, similarly as its sister species, the pied flycatcher, blue-green eggs. As the collared flycatcher is a hole nester the potential confounding effect of both predation and interspecific brood parasitism on egg colour can be rejected a priori. To test for causality between egg colour and paternal effort, we cross-fostered freshly laid eggs between pairs of nests. Consequently, egg colour was randomized with respect to parental and territory quality. We have included UV-range in measuring of egg spectra to deal with presumed UV-vision of flycatchers (see Cuthill et al., 2000). We have scored three measures of paternal effort (feeding frequencies to young and old nestlings and nest defence against a nest predator) to obtain its more detailed estimate. In addition we tested for relationships between egg colour and offspring survival, morphology, and T-cell mediated immunity. We paid also attention to female parents to test whether the egg colour is correlated with other female traits. If SSEC hypothesis (Moreno & Osorno, 2003) holds, egg colour of cross-fostered but not original eggs should predict paternal effort. If colour of original eggs predicted paternal effort, this would suggest one of the following: (1) males adjust their effort according to some female trait that is correlated with colour of her eggs, (2) high quality females laying eggs with more saturated colours mate assortatively with high quality males that are able to provide superior parental care, (3) the relationship is driven by territory quality: on superior territories it may be less costly both to lay eggs with saturated colours and feed young with greater frequency. If offspring quality was correlated with colour of cross-fostered eggs (i.e. eggs from which the young actually hatched), this would imply either genetic or environmental (i.e. egg composition) quality of eggs with saturated colour or differential male effort. In contrast, if colour of original eggs was predictive of offspring quality, this would imply that parental or territory quality is correlated with egg colour. METHODS Field methods We conducted the study in the Velký Kosíř area (49°32'N, 17°04'E, 300 – 400 m a.s.l.), the Czech Republic in 2005. In the study area there were about 300 nest-boxes in oak (Quercus petraea) forest. We conducted the experiment with collared flycatcher, a small migratory passerine that easily adopts nest-boxes for breeding. Females lay usually one egg per day and solely incubate clutches of 4 – 8 eggs. Similarly as in the sister species, the pied flycatcher (Moreno et al. 2005), eggs are unspotted and brightly blue-green in colour. Both collared flycatcher parents feed nestlings with invertebrate food for about 15 days until fledging. We have conducted cross-fostering experiments among 70 occupied nest-boxes. We cross-fostered eggs between pairs of nests in which laying began on the same day (n = 30 nest pairs) or which differed by one day in laying date (n = 5 nest pairs). We cross-fostered eggs on the day they were laid with the exception of 5 nests in which the first eggs were moved the day after they were laid. We have recorded both the time of removing of the original egg and that of adding the cross-fostered egg. We performed the exchange of first eggs of the clutches within two hours during which nest-box entrances were blocked by sticks to ensure that no parent recorded that its nest is empty. Subsequent exchanges were done within more variable time (0 – 10 hours) during which the nest-box entrances were not blocked. We have continued with exchanges on daily basis until clutches in both nests were completed. No experimental nest was abandoned in the course of laying.

3

The experimental design ensured that in the incubation stage there were only crossfostered eggs in a focal nest. However, in the laying stage there were both original and crossfostered eggs in focal nests. Under the assumption that eggs were visible from 6 a.m. to 8 p.m. in the laying stage (before the last egg was laid) the number of egg-hours for which original eggs were exposed in nests was 41.7 ± 9.9 (mean ± SD), while the respective figure for cross-fostered eggs was 195.0 ± 62.0. Some females began to incubate before their clutch was complete. As eggs are less visible when female incubate, we counted also number of egg-hours for which eggs were exposed in nests before females began continuous incubation. For original eggs this figure is 35.7 ± 10.9 while that for cross-fostered eggs is 143.8 ± 77.6. Taken together, these data suggest that colour of cross-fostered eggs should have overwhelming effect on paternal behaviour when compared with that of original eggs. Moreover, we can reasonably expect males to examine egg colours after the clutch completion to avoid any assessment bias resulting from assessing only partially laid clutch. In another system where birds assess egg appearance (hosts of brood parasites) this happens only after the clutch completion (Davies & Brooke, 1988). We began to check nests two days before the presumed hatching to determine hatching date and hatching success. To estimate cell-mediated immune response of nestlings, we injected them with 0.1 mg phytohaemagglutinin in 20µl of physiological saline solution into right wing web when they were 12 days old. Before injection we took two measurements of wing web thickness (to the nearest 0.01 mm) with a thickness gauge (Mitutoyo Quick-Mini) that was adjusted to push with constant pressure of one Newton. We re-measured wing web thickness 24 h (± 2 h) after injection. Both measures were highly repeatable (before injection: r = 0.797, F251,252 = 8.84, p < 0.001; after injection: r = 0.972, F249,250 = 70.38, p < 0.001). Therefore we calculated cell-mediated immune response for each chick as the difference in average thickness of wing web after injection minus the average thickness of wing web before injection. At age 13 we also weighed nestlings with Pesola spring balance (to the nearest 0.25 g), measured their tarsus with digital calliper (to the nearest 0.01 mm), and recorded whether ectoparasitic mites (Dermanyssus gallinoides) are present in the nest. We have also captured adults while feeding nestlings and weighed and measured them in the same way as nestlings. Measures of parental investment To investigate parental investment in the current brood, we recorded two principal types of investment. First, we recorded feeding frequencies by parents at two ages of young. We video-taped nest-boxes with cameras for 70 minutes on day 6 and 12 of the nestling period (hatching day = day 0). Feeding frequency was determined for each sex separately as the number of visits of nest-box per hour that started ten minutes after beginning of the taperecording. We have disregarded the first ten minutes of record to minimize the effect of disturbation due to installation of cameras. Collared flycatchers in our study area readily resume feeding regimes within minutes after next-box checks (personal observations). In a few cases only one parent fed the young. We suppose that in majority of these cases the other member of the pair was dead rather than investing little in the brood. Therefore we have excluded such nests from the analyses. Second, we have recorded parental nest defence against a predator of eggs and young. At variable age of young (5 – 13) we simulated intrusion of the great spotted woodpecker (Dendrocopos major), which is a common nest predator on our study plots, in the close vicinity of flycatchers’ nests. We used a stuffed specimen of female woodpecker in a posture simulating foraging on the bark. We attached the dummy to the bark of the tree about 0.5 m below the nest-box of a tested pair. The dummy was oriented (“looked”) towards the nestbox. Before the dummy had been placed, we installed also a camera that video-taped vicinity of the nest-box. After experiment started the observer retreated into the shelter that was installed about 20 m from the focal nest-box. After a member of the pair approached to a close vicinity of the nest and so presumably noticed the dummy, observation of this individual began. Observations for the later-arriving individual began after its arrival. Observations lasted 5 min for each focal bird. During that time we recorded for each individual the number of dive/contact attacks against a dummy and the latency from arrival to the first dive/contact attack. We focused our attention on dive/contact attacks because these are presumably most risky behaviour with the greatest efficiency in deterring nest

4

predators (see also Krist, 2004; Grim, 2005). Further we scored the overall response of each individual on an ordinal scale: (1) normal feeding or silent watching on the dummy, (2) fluttering against a dummy and/or vocalization, (3) dive or contact attack against a dummy. We have checked our notes in a field diary against a video record to ensure that no behaviour was overlooked when quantifying parental response for analyses. Egg colour measurements We measured colour of eggs on the day they were laid before their transfer to foster nests. Egg colour was measured by Avantes spectrometer (AvaSpec-2048) which was configured for measurements in UV-visible part of spectrum. The light source for measurements was PX-2 pulsed xenon lamp. The spectrometer and the lamp were connected by bifurcated fiber optic cable to reflection probe which consisted from seven optic fibres. Six of them transfer light from PX-2 lamp to measured surface and the seventh one transfers reflected light to spectrometer. Reflection probe was fixed in a probe holder at an angle of 45º and a distance of approximately 1cm to measured surface. We have placed eggs by side on the measurement port (ellipse with axes 7 and 9 mm) in the probe holder. We have covered eggs by black cap to prevent ambient light to confuse measurements. To improve the signal/noise ratio, each spectrum was obtained as the mean of 100 readings with one reading over 10-ms integration time. We have measured each egg twice (on the opposite sides) and the mean from the two measurements used in subsequent analyses. All data were generated relative to white standard (WS-2 Avantes). We have calculated three values from reflectance spectra. (1) Brightness or total reflectance was obtained as the summed reflectance at each 1-nm interval from 301 to 700 nm. (2) Blue-green chroma (BG chroma hereafter) was calculated as the reflectance between 401–600 nm divided by the total reflectance. We have focused on this part of the spectrum because biliverdin, the main eggshell pigment of blue-green eggs (Mikšík et al 1996) absorbs light weakly at this spectral range but strongly at shorter (<400 nm) and longer (>600 nm) wavelengths (Falchuk et al. 2002) which causes the blue-green appearance of eggs. As BG chroma should be most closely linked to biliverdin content of eggshells (see Saks et al 2003 for an analogous case with carotenoid pigments), we would expect that it should be the main cue for male assessment of female quality and therefore the main predictor of paternal behavior. (3) Hue was calculated as arctan{[(QG-QUV)/QT]/[(QR-QB)/QT]}. In this equation QT denotes brightness (summed reflectance between 301–700 nm), QUV is summed reflectance in the UV area of the reflectance spectrum (301–400 nm), QB is summed reflectance in blue area of the spectrum (401–500 nm), QG is summed reflectance in green area of the spectrum (501–600 nm), and QR is summed reflectance in the red area of the spectrum (601–700 nm). The method of calculating hue is basically the same as that used by Saks et al. (2003) with the exception that we used the whole spectral range visible to birds (i.e. 301–700 nm) in the calculations (for similar approach see McNett & Marchetti, 2005). Eggs with greater values of the hue (i.e. less negative values) have the peak of reflectance at shorter wavelength than those with more negative values. Mean brightness of eggs in a clutch was strongly correlated with mean BG chroma (r = – 0.835, n = 70, p < 0.001) while mean hue was not correlated with either mean brightness (r = 0.009, n = 70, p = 0.944) or mean BG chroma (r = – 0.179, n = 70, p = 0.138). The strong correlation between mean brightness and mean BG chroma means that these two measures are from the large part the same trait. We fit, however, separate models for these two spectral qualities, as the use of hue, saturation (chroma) and brightness is a standard way of description of spectra (Endler, 1990; Hill, 2002). However, we focus more attention on BG chroma which should be more indicative of biliverdin content in eggshells (see above). Statistical analyses We fitted separate models for effects of each spectral quality (i.e. brightness, BG chroma, and hue) on paternal effort and nestling performance. In each of these models, the spectral quality in question of both original and cross-fostered eggs was the main factor of interest and therefore always retained in final models. To reduce unexplained variation and increase the power of test, we added potentially important covariates as predictors in the initial models. These covariates were backward eliminated on the base of their significance. Thus

5

only significant covariates retained in final models. Different class of covariates entered models with different response variable. In models where the response variable was feeding rate, we controlled for hour, ambient temperature, and brood size when feeding frequency was recorded. We fitted separate models for paternal feeding frequency at the two ages of young, as these were only weakly correlated (r = 0.222, n = 45, p = 0.143). In models where the response variable was nest defence we controlled for age and brood size on the day of experiment. In models where the response variable was hatching/fledging success or nestling immunity we controlled for laying date of original clutch (1st May = 1), total feeding frequency at age 12, number of parents attending the nest, presence of mites (binary variable), and the difference between original and cross-fostered clutch size. We included the difference in clutch sizes in the model to control for potential effect of enlargement or reduction of clutch size caused by our cross-fostering design. We included number of parents attending the nest in addition to total feeding frequency because we assumed that widowed parent may reduce the quality of delivered food or quantity of food delivered per visit. In models where the response variable was fledging weight or tarsus length, we included all covariates as in the former models and additionally also midweight or midtarsus of genetic parents respectively. Weight of parents was adjusted for age of nestlings when parent was captured by including residuals from regression of parental weight on nestling age rather than the actual parental weight into the model. The reason for inclusion of parental characteristic into these models is that morphological traits are usually highly heritable (Merilä & Sheldon 2001) which means that they would affect the response variable. If parental morphology was correlated with the colour of original eggs, its inclusion among predictors would be particularly needed to avoid spurious results (Krist & Remeš 2004). We used brood means in models investigating effect of predictors on nestling performance (immunity, weight, tarsus length). Only the young that subsequently fledged were included in computation of the brood means. Only the broods in which at least one young hatched/fledged were included in the analyses of hatching/fledging success respectively. To investigate the relationship between female characteristics and colour of eggs she lays we computed correlations between spectral qualities of original eggs and female age, morphology (tarsus length, condition), and some reproductive parameters (egg size, clutch size, laying date). We knew exact age of 37 females that we had ringed as nestlings in previous years. As this variable was not normally distributed, we assessed the relationship between female age and egg colour by non-parametric Spearman rank correlation. We decided to use female condition rather than weight for analyses as the latter was strongly correlated with tarsus length (r = 0.371, n = 54, p < 0.001). Thus condition was determined as residuals from regression of female weight on tarsus length and age of young at female capture (weight = –2.748 + 0.857 × tarsus – 0.155 × age; F2,51 = 20.89, p < 0.001). To test for potential association between egg colour and maternal effort (intensity of nest defence, feeding frequency) we fitted models where egg colour was independent factor of interest and maternal effort the response variable. We fitted models instead of computing correlations since it is more easy to control for covariates in the former statistical design. In females, feeding frequencies recorded at the two ages of young were strongly correlated (r = 0.465, n = 45, p = 0.001). Therefore we performed PCA and used first principal component (73.3% of variation explained) as a measure of maternal feeding frequency. Consequently, we included in initial models hour (daytime), ambient temperature, and brood size recorded at both days of measuring of feeding frequency. In the model for female intensity of nest defence we included the same class of covariates as for male nest defence. To investigate the effect of laying order on the three spectral measures, we have first subtracted clutch mean from the actual value of the spectral measure (e.g. actual brightness of the sixth egg minus average brightness of eggs in that nest). By this method (centering) we have received values that are more comparable between clutches and consequently the power of statistical tests relating to intraclutch variation in egg colour has been increased. We fitted two types of models. In the first group we used actual laying order as the predictor variable. In the second we used relative laying order which might be more appropriate when different clutch sizes are pooled into single analysis. In our sample clutch size varied between 4 – 8. We therefore categorised eggs into the following categories: last, penultimate, pre-penultimate, and precedent eggs. In the precedent category 1 – 4 eggs are pooled depending on clutch size.

6

We fitted models with continuous response variables in JMP (SAS Institute 1995) and models with categorical or ordinal response in PROC GENMOD (SAS Institute 1999). We determined repeatabilities from variance components (PROC VARCOMP; SAS Institute 1999); the accompanying F-tests are the results of one-way ANOVAs. We determined prospective power of our tests in PROC POWER (SAS Institute 1999). When hatching and fledging success were the response variables, these were included in the models in event/trial syntax where the number of eggs/hatchlings in individual nests were trials and number of hatchlings/fledglings in individual nests were events in models for hatching and fledging success respectively. These models had binomial error structure, logit link function, and statistics corrected for overdispersion by Pearson chi-square/DF. We assessed the paternal effort in nest defence in two models: (1) The response variable was the overall intensity of nest defence (ordinal variable; multinomial error structure and cumulative logit link function). (2) In the subgroup that attacked the dummy by dive or contact, the number of attacks and the latency to the first attack were correlated (r = – 0.431, n = 23, p = 0.040). Therefore we performed PCA and used first principal component (71.6% variability explained) for subsequent analyses. Greater values of PC1 indicate more attacks and shorter latency to the first attack. Colour of original and cross-fostered eggs were correlated within individual nests (brightness: r = 0.259, n = 70, p = 0.031; BG chroma: r = 0.278, n = 70, p = 0.020; hue: r = 0.175, n = 70, p = 0.147). This might be caused by the fact that we cross-fostered eggs between pairs of nests with the same laying date (which is unavoidable in this sort of experimental study) and, due to time constraints, usually also between nests that were located in the same of the four study plots. Correlations among predictors in multiple regression (collinearity) might reduce the power of statistical tests (Quinn & Keough 2002). To assess the effect of collinearity on our results, we looked at variance inflation factors (VIF) for our predictors in individual models with continuous response variable. Predictors with VIFs < 10 are generally accepted as giving unbiased results (Quinn & Keough 2002). Recently, it has been suggested that VIF as small as two might bias results (Graham 2003). Anyway, all VIFs were smaller than 1.2 in our analyses. Therefore collinearity did not bias our results. RESULTS The reflectance spectra of eggs of the collared flycatcher have a bimodal shape. The major peak of reflectance lays in the blue-green part of the spectrum which agrees with humanperceived colour of these eggs while the minor one lays in UV part of the spectrum which is invisible to humans (Figure 1a). All three spectral measures changed significantly throughout the laying order. Relative brightness and relative hue increased linearly (Brightness: F1,445 = 783.9, p < 0.001, R2 = 0.638, Figure 1c; Hue: F1,445 = 100.2, p < 0.001, R2 = 0.184, Figure 1d), while relative BG chroma decreased non-linearly in the laying order (quadratic regression; F2,444 = 241.1, p < 0.001, R2 = 0.520, Figure 1b). These relationships hold even when the relative laying order is used for these tests (Brightness: F1,445 = 645.8, p < 0.001, R2 = 0.592; Hue: F1,445 = 95.4, p < 0.001, R2 = 0.177; BG chroma – quadratic regression: F2,444 = 365.1, p < 0.001, R2 = 0.622) Despite the significant intra-clutch trends, all three spectral measures were also significantly repeatable within clutches (brightness: r = 0.437, F69,377 = 5.52, p < 0.001; BG chroma: r = 0.509, F69,377 = 6.71, p < 0.001; hue: r = 0.315, F69,377 = 4.05, p < 0.001) which is a premise of signalling function at a clutch-level. Therefore, we used clutch means of spectral measures in subsequent analyses. The summed reflectance in the UV part of spectrum was most variable between clutches (CV = 12.6%), followed by summed reflectance in red (7.9%), blue (5.3%), and green (4.9%) part of spectrum. The above differences in coefficients of variation should be only little affected by different precision of measuring apparatus in particular part of spectra, since are based on repeated measurements of eggs in a clutch (8–16 measurements depending on clutch size). Egg colour and paternal effort Paternal feeding frequency at nestling age 6 was not related to any spectral quality of original or cross-fostered eggs while it was negatively affected by hour of observation and

7

positively affected by brood size [BG chroma original (BGOR): F1,49 = 0.02, p = 0.899, BG chroma cross-fostered (BGCF): F1,49 = 2.76, p = 0.103, hour: F1,49 = 5.50, p = 0.023, Figure 2a; brightness original (BOR): F1,48 = 0.65, p = 0.425, brightness cross-fostered (BCF): F1,48 = 2.46, p = 0.123, hour: F1,48 = 7.35, p = 0.009, brood size : F1,48 = 4.05, p = 0.050; hue original (HOR): F1,49 < 0.01, p = 0.993, hue cross-fostered (HCF): F1,49 = 0.77, p = 0.383, hour: F1,49 = 5.15, p = 0.028). Similarly, no spectral quality predicted paternal feeding frequency at age 12 while this was positively affected by brood size (BGOR: F1,41 = 3.57, p = 0.066, BGCF: F1,41 = 0.70, p = 0.409, brood size : F1,41 = 7.08, p = 0.011, Figure 2b; BOR: F1,41 = 2.23, p = 0.143, BCF: F1,41 = 1.19, p = 0.282, brood size : F1,41 = 7.59, p = 0.009; HOR: F1,41 = 1.89, p = 0.177, HCF: F1,41 = 0.23, p = 0.633, brood size : F1,41 = 7.82, p = 0.008). The overall intensity of nest defence by males was not affected by any spectral quality of original or cross-fostered eggs (BGOR: χ21,47 = 0.92, p = 0.336, BGCF: χ21,47 = 2.00, p = 0.157, Figure 2c; BOR: χ21,47 = 0.55, p = 0.457, BCF: χ21,47 = 0.05, p = 0.820; HOR: χ21,47 = 0.04, p = 0.850, HCF: χ21,47 = 0.02, p = 0.898). In the subgroup of males that attacked the dummy, BG chroma of original eggs and brood size negatively affected the intensity of attacks (BGOR: F1,19 = 10.48, p = 0.004, BGCF: F1,19 = 0.64, p = 0.435, brood size: F1,19 = 5.34, p = 0.032, Figure 2d) while the other two spectral qualities did not predict the intensity of attacks (BOR: F1,20 = 0.74, p = 0.400, BCF: F1,20 = 3.01, p = 0.098; HOR: F1,20 = 0.06, p = 0.813, HCF: F1,20 = 0.55, p = 0.469). Egg colour and offspring performance Out of 70 experimental clutches 56 hatched. In these 339 eggs hatched and 20 failed to hatch. No variable was predictive of hatching success (BGOR: F1,53 = 1.74, p = 0.192; BGCF: F1,53 = 0.66, p = 0.419; BOR: F1,53 = 1.63, p = 0.207, BCF: F1,53 = 0.41, p = 0.526; HOR: F1,53 = 0.25, p = 0.619, HCF: F1,53 = 0.90, p = 0.347). Out of 56 hatched broods 50 fledged. In these nests 249 young fledged and 52 died. Fledging success was lower in nests that were attended by only one parent, infested by mites, and initiated late in the season. In contrast to these significant effects of covariates, no spectral quality predicted fledging success (BGOR: F1,45 = 0.64, p = 0.426; BGCF: F1,45 = 0.04, p = 0.838; laying date: F1,45 = 7.51, p = 0.006; number of attending parents: F1,45 = 18.41, p < 0.001; BOR: F1,45 = 1.65, p = 0.206; BCF: F1,45 = 0.03, p = 0.867; laying date: F1,45 = 8.43, p = 0.006; number of parents: F1,45 = 18.71, p < 0.001; HOR: F1,36 = 0.22, p = 0.641; HCF: F1,36 = 2.51, p = 0.122; number of parents: F1,36 = 12.01, p = 0.001; presence of mites: F1,36 = 4.27, p = 0.046). Fledgling tarsus length was positively affected by midtarsus length of genetic parents and negatively affected by laying date in all models. From the spectral qualities only BGOR had significant and positive effect on fledgling tarsus length (BGOR: F1,32 = 6.67, p = 0.015; BGCF: F1,32 = 0.59, p = 0.449; laying date: F1,32 = 15.47, p < 0.001; parental midtarsus: F1,32 = 9.60, p = 0.004, Figure 3b; BOR: F1,32 = 2.78, p = 0.105; BCF: F1,32 = 0.08, p = 0.772; laying date: F1,32 = 16.17, p < 0.001; parental midtarsus: F1,32 = 10.25, p = 0.003; HOR: F1,32 = 1.30, p = 0.264; HCF: F1,32 = 1.73, p = 0.198; laying date: 12.33, p = 0.001; parental midtarsus: F1,32 = 9.67, p = 0.004). Fledgling weight was not affected by either covariates or spectral qualities (BGOR: F1,47 = 0.98, p = 0.326; BGCF: F1,47 = 1.77, p = 0.190, Figure 3a; BOR: F1,47 = 0.16, p = 0.688; BCF: F1,47 = 1.28, p = 0.263; HOR: F1,47 = 2.52, p = 0.119; HCF: F1,47 = 1.02, p = 0.319). Similarly to weight, T-cell mediated immunity of fledglings was also unaffected by any predictor in any of our models (BGOR: F1,47 = 0.33, p = 0.566; BGCF: F1,47 < 0.01, p = 0.952, Figure 3c; BOR: F1,47 = 2.04, p = 0.160; BCF: F1,47 = 0.28, p = 0.603; HOR: F1,47 = 2.81, p = 0.100; HCF: F1,47 = 2.78, p = 0.102). Original eggs and female traits Female age was positively associated with BG chroma of original eggs (rs = 0.399, n = 37, p = 0.014, Fig. 4a) while brightness (rs = –0.254, n = 37, p = 0.128) and hue of original eggs were unrelated to female age (rs = 0.214, n = 37, p = 0.204). There was marginally non-significant relationship between female morphology and colour of eggs she laid (condition: BGOR: r = 0.248, p = 0.070, Figure 4b; BOR: r = –0.213, p = 0.122; HOR: r = –0.136, p = 0.328; tarsus

8

length: BGOR: r = 0.251, p = 0.068; BOR: r = –0.269, p = 0.049, HOR: r = 0.169, p = 0.223; n = 54 in all cases). Similarly to female morphology, there was also marginally non-significant negative relationship between BOR and female feeding frequency (BOR: F1,42 = 3.18, p = 0.074, number of nestlings at age 6: F1,42 = 23.08, p < 0.001). BG chroma (BGOR: F1,42 = 1.17, p = 0.286, number of nestlings at age 6: F1,42 = 21.99, p < 0.001, Figure 4c) and hue (HOR: F1,41 = 0.24, p = 0.629, number of nestlings at age 6: F1,41 = 17.43, p < 0.001, hour of feeding at age 6: F1,41 = 4.47, p = 0.041) were not predictive of female feeding rate. No spectral quality of original eggs predicted the overall intensity of nest defence by females (BGOR: χ21,52 = 0.06, p = 0.809; BOR: χ21,52 = 0.04, p = 0.850; HOR: χ21,52 = 0.24, p = 0.625). Egg colour was associated with mean egg volume of the clutch; BGOR increased with egg volume (r = 0.261, n = 70, p = 0.029, Figure 4d), BOR decreased with egg volume (r = –0.442, n = 70, p < 0.001), and hue was unrelated to egg volume (r = 0.114, n = 70, p = 0.348). In contrast to mean egg volume, neither clutch size (BGOR: r = –0.035, n = 70, p = 0.772; BOR: r = 0.118, n = 70, p = 0.330; HOR: r = 0.113, n = 70, p = 0.350) or laying date (BGOR: r = 0.156, n = 70, p = 0.199; BOR: r = –0.093, n = 70, p = 0.443; HOR: r = 0.023, n = 70, p = 0.853) were related to egg colour. The (marginally) significant relationships that we found between egg colour and female morphology, feeding rate, and egg volume were probably not mediated through female age as the age was not related to any of the former variables (condition: rs = 0.053, n = 37, p = 0.758; tarsus length: rs = –0.002, n = 37, p = 0.989; feeding rate: rs = 0.172, n = 29, p = 0.371; egg volume: rs = 0.069, n = 37, p = 0.686). DISCUSSION We have found that egg colour changes in the laying order; brightness and hue increased linearly and BG chroma decreased non-linearly in the course of laying. We have also found correlations between egg colour and female traits. Eggs with more BG chroma were laid by older females and independently of this also by females with a tendency to be in better condition and to have longer tarsi. Saturation of egg colour was also positively associated with mean egg volume females laid. More pigmented (darker) eggs were laid by females that tended to feed the young with greater frequency. These findings supports the assumptions of sexually selected egg colour (SSEC) hypothesis (Moreno & Osorno, 2003). First, the blue biliverdin pigment colouring eggs seems to be limited source for laying females. Second, female quality is positively associated with egg colour. However, in contrast to support for the assumption of SSEC hypothesis, we have found no evidence in support for its prediction. Neither of our three measures of paternal care was associated with the colour of cross-fostered eggs. Furthermore, we have also found no association between paternal care and colour of original eggs. Consequently, our results are in contrast to previous correlative findings in the pied flycatcher (Moreno et al. 2004). Similarly to our study of the collared flycatcher, Moreno et al. (2005) have previously found nonlinear decrease in saturation and linear increase in brightness and hue of eggshells in the laying order in the pied flycatcher. Both adaptive and non-adaptive explanations were previously proposed to explain the pale colour of last eggs. Yom-Tov (1980) have suggested that laying pale last egg may be adaptive for females to prevent conspecific brood parasitism. By laying pale last egg females might indicate to conspecific females that they are already incubating the clutch. Conspecifics then should avoid to lay parasitic egg in the nest already containing pale egg since the probability of successful hatching would be diminished for such late-laid parasitic egg. Ruxton et al. (2001) formalized this idea and showed that such signalling may indeed theoretically work under some conditions. An alternative, non-adaptive explanation for the occurrence of pale eggs at the end of laying sequence is that pigments deposited in eggshells are depleted in the course of laying (Nice 1937). As we have detected gradual changes in brightness and hue, our data do not provide support for the adaptive scenario of Yom-Tov (1980) since only the last egg is predicted to be of different appearance under his hypothesis (Ruxton et al., 2001). BG chroma decreased at accelerating rate in the laying order, however, also in this case the decrease was evident as early as in the pre-penultimate egg (results not shown). Moreover, in our population there is no evidence for conspecific brood parasitism (see Krist et al., 2005) which suggests little opportunities for evolution of such anti-parasitic signalling

9

system. Taken together, our data suggest that eggshell pigments may be limited source for laying females which provide support for the assumption of SSEC hypothesis (Moreno & Osorno, 2003). To be honest signal for males, egg colour should be correlated with female and/or offspring quality (Moreno & Osorno, 2003). We have found that females laying presumably more costly (darker and more saturated) eggs tended to be in better condition, to have longer tarsi, and to feed the young with greater frequency; they also laid larger eggs, and were older than females laying less saturated and brighter eggs. These results are generally in the direction expected under the SSEC hypothesis and as such provide some support for its assumption. For example, egg size has been suggested to be indicator of female quality in another hole-nester with altricial young, the great tit (Hörak et al., 1997). Old individuals may be regarded as superior to young ones either due to increased breeding experiences (Cichoń, 2003) or because they have already proved their survival abilities (Mauck et al., 2004). Furthermore we have found that young that hatched in nests where originally laid eggs were more saturated had longer tarsi which is a trait positively associated with fitness in this species (Kruuk et al., 2001). As this finding was independent of total feeding frequency to young, it suggests that some unmeasured sort of parental or territory quality is associated with egg colour. However, as colour of cross-fostered eggs did not affect offspring performance, we can exclude the possibility that egg colour was associated with genetic quality or superior egg composition which would be more direct support for the SSEC hypothesis. Moreno et al. (2005) tested in a correlative but more detailed (especially with respect to immunocompetence) study for association between egg colour and female/offspring quality in the pied flycatcher. They found that females laying darker eggs had better both cellmediated and humoral immune response and concluded that their results fully support the signalling hypothesis for the blue eggs of the pied flycatcher (Moreno et al., 2005). However, their data seem to us to be more contradictory than authors suggested. Some correlations that they found were in fact in opposite direction than predicted under the SSEC hypothesis. Nestlings from eggs reflecting more toward the green part of the spectrum tended to be in poorer condition and had lower levels of immunoglobulins and females laying such eggs tended to be in poorer condition at laying (Moreno et al., 2005). The last result is also in discrepancy to that of ours. These contrasting results may be explained by different methods of assessment of female condition. Although in both studies female condition was similarly estimated as mass adjusted for tarsus length, Moreno et al. (2005) obtained two estimates of condition (on day 0 and day 11 of nestling period). They found negative association of condition on day 0 with saturation of blue-green part of egg spectra. However, they controlled in this test statistically for female condition on day 11. Consequently, their results may reflect more intra-individual changes in female condition rather than its actual condition. In a striking contrast to our results, Moreno et al. (2005) also found that younger females laid darker eggs. The assessment of female age was also somewhat different in the two studies. We used only individuals of exact known age (individuals ringed as nestlings) whereas Moreno et al. (2005) used for most females minimal estimate of their age. However, this difference do not seem to be so dramatic to cause such contradictory results. So this discordance remains unexplained for the present. We did not find support for the main prediction of SSEC hypothesis as no type of paternal care was related to colour of cross-fostered eggs. Neither was paternal care positively related to saturation of original eggs. The only significant relationship was in the unexpected direction as males defended more intensely nests in which original eggs were less saturated. Thus our experimental results are in contrast to previous correlative study in which males pied flycatchers fed with higher frequency young hatching from more saturated eggs (Moreno et al., 2004). There are several potential explanations for these contradictory results. First, the two studies were done on different species and it is possible that differential allocation of paternal effort evolved in one but not in the other species. However, we consider such explanation unlikely. Collared flycatcher is a sister species of the pied flycatcher from which it divided in relatively recent past and the reproductive isolation between the two species is still not completed (Sætre et al., 1997). Second, the discordance might be caused simply by sampling effect, either by type II error in our study or type I error in Moreno et al.’s (2004) study. However, as we have used

10

somewhat larger sample than Moreno et al. (2004), statistical power to detect effect of the magnitude detected by these authors (r = 0.38 and r = 0.41) was relatively high in our study (feeding frequency at age 6: power = 0.82 and 0.88 for the two effect sizes respectively; feeding frequency at age 12: power = 0.75 and 0.82 respectively). Type I error rate in the study of Moreno et al. (2004) has been controlled at the conventional level (α < 0.05). So also this explanation for the discordance results seems to be unlikely. Third, it may be possible that males increase paternal care only when superior egg colour is accompanied by superior female quality. In natural situation these traits may covary while our cross-fostering approach separated them. However, this would also mean that in such a scenario egg colour per se is not predictive of paternal effort. Finally, the description of egg colour differed between the two studies. Moreno et al. (2004) recorded egg reflectance in human-visible spectrum only, while we measured also egg reflectance in UV (301–400 nm). This difference in measured spectral range might potentially explain the different results. Recently, much evidence were found that UV vision plays important role in birds’ communication (Cuthill et al., 2000). This is also the case in the pied flycatcher as females in this species mate preferentially with males with high UV reflectance of plumage (Siitari et al., 2002) which is a trait correlated with male quality (Siitari & Huhta, 2002). Moreover, differences in UV reflectance of parental plumage (Limbourgh et al., 2004) and nestling skin (Jourdie et al., 2004) has been recently found to affect parental effort. The view that UV reflectance of eggs may have signalling function is further supported by our observation that high variability in this spectral range exists between females as well as between successive eggs in a clutch (see Figure 1a). Such signalling may be especially important for cavity nesters which face poorer light conditions at their nests. This idea is supported by the fact that eggs of cavity nesters reflect more in UV than those of open nesters (Soler et al., 2005). Moreover, UV reflectance of nestlings predicted parental provisioning in another cavity nester, the starling Sturnus vulgaris (Jourdie et al., 2004). Female ornamentation in species with conventional sex roles has been neglected both in studies of mate choice (Amundsen, 2000; Hill, 2002) and differential allocation (Sheldon, 2000). At present an interest in this topic is growing and some evidence have been already found to suggest that female ornaments are important for male mate choice (Hill, 2002; Griggio et al., 2005), paternal effort (Hill, 2002; Pilastro et al., 2003), and even sperm allocation (Pizzari, 2003). Moreno & Osorno (2003) suggested that colour of eggs might be viewed as external female ornaments, a special sort of extended phenotype (Dawkins 1982). Consequently entirely new area of research is opening at present as egg colour may signal different qualities than female plumage. Although many types of pigments appearing in plumage may have, similarly as the egg pigment biliverdin, antioxidant capacities (McGraw, 2005), they may be indicative of either health or pigment availability at the time of molting. However, molting is usually separated from breeding by considerable time periods. Consequently, at the time of breeding males can assess only genetic or persistent environmental quality from female plumage. In contrast to plumage ornaments, egg colour may be indicative also of female current physiological state, condition, and immunity. Given potentially large information content of egg colour and contrasting results of the few studies performed so far, testing for differential paternal effort in relation to egg colour provides fruitful area for further research. REFERENCES Amundsen T, 2000. Why are female birds ornamented? Trends Ecol Evol 15:149–155. Birkhead TR, 1978. Behavioural adaptations to high density nesting in the common guillemot Uria aalge. Anim Behav 26:321–331. Burley N, 1986. Sexual selection for aesthetic traits in a species with biparental care. Am Nat 127:415–445. Cichoń M, 2003. Does prior breeding experience improve reproductive success in collared flycatcher females? Oecologia 134:78–81. Cuthill IC, Partridge JC, Bennett ATD, Church SC, Hart NS, Hunt S, 2000. Ultraviolet vision in birds. Adv Stud Behav 29:159–214. Davies NB, Brooke ML, 1988. Cuckoos versus reed warblers: adaptations and counteradaptations. Anim Behav 36:262–284.

11

Dawkins R, 1982. The extended phenotype: the long reach of the gene. Oxford: Oxford University Press. Endler JA, 1990. On the measurement and classification of color in studies of animal color patterns. Biol J Linn Soc 41:315–352. Falchuk KH, Contin JM, Dziedzic TS, Feng ZL, French TC, Heffron GJ, Montorzi M, 2002. A role for biliverdin IX alpha in dorsal axis development of Xenopus laevis embryos. Proc Natl Acad Sci USA 99:251–256. Gosler AG, Higham JP, Reynolds SJ, 2005. Why are birds' eggs speckled? Ecol Lett 8:1105– 1113. Graham MH, 2003. Confronting multicollinearity in ecological multiple regression. Ecology 84:2809–2815. Griggio M, Valera F, Casas A, Pilastro A, 2005. Males prefer ornamented females: a field experiment of male choice in the rock sparrow. Anim Behav 69:1243–1250. Grim T, Honza M, 2001. Differences in behaviour of closely related thrushes (Turdus philomelos and T. merula) to experimental parasitism by the common cuckoo Cuculus canorus. Biologia 56:549–556. Grim T, 2005. Host recognition of brood parasites: Implications for methodology in studies of enemy recognition. Auk 122:530–543. Hill GE, 2002. A red bird in a brown bag. New York: Oxford University Press. Hörak P, Mand R, Ots I, 1997. Identifying targets of selection: A multivariate analysis of reproductive traits in the great tit. Oikos 78:592–600. Jourdie V, Moureau B, Bennett ATD, Heeb P, 2004. Ultraviolet reflectance by the skin of nestlings. Nature 431:262. Kaur H, Hughes MN, Green CJ, Naughton P, Foresti R, Motterlini R, 2003. Interaction of bilirubin and biliverdin with reactive nitrogen species. FEBS Lett 543:113–119. Krist M, 2004. Importance of competition for food and nest-sites in aggressive behaviour of Collared Flycatcher Ficedula albicollis. Bird Study 51:41–47. Krist M, Remeš V, 2004. Maternal effects and offspring performance: in search of the best method. Oikos 106:422–426. Krist M, Nádvorník P, Uvírová L, Bureš S, 2005. Paternity covaries with laying and hatching order in the collared flycatcher Ficedula albicollis. Behav Ecol Sociobiol 59:6–11. Kruuk LEB, Merilä J, Sheldon BC, 2001. Phenotypic selection on a heritable size trait revisited. Am Nat 158:557–571. Limbourg T, Mateman AC, Andersson S, Lessels CM, 2004. Female blue tits adjust parental effort to manipulated male UV attractiveness. Proc R Soc Lond B 271:1903–1908. Mauck RA, Huntington CE, Grubb TC, 2004. Age-specific reproductive success: evidence for the selection hypothesis. Evolution 58:880–885. McDonagh AF, 2001. Turning green to gold. Nat Struct Biol 8:198–200. McGraw KJ, 2005. The antioxidant function of many animal pigments: are there consistent health benefits of sexually selected colourants? Anim Behav 69:757–764. McNett GD, Marchetti K, 2005. Ultraviolet degradation in carotenoid patches: Live versus museum specimens of wood warblers (Parulidae). Auk 122:793–802. Merilä J, Sheldon BC, 2001. Avian quantitative genetics. In: Current Ornithology (Nolan V, ed). New York: Kluwer Academic/Plenum Publishers; 179–255. Mikšík I, Holáň V, Deyl Z, 1996. Avian eggshell pigments and their variability. Comp Biochem Physiol B 113:607–612. Moksnes A, Røskaft E, Tysse T, 1995. On the evolution of blue cuckoo eggs in Europe. J Avian Biol 26:13–19. Moreno J, Osorno JL, 2003. Avian egg colour and sexual selection: does eggshell pigmentation reflect female condition and genetic quality? Ecol Lett 6:803–806. Moreno J, Osorno JL, Morales J, Merino S, Tomás G, 2004. Egg colouration and male parental effort in the pied flycatcher Ficedula hypoleuca. J Avian Biol 35:300–304. Moreno J, Morales J, Lobato E, Merino S, Tomás G, Martínez-de la Puente J, 2005. Evidence for the signaling function of egg color in the pied flycatcher Ficedula hypoleuca. Behav Ecol 16:931–937. Nice MM, 1937. Studies in the life history of the song sparrow, volume I. A population study of the song sparrow. Trans Linn Soc New York 4:1–247.

12

Øien IJ, Moksnes A, Røskaft E, 1995. Evolution of variation in egg color and marking pattern in European passerines: adaptations in a coevolutionary arms race with the cuckoo, Cuculus canorus. Beh Ecol 6:166–174. Pilastro A, Griggio M, Matessi G, 2003. Male rock sparrows adjust their breeding strategy according to female ornamentation: parental or mating investment? Anim Behav 66:265– 271. Pizzari T, Cornwallis CK, Lovlie H, Jakobsson S, Birkhead TR, 2003. Sophisticated sperm allocation in male fowl. Nature 426:70–74. Quinn GP, Keough MJ, 2002. Experimantal design and data analysis for biologists. Cambridge: Cambridge University Press. Ruxton GD, Broom M, Colegrave N, 2001. Are unusually colored eggs a signal to potential conspecific brood parasites? Am Nat 157:451–458. Saetre GP, Moum T, Bureš S, Král M, Adamjan M, Moreno J, 1997. A sexually selected character displacement in flycatchers reinforces premating isolation. Nature 387:589– 592. Saks L, McGraw K, Hörak P, 2003. How feather colour reflects its carotenoid content. Funct Ecol 17 :555–561. SAS Institute Inc., 1995. JMP Statistics and Graphics guide. Cary: SAS Institute Inc. SAS Institute Inc., 1999. SAS Online Doc, version 8. Cary: SAS Institute Inc. Sheldon BC, 2000. Differential allocation: tests, mechanisms and implications. Trends Ecol Evol 15:397–402. Siitari H, Honkavaara J, Huhta E, Viitala J, 2002. Ultraviolet reflection and female mate choice in the pied flycatcher, Ficedula hypoleuca. Anim Behav 63:97–102. Siitari H, Huhta E, 2002. Individual color variation and male quality in pied flycatchers (Ficedula hypoleuca): a role of ultraviolet reflectance. Behav Ecol 13:737–741. Sokal RR, Rohlf, FJ, 1995. Biometry: the principles and practice of statistics in biological research. New York: W. H. Freeman and Company. Soler JJ, Møller AP, 1996. A comparative analysis of the evolution of variation in appearance of eggs of European passerines in relation to brood parasitism. Behav Ecol 7:89–94. Soler JJ, Moreno J, Aviles JM, Møller AP, 2005. Blue and green egg-color intensity is associated with parental effort and mating system in passerines: support for the sexual selection hypothesis. Evolution 59:636–644. Tinbergen N, Broekhuysen GJ, Feekes F, Houghton JCW, Kruuk H, Szulc E, 1962. Egg shell removal by the black-headed gull, Larus ridibundus L.: a behaviour component of camouflage. Behaviour 19:74–117. Underwood TJ, Sealy SG, 2002. Adaptive significance of egg coloration. In: Avian incubation: behaviour, environment, and evolution. (Deeming DC, ed). New York: Oxford University Press; 280–298. Weidinger K, 2001. Does egg colour affect predation rate on open passerine nests? Behav Ecol Sociobiol 49:456–464. Yom-Tov Y, 1980. Intraspecific nest parasitism in birds. Biol Rev 55:93–108.

13

FIGURE LEGEND Figure 1 Relationship between laying order and (a) shape of mean reflectance spectra; laying order is indicated by numbers alongside the reflectance curves and (b–d) individual spectral qualities (centered within clutches); lines of best fit are displayed. See text for statistical tests. Figure 2 Relationship between BG chroma of original (solid diamonds, solid lines) and cross-fostered (open circles, dashed lines) eggs and four measures of paternal effort. Regression lines are displayed. The displayed data and fitted regression lines are uncontrolled for effects of covariates. See text for statistical tests. Figure 3 Relationship between BG chroma of original (solid diamonds and lines) and crossfostered (open circles, dashed lines) eggs and three measures of nestling performance. The displayed data and fitted regression lines are uncontrolled for effects of covariates. See text for statistical tests. Figure 4 Relationship between BG chroma of original eggs and female age (a), condition (b), feeding frequency (c), and mean egg volume (d). The displayed data and fitted lines are uncontrolled for effects of covariates. The standard major axes from model II regressions (Sokal & Rohlf, 1995) are fitted instead of ordinary regression lines as there is no clear causality between the pairs of variables. See text for statistical tests.

14

Fig. 1

0.06

60 (a)

(b)

0.04 7 5 3 1

40 30

Relative BG chroma

Egg reflectance (%)

50

20

0.02 0.00 -0.02 -0.04

10

-0.06 300

400

500

600

700

1

2

Wavelength (nm) 4000

4

5

6

7

8

7

8

Laying order 0.12

(c)

(d)

0.08

2000 Relative hue

Relative brightness

3000

3

1000 0

-1000

0.04 0.00 -0.04

-2000 -0.08

-3000 -4000

-0.12 1

2

3

4

5

6

Laying order

7

8

1 Laying order

2

3

4

5

6

Laying order

15

Fig. 2

50

35 (a)

(b) Feeding frequency (age 12)

Feeding frequency (age 6)

30 25 20 15 10 5 0

40

30

20

10

0 4 3

3

(d)

2

PC1 attacks

Intensity of nest defence

(c)

2

1 0

-1

1

-2

0.56

0.58

0.60

BG chroma

16

0.62

0.64

-3 0.56

BG chroma

0.58

0.60 BG chroma

0.62

0.64

Fig. 3 17 (a)

Nestling weight (g)

16 15 14 13 12 11 21

Nestling tarsus length (mm)

(b)

20

19

0.8 (c) Nestling immunity (mm)

0.7 0.6 0.5 0.4 0.3 0.2 0.56

0.58

0.60

0.62

0.64

BG chroma

17

Fig. 4

7 (a)

(b)

1

5

Female condition

Female age (years)

6

2

4 3 2

0

-1

1 -2 2.0

3 (c)

1.9

(d)

3

Mean egg volume (cm )

PC1 feeding frequency

2 1 0 -1

1.8 1.7 1.6 1.5

-2

1.4

-3 0.56

1.3 0.56

0.58

0.60

0.62

0.64

BG chroma

18

0.58

0.60

0.62

0.64

26. Grim T. 1999: O holé kryse krtčí a jiné zvířeně. To by se překladateli stát nemělo. 26. Vesmír 78(8): 464–467. Grim T. 1999: O holé kryse krtčí a jiné zvířeně. To by se překladateli stát nemělo. Vesmír 78(8): 464–467.

TOMÁŠ GRIM

O holé kryse krtčí a jiné zvířeně To by se překladateli stát nemělo ROBERT FOLEY: Lidé před člověkem Nakladatelství Argo, Praha 1998, 256 stran, náklad 1500 výtisků, 298 Kč KONRÁD LORENZ: Základy etologie Academia, Praha 1993, 254 stran, náklad neuveden, 175 Kč DIERK FRANCK: Etologie Vydavatelství Karolinum, Praha 1996, 324 stran, náklad a cena neuvedeny LYALL WATSON: Temné síly přírody Nakladatelství Fontána, Olomouc 1996, 272 stran, náklad a cena neuvedeny JOHN BROCKMAN, KATINKA MATSONOVÁ: Jak se věci mají. Průvodce myšlenkami moderní vědy Nakladatelství Archa, Bratislava 1996, 246 stran, náklad a cena neuvedeny Veliký záhon pivoněk zářil v horkém slunci a docela vzadu ve větvích vysokých topolů krákaly vrány, rozplývá se ve svém romantickém popisu populární spisovatel James Herriot na 17. straně své knihy To by se zvěrolékaři stát nemělo (Svoboda, Praha 1991). Přesněji řečeno se rozplývá překladatelka, neboť v Herriotově originále krákají havrani, a to na jilmu.1) Herriotova knížka je beletrie, čistě oddechová věc, a tak je záměna uvedených tvorů v podstatě nedůležitá.2) Navíc Eva Marxová knížku přeložila výtečně.

Kompetentní (?) lektor Populárněvědecká literatura je ovšem něco docela jiného než beletrie. Dlouhá série „polistopadových“ překladů je hmatatelným důkazem toho, že si to z překladatelů v podstatě nikdo neuvědomuje. Podivnější je fakt, že to neberou v úvahu ani redaktoři. Největším překvapením však pro mě bylo zjištění, že jedna z nejnovějších přírodovědných knih Lidé před člověkem od Roberta Foleyho údajně prošla lektorským posouzením (Jan Vesták). Přesto pěkně ilustruje řadu omylů, které jsou mezi překladateli obecně oblíbené. Pokud je citován titul, který již byl poctěn českým překladem, je překladatel povinen uvést jeho český název (totéž se samozřejmě týká citátů). Možná není lehké zjistit název titulu vydaného v nějakém polozapomenutém nakladatelství. Stopařův průvodce po galaxii od Douglase Adamse je ale tak notoricky známá kniha, že nechápu, proč se z ní ve Foleym stalo Autostopem po galaxii (s. 33). O Darwinově Výrazu emocí u člověka a u zvířat (na s. 52 jako Vyjadřování emocí) nemluvě (totéž se týká názvů kapitol z Darwinova O vzniku druhů na s. 39 a 40). Na s. 170 je silně zavádějící překlep. Foley cituje Dawkinsův příklad jednostupňové selekce, ilustrovaný otázkou, s jakou pravděpodobností by opice náhodně bušící do psacího stroje napsala napoprvé frázi z Hamleta o délce 28 znaků (je to 1/27 28 – počítáno s abecedou o 27 znacích – což nedává šanci 1 ku 1000, jak je uvedeno ve Foleym, ale 1 ku číslu se 40 nulami). Vzhledem k dalšímu obsahu odstavce mohlo překladatelku trknout, že tady něco nesedí (pokud je to dílo šotkovo, tak se omlouvám). V nadpisu uvedená holá krysa krtčí je doslovný překlad anglického jména naked mole rat čili rypoš lysý. Kdyby šlo o nějakou obskuritu, tak budiž. Ale rypoš je zvíře tak slavné a opředené tolika (pravdivými) legendami, že ho žádný student biologie, který prošel základními kurzy etologie, evoluční biologie či behaviorální ekologie, prostě nemůže neznat. Nastávají dvě možnosti: buď byl lektor záškolákem, nebo nestudoval biologii. Uznávám, že setkání s rypošem lysým na katedře jaderné fyziky či teologie je poněkud méně pravděpodobné, ale to není omluva (stejně tak ani záškoláctví). O stránku dále se z rypoše stává bezsrstá krysa krtčí. Název živočicha ale neslouží k prezentaci překladatelo-

464

VESMÍR 78, srpen 1999 l http://www.cts.cuni.cz/vesmir

vy rozvité slovní zásoby – je prostě jednou dán a basta. Takže třeba bonobo je prostě bonobo a ne šimpanz trpasličí, pod guenonem se zase skrývá kočkodan. Zkomolení neunikly ani některé odborné termíny: např. z ekologie společenstev se stala ekologie komunit (s. 167), z apomorfií jsou apomorfy (s. 106), z uniformizmu zas uniformitářství (s. 67). Nevalná stylistická úroveň překladatelčina projevu spojená s nepochopením odborné stránky textu vede někdy k značně podivným formulacím, např. lidský mozek je tak mocný, že jsme se coby živočišný druh úplně vymkli ze záměru (s. 172). Na s. 59 dokonce šimpanz tluče kladivem.3) Ach jo… Antropologové říkají patrilokalita, ne patrilokálnost (s. 203) a magdalénien není magdalén (jak se uvádí ve Wilsonově Konsilienci, viz dále). (Nejen) na to se mohl překladatel podívat sám do naučného slovníku a nemusel s tím obtěžovat lektora, který mohl svůj čas věnovat jiné bohulibé činnosti. V knize je omylů podstatně více, ale za zmínku jistě stojí i pár ukázek z několika dalších překladů.

Colobiny, nebo kolobuské opice? S druhovými jmény bývají potíže. Spoléhat se na intuici je zrádné. Black-headed gull je racek chechtavý (Larus ridibundus) a ne racek černohlavý, jak by se mohlo zdát. Ten se místo toho řekne Mediterranean Gull (L. melanocephalus), zatímco doslovný překlad racka chechtavého (Laughing Gull) je zase racek atlantický (L. atricilla).4) Tak by se dalo pokračovat až do úplné ztráty způsobilosti (tento ohavný termín pro fitness čili reprodukční úspěšnost je kupodivu stále používán s neobyčejnou zavilostí). Kromě toho nemusí robin znamenat červenku, ale i drozda, oriole je zase žluva nebo vlhovec – záleží na tom, z které strany Atlantiku text pochází. Podívejme se třeba na taková Zvířátka v mé posteli od Jacquie Durrellové (první manželky G. Durrella). Hned na začátku první kapitoly se autorka ptá Proboha, co jsou to ary… Také já se musím ptát: Proboha, co to je pták se jménem Picathartes z rodiny vran? Kdyby paní překladatelka nahlédla do slovníku, zjistila by, že family je také taxonomická jednotka zvaná čeleď. Kromě toho Picathartes je česky vranule (hnízdí v jeskynních koloniích v západní Africe) a s vránou má společného pramálo. Tradičně nešťastný je osud afrických opic gueréz: ve Foleyho knize vystupují převlečeny za colobiny a v knížce Jacquie Durrellové se z nich staly kolobuské opice. To je poněkud zarážející, poněvadž překladatel by měl mít jisté povědomí o tom, co od uvedeného autora česky vyšlo, a také o kontextu, ze kterého je kniha … vytržena (v tomto případě). Durrellova kniha Catch me a colobus u nás totiž vyšla přímo pod doslovným překladem svého názvu (Chytněte mi guerézu) v nákladu sto tisíc výtisků a každý, kdo jen koutkem oka zavadil o řádky napsané Geraldem Durrellem, ji zná. 1) If Only They Could Talk, Pan Books, London & Sydney 1978, s. 21. 2) …provádět botanické a zoologické opravy v básnickém líčení přírody… je absurdní, píše Jiří Levý ve svém Umění překladu (s. 45, vydání z r. 1983). Ono ovšem záleží na tom, o jaké omyly jde – některé nesprávnosti básníkovy mohou vést k narušení toho, co je na překladu nejdůležitější – funkční ekvivalence. Na čtenáře znalého reálií pak takový překlad může v některých případech působit ne básnicky, ale komicky. 3) Anglické hammer znamená nejen kladivo, ale cokoliv, čím se dá tlouci – kyj, palici apod. 4) Navíc, jak trefně poznamenala známá ekoložka prof. Rychnovská, Luční kobylka není totéž co kobylka luční.

Mémy jako gény? Další stálicí jsou pářicí systémy (mating systems). Termín nemá co dělat s pářením. Důležité je, kolik má kdo partnerů, proto jde o partnerský nebo párovací systém. Pro zavedené české termíny není důvod vymýšlet nové ekvivalenty. Pěkně to ilustruje Wilsonova Konsilience: mem je zde uváděn jako mém. Překladatelky sice citují českou mutaci Sobeckého genu, kdyby do něj ale nahlédly, bylo by jim jasné, že mem je kulturní analogie genu a ne „génu“. Stejně tak je místo nadnormální podnět uvedeno mimořádný stimul (s. 250). Autorky překladu s jim neznámými termíny žonglují jak s horkými brambory: místo heritability čili dědivosti (tj. podílu geneticky podmíněné variability na celkové variabilitě) si pohrávají s dědičností či dědictvím (s. 175), partikulární dědičnost je podle nich částečná (s. 86)! Mluví dokonce o vývojové a genetické evoluci (s. 107; o termínech viz ještě dále). Kromě toho je Wilsonova kniha prošpikována řadou pravopisných archaizmů jako mythologický, theologie, atheisté, poesie apod. Tradiční školskou chybou jsou americké či britské miliardy a biliony. Přehlížení tohoto rozdílu vede k absurdním tvrzením – lidský mozek je protkán soustavou stovek biliónů nervových buněk (s. 110; celé lidské tělo tvoří řádově asi 10 bilionů buněk). Přírodní vědci (s. 142) zjistili, že život se vyvíjel několik biliónů let (s. 146). Takový nesmysl musí okamžitě zapnout poplašnou sirénu v hlavě každého člověka. Překladatelky si svého omylu mohly všimnout např. na s. 57, kde jsou charakteristiky elektronu uvedeny jak číselně, tak slovně. Dalším zlozvykem je nepřevádění cizích měr a jednotek na české: k čemu je mi sdělení, kolik čtverečních palců má mozková kůra (s. 119), nebo že Einsteinův mozek vážil 2,75 libry (s. 110)?

A co učebnice? Obrozenecká produkce novotvarů laickými překladateli je snad do jisté míry pochopitelná, ale je neomluvitelná, jestliže knihu přeložil odborník.5) Velmi ilustrativní je v tomto směru učebnice Etologie (Dierk Franck). Překladatel Leo Sigmund přehlíží zavedený a výstižný český termín pro insight learning (učení vhledem) a překládá ho jako chovat se rozumně.6) Rozumně se chová i ploštěnka, když je zalezlá v potoce pod kamenem. Je však schopna pochopit souvislosti a předvídat? V uvedené učebnici také rušivě působí nekonzistentní uvádění ekvivalentů, např. pro costs-benefits (tj. výdaje a zisky) jako výdaje, náklady, resp. prospěch, zisk a užitek na různých místech v knize. Ze správně přeloženého přeorientovaného chování (s. 45) se o stranu dále stává přeadresované chování. Fitness je překládána jako způsobilost (místo zdatnost), ve finále knihy dokonce metamorfovala na působnost! Takové zmatenosti ztěžují orientaci.7) Snahou překladatele by také měla být stručnost: myslím, že dědičně fixované je vrozené a učením získané je prostě naučené, výškovým liniím u nás říkáme vrstevnice (s. 249). Pachově infanticidní efekt u myší je efekt 5) To se týká i původní české literatury. Např. kin selection je příbuzenský výběr a ne výběr příbuzného (kterého?), jak se uvádí v Úvodu do etologie člověka (s. 178). 6) Totéž se týká home range (domovského okrsku), zde obytných prostorů! 7) To se netýká jen těchto termínů. V naší etologické literatuře najdeme pro fixed action pattern ekvivalenty jako dědičně fixované koordinační schéma (Franck), fixní motorický projev (Gaisler: Zoologie obratlovců), stereotypní pohybový vzorec nebo dědičně koordinovaný způsob pohybu (Lorenz). Milí studenti, upřímnou soustrast... Nebylo by jednodušší zůstat u konečného chování? (Tak uvedeno např. v knize Veselovského Chováme se jako zvířata?) 8) Někteří čeští etologové možná získali dojem, že si myslím, že Lorenz si ten mizerný překlad zasloužil, neboť vyšel před rokem 1990, a proto je to veteš. Není tomu tak. Proč by upozornění na absenci nového mělo znamenat zavrhování starého? 9) Zatím se zdá, že překladatelé se ze všech sil (úspěšně) snaží potvrdit Konvičkovu „paranoidní hypotézu“ o spiknutí anti-sociobiologické lobby (Vesmír 78, 405, 1999/7).

Bruceové (ne Bruceův, s. 158). Překlad constraints (omezení) jako hranic systému či systémových tlaků je nešťastný, stejně jako žebrání hnízdošů (snad jde o žadonění mláďat), žádostiví termiti (s. 235) či tvrzení, že intenzita chování varírovala (s. 93). Místy je text poněkud nejasný (Pro samičky je důležitější investovat do rozmnožovacích schopností než do sexuálních partnerů, s. 211), až zavádějící – pro rákosníky jsou prý nejvhodnější teritoria rákosiny s pozadím ze starých stromů – tam jsou ale nejvíce parazitováni a predováni. Také není pravda, že konflikt rodič-potomek se týká jen momentu odstavení mláďat (s. 228), týká se celého období rodičovské péče. Místo nepřátelské riziko (s. 279) by mělo být riziko predace. Predátor není totéž co nepřítel! Překladatel měl také věnovat pozornost seznamu literatury, který je doslovně opsán včetně německých (!) poznámek. Titel der Originalausgabe (s. 303, 312) měl být překladatelem uveden místo titulu německého vydání automaticky – vždy je myslím lepší studovat z primárních zdrojů než z překladů do třetího jazyka. Kvalita Etologie je poněkud nevyrovnaná. Na jedné straně učebnici uškodily uvedené nedostatky (včetně neobvykle vysokého počtu překlepů v „latinských“ názvech živočichů a anglických termínech). Na druhé straně je například vysvětlení proximátní a ultimátní roviny pohledu na chování (s. 147–149) perfektní (je dokonce lepší než v excelentní učebnici An Introduction to Behavioural Ecology od Krebse & Daviese – a to už je co říct!). Stejně tak je velkým kladem této publikace uvádění anglických termínů v závorkách. Navíc je učebnice tematicky vyrovnaná – autor (nezodpovědně) uvádí i zbrusu nové (tj. méně než čtvrtstoletí staré) poznatky, dosud neověřené desítkami let přepisování z jedné učebnice do druhé (stejně buřičsky se zachoval i Lorenz ve svých Základech etologie: téměř 50 citovaných literárních pramenů je ze sedmdesátých let, kdy kniha vyšla).

To si Lorenz nezasloužil8) Kapitola sama pro sebe jsou Lorenzovy nebohé Základy etologie, proslavené svým bídným překladem stejných „kvalit“ jako Wilsonova kniha O lidské přirozenosti.9) V mnohde neuvěřitelně kostrbatém a otrockém překladu se např. píše o zavedení představy o překřížení pudu a drezury překřížené (s. 17), o aktivním uchovávání prostorově orientovaného bytí (s. 147) či o silném krátkodobém a zadržovaně pomalu odeznívajícím zvyšování vzrušivosti (s. 129). Takový paskvil vážně ohrožuje čtenářovo duševní zdraví. Označení nadnormálního podnětu na s. 113 a 117 jako nadoptimálního podnětu je velmi populární protimluv. Prokousávat se textem, který má vyloženě německou syntax s řadou vedlejších vět vložených (jako je tato), je značně únavné. Nakonec se ještě čtenář doví, že Samička kanára http://www.cts.cuni.cz/vesmir l VESMÍR 78, srpen 1999

465

… se s ním pokoušela provádět pohyby stavby (s. 95) nebo že vzrušivost stíracího pohybu odpovídá trofickému stavu koordinujícího jej vřazeného neuronu (s. 92). Na s. 40 se píše o načrtávání tak zvaného průtokového diagramu […], který ve svých tak zvaných black boxies předvídá funkce… Zavedený ekvivalent je „blokový diagram“, proč nepřeložit i „černé skříňky“? (Na s. 205 se to kupodivu podařilo.) Nebo na s. 65: Jako stejně samozřejmé se někomu, kdo stojí mimo, zdá vedle toho, že všichni obratlovci mají hlavu a dvě oči, i to, že mají spodní čelist a pár předních a zadních končetin. Ať se na mě nikdo nezlobí, ale tato věta není česky. Nemohu se ubránit pocitu, že mluvím (a rozumím) jiným jazykem, než který používá překladatel. Na s. 58 se Lorenz (nebo překladatel?) snaží oddělit živočichy kloakovité (míní se tím chudáci ježury…) od vačnatých a placentálních. Marná snaha – vačnatci mají kloaku také. Této publikaci (a ani žádné jiné) neprospívá používání různých ekvivalentů pro tentýž termín. Popis infanticidy u hulmanů je korunován následujícím sdělením: Protože dětí zbavená samice otěhotní znovu dříve než ty, které vychovávají mláďata, má tak nový tyran lepší šanci uplatnit svůj genom. Tento zvláštní jev byl označen jako egoistický gen (s. 34). Panebože proč?!?

NADOPTIMÁLNÍ PROTIMLUV – KDYŽ MÉNĚ JE VÍCE Jako příklad fenoménu nadnormálního podnětu se v Základech etologie uvádí bezpočtukrát citovaný pokus s ústřičníky: při možnosti simultánního výběru mezi vlastními vejci a obrovskými atrapami vajec dávají ptáci přednost atrapám. Proč? Velké vejce je adaptivní – vylíhne se z něj velké mládě, které má větší šanci na přežití. Čím je ale vejce větší, tím víc začínají nad výhodami (zisky v podobě fitness) převažovat nevýhody (tj. výdaje). Ty musíme od zisků odečíst (viz obrázek). Na velkém vajíčku se musí dlouho sedět, a navíc značně fyziologicky zatěžuje samici – vyrobit vejce dá pořádnou fušku. Jediná optimální velikost vejce je proto ta, kterou pozorujeme v přírodě – je kompromisem mezi ziskem z velkého vejce a výdaji, které jsou s ním spojeny. (Totéž se týká velikosti snůšky – když do hnízda přidáme vejce, pěstouni odchovají méně mláďat než z původního menšího počtu vajec.) Atrapa je sice větší než normální přírodní vejce (je tedy nadnormální), není však nadoptimální. Vzhledem ke zvýšeným výdajům a nezvyšujícím se ziskům by takové vejce (pokud by ho samička vůbec snesla) bylo naopak suboptimální! Nic nemůže být optimálnější než to, co je optimální.

Perlička na dně A ještě k odborné literatuře: Na terminologické revizi učebnice Ekologie (Begon, Harper & Townsend) se podílely více než tři desítky odborníků, přesto se hned v Úvodu (s. xxiii) setkáváme s nepochopením jednoho ze základních metodologických pojmů – proximátního vysvětlení (viz rámeček). Také ekvivalent pro hojně používaný termín tradeoff – směna (s. 481) se mi nezdá vhodný, podstatu problému10) lépe vystihuje výraz kompromis, který ostatně někdy používá třeba Dawkins (např. The Blind Watchmaker, Řeka z ráje). Tyto detaily nic nemění na jinak vysoké kvalitě překladu.

Temné síly překladatelské „Elegantním“ řešením překladatelského oříšku je ho jednoduše nepřeložit. V knize Temné síly přírody (Lyall Watson) se tak setkáme s lemmingy a bowerbirdy. Přitom jak lumíky, tak lemčíky najdeme ve standardním čtyřdílném anglicko-českém slovníku (bohužel v něm také najdeme pod heslem bee-eater včelojeda; to doslovně sedí, naneštěstí je bee-eater označení pro vlhu). V této jinak velmi zajímavé knize také autor, ač povoláním mořský biolog, vychází do pole (alespoň u nás na východě „deme do fíldu“ čili do terénu). Docela mile působí i Jungovy archeotypy, evoluční předadaptace nebo replikátory, které na sebe vzaly podobu opakovatelů. 10) Jde o to, že každý organizmus má omezené množství času a energie, které může investovat do různých aktivit. Každé rozhodnutí investovat je tedy kompromisem: buď se půjdu najíst, nebo se půjdu množit, nebo uteču predátorovi, ale v plném sprintu před gepardem, který je mi v patách, povečeřím jen stěží.

PROXIMÁTNÍ A ULTIMÁTNÍ Každé chování lze vysvětlovat ze dvou základních pohledů, v anglické literatuře označovaných jako proximate a ultimate. Nerozlišování těchto pohledů (a následný zmatek v odborných diskusích) je kupodivu docela běžné. Výraz proximate je v Ekologii přeložen jako proximativní s poznámkou, že jde o přibližné vysvětlení. Proximátní vysvětlení jevu se však týká jeho mechanizmu a není přibližné, ale bezprostřední – vztahuje se k okamžité, empiricky a experimentálně ověřitelné příčině daného jevu (např. „Špaček Tonda je agresivní, protože má vysokou hladinu testosteronu“). Naopak ultimátní (konečné nebo evoluční) vysvětlení se týká funkční příčinnosti – vlivu na reprodukční úspěšnost, tj. „Tonda zahnal souseda Pepíka, pořádně se nacpal larev tiplic a má dost energie jít na zálety, tj. přilepšit si mimopárovou kopulací“. V jistém smyslu je právě toto druhé vysvětlení přibližné – svými evolučními bajkami o vzniku znaku a jeho příspěvku k fitness jedince si můžeme být jisti jen přibližně.

466

VESMÍR 78, srpen 1999 l http://www.cts.cuni.cz/vesmir

Autor vysvětluje formy mimikrů – mimikry je ale nesklonné neutrum. Na s. 144 se dočteme, že jeden z vězňů… má zvláštní shluk činností v oblasti thalamu. Od takových výtvorů odchází čtenář v nezáviděníhodném duševním rozložení. Temné síly omyly přímo perlí. Překladatel vládne kouzelnou hůlkou, a tak dokáže i věci dosud vědě neznámé, třeba změnit pacienta pomocí lobotomie ve zvládnutelnou vegetaci! A Jiří Levý se spolu se svým Uměním překladu obrací v hrobě… Mnohdy by překladateli pomohl sedlácký rozum (s. 20), když na něj zaútočí netopýři upíří (s. 77). Tamtéž se píše, že samice hyen jsou velké, kompetentní a agresivní. Otázkou zůstává, nakolik byl kompetentní překladatel.

Průvodce bezmyšlenkovitostí moderního překladatele Překládání předpokládá znalost nejen výchozího, ale i cílového jazyka. To zní jako samozřejmost, ale mnohdy se opravdu nelze ubránit dojmu, že překladatel neumí česky. V tomto směru je skutečným monumentem překladatelské neschopnosti kniha Jak se věcí mají (autoři Brockman & Matsonová). Není to totiž ani tak Průvodce myšlenkami moderní vědy (jak hrdě hlásá název) jako spíš průvodce labyrintem překladatelských omylů, neznalostí a prosté lidské hlouposti. Docela mě fascinuje zjištění, že někdo je schopen překládat text, kterému absolutně nerozumí. „Překladatelka“ Eva Lacinová píše o hemoglobinu v červených buňkách (s. 61), určení genů, odpovědných za vzorkování raných embryí (s. 63), elektrónech a pozitrónech (s. 196) a domnívá se, že elementární nářadí je možno pozorovat i u jiných primátů (s. 20). Teď už také konečně víme, že Známí protoktisti zahrnují améby-měňavky, eugleny, ciliaty, diatomy, červené mořské řasy a všechny další řasy slizovky a řasohouby (s. 65) a že organizmus ohrožuje, když se zmnoží otravný plyn (s. 66) nebo mikrobi (archebakterie podobné termoplasmě) (s. 70). Lidé trpící depresí prý mají porušený noradrenalin (s. 167). Co si myslet

Umění, nebo karikatura překladu? To bylo bohužel jen pár ukázek z mnoha potenciálně citovatelných. Jaký z toho plyne závěr? Mrzká úroveň mnoha překladů je odrazem zarážející neschopnosti překladatelů, odpovědných redaktorů a někdy i odborného lektora. Produkce titulů podobných kvalit je ostudou všech zúčastněných. Na práci překladatele jsou kladeny tři základní požadavky: pochopení předlohy, interpretace předlohy a přestylizování předlohy. Výše uvedené překlady v různé míře neuspěly ani v jedné z těchto podmínek pro vznik dobrého překladu. Mnozí překladatelé se nezatěžují čtením, uvažováním, natož pak lektorováním, nahlížení do slovníků je zdržuje, a hlavně si bezmezně věří; svým diletantstvím tak špatný překladatel neškodí jen čtenáři, ale škodí i autorovi, jak říká nederlandistka Olga Krijtová. Za velice podstatný pokládám také následující problém: jako student biologie se mnohdy dovtípím o co jde, když se setkám s rackem sleďovým (v Konsilienci) či škorpióní muchou (v Temných silách), ale jak poznám, kde končí pravda a kde začíná překladatelovo blouznění v textu o fyzice či chemii? Kvalita překladu není nepodstatná ani v dalším ohledu. Mnozí daňoví poplatníci by jistě rádi věděli, kde končí jejich peníze. Jednou z mála možností jak se to dovědět je populárněvědecká literatura. K jakému poznání je asi dovede Průvodce myšlenkami moderní vědy? U žádné z uvedených knih neznám originál. A tak se jen můžu dohadovat, kolik dalších nesmyslů by se dalo odhalit, kdyby ho měl člověk po ruce. Za pozoruhodné také pokládám zjištění, že mezi nakladatelství produkující nekvalitní překlady se zařadila i Academia (Lorenz). A to je opravdu smutné. Občas se ale naštěstí objeví nějaká ta perlička na dně. O tom, že skutečně kvalitní překlad lze vyrobit i po revo11) Z produkce nakl. Archa znám už jen Dawkinsovu Řeku z ráje, která je až na pár ovocných netopýrů, tenkozubců opačných a racků černohlavých s kamerovýma očima přeložena docela slušně. Další ediční plán Archy ve mně ale vzbuzuje obavy – jestli mají skončit vynikající knížky J. Diamonda, G. C. Williamse nebo S. Jonese jako Průvodce myšlenkami, to už je lepší je nepřekládat vůbec.

Kresba © Vladimír Renčín

o duševních pochodech člověka, který napíše: Replikace tvorů (obzvláště tvorů, jejichž vlastní vnitřní organizace vyžaduje stálé vzorky chování a struktury) může vzniknout pouze v masově vytvořených vesmírech (s. 202)? Překladatelka zaměňuje, co se jen zaměnit dá: chlor za chlorid, vajíčko za vejce, páry bází za základní páry, lidoopy za opice, teorii mitochondriální Evy za teorii „Eve“, křížové bratrance a sestřenice za křížící se bratrance a sestřenice (vskutku pikantní), obrázek za obrazec (!), konzervativní za konzervované sekvence DNA, dočteme se i o dvojité závitnici s trojčaty DNA! Malthus napsal Pojednání o růstu obyvatelstva (s. 211) a Genese obsahuje mnoho příběhů a také dvaapůl verze o tom, jak vznikl vesmír (s. 225)! Lidé mají nezřízenou touhu umístit se na špičce přírodní hromady (s. 76)… Na frak dostali i autoři jednotlivých esejů: S. J. Gould je genetik v oboru hlemýžďů a S. Jones se při své venkovní práci věnuje genetickému dosahu lidských fosilií na biologickou povahu lidské rasy (s. 104). Chudák P. D. Ward zase předsedá mezinárodnímu grémiu o vymření za období křídy ve třetihorách – zřejmě předběhl svou dobu (s. 109) a A. Faustová-Sterlingová bádá v oboru Drosophila (s. 114). Na s. 69 zjevená Kingdom Monera (bakteria) a Kingdom Fungi mohou směle konkurovat Lordu Ježíši Kristovi, kterého už jakýsi překladatel stvořil také (uvádí O. Krijtová v Pozvání k překladatelské praxi, s. 68). K překladu byl zřejmě použit počítač automaticky zaměňující slovo za slovo. Obzvláště ironicky pak působí tvrzení v úvodu, podle kterého jsou tyto eseje jedinečným a bohatým zdrojem poznání. Za takové zprznění vynikajících myšlenek předních světových myslitelů by si překladatelka zasloužila být penalizována imperativem smrti (s. 72).11)

luci, svědčí nejen zmíněná Ekologie, ale z populárněvědeckých knížek třeba Jazyk genů (S. Jones), O zániku druhů (D. M. Raup), Triumf embrya (L. Wolpert) nebo Wrightovo Morální zvíře (i když také zde na čtenáře vybafne něco tak obludného jako klunatka kopinatá, s. 53). Většina překladů ale takové štěstí bohužel neměla. Jak řešit tuto politováníhodnou situaci? Lék je jediný: překladatel musí perfektně ovládat výchozí jazyk i češtinu, musí znát kontext, ze kterého autor a jeho dílo pocházejí, potřebuje se průběžně seznamovat s ostatními autorovými díly a nemůže spoléhat na svou intuici při setkání s neznámým termínem – překlad musí projít odborným recenzním řízením (J. Levý 1983: Umění překladu, Z. Kufnerová ad. 1994: Překládání a čeština, D. Knittlová 1995: Teorie překladu).

A co lidé před člověkem? Vraťme se ke knize R. Foleyho. Čím je zajímavá? Především je autorův přístup konsilientní: dobře ví, že na otázku, kde jsme se tu vzali a proč jsme takoví, jací jsme, lze odpovědět pouze při doplnění tradičního antropologického pohledu pohledem evolučním. Ukazuje, že v evoluci hominidů šlo o celou sérii adaptivních radiací, o působení obecných evolučních mechanizmů v unikátním kontextu. (Čili se dělo to, co se dělo vždycky předtím a co se dále děje všude kolem nás.) V našem konkrétním případě se prostě ukázalo, že je zrovna lepší mít místo velkých zubů velký mozek. R. Foley dále poukazuje na fakt, že dřívější systematické oddělení člověka od lidoopů byla záležitost ideologického zbožného přání a ne vědeckého přístupu. Zajímavá, a především adekvátní je analýza evoluce znaků (evolučních novinek) na základě výdajů a zisků s nimi spojených. Velice příjemný a přínosný je autorův filozofický nadhled a občas i kapka dobře mířeného humoru. Čili obsah kvalitní, forma českého překladu už méně. Výše uvedené příklady některých nedostatků této knihy nejsou až tak velké hrůzy – ale to pouze díky tomu, že jiná překladatelská veledíla podsadila laťku proklatě (proklatě!) nízko. Nakonec z toho cambridžský evoluční antropolog ještě vyvázl docela slušně. Avšak ceně knihy (baťovských 298 Kč) by měla odpovídat kvalita zakoupeného zboží, tj. především by překladatel měl mít alespoň elementární znalosti českého jazyka a překlad by si měl také přečíst člověk, který se v daném oboru vyzná. Jinak se uvedení odborného recenzenta v tiráži stává něčím podobným jako inflace nálepek „eko“ a „bio“ – v obou případech máme pocit, že si kupujeme něco kva¨ litního. http://www.cts.cuni.cz/vesmir l VESMÍR 78, srpen 1999

467

27. Grim T. 2000: Paralelní vysvětlení. Proč a jak se ptát „proč“ a „jak“. 27. Vesmír 79(2): 92–93. Grim T. 2000: Paralelní vysvětlení. Proč a jak se ptát „proč“ a „jak“. Vesmír 79(2): 92–93.

Paralelní vysvětlení Proč a jak se ptát „proč“ a „jak“?

náležitě hrdí na své obory a vědí, že ti z ostatních oborů je stejně jen obírají o grantové peníze, debata probíhá dál. Hladina adrenalinu kamarádům (teď už bývalým) stoupá, zasahuje policie, kos přestává zpívat. Richard, Pepa, Milan a Vašek se – izolováni ve svých samotkách – už asi nedomluví, ale my bychom se mohli zamyslet, proč se vlastně nedomluvili. Poněkud ironickým aspektem zkarikované debaty je, že pravdu měli samozřejmě všichni. Proč nám ten kos tak krásně zpíval?

TOMÁŠ GRIM

“Myslíte, že tohle chování je adaptivní? Já nevím. Co když se to těm zvířátkům jenom tak líbí?“ Asi nebudu mířit úplně mimo, když budu hádat, že většina z nás (ať už jsme nebo nejsme biologové) se s takovou otázkou setkala. Pokud ji vysloví nebiolog, není na tom nic divného. Pokud ji uslyšíme z úst biologových, je na tom něco smutného. Ta otázka je totiž špatně položena a je odrazem jednoho docela závažného a rozšířeného nepochopení. Proč nám ten kos tak krásně zpívá? Na otázku, proč živý tvor dělá to či ono, se dá odpovědět řadou způsobů. Představme si debatu o ptačím zpěvu. Scéna: letní zahrádka, piva jsou čerstvě načepována, sluníčko svítí. Obsazení: Richard (fyziolog), Pepa (vývojový biolog), Milan (behaviorální ekolog), Vašek (evoluční biolog). Drama začíná. V poklidné atmosféře letního odpoledne prosyceného vůní kvetoucí lípy jeden z kamarádů vznese onu osudovou otázku: „Proč nám ten kos tak krásně zpívá?“ Richard se chopí slova jako první: „No to je přece jasné – na jaře se prodlužuje délka světelného dne, hladina testosteronu (nejen u kosů) stoupá a syrinx začíná pracovat na plné obrátky.“ Pepa na to: „Ale ne – kos zpívá jen proto, že se to naučil od svého otce nebo ostatních samečků v okolí, když byl ještě mladý.“ Milan to tak nenechá: „To nemyslíš vážně! Ten frajírek v černém fraku si přece zpěvem ohraničuje teritorium, láká samičky atd.“ Vašek nesouhlasí: „Hoši, tak na to zapomeňte. Předci dnešních kosů zpívali samozřejmě taky, tenhle kos prostě jen zdědil zpěv, který vznikl v hlubinách evoluční minulosti z nějakého ancestrálního skřehotání.“ Protože jsou naši hrdinové

Každý vědec hýčká své hypotézy a mnohdy je žárlivě střeží před vyvrácením tak dlouho, dokud nejsou úplně zkostnatělé. Řada autorů považuje jiná vysvětlení za konkurenční alternativy1) svých teorií, které pak se zaslepenou zavilostí chrání v debatách, většinou zcela sterilních. Dnes už klasickým příkladem je diskuse z počátku padesátých let, v níž se na kolbišti pod vlajkou „nature-nurture“ utkali K. Lorenz a D. Lehrman. Spor byl o to, zda je určité chování vrozené, nebo získané v průběhu ontogeneze. Po dvou desetiletích bezúspěšných debat se nakonec ukázalo, že vlastně o žádný spor nejde. Roku 1970 Lehrman napsal: „Ten nejjasnější možný důkaz, že daný znak živočicha je geneticky podmíněn, ani v nejmenším neřeší otázku, které vývojové procesy zprostředkovaly vznik fenotypového znaku během ontogeneze.“ Neshoda nebyla způsobena nelegitimností navrhovaných vysvětlení, ale konceptuálním problémem. Kamenem úrazu byla paralelní vysvětlení (multiple explanations). Určité chování můžeme zcela pochopit pouze a jen tehdy, když se na něj podíváme ze čtyř různých rovin pohledu – fylogenetické, ontogenetické, příčinné a funkční. Proximátní roviny vysvětlení popisují blíže jev samotný, jak chování vzniká během fylogeneze a ontogeneze a jaký je jeho konkrétní neurofyziologický mechanizmus, tj. jeho bezprostřední příčina.2) Tato proximátní vysvětlení ale nedokážou objasnit, proč se od sebe různé jevy liší, proč třeba příbuzné druhy žijí různými způsoby. Smysluplnou odpověď nám dá jen pohled z roviny ultimátní, tj. proč dané chování v konkrétním ekologickém kontextu přispívá k fitness (funkční rovina). Alternativní, nebo komplementární? Na proximátní rovině se ptáme jak: jak dané chování probíhá na neuronální, hormonální a biochemické úrovni, jak se formuje během ontogeneze a jaká je jeho evoluční historie. Na druhé straně tu máme rovinu ultimátní a ptáme se proč: tady mluvíme o adaptivním významu daného chování3) (či jiného morfologického nebo fyziologického znaku). Jednotlivá vysvětlení pocházejí z různých oborů a vzájemně se doplňují. Všechna vysvětlení jsou samozřejmě stejně důležitá a je třeba je samostatně testovat. Potvrzení či zamítnutí hypotézy na jedné 1) Alternativa je zde chápána jako výběr mezi více navzájem se vylučujícími možnostmi (viz Akademický slovník cizích slov). 2) Terminologie je zde poněkud zavádějící: proximátní i ultimátní procesy jsou příčinné, ale v jiných časových měřítkách. Na tom není nic divného: přírodní výběr je zpětnovazebný proces, ve kterém se následek stává příčinou (viz obrázek). 3) Tj. proč znak přispívá, či naopak nepřispívá k reprodukci. Rozdělení na „how“ a „why“ questions, zavedené v anglické literatuře, je trochu nešikovné (můžu se také zeptat „jak znak zvyšuje fitness“).

Mgr. Tomáš Grim (*1973) vystudoval systematickou zoologii a ekologii na Přírodovědecké fakultě MU v Brně. Na katedře zoologie Univerzity Palackého se zabývá etologií hnízdního parazitizmu.

92

VESMÍR 79, únor 2000 l http://www.cts.cuni.cz/vesmir

rovině neříká nic o hypotézách na ostatních třech rovinách. Zároveň má ale komunikace mezi proximátním a ultimátním pohledem velký význam. Analýza adaptivního problému, kterému organizmus čelí, může vést k hledání dosud neznámého mechanizmu jeho řešení. Naopak detailnímu poznání určitého mechanizmu může pomoci, když se budeme ptát, proč je utvářen zrovna takto. Paralelní vysvětlení se tedy nevylučují, ale doplňují. Málokdo pracuje ve více oborech zároveň, a tak je do jisté míry pochopitelné, že vysvětlení jiného oboru považuje za alternativu, která může jeho vysvětlení vyřadit ze hry. To je omyl. Dvě vysvětlení si mohou konkurovat, pouze pokud se nacházejí na stejné explanační rovině.4) Slovo je dobrý sluha, ale špatný pán Aby slovo bylo dobrým sluhou, je třeba si ho definovat – samozřejmě podle potřeby daného oboru. Je třeba nepřehlédnout, že otázky proč a otázky jak mohou mít jiné odpovědi v evolučním kontextu než v běžné konverzaci. Pokud bych na otázku „proč kos zpívá?“ odpověděl „protože se to naučil“, mohl by někdo (oprávněně) namítnout, že se mýlím. Na ultimátní otázku jsem totiž dal proximátní odpověď. Správná odpověď by byla Milanova (viz výše). Normální občan (tj. ne evoluční biolog) pochopitelně uvažuje na rovině proximátní (viz úmysly, motivy, apod.). Průšvih je v tom, že běžně užívaná slova (podvádění, rozhodování, sobecký, altruistický) jsou v evoluční biologii používána na ultimátní rovině – nepopisují příčinné mechanizmy chování, ale jeho funkční následky. Používání pseudo-antropomorfického „těsnopisu“ je na ultimátní rovině zcela legitimní. Teleologickou formulaci „vlaštovky táhnou na jih proto, aby je nezastihla zima“ lze vždy převést na bezproblémové tvrzení „vlaštovky táhnou na jih proto, že (jejich předky, kteří migrovali) nezastihla zima“. Nerozlišování dvou rovin pohledu vedlo v etologii např. k nemístnému používání termínu „motivace“. Šimpanz spatří predátora, začne mávat větví, a tím upozorní členy tlupy, že se něco děje. Lidé tomu říkají varování. Stačí drobný mentální zkrat, abychom tvrdili, že motivací jeho chování bylo varovat členy tlupy. V důsledku teleologické záměny příčiny a následku jsme sklouzli z roviny ultimátní (varování jako efekt) na rovinu proximátní (varování jako motivace). Falešné explanační dichotomie Tendence postavit proti sobě kompatibilní vysvětlení v domnění, že jde o alternativy, vedla k řadě explanačních dichotomií.5) Řada autorů se např. ptala, zda je dané chování dáno geny, nebo prostředím. Každý gen je ale v nějakém prostředí a v prostředí bez genů zase není život. Geny a prostředí od sebe nelze oddělit. Položit si otázku geny versus prostředí je stejně „smysluplné“ jako ptát se, jestli je pro přežití jedince důležitější kyslík, nebo červené krvinky. Stejně scestné je domnívat se, že něco je dáno spíš geny a něco spíš prostředím (analogicky: je chleba determinován spíš receptem, nebo spíš použitými surovinami?). Jediné, co můžeme připsat genům nebo prostředí, je variabilita mezi jedinci. Variabilita samozřejmě není vlastnost jedince, a tak nemá smysl tvrdit, že jeden je takový díky prostředí a druhý zase díky genům. Je snad také jasné, že každé chování je podmíněno a umožněno geny. Nelze jinak. Živočich se nějak chová jen kvůli tomu, že si ho geny postavily. Nikdo si nemyslí, že existuje gen pro stoj na hlavě, ale to neznamená, že tato schopnost není umožněna geneticky – mnozí z nás stojí na hlavě docela dobře, ropucha již hůře.

Populárním problémem byla také agresivita. Podle Lorenze vrozená a nemodifikovatelná zkušeností, podle jeho oponentů naučená, a tudíž nevysvětlitelná evolučně. Obě tvrzení vyplývají spíše z rozšířené snahy zastávat extrémní názory a ukázat tak jejich odlišnost od dosavadní doktríny (efekt kyvadla). Lorenz i jeho oponenti v tomto sporu ztroskotali proto, že si neuvědomili možnost, ba nutnost vysvětlení z více úhlů. Proč se mýlili? Selekce pracuje s fenotypem bez ohledu na to, jaká je jeho ontogeneze (viz Lehrman výše). Proximátní mechanizmy chování byly designovány selekcí tak, „aby“ se optimalizovala jejich funkce, tj. maximalizace genového příspěvku do dalších generací. Pokud agresivita přispívá k fitness, je úplně jedno, zda je vrozená či naučená. Je zásadně mylné se domnívat, že naučené chování není produktem evoluce (vzpomeňte si třeba na ptačí zpěv). Ontogenetické procesy jsou také adaptace. Stejně nesmyslná je představa, že každý produkt evoluce je „v genech“ a není ovlivněn prostředím během ontogeneze. Další zajímavou diskusí byla debata mezi S. J. Gouldem a J. Alcockem o samičích klitorisech. Podle Goulda je tento orgán vedlejším produktem ontogenetických procesů, které jsou pozitivně selektovány, protože vedou ke vzniku samčích penisů, a tak je zbytečné hledat nějaký adaptivní význam klitorisu. Na to reagoval Alcock argumentem, že klitoris sice nějak vznikl, ale dnes by snad mohl samici umožňovat, aby prostřednictvím orgazmu kontrolovala paternitu (a zdá se, že tomu tak opravdu je). Podle něj by tedy klitoris mohl být adaptace. Rozpor byl opět zdánlivý. Pokud znak vzniká jako vedlejší produkt jiných procesů, neznamená to, že se nemůže stát předmětem selekce. To je ovšem třeba testovat. Adaptace je znak, který relativně zvyšuje fitness jedince vůči jedinci, který ho nemá – zda znak je, či není adaptací, lze tedy zjistit manipulativním experimentem. Provádět v tomto případě pokusy na zvířatech by bylo nehumánní. Výzkumník posedlý tímto orgánem však naštěstí může směle vyrazit do afro-asijských oblastí, kde je dodnes dobrým zvykem klitoridektomie (ženská obřízka). Pro jistotu ještě zdůrazňuji, že zdaleka ne všechny znaky jsou adaptace. Adaptivní je pouze ten znak, který byl selektován proti alternativní formě téhož znaku a více přispíval k reprodukčnímu úspěchu. To, že jsou kosti bílé, lze zdůvodnit proximátně (obsahem minerálů), ale položit si můžeme i otázku o funkčním významu jejich barvy. Kosti ovšem nebyly selektovány kvůli své barvě (aby třeba umrlcova vybělená kostra na poušti signalizovala dalším jedincům „Tudy ne, přátelé“) – své podpůrné a další funkce by kosti plnily, i kdyby byly třeba fialové nebo růžově tečkované. Barva kostí sama o sobě není adaptací. Je vedlejším produktem selekce chemického složení kostí (které naopak adaptivní je). Jaké z toho plyne poučení? Na jednu otázku se dá odpovědět mnoha způsoby (to všichni víme). Správných otázek ale může být více (a to zjevně všichni nevíme). Která odpověď je pokládána za nejzajímavější a nejuspokojivější, je pak otázkou odborného zaměření a vkusu. Přeji vše nejlepší do příštích zajímavých diskusí! 4) Mohou, ale nemusí. Funkční vysvětlení pozitivní selekce ptačího ocasu může být jak samičí výběr (intersexuální selekce), tak predace (alespoň na počátku se s delším ocasem před dravcem uprchne líp). 5) Explanace je vysvětlení, dichotomie je razantní (ř. tomé – řez) třídění do dvou skupin. Pozn. red.: K dalšímu uvažování nad tématem viz článek Ivana M. Havla: Molekuly na smutek, Vesmír 78, 612, 1999/11

http://www.cts.cuni.cz/vesmir l VESMÍR 79, únor 2000

93

28. Grim T. 2000: Poučení z krizového vývoje společenských věd. 28. Vesmír 79(9): 524–527. Grim T. 2000: Poučení z krizového vývoje společenských věd. Vesmír 79(9): 524–527.

NAD KNIHOU

Poučení z krizového vývoje společenských věd J. BARKOW, L. COSMIDESOVÁ, J. TOOBY: The Adapted Mind: Evolutionary Psychology and the Generation of Culture

TOMÁŠ GRIM

na, ale minulé chyby jsou od toho, abychom se z nich poučili.

Oxford University Press, New York 1992, 666 stran, cena 22,95 libry

Sociální vědy versus přírodní vědy

Proč se zabývat knihou bezmála deset let starou? Adapted Mind sice není úplná novinka, ale vzhledem k obecnosti povědomí o evolučním přístupu ke zkoumání člověka u nás je dosud více než aktuální. Navíc se věnuje důležitému tématu komunikace mezi sociálními a přírodními vědami. Nejdříve stručné shrnutí. The Adapted Mind je bez nadsázky povinné čtení pro každého, kdo se zajímá o člověka a o smysluplné zkoumání toho, „proč jsme takoví, jací jsme“. Kniha je souborem samostatných prací různých autorů, z nichž většina je nadnormálně dobře napsána. Při čtení kapitol sepsaných L. Cosmidesovou a J. Toobym, M. Profetovou nebo D. Symonsem se lze jen stěží ubránit dojmu, že tito psychologové mají v evolučním myšlení více jasno než řada biologů. Nemálo tvrzení je samozřejmě diskutabilních a některá východiska se časem jistě ukáží jako mylná, ostatně jako ve kterékoliv publikaci podobně širokého záběru. Klíčovým momentem celé knihy je kapitola Psychologické základy kultury, která je kritikou tradičních přístupů ke zkoumání věcí lidských. Více než stostránková kapitola je psána velice hustě, ale i tak bych se pokusil vybrat některé základní myšlenky a doplnit je postřehy z dalších publikací a několika vlastními názory. Kritika Standardního společenskovědního modelu (viz dále) může našemu čtenáři připadat nemístně tvrdá (což je ovšem způsobeno abnormálností našeho vědeckého ovzduší, kde je kritika mnohdy pokládána za osobní urážku). Je třeba zdůraznit, že kritika napsaná Cosmidesovou a Toobym je konstruktivní kritika a ne snaha diskvalifikovat někoho, z něhož jsme si uměle a hloupě udělali protivníka. Ano, narážím na sociálněvědní rádoby kritiky evolučních názorů na člověka ve stylu „(socio)biologie nemá sociologii co říct“. Netvrdím, že všechny kritiky sociobiologie spadají do dvou (bohužel enormně rozbujelých) kategorií „blábolení hysterického moralisty“ a „božský Stephen Jay Gould* si ale myslí tohle“ (viz třeba J. Skupnik, Cargo, s. 67, 1999/1 a s. 232, 1999/3, 4). Existují samozřejmě i kvalitní a konstruktivní výtky vůči sociobiologii, třeba druhá kapitola v Adapted Mind od D. Symonse nebo některé články Davida S. Wilsona, které reflektují názory uváděné v Adapted Mind. Nakonec je třeba dodat, že řada myšlenek v této knize kritizovaných již byla v některých proudech sociálních věd opuště-

Leda Cosmidesová a John Tooby se pokusili o analýzu metodologického rámce tradičních sociálních věd, kterou označili za Standardní společenskovědní model (SSSM). Pozor: zdaleka ne všechny sociální vědy pracovaly v rámci SSSM (viz srovnávací psychologii či psycholingvistiku) a ne všechny vědy fungující pod praporem SSSM byly vědy sociální (např. behaviorizmus). SSSM se opírá o představu, že pokud se děti všude na světě rodí stejné, zatímco chování dospělých se v různých kulturách liší, nemohou být tyto odlišnosti podmíněny biologicky, ale kulturně. Lidská mysl je téměř nekonečně tvárná a sociální svět utváří mentální organizaci člověka. Klíčovými faktory, jimiž se sociální vědy zabývají, pak jsou socializace a učení, tj. procesy, kterými kultura dělá ze stejných dětí různé dospělé. Základní představou je domněnka, podle níž se lidská mysl skládá z univerzálních psychických mechanizmů, které umožňují naučit se prakticky cokoliv bez omezení. Autoři ukazují, že sociální vědy se již ve svých počátcích vydělily od ostatních oborů a deklarovaly své výsostné právo na zkoumání sociálních záležitostí člověka. Otcové-zakladatelé sociálních věd a priori vyloučili potenciální použitelnost biologických vysvětlení pro sociální jevy (těžko si představit, co by mohlo rozvoji poznání škodit víc než podobné iracionální argumenty). Rád bych zdůraznil, že nikdy nelze o ničem předem říci, že to není biologicky nebo jakkoliv jinak vysvětlitelné – přinejmenším do doby, než pokusy o vysvětlení daného jevu odmítaným vysvětlovadlem ztroskotají. „Díky“ svému intelektuálnímu uzurpátorství se sociální vědy připravily o možnost čerpat z poznatků ostatních oborů. Následná izolace (nejen vůči přírodním vědám, ale i mezi sociálními vědami navzájem) zavedla nepřírodovědné obory do slepé uličky, v níž se tyto vědy ocitly zcela mimo vědu. [1] To bylo mnohdy zamaskováno lidovou představou, že když něco změříte, a pak spočítáte průměr, je to věda.

* Bezmála obsedantní fixace na Gouldovy názory „nefandící“ sociobiologii je typická pro řadu kritiků. Ti Goulda evidentně považují za kapacitu, kterou je dobré se zaštítit. Opírat se o jeho tvrzení je však spíše nejlepší způsob, jak si uříznout pořádnou ostudu v případě, že si kritiku přečte někdo, kdo má o evoluční biologii aspoň trochu páru. Jak napsal klasik současné evoluční biologie John Maynard Smith: Vzhledem k vynikajícím kvalitám Gouldových esejů ho začali nebiologové vnímat jako jednoho z předních evolučních teoretiků. Naproti tomu evoluční biologové, se kterými jsem diskutoval o jeho práci, ho berou spíše jako člověka, jehož myšlenky jsou tak zmatené, že sotva stojí za to, aby se jimi někdo obtěžoval. Zároveň se však domnívají, že by Gould neměl být veřejně kritizován, protože je aspoň na naší straně proti kreacionizmu. Na tom by ani tak nezáleželo, kdyby Gould nepodával nebiologům velmi pokřivený obraz současného stavu evoluční biologie (The New York Review of Books, 30. 11. 1995). Uváděním Gouldových tvrzení na pravou míru se zabývá také dvanáctá kapitola v Adapted Mind (Pinker, Bloom).

524

VESMÍR 79, září 2000 l http://www.cts.cuni.cz/vesmir

Kde se stala chyba? Představte si, že by chemik formuloval hypotézu, která by byla v rozporu se známými fyzikálními zákony, či biolog postuloval evoluční mechanizmy, které by ignorovaly fyzikální a chemické procesy. Psychologové se podobných omylů dopouštěli bohužel často. Jak poznamenali významní evoluční psychologové Daly a Wilsonová: Kdyby Freud lépe rozuměl Darwinově teorii, mohli jsme být ušetřeni jeho neplodných představ o pudu smrti. [5] Na druhé straně je třeba si uvědomit, že heuristický význam Freudova omylu je stejně zásadní jako Wynneova a Edwardsova představa o skupinové selekci. Oba autoři upozornili na nesmírně zajímavé a opomíjené fenomény, ale řešení, která navrhli, byla mylná. Tooby a Cosmidesová uvádějí několik závažných nedostatků Standardního modelu: ústřední logika SSSM je založena na naivních a mylných koncepcích vypůjčených od zastaralých ontogenetických teorií, SSSM spočívá na chybné analýze problému vrozenézískané a vyžaduje nemožnou psychologii, která nemohla vzniknout přírodním výběrem. [1] Podle SSSM

Za co může příroda? Tooby a Cosmidesová považují za zásadní i řadu dalších připomínek. Vzhledem k tomu, že podle SSSM je každá kultura unikátní, nemůže ze samé podstaty tento model vést k testovatelným predikcím. Heuristická hodnota takového přístupu se významně neliší od nuly. Další chybou (zdaleka nejen SSSM) je běžné zaměňování popisu s vysvětlením. Tvrzení sociálního vědce, že za něco

může kultura, je přijímáno s respektem k vzdělanému kolegovi – co byste ale řekli tomu, kdyby vám biolog zcela analogicky sdělil, že za něco může příroda? Někteří psychologové se tedy domnívají, že přístup založený na SSSM je značně povrchní, což není míněno pejorativně, ale doslova. Zaměřuje se totiž na variabilitu a odmítá vidět podobnosti a zákonitosti, které jsou pod povrchem. Jenže do jedné kategorie nelze zařadit ani dva lidi – mají různou barvu vlasů, očí, výšku. Nepopírám, že každá kultura je unikátní – kdyby nebyla unikátní, tak ji ani nepoznáme jako samostatnou kulturu. Jenže zvolení jednotlivých kultur za jednotku unikátnosti je arbitrární a opomíjení společných atributů logicky neospravedlnitelné. Již v Exupéryho Citadele se dočteme, že věda se zabývá jen tím, co se opakuje; sociální věda zabývající se jednotlivostmi je beletrie. Pikantní je i fakt, že samotná činnost kulturních antropologů (komunikace s příslušníky cizích kultur) vyvrací jejich předpoklady – bez ohledu na výraznou kulturní variabilitu nám nejsou pocity obyvatel exotických společností cizí. [5] Pokud sdílíme společné poznávací mechanizmy s příbuznými „subhumánními“ primáty, tak by bylo přinejmenším podivné, kdybychom je nesdíleli s jedinci uvnitř svého druhu.

Proč věnovat pozornost evoluci? Jako alternativu k SSSM navrhují Cosmidesová a Tooby aplikaci evolučního pohledu na člověka. Vycházejí z toho, že (lidská) mysl vykazuje všechny znaky komplexní adaptivní struktury, a jedinou vědeckou teorií, která dokáže vysvětlit vznik takových struktur, je teorie přírodního výběru (genů i memů) [2] – žádná alternativa zatím není známa. [6] Evoluční přístup při zkoumání lidských psychických vlastností je stejně legitimní jako fyzikální přístup při zkoumání Saturnových prstenců. Testovat hypotézy o lidské psychice, které jsou v rozporu s poznatky evoluční biologie a memetiky, [2] je ztráta času. Takové hypotézy jsou falzifikovány a priori. Přehlížení biologických základů lidského chování nemůže vést k odhalení mechanizmů formujících kulturu a vede jedině k formulování hypotéz postavených na tekutých píscích neznalosti. Zkoumat sociální jevy a neznat psychologii a biologii je stejně absurdní jako myslet si, že v molekulární biologii se vystačí bez znalostí chemie a fyziky. Psychologie je roztříštěna na řadu oborů – kognitivní, sociální či vývojovou psychologii a mnoho dalších. Evoluční psychologie je ale jedinou životaschopnou metateorií, která je schopna integrovat všechny tyto disciplíny. [3] Výhoda evolučního přístupu je v tom, že věnuje pozornost jak variabilitě (která evidentně existuje jen ve vymezených hranicích), tak společným vlastnostem a obecným jevům na všech hierarchických úrovních. Kritická analýza SSSM a zdůraznění významu evolučního přístupu Cosmidesovou a Toobym byly přijaty celou řadou psychologů a vedly k objevení nových a dosud netušených psychologických mechanizmů (viz brilantní kapitoly o těhotenské ranní nevolnosti nebo o kognitivních adaptacích pro sociální výměny). [1] Heuristický přínos nové evoluční psychologie je evidentní. [3, 4, 5] Na variabilitu sociálního chování má vliv řada faktorů – nejen unikátnosti konkrétní kultury, ale třeba i antigeny hlavního histokompatibilního komplexu. Ženy si vybírají muže s takovými antigeny, které jsou jiné než jejich vlastní – poznají to jednoduše čichem (že by to byla sociální role daná jejich kulturou?). Pokud při analýze variability lidského sexuálního chování vezmeme v úvahu tento faktor, je pravděpodobné, že tím snížíme podíl nevysvětlené variability (což je jádrem každého vědeckého počínání). Jistě je také http://www.cts.cuni.cz/vesmir l VESMÍR 79, září 2000

525

NAD KNIHOU

by mohl být nekonečně tvárný člověk sexuálně přitahován nejen partnerem opačného pohlaví, ale třeba i pařezem v lese (resp. protože je variabilita neomezená, měli bychom najít i takovou kulturu). Představa univerzálního mechanizmu mysli je nelogická – každý organizmus se setkává se specifickými adaptivními problémy a ty jsou řešitelné pouze specializovanými psychickými mechanizmy. Neexistují žádné univerzální problémy, a tak nemohou existovat ani žádné univerzální mechanizmy jejich řešení. [1] Některá tvrzení sociálních věd zůstala ideologicky motivovanými mýty, které nikdy nebyly v souladu s důkazy – řada závěrů je v přímém rozporu s citovanými zdroji informací. [1, 5] Není divu: vinou malé (nebo žádné) komunikace mezi sociálními vědami se často setrvačně lpělo na poznatcích, které náhodou proklouzly ze sousedního oboru i poté, co byly ve „zdrojovém“ oboru odhaleny jako mylné. Klasickým příkladem jsou některé „poznatky“ Margaret Meadové, které běžně přežívají mimo antropologickou literaturu. [5] Meadová se jako řada dalších vědců pod praporem SSSM snažila dokázat, že pohlavní rozdíly v chování jsou arbitrární a reverzibilní kulturní artefakty. Jedním z jejích trumfů byl novoguinejský kmen Arapešů, kde se údajně muži chovají jemně a zženštile. D. Tuzin však tuto kulturu podrobně prostudoval a dospěl k překvapivému výsledku: aby se „jemní“ mladíci Arapešů stali plnohodnotnými dospělci, mají povinnost odeslat na věčnost jiného muže. Vskutku pozoruhodný projev zženštilosti… V kritice SSSM se dále uvádí, že sociální vědy ignorovaly explanační kompatibilitu (Vesmír 79, 92, 2000), která je jednou z nutných podmínek rychlého rozvoje ostatních věd. Dichotomie mezi sociálními a přírodními vědami je však zbytečná (kdo, jak a čím změří, jestli je větší rozdíl mezi fyzikou a biologií, nebo mezi biologií a sociologií?). Je třeba mít stále na paměti, že sociální vědy jsou odvětvím biologie. Biologie je souborem přírodních věd, které se zabývají životem, a má tak celou řadu oblastí výzkumu. Jedna z nich, sociální vědy, se zabývá jen těmi aspekty lidské psychiky, které nás činí unikátními a odlišují nás od ostatních tvorů. To pochopitelně nevyjímá sociální vědy ze sféry biologie (tím se nemíní zrušení sociálních věd, ale potřeba slučitelnosti jejich poznatků s ostatními obory – biologie je v jistém smyslu také odvětvím fyziky, ale není to jen fyzika). Navíc řada fenoménů, které psychologové studují, nejsou čistě lidské atributy, a proto nemohou být v žádném smyslu striktně definovány jako doména sociálních věd, jak píše psycholog z londýnské univerzity Henry Plotkin. [7] Pokud budete zkoumat pisivku, dub nebo člověka a nebudete brát v úvahu jejich biologickou podstatu, s vysokou pravděpodobností budou vaše závěry „mimo mísu“. Psychologické, sociální a kulturní vlastnosti jsou produktem historie v konkrétním prostředí, ve kterém platí fyzikální a chemické zákony a probíhají evoluční procesy – sociální jevy nelze pochopit bez reflexe poznatků přírodních věd. Fyzikální a chemické zákony umožňují posoudit přijatelnost biologických teorií. Stejně tak poznatky biologie mohou posoudit přijatelnost teorií psychologických, antropologických a sociologických.

NAD KNIHOU

možné, že výběr podle antigenů nevysvětlí nic – v naší kultuře používáme deodoranty. V jiných kulturách to může dopadnout jinak. Sociální a biologická vysvětlení tedy nemusí být alternativní, mohou být i kompatibilní (Vesmír 79, 92, 2000/2). Domnívám se, že častý nesoulad mezi „psychology SSSM“ a evolučními psychology je nevyhnutelným důsledkem samotných podstat obou směrů. Psycholog řídící se SSSM je na rozdíl od evolučního psychologa nebo přírodovědce vyučen také k tomu, aby mohl pomáhat konkrétním jednotlivcům. Evoluční psycholog však mluví o lidech, ne o jednotlivcích – věda se zabývá obecnostmi, ne jednotlivostmi. Přesto se lze setkat i s argumentem ve stylu „vy sice tvrdíte, že muži jsou vyšší než ženy, ale moje sousedka má dva metry“. Představa, že takové tvrzení něco vyvrací či dokazuje, je projevem optimizmu hraničícího s blbostí. Mějme tedy stále na paměti, že tradičního psychologa zajímá člověk, zatímco evolučního psychologa zajímají lidé. Právě zmíněný oblíbený omyl sociálních věd (odvozování obecností z jednotlivostí) se odráží i v kritice sociobiologie, která do omrzení vychází z mentálního zkratu „E. O. Wilson si myslí XY; E. O. Wilson je sociobiolog; sociobiologové jsou hloupí, poněvadž si myslí XY“. Ach jo…

Proč byl SSSM úspěšný? Otázkou zůstává, proč byl SSSM obecně akceptován a přes svou vysvětlovací neschopnost se udržel tak dlouho. Důvodem byl zřejmě jeho silný morální apel: přijetí SSSM bylo obecně pokládáno za automatické odmítnutí rasizmu a sexizmu, naopak jeho kritika, třeba sociobiology, zaváněla rasistickými choutkami. Jistěže se najdou i rasističtí sociobiologové, stejně jako rasističtí antropologové, homofobní teologové či pedofilní etici. Standardní model přináší také jednu nezanedbatelnou morální oporu. Pokud ho přijmete jako biolog (třeba S. J. Gould), máte jistotu, že mezi biologickým a duchovním (lidským) světem zeje propast – ať zjistíte cokoliv, nebude to mít vliv na dlouho hýčkané a posvěcené iluze vznášející se v nadoblačných morálních výšinách. Standardní model – stejně jako módní pokrytecká morálka zvaná politická korektnost – imunizuje před potenciálně nepříjemnou realitou. Podle evolučních psychologů naplňují mýty sociálních věd potřebu sociálních vědců po tvárném světě, který budou vylepšovat hodní lidé s těmi nejlepšími úmysly a ne ti hnusní, sobečtí despotové jako dosud. [5] Řadě sociálních vědců však zjevně uniká, že vzhledem k dědičně fixované komponentě lidské přirozenosti je velmi iluzorní domnívat se, že strukturu osobnosti kteréhokoli člověka lze podstatným způsobem ovlivnit jinak než destruktivně (S. Komárek: Lidská přirozenost, s. 87). Nepřátelský vztah k biologii vyvěrá z představy biologického determinizmu – jenže veškerá věda je hierarchicky redukcionistická. [6, 7] Pro ilustraci si představme archeologa, který odmítne poznatky geologie jako deterministické a redukcionistické. Jevy, které studuje, jsou částečně produktem dobře známých geologických procesů, a tak důsledkem jeho ignorantství bude handicap, který si sám zavinil. Samozřejmě že žádný archeolog by nebyl tak nerozumný. Jenže biofobní sociální vědec je v analogické situaci. [5] Škodí sám sobě.

Ideologie ano – ale čí? C. G. Jung by měl zajisté neobyčejné povyražení z obviňování biologů z ideologičnosti ústy sociálních vědců (hezčí příklad projekce aby člověk pohledal). Stěží si lze představit krásnější ukázku ideologie než tu, která vyvěrá z politicky korektních sociálních věd. Tu, která se pokouší (přírodo)vědcům říkat, co se zkoumat smí a co ne, a která předem ví, co se zjistit ne-

526

VESMÍR 79, září 2000 l http://www.cts.cuni.cz/vesmir

smí. V souvislosti s tím je třeba připomenout, že je naprosto v pořádku, pokud E. O. Wilson dostane na frak za své teoretizování o sociálním inženýrství. Poněkud méně v pořádku je fakt, že zástupy postmoderních sociálních vědců, kteří sociální inženýrství aktivně praktikují (viz A. a B. Moirovi: Proč muži nežehlí), jsou tiše zameteny pod koberec. Obzvláště milé je, že se k sociálním vědcům v tomto směru přidali i někteří biologové, třeba věhlasní S. J. Gould nebo R. Lewontin. Ti se nikdy netajili svým zalíbením v rudých oblastech politického spektra – není proto divu, že jeden z článků reflektujících jejich známou publikaci The Spandrels of San Marco and the Panglossian paradigm se jmenuje Patolízalové svatého Marxe a panglosiánský paradox: kritika krasomluvného programu (The Spaniels of St. Marx and the Panglossian Paradox, Q. Rev. Biol. 70, 486–489, 1995). Jejich problém je právě v morálce politické korektnosti, která v podstatě tvrdí, že něco nějak je, protože by to tak mělo být (čili pravý opak tzv. naturalistického omylu). Politická korektnost vyvěrá z fobie, že případné rozdíly mezi pohlavími, národy a sociálními třídami vedou k diskriminaci. Proto se a priori tvrdí, že žádné takové rozdíly být nemohou – kdybychom je zjistili, bylo by to tak nehezké, že by to snad ani nebylo možno považovat za vědu (J. Zrzavý, doslov k Sobeckému genu). Rozhodně si nemyslím, že všichni sociální vědci jsou v tomto ohledu tak mdlého rozumu jako Gould či Lewontin. Na druhé straně by se jistě mohli vyvarovat mnoha nesmyslných tvrzení, kdyby se častěji zamysleli nad tím, proč tvrdí to, co tvrdí.

Intelektuální harakiri a pohlcovací paranoia Nikdo netvrdí, že tradiční sociální vědy nic nepřinesly. Problém je ale v tom, že intelektem schopných a pracovitých lidí se někdy plýtvá na výzkum, který nemá žádný smysluplný jednotící rámec. Kam vede ignorování např. fylogenetického přístupu k poznání, lze lehce ilustrovat. Mezi základní problémy sociálních věd patří otázka „Jak vznikla lidská společnost?“. Při hledání odpovědi přišlo vniveč hodně papíru. Dnes ovšem rousseauovští vznešení divoši zkažení civilizací či lidé hobbesovsky zkrocení Leviatanem dávno tlejí na smetišti intelektuálních dějin – jednoduše proto, že otázka byla chybně položena. Všichni předci člověka žili sociálně (Vesmír 78, 109, 1999/2), a tak lidská společnost vzniknout nemohla – společnost je starší než lidé. Dokonce ještě dnes však může být člověk svědkem debaty, na které přední česká socioložka Hana Librová spáchá veřejně intelektuální harakiri tvrzením, že sociobiologie nemá sociologii co říct. Pokud má zájem, aby se poznatky jejího oboru zabývali pouze kulturní paleontologové, pak je na správné cestě. A. Markoš v recenzi Wilsonovy Konsilience (Vesmír 78, 285, 1999/5) upozorňuje na hrozbu pohlcení humanitních věd přírodními. Domnívám se, že tato obava je neoprávněná. Pokud je mi známo, kdykoli se setkaly dva vědní obory, nebyl jeden pohlcen druhým, ale naopak vznikl třetí, hybridní obor (např. behaviorální ekologie vznikla smícháním tradiční etologie, ekologie, evoluční a populační biologie). Dále si uvědomme, že Wilson o konsilienci píše. Evoluční psychologové o konsilienci píší, a navíc ji dělají. K propojení evoluční biologie a psychologie došlo, nikdo pohlcen nebyl a zjistilo se mnoho nového a zajímavého. Hrozba interdisciplinárního imperializmu by měla zajisté vyplývat právě z takových publikací, jako je Adapted Mind. Problém je ale v tom, že žádný významný představitel evoluční psychologie není biolog – všichni jsou z řad sociálních vědců. Poznatky evoluční biologie do sociálních věd nepřinášejí biologové, ale sociální vědci sami. Evoluční psychologie není náhra-

Proč nemáme humánní pisivkologii? Domnívám se, že rozlišování věd na přírodní a sociální je zbytečný alibizmus, za kterým se skrývají přežívající dichotomické představy ve stylu „hádala se duše s tělem“. Předsudek, že pro sociální či psychický život člověka neplatí fyzikální, chemické a biologické zákony, byl důsledkem zamoření evropského intelektuálního světa odvěkými dichotomiemi materiální/spirituální, člověk/zvíře, biologické/sociální. [1] Trvání na těchto dichotomiích přineslo víc škody než užitku a stalo se spolehlivou brzdou vývoje poznání. Myslím si, že odpor sociálních věd vůči biologii je evidentně přežitkem z doby, kdy si lidé ještě mysleli, že jsou něco

Případ hagahajské krve

Hagahajové jsou malý kmen novoguinejských horalů z okolí řeky Yuat v odlehlém a nepřístupném Schraderově pohoří. Není jich více než tři sta a až do časných 80. let žili sami pro sebe, aniž by věděli o okolním světě a aniž by okolní svět věděl o nich. Teprve stoupající dětská úmrtnost, zejména na malárii, je přiměla, aby vyslali průzkumníky za hranice svého území a ověřili, co je pravdy na pověstech o mocných civilizacích daleko za kopci. A tak Hagahajové v roce 1983 objevili svět. Dnes, po necelých dvaceti letech, jim je na internetu věnována řádově větší pozornost než třeba Hanákům. Kýžená lékařská pomoc se na území Hagahajů dostavila v osobě Carol Jenkinsové, americké lékařky a vědkyně, která pracovala v Ústavu pro lékařský výzkum v provinčním novoguinejském městě Goroka. Jenkinsová Hagahaje přes deset let nejen léčila, ale také studovala. Při tom objevila (ve spolupráci s týmem Carletona Gajduska z Národního ústavu zdraví USA) v T-leukocytech hagahajské populace dosud neznámou benigní formu leukemického retroviru. Jelikož takový virus může být užitečný při vývoji vakcin nebo testů na přítomnost leukemie, rozhodla se Jenkinsová se svými kolegy linii leukocytů infikovanou retrovirem patentovat. Zdálo se, že tento postup je zároveň vhodnou formou ochrany ekonomických zájmů Hagahajů. Po deseti letech práce u nich Jenkinsová dosáhla téměř polobožského statutu a měla přirozeně starost o jejich budoucnost. Hagahajům tak

jiného než okolní příroda. V okamžiku, kdy zjistíte, že neexistují sociální sumýšologie a humánní pisivkologie jako samostatné obory vyčleněné oproti přírodním vědám o těch stejných sumýších a pisivkách, tak vám je jasné, z jakého důvodu jsou vyčleněny sociální a humanitní vědy o člověku vůči přírodním vědám o tom stejném člověku. Máte pravdu. S vědou to opravdu nemá co dělat.

O slonech, slepcích a vědcích Známý indický příběh o slonu a slepcích poukazuje na relativitu (pravdivého) poznání přirozeného světa. Jeden slepec uchopil nohu slona, řka, že slon je jako strom. Druhý vzal slona za chobot, a ejhle: slon vypadá jako had! Nepomohlo by, nejen slepcům, kdyby si podali ruce? ¨ LITERATURA [1] The Adapted Mind. [2] Blackmore S.: The Meme Machine. OUP, Oxford 1999 [3] Buss D. M.: Evolutionary Psychology. The New Science of the Mind. Allyn & Bacon, Boston 1999 [4] Daly M., Wilson M.: Sex, Evolution, and Behavior. Willard Grant Press, Boston 1983 [5] Daly M., Wilson M.: Homicide. Aldine de Gruyter, New York 1988 [6] Dawkins R.: The Blind Watchmaker. Penguin, London 1986 [7] Plotkin H.: Evolution in Mind. An Introduction to Evolutionary Psychology. Penguin, London 1997

VOJTĚCH NOVOTNÝ

bylo o retrovirech a patentování vysvětleno, co se vysvětlit dalo, byl získán jejich souhlas s patentem a uzavřena smlouva, která jim zaručuje neobvykle vysoký (padesátiprocentní) podíl na jakémkoli budoucím zisku. Patent byl nakonec pod číslem 5397696 udělen dne 14. března 1995. Pokud se domníváte, že toto je šťastný konec celého příběhu, jenž by mohl být vylepšen nanejvýš využitím patentu, které by vedlo k vyléčení leukemie a zbohatnutí všech zúčastněných, je to známkou značné naivity. Neboť těžko si lze představit něco politicky výbušnějšího než patent americké vlády na genetický materiál z krve domorodého kmene, právě se vynořivšího v rozvojové zemi z doby kamenné. Však se také případu promptně chopila kanadská nevládní organizace RAFI (Rural Advancement Foundation International čili Mezinárodní organizace pro pokrok venkova). Ta v článku „Domorodec z Nové Guineje patentován vládou USA“ konstatovala, že „14. březnem 1995 přestali Hagahajové vlastnit svůj genetický materiál“, a tím zahájila proti patentu rozsáhlou kampaň, jež potom pokračovala slogany o vědeckých upírech a genetickém kolonializmu. Kontroverze se nevyhnutelně přenesla i na Papuu-Novou Guineu, kde byla Jenkinsová nejprve obviněna ze zločinů biopirátství, a poté díky Hagahajům, kteří se jí zastali, zase zcela očištěna. A jaká je dohra celé tragikomedie dnes, po pěti letech? Jenkinsová opustila Novou Guineu a pracuje v Bangladéši. Vláda USA kontroverzní patent stáhla, oficiálně pro jeho malý komerční potenciál. RAFI se stále ze svých úřadů na ottawské Bank street bije hlava nehlava za práva domorodců celého světa. A Hagahajové? Ti nadále loví v hloubi tropického pralesa a pěstují batáty, taro a banány na jeho mýtinách. Lze jen doufat, že s hřejivým pocitem, jak ten retrovirus v jejich T-leukocytech je už zase jenom ¨ a výlučně jejich. http://www.cts.cuni.cz/vesmir l VESMÍR 79, září 2000

527

NAD KNIHOU

da psychologie biologií, ale psychologií, která je informována poznatky evoluční biologie [1] (není mi jasné, proč by chtěli sami psychologové rušit svůj obor). Nutno dodat, že evoluční biolog postulující kognitivní procesy, které nemohly vyřešit uvažovaný adaptivní problém, se ocitá mimo vědu úplně stejně jako psycholog navrhující psychický mechanizmus, který nikdy nemohl evolucí vzniknout. [1] Konsilience neznamená jednostranný tok informací z přírodovědných oborů do sociálních věd, ale oboustrannou komunikaci. Pokud dojde podle Wilsonových představ k sjednocení přírodních a ostatních věd, nevzniknou tím vědy přírodní. Vznikne tím prostě věda.

29. Grim T. 2001: Etologie versus sociobiologie? 29. Vesmír 80(4): 231–233. Grim T. 2001: Etologie versus sociobiologie? Vesmír 80(4): 231–233.

Snímek © Vladimír Voráček

nické syntaxonomie). Takže určitě to není tak, že by obě strany neznaly své mantinely. Naopak – těmi mantinely je ona vzájmená nekompatibilita, a tu všichni znají až příliš dlouho a dobře. Kdy je lepší bránit se šátečkem bílým a kdy kalašnikovem, je věcí podmínek, ne odhodlání. Odhodlání je totiž taky věcí podmínek. A ještě k těm citacím – není to přece jen tuze velká přísnost? Ono je samozřejmě těžko odporovat a tvrdit, že tam současné světové literární odkazy být

Etologie versus sociobiologie? S. FRAŇKOVÁ, V. BIČÍK: Srovnávací psychologie a základy etologie NAD KNIHOU

Karolinum, Praha 1999, 296 stran, náklad 700 výtisků, cena 165 Kč

Jestliže student není informován o aktuálním stavu svého oboru, stěží se může podílet na výzkumu, který by přinesl něco zajímavého. Zkreslování koncepcí a vědeckých poznatků nevydařenými překlady (Vesmír 78, 464, 1999/8) či učebnicemi (Vesmír 78, 335, 1999/6) rozvoji žádného oboru nepřispěje. To vše je myslím dostatečný důvod pro pár poznámek o nových etologických skriptech, jež ilustrují několik běžně rozšířených nedostatků literatury určené studentům. S. Fraňková a V. Bičík jsou zkušení pedagogové, a proto není divu, že řada kapitol je napsána kvalitně, přináší zajímavé poznatky, upozorňuje na tradované omyly a rozhodně stojí za přečtení. Autoři poukazují na důležitost všech čtyř explanačních rovin při vysvětlování chování (Vesmír 79, 92, 2000),

230

VESMÍR 80, duben 2001 l http://www.cts.cuni.cz/vesmir

nemají. Nejsou tam, a to je velká škoda, i když připustíme, že potenciální čtenář je hlavně člověk z praxe (lesák, úředník, ochranář), a ten valnou většinou stejně do pramenů nepůjde. Přesto to mám spíš za problém štábní kultury. Provinciální by to bylo, kdyby se pracně objevovala Amerika nebo se nevědomky tvrdilo něco zásadně odlišného, než co se o věci říká mimo ČR. Ale ani tak, ani onak. Pes má pořád čtyři nohy, ať už to spočítám sám a nechám bez citace, či to ocituju dle fosilního Brehma, anebo podle čtrnáct dní starého článku v Mammalogy Today. V této knize se netvrdí nic o tolik složitějšího. Jiří Sádlo, BÚ AV ČR

TOMÁŠ GRIM

ale sami nedocenili potřebu vyrovnaného a všestranného pohledu. Funkční (evoluční) přístup totiž buď zcela chybí, nebo je založen na překonaných koncepcích. To je významný nedostatek – učebnice je určena psychologům, pro které má funkční pohled zásadní význam (viz též Vesmír 79, 524, 2000/9). Pokud se pak navíc student v psychologické literatuře dočte, že sociobiologie je směr sociálního darwinismu (B. Geist: Psychologický slovník, Vodnář, Praha 2000, s. 272), je to dost na pováženou. Také tvrzení, že Řada psychologů nesouhlasí se základními tezemi evoluční psychologie, neboť lidské chování je velmi pružné a nelze jej odvozovat od strohé, jednostranné závislosti na genetickém dědictví (tamtéž s. 217), svědčí o tom, že psychologové mají o tématu dosti mlhavé vědomosti. Pokud psychologové (až na pár světlých výjimek) nemají o evoluční biologii a příbuzných oborech ani ponětí, kdo jiný než biologicky vzdělaní autoři skript by jim to měl vysvětlit?

Dalším problémem je úsilí přinést nové poznatky (s. 3) a odkazovat na prameny staré desítky let. Taková snaha je přímo z definice odsouzena k nezdaru. Také nevím, proč nebylo možno uvést práce všech autorů citovaných v textu (s. 3) – seznam literatury je tištěn s velkými mezerami mezi citacemi, navíc citované literatury není nikdy dost, zvláště když je text určen studentům. Jistě by se také dalo uvést pár desítek vědců, kteří přispěli k poznatkům o chování živočichů více než např. pan Akimuškin. O těch ale není v knize ani zmínka. Ve skriptech jsou převážně uváděny příklady z laboratorního (většinou behavioristicky laděného) výzkumu. Potkan se ale v přírodě skutečně jen zcela výjimečně dostává k potravě mačkáním páčky a pes se za normálních podmínek nemůže zrovna moc spoléhat na to, že ho na přítomnost kořisti upozorní žárovka a zvonek. Jinak řečeno, psychologie je věda o člověku. O jeho chování a psychických adaptacích nám toho více řeknou studie současných lovecko-sběračských společností spolu s analýzou adaptivních problémů, jimž lidé čelili během dvou milionů let v pleistocénu, než chování potkanů a psů v laboratoři. Aby psychologové udrželi behaviorální fenomény v souladu s teorií, museli sledovat lidi a další pokusné objekty mimo ekologicky významné podmínky, nepředkládali jim biologicky relevantní podněty a testovali je na umělých problémech, se kterými se tyto subjekty nemohly setkat v prostředí evolučních adaptací [1]. Ovšemže tímto netvrdím, že by v učebnici pro psychology neměly být uváděny výsledky laboratorního výzkumu (ostatně jen sledování organizmu v kontrolovaných laboratorních podmínkách může poskytnout informaci např. o jeho senzorických kapacitách či schopnosti učení). Jde spíše o to, že výsledky laboratorního výzkumu nejsou to jediné, co je relevantní pro pochopení lidské psychiky (navíc student získává dojem, že většina poznatků o chování plyne z vhledu F. B. Skinnera, který si uvědomil, že výslechové metody CIA a FBI lze aplikovat i na zvířátka). Stejně tak je doufám jasné, že se nesnažím nahradit tradiční etologii či kognitivní vědy poznatky evolučně inspirovaného pohledu na věc. Každá rovina pohledu řeší jiné otázky, a právě proto je důležité poskytnout studentovi vyváženě poznatky z co nejvíce oborů a také ho nasměrovat ke čtení odpovídající literatury. Vzhledem k obsahu učebnice lze recenzovaný počin chápat jedině jako upřímnou a vydařenou snahu autorů podat obraz historie svých oborů. (V tom je cosi esenciálně lorenzovského. Už v roce 1982 napsal známý evoluční biolog Mark Ridley v recenzi na anglický překlad Lorenzových Základů etologie: Lorenz mohl napsat knihu v duchu Základů etologie před 30 lety. Kdyby ji vydal tehdy, mohla to být klasika jako Tinbergenova Study of Instinct. Její vydání dnes působí dojmem trapně předpotopním [8].) To je ale zbytečný luxus v situaci, kdy student nemá přístup k aktuálním poznatkům. S. Fraňková má bohužel stále pravdu v tom, že v České republice dosud nebyl publikován materiál, z nějž by bylo možno vycházet při přednáškách z etologie člověka (a s moderní vědou o chování zvířat to není o moc lepší). Oboje skripta napsaná pod vedením S. Fraňkové jsou zatížena podobnými nedostatky – nejsou reprezentativním obrazem současného stavu oboru (viz recenzi na skripta o etologii člověka ve Vesmíru 78, 335, 1999/6). Co je příčinou? Především vytváření umělé demarkační linie mezi etologií a sociobiologií (či behaviorální ekologií, etoekologií, ekologií chování apod. – říkejte si tomu jak

Kritika by měla být normální součástí vědy, protože jen skrze kritickou výměnu názorů můžeme využít to, že věda se stala kolektivním podnikem. Měli bychom bojovat otevřeně, vyříkat si věci, vyhledat si navzájem slabosti ve výsledcích, teoriích, tvrzeních. Málokdy tak ale činíme. A pokud se vůbec někdo odváží kritizovat nás, bereme to jako osobní útok a odpovídáme iracionálně. A všechno to je známkou nejistoty, a nejistota je známkou malosti a průměrnosti. Jan Klein, Vesmír 77, 109, 1998/2

chcete). Jediným smyslem takového škatulkování je obrana proti kritice vedené na domácí půdě. Křečovité vytváření dichotomie mezi oběma obory lze ukázat na vývoji publikační činnosti asi nejznámější osobnosti evoluční biologie – Richarda Dawkinse. Ten svou první práci (o ontogenezi klovacích preferencí u kuřat) publikoval v Zeitschrift für Tierpsychologie [2], klasickém etologickém časopise, který dnes vychází pod názvem Ethology. Editorem tehdy nebyl nikdo jiný než Konrad Lorenz. Dawkins (žák dalšího slavného etologa Niko Tinbergena) ve stejném časopise publikoval i své dvě nejbrilantnější práce [3, 4], a to až po sociobiologické revoluci. Proč revoluci? Během relativně krátkého období v sedmdesátých letech se výzkum chování zvířat změnil tak radikálně, že se to dá nejlépe popsat jako revoluce. Klasická etologie v podstatě přestala existovat a byla nahrazena plodnou směsicí sociobiologie, která je populárně známa jako teorie sobeckého genu, behaviorální ekologie a teorie her. Dlouho zavedené kvalitní učebnice byly rychle nahrazeny knihami, které reagovaly na tyto nové přístupy. …starší, více popisné srovnávací studie prakticky vymizely. Výuka se změnila tak zásadně, že kurz o chování zvířat z roku 1973 jednoduše neměl nic společného s touž přednáškou o pět let později. Tyto fundamentální změny v myšlení souvisely s posunem v interpretaci a chápání významu evoluční teorie [7]. Pěkně to shrnuje legenda české etologie Z. Veselovský: Nutností pro zoologa je, aby si kladl otázku „proč“. Proto nemám rád práce pouze popisné (Vesmír 77, 690, 1998/12). To dobře věděl N. Tinbergen, jeden ze zakladatelů etologie. Ve svém klasickém článku o cílech a metodice etologického výzkumu [9] zdůrazňuje nutnost vyrovnaného pohledu na chování prizmatem všech čtyř explanačních rovin (Vesmír 79, 92, 2000). Ostatně jeho vlastní práce z podstatné části probíhala v rámci behaviorálně-ekologického myšlení už před druhou světovou válkou, jak se může každý přesvědčit v jeho vynikající knížce Zvědaví přírodovědci, která vyšla i u nás, či v klasické, dodnes hojně citované monografii The Study of Instinct (1951). (Nejen) proto jsou tak bizarní známé povzdechy „co je to ta behaviorální ekologie zase za novotu“ a jak „ti sociobiologové opovrhují vším, co je starší než deset let“. To, že jistému přístupu ke zkoumání věcí zvířecích a lidských říkáme behaviorální ekologie až od sedmdesátých let dvacátého století, pochopitelně neznamená, že obor je starý 30 let. Kdepak. Přidejte jednu nulu – klasické behaviorálně ekologické práce od Lottingera a Jennera pocházejí z konce osmnáctého století a jsou stále citovány. Kulturní svět si letos připomene 225. výročí vzniku behaviorální ekologie. Jistěže existuje rozdíl mezi klasickou, spíše popisnou a proximátně orientovanou etologií a relativně mladší, spíše evolučně a funkčně, tj. ultimátně zaměřenou behaviorální ekologií. (Velmi významné jsou i další dva rozdíly. Klasická etologie uvažuje o adaptacích pro dobro skupiny či druhu a v důsledku toho považuje chování za druhově typické – variabilitě uvnitř druhu není věnována žádná zvláštní pozornost. Naopak pro behaviorální ekologii je vnitrodruhttp://www.cts.cuni.cz/vesmir l VESMÍR 80, duben 2001

231

Opačný byl názor na úroveň našich vysokých škol, jejichž styl se podle Barlowa od středoškolského liší jen málo. Například byl Barlow překvapen existencí skript – americký student si faktografii musí vyhledávat sám, nedozví se ji ani na přednáškách. Tento způsob práce podněcuje k dobrému výkonu i průměrného studenta. Klade se důraz na dialog mezi učitelem a studentem, a také na zpracování písemných elaborátů, ale tráví se mnohem méně času při výuce. […] Často musí učitel kvůli zanedbanému stavu našich knihoven věnovat neúměrně mnoho místa přednášení. Mnohým pedagogům to ale vyhovuje, neboť stačí „sehnat“ pár zahraničních titulů a napsat podle nich skripta... RNDr. Blažena Švandová, Ph.D. viz Vesmír 79, 424, 2000/8

hová variabilita chování klíčovým tématem a za jednotku selekce není považována skupina, ale jedinec, respektive gen.) Na druhé straně jde ovšem pořád o vědu o chování zvířat; dnes se jen více než dříve ptáme, proč živočich vůbec něco dělá. Letmé prolistování aktuálních čísel libovolného časopisu o chování zvířat jasně ukazuje, že termíny „etologie“ a „behaviorální ekologie“ jsou volně používány jako zaměnitelné ekvivalenty. Ostatně behaviorálně-ekologický článek mých katedrových kolegů byl publikován v časopise Ethology [6], zatímco můj spíše etologický článek vychází v časopise Behavioral Ecology and Sociobiology [5]. Krajně pochybuji, že by některý současný zahraniční etolog dokázal strávit větu, že určitým problémem se zabývali sociobiologové, …avšak i etologie nashromáždila mnoho dokladů, jak se píše ve skriptech. Věda o chování zvířat vždy byla a je jen jedna. Tuto kontinuitu učebnice S. Fraňkové nereflektují. Dichotomické uvažování autorku nakonec přivedlo k tvrzení, že přínos sociobiologie k poznání základů sociálního chování člověka mohou kvalifikovaně posoudit lépe sociální psychologové (Vesmír 78, 336, 1999). Problém mohou samozřejmě nejlépe posoudit ti psychologové, kteří si něco přečetli také o evoluční biologii – ti znají problematiku z obou stran. Přínos sociobiologie byl psychology doceněn před drahnou dobou – všechny velké postavy evoluční psychologie (J. Barkow, D. M. Buss, L. Cosmidesová, M. Daly, H. Plotkin, D. Symons, J. Tooby, M. Wilsonová) jsou psychologové a antropologové. Všichni se snaží upozornit na to, že neúspěch standardního společenskovědního modelu, na kterém spočívala tradiční psychologie, vyplývá z ignorování evoluční biologie (Vesmír 79, 524, 2000/9). S heuristickým významem evoluční psychologie se český čtenář již dříve mohl seznámit v článku J. Madlafouska (Československá psychologie 38, 53, 1994/1). Na druhé straně bychom se výše citovaným tvrzením dostali do rozporu se samotným zakladatelem etologie K. Lorenzem. Ten neustále zdůrazňoval, že studium chování zvířat je jediným a konečným zdrojem pochopení nás samých (citováno z dopisu Margaretě M. Niceové). V důsledku uvedeného autorčina přístupu pak student nemůže získat obrázek o aktuálním stavu oboru. K tomu přispívá i způsob argumentace některých kritiků, kteří vycházejí ze stejného myšlenkového mycelia jako autorka. Např. Z. Veselovský píše, že kromě nepřesných informací vysílaných o přírodě např. v české televizi považuje za nebezpečné i velmi zjednodušené názory některých módních sociobiologů, kteří vykládají chování člověka i vyšších obratlovců ekonomickými metaforami jedině jako přenos svých egoistických genů (Člověk a zvíře, s. 225). Pokud se ovšem máme držet pravidel vědecké komunikace, měly být připomínky uvedeny jasně a explicitně. Jediným smysluplným a přijatelným argumentem by

232

VESMÍR 80, duben 2001 l http://www.cts.cuni.cz/vesmir

bylo uvést jiné myšlenkové schéma, které vysvětluje chování živočichů lépe než současná ekologie chování (to, že se mi něco nelíbí nebo to „považuji za nebezpečné“, lze za argument považovat stěží, resp. takové argumenty se mi nelíbí a považuji je za nebezpečné). Zde jen připomínám, že téměř všichni vědci zkoumající chování živočichů pracují v myšlenkovém prostředí „sobeckých genů“ [7] už od sedmdesátých let. Z další poznámky (sociobiologie přinesla mnoho závažných poznatků zejména u sociálního hmyzu a nižších obratlovců…, s. 222) a následujícího textu vzniká dojem, že sociobiologická ultimátní vysvětlení (Vesmír 79, 92, 2000/2) pro vyšší obratlovce neplatí. Bylo by tedy vhodné uvést jasné teoretické důvody, proč by tomu tak mělo být. Sociobiologie se, jak přímo z názvu vyplývá, zabývá chováním sociálních živočichů, není tedy divu, že většina sociobiologických článků zabývajících se obratlovci se týká nikoli ryb či obojživelníků (nižších obratlovců), ale ptáků a savců (tedy vyšších obratlovců), což zrovna není v nejlepším souladu s citovaným tvrzením. Ostatně stačí si prolistovat kterékoli číslo v současné době nejprestižnějších (pokud budeme hodnotit dle impakt faktoru) zoologických časopisů, tj. Behavioral Ecology and Sociobiology, Behavioral Ecology, Animal Behaviour nebo Ethology (z 300 náhodně vybraných článků z těchto časopisů z osmdesátých a devadesátých let se 18 % věnovalo nižším a 54 % vyšším obratlovcům a 28 % bezobratlým živočichům). Navíc mi připadá přinejmenším podivuhodné argumentovat proti obecné koncepci sociobiologie, která velice dobře vysvětluje chování desítek modelových druhů od švábů po sojky křovinné, tím, co dělá samice makaka japonského nazvaná Mozu z parku Jigokudani v Japonsku (s. 222). Proč to rozebírám? Z. Veselovský je v pravém slova smyslu legenda (plným právem), čili člověk, jenž má obrovský potenciál ovlivňovat mínění především laiků, kteří o ekologii chování a sociobiologii nic nevědí. Ti pak zcela zbytečně na základě mlhavých a nepodložených argumentů z populární literatury získávají apriorní nedůvěru k tomu, o čem by si mohli přečíst leccos zajímavého. To jistě k rozvoji vědeckého bádání a jeho popularizaci příliš neprospěje. Skriptům, která mohla přinést vyváženější pohled na současnou situaci v behaviorálních vědách, by nebylo třeba věnovat nijak zvláštní pozornost, kdyby těch skript existovalo více a aspoň některá z nich by byla svou kvalitou rovnocenná zahraničním dílům. Tradičním procesem kulturní evoluce se studenty a jejich učiteli jako selekčním agens by se nakonec v memofondu udržela jen díla skutečně kvalitních psavců – „přežili“ by jen ti autoři, jejichž skripta by se kupovala a kteří by dostali grant na jejich další vydání. A to se opravdu netýká jen skript o etologii. ¨ LITERATURA [1] Barkow J., Cosmides L., Tooby J.: The adapted mind, OUP, New York 1992 [2] Dawkins R.: The ontogeny of a pecking preference in domestic chicks, Z. Tierpsychol. 25, 170–186, 1968 [3] Dawkins R.: Replicator selection and the extended phenotype, Z. Tierpsychol. 47, 61–76, 1978 [4] Dawkins R.: Twelve misunderstandings of kin selection, Z. Tierpsychol. 51, 184–200, 1979 [5] Grim T., Honza M.: Does supernormal stimulus influence parental behaviour of the cuckoo’s host? Behav. Ecol. Sociobiol. 49, 322–329, 2001 (http://link.springer.de/link/service/ journals/00265/contents/00/00295/) [6] Pavel V. a kol.: Distraction displays in meadow pipit (Anthus pratensis) females in Central and Northern Europe, Ethology 106, 1007–1019, 2000 [7] Plotkin H.: Evolution in mind. An introduction to evolutionary psychology, Penguin, London 1997 [8] Ridley M.: Tasteless behaviour, Nature 295, 439–440, 1982 [9] Tinbergen, N.: On aims and methods of ethology, Z. Tierpsychol. 20, 410–433, 1963

30. Grim T. 2001: Mafiánské kukačky a tyranští mravenci. Kukačky nutí hostitele 30. vychovávat parazitická mláďata. Grim T. 2001: Mafiánské kukačky tyranští mravenci. Kukačky nutí hostitele Vesmír a80(9): 488–490. vychovávat parazitická mláďata. Vesmír 80(9): 488–490.

Mafiánské kukačky a tyranští mravenci Kukačky nutí hostitele vychovávat parazitická mláďata TOMÁŠ GRIM

O vědních oborech, které se věnují spíše mechanizmům biologických jevů (např. molekulární nebo vývojová biologie), se na stránkách populárního tisku většinou žádné adrenalinové debaty nerozvíjejí (genetika klonování budiž výjimkou). Poznatky těchto odvětví přicházejí ze světů, které jsou tomu našemu lidskému příliš cizí a vzdálené, takže v laikovi bohužel často vyvolávají jen náznak zájmu smíšeného s podivem. Na druhé straně evoluční biologie či behaviorální ekologie ze samé své podstaty nutně drnká na skryté struny našich myslí – jsme sociální tvorové posedlí vyprávěním příběhů; co není prezentováno formou příběhu, je nestravitelná nuda. (Jak jste si ve škole zapamatovali, ve kterém ovoci se který vitamin vyskytuje, a jak vám utkvěl v hlavě popis Kolumbova pachtění do západní Indie?) V hlavě evolučně orientovaného biologa se tak mohou zrodit kdejaké podivnosti poutající pozornost.* Kukaččí mafie Jeden z nápadů na první pohled šílených vypučel v mozkových závitech Amotze Zahaviho – legendární postavy evoluční biologie, nejdříve haněného, a nakonec oslavovaného autora hendikepového principu (Vesmír 79, 625, 1999/11). V roce 1979 ho (zřejmě při sledování napínavých izraelských thrillerů) napadlo, že existence mafie (tj. struktury, která postihuje nedostatečně loajální jedince) nemusí být unikátním lidským znakem, ale mohla vzniknout i u jiného druhu žijícího v podobných ekologických Kukačka chocholatá (Clamator glandarius), kresba © Jan Hošek

podmínkách. Co třeba takové kukačky? Tvorové s ne zrovna nejlepším morálním profilem jako by se pro úlohu zvířecích mafiánů přímo vylíhli. Tito paraziti dotáhli reprodukční dělbu práce mezi plozením potomstva a staráním se o potomstvo do naprostého extrému – kukačky plodí, pěstouni se starají. Osvobozeny od rodičovských povinností mají kukačky spoustu volného času, takže by podle mafiánské hypotézy velkého Amotze mohly provádět „kontrolní návštěvy“ hnízd svých hostitelů a pěkně po mafiánsku trestat ty, kteří nebyli hodní a bránili se tím, že kukaččí vejce vyhodili. Zahavi se domníval, že takovým chováním by parazitický pták mohl donutit svého hostitele, aby mláďata parazita vyseděl a vychoval, i kdyby byl schopen parazitická vejce rozpoznat a diskriminovat je. Tato hypotéza je zjevně tak přitažená za vlasy, tak antropomorfizující, a navíc sociomorfní, že snad ani nikoho nepřekvapí, že se bezmála dvě desetiletí po publikování potvrdila, a to u kukačky chocholaté (Clamator glandarius). Tento původně africký druh je jihoevropským protějškem naší kukačky obecné (Cuculus canorus). Svá vejce klade hlavně do hnízd straky obecné (Pica pica). Výzkumná skupina Manuela Solera prohání kukačky chocholaté po andaluských mandloňových sadech už dobrých dvacet let. Testovala mimo jiné i mafiánskou hypotézu. Solerovci v některých parazitovaných hnízdech kukaččí vejce nechávali, v jiných je odebírali. Tato z našeho pokřiveného humánního pohledu chvályhodná činnost však odparazitovaným hostitelům nic dobrého nepřinesla – hnízda, kde byla vejce odebrána, byla pětkrát častěji vypleněna než hnízda ponechaná „neblahému“ osudu. Ukázalo se, že žádná smysluplná vlastnost prostředí (např. nalezitelnost hnízda) nedokáže vysvětlit rozdíl v intenzitě predace (není divu – mandloňové sady jižního Španělska jsou mimořádně stejnorodým prostředím). Tento výsledek hypotézu o mafiánských manýrách kukaček chocholatých podporuje, ale důkaz je to nepřímý. Proto byly kukačky sledovány pomocí vysílaček a skutečně se podařilo zaznamenat samice navštěvující stračí hnízda a plenící ta, která byla kukaččích vajec zbavena. Kukačky chocholaté jsou čistě hmyzožravé a při likvidaci hnízd hostitelů jejich vejce pouze naklovou – nežerou je, jako to dělá kukačka obecná. (Zda snůšku zlikvidovala kukačka chocholatá, nebo nějaký predátor, lze tedy snadno poznat i zpětně.) Jaký je smysl kukaččího chování? Paraziti využívají toho, že straky po zničení první snůšky zahnízdí znovu, a vytvářejí si tak příležitost napravit neúspěch prvního parazitického pokusu. Co ale dělat, když je parazitována i náhradní snůška? Jasná odpověď zní: starat se o kukaččí mládě, a to i v případě, že hostitel cizí vejce rozpozná (což je vysoce pravděpodobné a závisí to pouze na kvalitě mimikry kukaččích vajec – rozpoznávací schop* Ne že by se nerodily i v hlavě molekulárníkově – to, že se v procesu translace v eukaryotické buňce „vazbou přediniciačního komplexu k mRNA za spolupůsobení faktorů eIF1, eIF4A a eIF4B vytvoří za hydrolýzy ATP iniciační komplex GTP.eIF2.Met~tRNAiMet.40S. mRNA.eIF3.eIF4C.eIF4A.eIF4B.eIF1“ (viz S. Rosypal a kol.: Molekulární genetika, SPN, Praha 1989, s. 177), je příběh dosti podivuhodný a žádný student biologie ho zajisté nezapomene, ale vyprávění je bez zápletky a pointě nikdo nerozumí.

Mgr. Tomáš Grim, Ph.D., (*1973) vystudoval zoologii na Přírodovědecké fakultě MU v Brně. Na katedře zoologie Univerzity Palackého v Olomouci se zabývá etologií hnízdního parazitizmu. (www.zoologie.upol.cz/osoby/grim.htm)

488

VESMÍR 80, září 2001 l http://www.cts.cuni.cz/vesmir

nost hostitel nutně má, jinak by nevyhodil parazitické vejce při prvním hnízdění). Taková na první pohled absurdní odpověď vyplývá z toho, že při vyplenění druhé snůšky je už příliš pozdě na třetí hnízdní pokus. Druhý pokus parazitovat hostitele bývá skutečně úspěšnější než pokus první – hostitelé nejsou hloupí, vědí, co by je čekalo, kdyby zase kukaččí vejce vyhodili, a tak mění své chování v průběhu jediné sezony – jednoduše přepnou program „vyhazuj“ na program „akceptuj a buď hodným pěstounem“. (Straky, které nebyly parazitovány kukaččími vejci, ale jejich hnízda byla zničena, své chování během sezony nemění!) Tato strategie funguje pouze proto, že mládě kukačky chocholaté na rozdíl od kukačky obecné nevyhazuje hostitelova vejce nebo mláďata. Přestože parazitické mládě zlikviduje konkurencí o potravu skoro všechny nevlastní bratříčky a sestřičky, je pro straky pořád výhodnější kukačku akceptovat než vyhodit. Straky, které kukaččí vejce přijmou, nějaká vlastní mláďata občas vychovají. Straky, které vejce vyhodí, nevychovají nic. V prostředí zamořeném mafiány je lepší platit a něco málo si uchovat než neplatit a ztratit vše. Na pravděpodobnost přijetí kukaččího vejce v náhradní snůšce má rozhodující vliv intenzita parazitace. V oblastech, kde je parazitováno jen málo stračích hnízd, změní své chování zanedbatelné množství hostitelských párů. Naopak v populacích zatížených vysokou frekvencí parazitizmu přejde z odmítacího na přijímací chování 90 % strak. V oblastech s vyšším výskytem kukaček je riziko parazitace náhradní snůšky samozřejmě větší, takže rozdílné změny v chování hostitelů mezi různě intenzivně parazitovanými populacemi dávají jasný adaptivní smysl. Brilantní práce M. Solera a jeho spolupracovníků ukázaly, že straky akceptující kukaččí vejce mají mnohdy vyšší sezonní reprodukční úspěch než straky, které odmítají vychovat parazitické mládě. Jak je pak ale možné, že se v populaci udržuje strategie vyhazování parazitického vejce? Důvodů je zřejmě více. Zda je pro konkrétní stračí pár výhodnější kukaččí vejce odmítnout, nebo přijmout, závisí na řadě faktorů, jako je třeba chování kukaček, frekvence parazitace, kvalita rodičů a jejich teritoria. Kromě toho ne všechna odparazitovaná hnízda jsou nalezena kukačkami a vypleněna. A v důsledku toho se v populaci udržuje stabilní kombinace obou strategií. Zkrátka pro některé hostitele je za určitých podmínek výhodnější, když kukaččí vejce přijmou, pro jiné, když ho vyhodí. Parazitace může dokonce hostiteli přinášet malou výhodu. Samice kukaček totiž tráví svůj volný čas také aktivní obranou hnízd, která parazitovaly – vydávají varovné hlasy a útočí na potenciální nepřátele ochomýtající se kolem parazitovaných stračích hnízd. Toto chování docela funguje. Hnízda, na jejichž obraně se podílely kukačky chocholaté, se stávala obětí predátorů méně často než hnízda hájená pouze strakami. Kukaččí mafie, jako každá jiná slušná zločinecká organizace, drží ochrannou ruku jen nad těmi, kteří za ochranu platí. Toto podivuhodné chování kukaččích samiček kompenzuje hostiteli výdaje spojené s parazitizmem, a tím snižuje selekční tlak na evoluci obranného chování strak. U kukačky obecné se mafiánské choutky zřejmě nevytvořily. Není divu – mafii zjevně svědčí teplý jih (takovým Laponcům nebo Hanákům se krev na bod varu rozehřívá dost ztuha).

Proč lákat parazita k hnízdu? Absurdní situaci, kdy v zájmu hostitele je nebránit parazitovi, najdeme i na druhé straně Atlantiku. Severoameričtí strnadi zpěvní (Melospiza melodia) se chovají podivně. Staří jedinci aktivně útočí na samice parazitických vlhovců hnědohlavých (Molothrus ater), které hledají hnízda potenciálních hostitelů. Mladí jedinci to ještě neumějí. Průšvih pro strnady je v tom, že samice vlhovců využívají jejich nápadné chování k lokalizaci parazitovatelných hnízd. Nakonec jsou starší a zkušenější strnadi parazitováni víc než mladší naivní jedinci. Pro vlhovce to má jasnou výhodu – starší páry strnadů jsou lepšími rodiči a vyhlídky mladých vlhovců na přežití jsou tedy podstatně růžovější, než kdyby se ocitli v hnízdech mladých strnadů. Ale proč se strnadi chovají tak hloupě? Paradoxní odpověď zní, že pro starší jedince je možná výhodnější být parazitován. Pokud se samici vlhovce podaří parazitovat strnada během doby, kdy klade vejce, tak samice parazita nezničí hostitelovu snůšku. To by udělala v případě, kdyby nalezla hnízdo hostitele příliš pozdě, tj. kdyby už hostitel seděl na vejcích. Staří strnadi (a pěvci obecně) začínají hnízdit dříve než mladí a nezkušení. Samice vlhovce během hnízdní sezony najde postupně téměř všechna hnízda strnadů ve svém teritoriu. Z těchto zjištění vyplývá, že pro staršího strnada je výhodnější odchovat mládě vlhovce i svá mláďata (mladí vlhovci nevyhazují obsah hostitelova hnízda) než riskovat pozdější nalezení hnízda a predaci celé snůšky. Opět je lepší zaplatit málo než neplatit a přijít reprodukčně na buben. Proč se ale podobně nechovají mladší páry strnadů zpěvných? Výdaje spojené s vychováním parazitického mláděte jsou pro ně samozřejmě vyšší než pro starší hostitele, takže adaptivnější strategií je spolehnout se na to, že je samice vlhovce třeba nenajdou. Mafiáni jsou všude kolem nás (i v nás) Podstata mafiánského efektu spočívá v tom, že jedinec, proti kterému je namířena nějaká obrana, činí svým chováním tuto obranu příliš ekonomicky nákladnou, a tudíž nevýhodnou. Je tedy možné, že mafiánský mechanizmus není omezen jen na hnízdní parazity. Podívejme se třeba na parazitické genetické elementy zvané retrovirové transpozony. Potomstvo hostitele, který se jich zbaví, má narušený ontogenetický vývoj. Mafiánskou hypotézou by bylo možné vysvětlit také fenomén cytoplazmatické inkompatibility (samice neinfikovaná bakteriemi má při křížení s infikovaným samcem méně potomstva, zatímco infikovaná je normálně plodná). Cytoplazmatická inkompatibilita u hmyzu a dalších bezobratlých je způsobena vnitrobuněčnými parazity (např. rodu Wolbachia). Potomstvo hostitele, který nepředá Wolbachie do další generace, má při křížení s infikovanou linií sníženou reprodukční úspěšnost. Mafiánsky se chovají i známé a jinak neškodné bakterie jako Escherichia coli. Ty se stávají silně virulentními při intenzivní léčbě antibiotiky. Hostitel, který nechtěně vyvolá „dojem“, že se brání, je penalizován zvýšenou patogenitou bakterie. Zvýšená intenzita obrany hostitele vede k zvýšené virulenci patogena. Mafiánský efekt zřejmě funguje i v případě, kdy je hostitel parazitován více parazity různě virulentními. Například parazitická houba Cryphonectria parasitica napadající kaštanovníky (Castanea) způsobuje blokádu výměny živin a vody mezi kořeny a listy. Existuje také nízkovirulentní kmen houby, http://www.cts.cuni.cz/vesmir l VESMÍR 80, září 2001

489

PODMÍNĚNÁ VIRULENCE PARAZITŮ Závody ve zbrojení mohly vést i k evoluci podmíněné virulence parazitů. Reakce hostitele a parazita nemusejí být konstantní, ale mohou záviset na reakcích druhé strany. Parazit nemusí být virulentní, dokud se hostitel nebrání. Jakmile se hostitel začne bránit, stane se parazit virulentním, a tím pokutuje hostitele, který se s parazitací „nesmířil“. Hostitel tak zbytečně investoval do obrany, která neměla žádný efekt, a zbylo mu méně zdrojů na zajištění ostatních životních funkcí. Oběť parazitace, která zvyšuje intenzitu obrany, nezvýší svou reprodukční úspěšnost, protože parazit včas změní své chování. Virulence podmíněná tím, co dělá hostitel, pomáhá parazitovi k vyššímu reprodukčnímu úspěchu a má výhodu i pro hostitele – v populaci napadené mafiánským parazitem, který má možnost volby, budou hostitelé méně zdrojů utrácet na obranné mechanizmy jednoduše proto, že se jim to nevyplatí. Pokud jsou reakce hostitele i parazita vzájemně podmíněné, ustaví se nakonec rovnovážná úroveň rezistence a virulence. Takový stav vede k vyšší reprodukční úspěšnosti hostitele i parazita než stav vysoké rezistence nebo vysoké virulence. Tolerovat tyranii se občas vyplácí.

který poskytuje rostlině odolnost vůči smrtícímu kmenu. Na případnou obranu proti nízkovirulentnímu obránci-parazitovi by ovšem kaštanovník těžce doplatil, kdyby byl parazitován virulentním kmenem. Jiným zajímavým příkladem mezidruhového „donucovacího“ chování může být vztah mezi mravenci a mšicemi. Ten bývá klasickým příkladem komenzalizmu (neškodného příživnictví): Mravenci brání mšice proti predátorům, za což mšice platí cukernatými šťávami, které vysávají z rostlin. Mravenci ovšem mšice někdy i žerou, a to právě tehdy, když se mšice pokoušejí podvádět (tj. když poskytují nekvalitní medovici nebo jí produkují málo). Každá rozumná mšice si pak zatraceně rozmyslí, jestli bude švindlovat. Podle mafiánských pravidel se mohou chovat nejen živočichové při mezidruhových interakcích, jak tomu bylo ve všech předešlých příkladech, ale i jedinci uvnitř druhu. U sociálně žijících druhů, ať už jde o sicilské mafiány či jiné primáty, mohou dominantní jedinci pokutovat nedostatečně loajální podřízené třeba ostrakizmem (vyloučením ze sociální jednotky). Tyranské choutky nedokázali bdělému zraku výzkumníků skrýt ani papoušci kea (Nestor notabilis). Ti byli testováni ve zvláštním zařízení. K získání potravy museli spolupracovat dva jedinci, ale pouze jeden mohl získanou potravu pozřít. Ukázalo se, že dominantní jedinec donutil podřízeného, aby otvíral zařízení skrývající potravu. Podřízenému jedinci to přineslo jedinou výhodu – nebyl potrestán dominantním mafiánem za nedostatečnou servilitu. Mafiánský mechanizmus obecně spočívá ve zvyšování výdajů spojených s obranou proti nějakému negativnímu vlivu. Zvyšování nákladnosti obrany je výsledkem zpětnovazebné interakce zúčastněných stran. Třeba v případě kukaček parazit „naučí“ hostitele, že je lepší obětovat část svých zdrojů parazitovi než nesmyslně utrácet mnohem víc za obranu, která k ničemu rozumnému nevede. Analogické procesy zřejmě probíhají v přírodě na mnoha úrovních, od molekulární (retrovirové transpozony) a buněčné (Wolbachia) až po chování individuálních organizmů mezi druhy a uvnitř druhu. Chapadla mafie pronikají celým světem, ve kterém žijeme – komisař Catani by se divil. o

490

VESMÍR 80, září 2001 l http://www.cts.cuni.cz/vesmir

Chudí a bohatí: triky a pověry Jak je rozděleno bohatství ve společnosti FRANTIŠEK SLANINA

K majetkové nerovnosti lze přistupovat různě. Někdo bude vysvětlovat a posvěcovat status quo vyššími ideologickými principy, jiný zapojí jiné vyšší principy do služeb ekonomické transformace, která má s dosavadní praxí skoncovat a přinést nové, spravedlivé a fungující rozdělení bohatství. Vědecký přístup se snaží příliš se do ideologické mely nezamíchat a brát bohatství a chudobu jako přírodní jevy, které chceme fenomenologicky popsat a jejich mechanizmus matematicky modelovat. Nejdříve se tedy musí provést statistická analýza bohatství ve společnosti. Jako první se do takové analýzy pustil Vilfredo Pareto, a to již roku 1897 ve svém Kurzu politické ekonomie. Šlo mu o to zjistit, jak velká část obyvatelstva patří do jednotlivých příjmových kategorií. Jelikož roční příjem každého občana lze (teoreticky) snadno zjistit podle toho, kolik zaplatí na daních, lze údaje z finančních úřadů velmi pohodlně využít k statistice příjmů. Rozdělíme si například občany České republiky do škatulek odstupňovaných po 10 000 Kč ročního příjmu, takže první škatulka bude zahrnovat všechny, jejichž příjem je od 0 do 10 000 Kč, druhá všechny s příjmem od 10 000 do 20 000, patnáctá od 140 000 do 150 000 atd. Funkce P(x), pravděpodobnostní hustota, bude říkat, jaká je relativní četnost lidí v x-té škatulce. Výsledek, ke kterému dospěl Pareto a po něm i další, byl překvapivý: Ve všech studovaných zemích a časových obdobích, po dobu dlouhých a bouřlivých sta let (jak konstatovali ve svém článku z roku 1997, tedy právě 100 let po Paretovi, Moshe Levy a Sorin Solomon) se matematická funkce popisující rozložení bohatství téměř nemění. Pravděpodobnostní hustota má tvar mocniny, čili P(x)=x –a. Tato závislost se na počest svého objevitele nazývá Paretův zákon. Proč je právě taková funkce tak zajímavá? Souvisí to se zájmem o fraktální objekty, tj. takové, které se skládají ze stále se opakujících, zmenšujících se do sebe vložených tvarů, jaké vytvářejí například obrysy listů kapradin, ledové květy na oknech a jiné. Tyto geometrické tvary se vyznačují tím, že když z nich vyřízneme detail, a pak jej zvětšíme do velikosti původního obrázku, dostaneme tentýž tvar (alespoň RNDr. František Slanina, CSc., (*1962) vystudoval Matematicko-fazikální fakultu UK. Ve Fyzikálním ústavu AV ČR se zabývá počítačovým modelováním komplexních jevů, například evoluce.

31. Grim T. 2002: Limity memetiky. 31. Cargo 2002(1, 2): 94–100. Grim T. 2002: Limity memetiky. Cargo 2002(1, 2): 94–100.

32. Grim T. 2003: Dobrá zpráva pro ornitology – špatná pro kukačky. Čím se prozrazují 32. mláďata hnízdních parazitů? Grim T. 2003: Dobrá zpráva pro ornitology špatná pro kukačky. Čím se prozrazují Vesmír 82(8):– 437–441. mláďata hnízdních parazitů? Vesmír 82(8): 437–441.

Dobrá zpráva pro ornitology – špatná pro kukačky Čím se prozrazují mláďata hnízdních parazitů? TOMÁŠ GRIM

Letmý pohled na kukaččí vejce a kukaččí mládě vzbuzuje tušení, že hostitelé přijmou parazitova vejce (která jsou jejich vlastním vejcím podobná) a svou protiparazitickou obranu zaměří až na vetřelcova mláďata (která se budou od jejich mláďat nápadně lišit). Chyba lávky – skutečnost je právě opačná. Hostitelé hnízdních parazitů – kukaček a vlhovců – běžně rozpoznávají a diskriminují1 i podobná (mimetická) vejce parazitů tím, že je vyhodí, zastaví hnízdním materiálem do hnízdní kotlinky, popřípadě parazitované hnízdo opustí a znovu zahnízdí jinde. Na druhé straně se tito hostitelé – až na naprosté výjimky – neubrání péči o parazitova mláďata, dokonce se o ně starají více než o své vlastní potomstvo.2 Tento paradox – diskriminace mimetických vajec a neschopnost diskriminovat nemimetická mláďata – zůstává jednou z největších záhad studia hnízdního parazitizmu a koevoluce vůbec. Oproti desítkám druhů hostitelů odmítajících parazitická vejce byly dosud známy jen tři hostitelsko-parazitické systémy, kde se hostitelé nenechali oblafnout parazitickým mládětem: afričtí astrildi parazitovaní vdovkami, jihoamerický vlhovec Molothrus badius parazitovaný dalšími dvěma vlhovci téhož rodu a jihoamerický drozd Turdus rufiventris parazitovaný také vlhovci. Pro všechny tři případy je společné, že parazitické mládě po vylíhnutí nevytlačí z hnízda hostitelova vejce ani jeho mláďata, a naopak je vychováváno s nimi. Zdálo se tedy, že pro diskriminaci parazitových mláďat je nezbytný srovnávací materiál – mláďata hostitele. Tato představa byla navíc podpořena jednoduchým matematickým modelem3 – pro hostitele, kde je mládě parazita vychováváno samotné (poté, co se svých spolubydlících samo zbavilo), je sice adaptivní naučit se vzhled vajec při prvním hnízdění, a pak odmítat jakákoli jiná vejce, ale totéž chování ve stadiu mláďat už adaptivní není. Záleží hostiteli na vzhledu parazitova mláděte? Pozorování osudu kukaček bronzových (Chrysococcyx basalis) a nádherných (Ch. lucidus) v hnízdech modropláštníků nádherných (Malurus cyaneus) v Austrálii však ukazuje, že předešlé představy byly zřejmě mylné.4 Oba druhy kukaček patří k „vyhazovačům“ – brzy po svém vstupu na tento svět se zbaví nepohodlné konkurence hostitelových mláďat a získají veškerou rodičovskou péči pro sebe. Ačkoli hostitelé přijímají všechna přirozeně nakladená parazi-

Necelý den staré mládě kukačky působí zcela nemohoucně, je však schopno zbavit se nepohodlných spoluobyvatel hostitelova hnízda již v tomto batolecím věku. Čerstvě vylíhlé mládě je růžové, vnitřek zobáku pak oranžový. Během několika příštích dnů kůže ztmavne a zobáková dutina nabude jasně červené barvy. Snímek © Tomáš Grim

tická vejce, opouštějí 40 % hnízd s kukačkou bronzovou a všechna hnízda s kukačkou nádhernou. Jak hostitelé kukačku rozpoznají, když jim chybí srovnávací materiál v podobě vlastního potomstva? Ukázalo se, že hostitelé-modropláštníci občas opouštějí i vlastní (experimentálně zmenšené) snůšky s jediným mládětem. To však vysvětluje chování hostitele jen částečně, samotná mláďata kukačky jsou opouštěna častěji než samotná mláďata hostitele. Pro diskriminaci parazitických mláďat se přímo nabízí jejich vzhled – mláďata modropláštníků jsou růžovo-žlutá, zatímco neopeřené potomstvo kukačky nádherné se líhne ve dvou barevných formách, růžovo-žluté a černé (mláďata kukačky bronzové jsou svým vzhledem někde uprostřed). Světlá forma kukačky nádherné (o které bychom snadno řekli, že je mimetická) je však opouštěna vždy, zatímco kukačka bronzová, která se mláďatům hostitele podobá méně, bývá občas přijata. Hostitelům je tedy vzhled mláďat lhostejný. Jak diskriminovat, když není s čím srovnávat? Dobrým kandidátem na podnět k diskriminaci zůstává hlasové žadonění mláďat, které vyvolává krmicí chování rodičů (popř. pěstounů, jestliže je hnízdo

1) Rozpoznávání je kognitivní schopnost organizmu odlišit dva podněty. Diskriminace je odlišná behaviorální reakce na dva či více podnětů. Rozpoznávání může, ale nemusí vést k diskriminaci, což je zjevné v lidském chování, je to však známo i u živočichů. Strakám, které po parazitaci kukačkou chocholatou vyhodí kukaččí vejce, se kukačka často pomstí zničením jejich snůšky. Jsou-li tyto straky parazitovány v náhradním hnízdění, akceptují parazitické vejce, přestože jsou evidentně schopny ho rozpoznat (viz Vesmír 80, 488, 2001/9). 2) Tak například rákosník obecný pečuje o kukačku obecnou, Behav. Ecol. Sociobiol. 49, 322, 2001. 3) Viz též Nature 362, 743, 1993. 4) Viz Nature 422, 157, 2003.

RNDr. Tomáš Grim, Ph.D., (*1973) vystudoval zoologii na Přírodovědecké fakultě MU v Brně. Na katedře zoologie Univerzity Palackého v Olomouci se zabývá etologií hnízdního parazitizmu. (www.zoologie.upol.cz/osoby/grim.htm) http://www.vesmir.cz l VESMÍR 82, srpen 2003

437

parazitováno). Skutečně se ukázalo, že struktura žadonění mláďat kukačky bronzové (která občas bývají akceptována) je poměrně podobná žadonění hostitelových mláďat, zatímco žadonění mláďat kukačky nádherné se od žadonění hostitelových mláďat během růstu čím dál víc odlišuje. (Tím by se dalo vysvětlit, proč hostitelé neopustí cizí mládě hned, ale 3 až 6 dnů po vylíhnutí.) Parazitické mládě se tedy „prozradí“ nikoli tím, jak vypadá, ale tím, co „říká“. Zásadní je pak zjištění, že hostitelé mohou parazita diskriminovat, přestože je v hnízdě vychováván sám. Starat se je pěkné, ale co je moc, to je moc Další zjištění schopnosti kukaččího hostitele diskriminovat parazitické mládě na sebe nenechalo dlouho čekat. Tentokrát šlo o kukačku obecnou, nejprozkoumanějšího hnízdního parazita vůbec, a rákosníka obecného, jejího nejběžnějšího hostitele. Na jihomoravských rybnících jsme zjistili,5 že hostitelé občas (asi v 16 % parazitovaných hnízd) opouštějí velká kukaččí mláďata, která vyžadují více rodičovské péče než celá průměrně velká hostitelova snůška v době, kdy opouštějí hnízdo (a kdy tedy bývá intenzita rodičovské péče největší). Na všech sledovaných hnízdech rákosníci zvyšovali frekvenci krmení kukaček až do 11. dne (v tomto věku mláďata rákosníků hnízdo opouštějí a rodiče o ně dále pečují v jeho okolí). Později už frekvence krmení významně nestoupala (u úspěšně vyvedených kukaček), někde i klesala (u opuštěných mláďat), a to přesto, že mláďata aktivně žadonila. V jednom případě dokonce hostitelé rozebírali staré hnízdo, v němž stále ještě žadonila kukačka, a začali v jeho blízkosti stavět nové. Jak toto bizarní chování vysvětlit? Je možné, že hostitelé nebyli ochotni zvyšovat své rodičovské úsilí nad míru, která je selekcí upravena pro potřeby jejich vlastních mláďat. Kukaččí mláďata tráví v hníz5) Proc. R. Soc. Lond. B (Suppl.) DOI: 10.1098/rsbl.2003.0017.

Dá se léčit roztroušená skleróza? Roztroušenou sklerózou trpí na celém světě zhruba milion lidí, dvakrát více žen než mužů. Při tomto autoimunitním onemocnění napadá tělo svou vlastní nervovou tkáň. Podle toho, která část mozku nebo míchy je zasažena, může „sebeútok“ způsobit až slepotu, ztrátu koordinace nebo celkové znehybnění. Terčem imunitní destrukce je lipoprotein myelin, který tvoří izolační pochvy kolem neuronálních výběžků. Destrukce je doprovázena zánětem, což pro imunitně privilegovanou mozkovou tkáň není příznivé. Poškozeny jsou také oligodendrocyty – pomocné buňky, které vyrábějí myelin a při vedení vzruchů neurony působí rovněž izolačně. Jiné pomocné buňky – astrocyty – mají tendenci poškozená místa „zacelit“, a tím situaci zhoršují. Vytvářejí jizvy, které přenosu elektrických impulzů brání ještě víc, až tkáň lokálně „zamrzne“. Dobrým modelem pro léčebné pokusy jsou myši trpící experimentální autoimunitní encefalomyelitidou. Zkoušely se různé strategie, které by zabránily destrukci tkáně – od snahy potlačit zánět přímo v místě až po zablokování adhezivních molekul, jež jsou pro sebepoškozující obranné buňky vstupní branou do mozku.

438

VESMÍR 82, srpen 2003 l http://www.vesmir.cz

dech rákosníka obecného více času (18 dní oproti 11 dnům hostitelových mláďat), a navíc od osmého dne po vylíhnutí potřebují více potravy než celá průměrná snůška hostitele v době opuštění hnízda. Oba tyto faktory (délka a intenzita péče) by mohly sloužit jako podněty pro reakci „starat se je pěkné, ale co je moc, to je moc“. Taková reakce dává adaptivní smysl – pro organizmus, který se množí ve více sezonách, je v každé konkrétní sezoně adaptivní investovat do množení jen část své energie a zbytek ušetřit na příště. (Současná investice pochopitelně snižuje pravděpodobnost přežití do další sezony i schopnost budoucí investice.) Regulačním mechanizmem může být prosté vyčerpání. Pro tuto hypotézu svědčí skutečnost, že hostitelé krmící velká kukaččí mláďata snižují selektivitu svého potravního chování, což je obecně známka zvýšeného úsilí a vyčerpání. Nechat se oblafnout, ale ne dlouhodobě Nesmírně zajímavé je, že podle předešlých experimentů nerozeznají rákosníci vlastní mláďata od cizích. Diskriminace parazitických mláďat tedy není založena na rozpoznávání! To může na první pohled působit nesmyslně, ale diskriminace v kauze rákosník – kukačka může fungovat prostě proto, že kukačka je v hnízdě déle než rákosníci. Opustit kukaččí mládě (místo vyhození či opuštění vejce) se může oprávněně jevit jako dosti nedostatečná obrana proti parazitizmu – hostitel smiřující se s vejcem a odmítající až mládě parazita investuje zbytečně čas i energii tam, kde se mu to rozhodně nevrátí, nanejvýš jako zvýšené riziko parazitace v další sezoně. Navíc v důsledku parazitova vyhazovacího chování přijde o veškeré vlastní potomstvo. Rozhodně je však lepší odmítnout kukaččí mládě ve dvou týdnech než o něj pečovat další čtyři týdny. Hostitel by tak mohl získat možnost náhradního zahnízdění, ale i kdyby nezahnízdil, byl by na tom celkově lépe než ten, kdo by se nechával kukačkou oblafnout dlouo hodobě, až do jejího osamostatnění.

LENKA DOUBRAVSKÁ

S novým postupem přišli S. Pluchino a jeho kolegové, kteří navrhují způsob, jímž by se poškozená tkáň dala opravit. Pluchinova skupina izolovala myší nervové prekurzory, které nejsou zcela diferencované, a mohou tedy umožnit vznik více typům nervových buněk. Skutečný původ těchto prekurzorů nelze přesně určit, jisté však je, že splňují kritéria nervových kmenových buněk. Shodně s obrannými leukocyty nesou tyto prekurzory na svém povrchu adhezivní molekulu a4 integrin coby vstupenku do mozku. Stačí je tedy injikovat do krve nebo míšní tekutiny, a do místa zánětu už doputují samy. Jakmile tam jsou, slouží jako zdroj nových neuronů a oligodendrocytů, které začnou tvořit myelin. Zdá se, že tyto prekurzory mohou dokonce vyhladit jizvy vytvořené astrocyty. Tkáň se opět probouzí, symptomy mizejí. Autoimunitním útokům tento přístup nezabrání, ale zmírní důsledky choroby. Možná by mohl být vodítkem v léčbě dalších nervových onemocnění, kde nervové buňky degenerují z jiných příčin. Nebylo by marné políčit prekurzory třeba na Parkinsonovu nebo Alzheimerovu chorobu. (Nature ¨ 422, 671, 2003)

33. Grim T. 2006: Kde jsou ochranářské prority? Medializace kontra ochrana přírody. 33. Vesmír 85(3): 140–147. Grim T. 2006: Kde jsou ochranářské prority? Medializace kontra ochrana přírody. Vesmír 85(3): 140–147.

Kde jsou ochranářské priority?

TOMÁŠ GRIM

Medializace kontra ochrana přírody

BIODIVERZITA

S

tvrzením, že globální ekosystém je v důsledku lidské činnosti v krizi, můžeme souhlasit i nesouhlasit (Vesmír 80, 573, 2001/10). Jistě se však shodneme na tom, že s přírodou se dějí věci nepěkné a že si příroda ochranu zaslouží, neboť je pěkná a občas i užitečná. Ideální by samozřejmě bylo chránit přírodní prostředí celé. Finance i čas jsou bohužel omezené, můžeme tedy chránit jen některé oblasti a druhy. Ale které? Ochranářských nástrojů je řada – od národních parků a biosférických rezervací přes síť mokřadů zaštítěných Ramsarskou úmluvou až po významné ptačí oblasti nebo dohodu o obchodování s ohroženými druhy (CITES). Pro určení priorit územní ochrany přírody v globálním měřítku je však nejčastěji používán koncept hotspots – „horkých míst biodiverzity“ (Vesmír 84, 30, 2005/1). Právě biodiverzita (dnes většinou chápaná jako synonymum druhové bohatosti) je jednoznačně nejfrekventovanějším zaklínadlem ochrany přírody. V souvislosti s ní se nejčastěji mluví o kácení amazonských deštných pralesů, oněch „zelených plic planety“ a „kolébky a pokladnice biodiverzity“. Média v souvislosti s ochranou přírody vytvářejí dojem, že je třeba chránit především biodiverzitu, deštné lesy („ničené kácením a fragmentací rychlostí x fotbalových hřišť za minutu“) a v rámci deštných lesů pak Amazonii, která dosud hostí „největší a druhově nejbohatší deštné lesy na světě“. Jsou ale tyto všeobecné „ochranářské pravdy“ v souladu s dostupnými daty? Odráží mediální obraz ochrany přírody skutečné ochranářské priority? Jistěže je třeba chránit lesy a kácení může škodit, ale jsou tohle opravdu nejzávažnější problémy? Horká místa: přihořívá, ale nehoří

Jako horká místa (www.biodiversityhotspots. org) jsou označovány velké geografické oblasti vybrané podle jednoho z následujících kritérií: 1) vysoká druhová bohatost, 2) zvýšený výskyt vzácných endemických druhů s malými areály, 3) největší počet ohrožených druhů bez ohledu na celkovou druhovou bohatost a endemizmus, 4) různé kombinace jmenovaných faktorů. Preferovaná kritéria jsou endemizmus a ohroženost – Norman Myers, který hotspoty vymyslel, je ostatně definoval jako oblasti, které obsahují nejméně 0,5 % (čili 1500) endemických druhů cévnatých rostlin z celosvětového počtu 300 000 a které ztratily přes 70 % své původní vegetace. Informace o rozšíření a biologii většiny organizmů jsou však velmi kusé. Není divu, když tři čtvrtiny taxonomů (lidí, jejichž pracovní náplní je popisovat biodiverzitu) pocházejí z mírného pásu. A nejen pocházejí, ale většinou z něj ani nevycházejí do tropů, kde žije asi 80 % všech organizmů. Většinu našich biologických znalostí jsme získali z menšiny druhů, které navíc velmi špatně reprezentují světovou biodiverzitu. Kromě to-

1 1 1. Podvečerní zátiší z oblasti Gran Sabana v jihovýchodní Venezuele. Všechny snímky na s. 140–147 © Tomáš Grim.

ho je vzájemné zastoupení taxonů a taxonomů velmi nerovnoměrné (obr. 2). S ohledem na naše chabé znalosti je koncepce hotspotů založena na dvou zásadních předpokladech: a) diverzita dobře prostudovaných indikátorových skupin dobře předpovídá diverzitu méně známých skupin, b) diverzita a endemizmus pozitivně korelují. Diverzita versus diverzita

Jak nás poučují učebnice ekologie a biogeografie, diverzita různých taxonů obecně pozitivně koreluje. Asi nejznámějším příkladem tohoto jevu je stoupající diverzita většiny taxonů směrem k rovníku (Vesmír 83, 508, 2004/9). Učebnicové představy poněkud nabouralo zjištění J. Prendergasta a kol., že překryv výskytu diverzity různých skupin organizmů (ptáků, motýlů, vážek, játrovek, vodních cévnatých rostlin) ve Velké Británii je oproti předpokladům minimální. To lze ovšem odbýt tím, že ostrovní království leží na okraji kontinentu, je atypické a spíš půjde o výjimku. Podobných prací však postupně přibývá a upozornily i na další problémy. Například diverzita primátů na Borneu hezky předpovídá biodiverzitní hotspoty na tomto ostrově – ovšem jen do té doby, než použijeme alternativní systematickou klasifikaci primátů; pak zmíněný vztah zmizí. Diverzita versus endemizmus a ohroženost

Jak je na tom druhý předpoklad, na němž koncept hotspotů stojí a s nímž padá? Zmiňovaná průkopnická práce J. Prendergasta se sice endemizmem nezabývala, ale ukázala, že korelace mezi diverzitou a ohrožeností (která často s endemizmem souvisí) různých skupin organizmů ve Velké Británii je slabá. I další práce opakovaně poukazovaly na chabou souvislost mezi diverzitou a endemizmem – nakonec i na tak obrovské ploše, jako je subsaharská Afrika. Argument, že tyto práce popisují výjimky, a koncept hotspotů nadále znamená, že

RNDr. Tomáš Grim, Ph.D., (*1973) vystudoval zoologii na Přírodovědecké fakultě MU v Brně. Na katedře zoologie Univerzity Palackého v Olomouci se zabývá etologií hnízdního parazitizmu. http: //www.zoologie.upol.cz/ osoby/grim.htm

2. Zastoupení taxonů a taxonomů. Zatímco nadpoloviční většinu druhů všech organizmů tvoří asi hmyz, jen 16 % taxonomů jsou entomologové. Poměry v grafu se však mohou dramaticky změnit, pokud se ukáže správným odhad, že nejméně čtyři pětiny živočichů jsou paraziti. O nich jsou naše znalosti podobně bídné jako o tropických tvorech. A to mluvíme jen o suchozemských organizmech – v 90 % mořského prostředí zatím nikdo jeho obyvatele nehledal, a tak o mořské biodiverzitě nevíme skoro nic.

�����

�������������������� ����������� ���������������������� ������������� ����������������������

���������

���������������������

���������� ����������������������� �����������������

����������

����������������� ��������������������������������������� �������������� ������������

������������� ������������

�������������

���������������������������

http://www.vesmir.cz | Vesmír 85, březen 2006

141

dou často vázány na vzácná prostředí, tedy nejčastěji na extrémní či okrajové ekologické podmínky (např. ostrovy s typicky vysokým endemizmem a často nízkou diverzitou). Naopak běžné druhy jsou převážně tam, kde jsou podmínky stabilní a umožňují dlouhodobou existenci početných populací. Jinak řečeno, biodiverzita a endemizmus jsou dvě odlišné věci. Vysoká diverzita je i v globálním měřítku způsobena překryvem areálů běžných druhů – vzácné, a tedy ochranářsky významné druhy žijí jinde. Jak se horká místa stávají ještě více horkými

3. Nahoře: Mangrove sice nebyly zahrnuty do výzkumu představeného na obr. 9, ale patří mezi nejohroženější biomy světa. Najdeme je i na ostrově Superagui v jihovýchodní Brazílii, odkud byl r. 1990 popsán lvíček Leontopithecus caissara – jeden z posledních skutečně objevených druhů primátů na světě. 4. Dole: Palmová savana s dominantní palmou moriche (Mauritia flexuosa) v Národním parku Chaco v severní Argentině. Na moriche je vázáno několik druhů ptáků, z nichž např. hrnčiřík Berlepschia rickeri na ní sbírá potravu, spí, páří se i hnízdí a nebyl nikdy pozorován mimo tuto rostlinu.

142

s diverzitou chráníme zároveň endemizmus a naopak, byl definitivně odeslán na věčnost až v pozdním létě roku 2005. David Orme a kol. ukázali na zatím nejobsáhlejší databázi rozšíření jakéhokoli taxonu na světě (tedy ptáků), že překryv hotspotů diverzity, endemizmu a ohrožených druhů je téměř nulový! Jako hotspot definovali 2,5 % nejbohatších buněk (1° zeměpisné šířky × 1° zeměpisné délky) a v nich hodnotili každé ze tří kritérií. Pro všechna je společných jen 2,5 % z těchto nejbohatších čtverců. Diverzita a endemizmus korelují pozitivně jen ve velkých měřítkách, což není ochranářsky důležité. Nikdo nemůže za chráněnou oblast vyhlásit celé tropy, a dokonce ani plošně mnohem omezenější hotspoty – tropické Andy nebo Nová Guinea – se rezervacemi nikdy nestanou. V jemném lokálním měřítku (plochy od stovek m2 po stovky km2), tedy právě v tom měřítku, které je pro ochranu přírody jediné významné, se oblasti výskytu diverzity a endemizmu neshodují. Se zmenšováním měřítka se nám tedy globální pozitivní korelace mezi diverzitou a endemizmem postupně vytrácí, až se nakonec stává v lokálním měřítku negativní (což je, dotaženo do důsledku, docela triviální: v mapovacím čtverci o hraně 1 m už kromě slona nemůže stát nic dalšího, takže korelace mezi výskytem slonů a čehokoli dalšího, snad kromě sloních tasemnic, musí být stoprocentně negativní). Otázkou zůstává, proč je překryv výskytu biodiverzity a endemizmu tak malý. Biologická intuice napovídá, že vzácné druhy bu-

Vesmír 85, březen 2006 | http://www.vesmir.cz

Na druhé straně můžeme očekávat, že diverzita různých skupin a diverzita s endemizmem nám časem začnou korelovat alespoň uvnitř hotspotů. Důvodem je tzv. taxonomická inflace. Módou posledních let se totiž stalo povyšování poddruhů na druhy – takže nám diverzita utěšeně roste. Potíž je v tom, že taxonomická inflace není jednotná – rychleji nám přibývá velkých zvířat, druhů z charizmatických skupin a zvířátek, která sdílejí svůj areál s taxonomy. Např. počet primátů se za posledních dvacet let zdvojnásobil téměř na čtyři stovky druhů, přičemž skutečné objevy spočítáte na prstech jedné ruky. Všechny ostatní „nové“ druhy jsou pouhé bývalé poddruhy. A není divu, že nejvíce se taxonomická inflace projevuje právě v hotspotech, které se svým vyhlášením automaticky stávají magnetem pro taxonomy. Takže nás případná vysoká diverzita v horkých místech nemusí překvapovat – vyrobili jsme si ji částečně sami. Žádný ze dvou klíčových předpokladů konceptu hotspotů tedy neplatí v prostorových měřítkách ochranářské praxe. Jinými slovy volba kritéria, podle něhož stanovíme hotspot (diverzita, endemizmus, ohrožení či nějaká jejich kombinace), významně ovlivňuje ochranářské priority. To je ale v příkrém rozporu s tradičními ochranářskými představami. Jak takové dilema řešit? Kvantita není kvalita – na diverzitě čili délce druhového seznamu nezáleží. Nejvyšší prioritu by měla mít společenstva složená převážně z druhů už ohrožených či snadno ohrozitelných – tedy těch endemických. Ty jsou ve frontě k odchodu na věčnost na řadě první. Ochranářská praxe s endemizmem počítá, a to nejen v hotspotech, ale třeba i v konceptu endemických ptačích oblastí (EBAs). Značné pochybnosti o reálných účincích tohoto povědomí ovšem vzbuzuje zjištění, že téměř všechny národní parky v Andách byly pečlivě naplánovány a zřízeny v místech mimo oblasti s vysokou koncentrací výskytu kriticky ohrožených endemitů, v USA se třetina ohrožených druhů vyskytuje výhradně mimo chráněná území a v Africe zase chráněná území nezahrnují ani část areálu nadpoloviční většiny ohrožených ptačích druhů. Druhový versus biotopový přístup

Jednu ze základních kontroverzí ochrany přírody, tedy zda chránit druhy, nebo ekosystémy, nakonec vyhrály ekosystémy. Oba koncepty však od sebe úplně oddělit nelze: druhy

5. Nahoře: Severské jehličnaté lesy patří k nejzachovalejším biomům. Národní park Waterton Lakes, Alberta, Kanada.

6. Dole: Plošně velmi omezeným biomem jsou deštné lesy mírného pásu. Vancouver Island, Britská Kolumbie, Kanada.

7. Nahoře: Atlantské deštné lesy ve východní Brazílii. Národní park Serra da Bocaina.

8. Dole: Vysokostébelná prérie u Head-Smashed-In Buffalo Jump v kanadské provincii Alberta je jedním z posledních zbytků travinných společenstev mírného pásu, nejohroženějšího biomu vůbec.

��������

��������

se chrání ochranou prostředí a každý ekosystém tvoří nějaké druhy. I proto jsou poněkud šokující zjištění vyplývající z nové analýzy ohroženosti biomů. Jako měřítko ohroženosti slouží poměr mezi procentem zničené plochy daného biomu a procentem jeho chráněné plochy (obr. 9). Navzdory tradiční ochranářské moudrosti, že nejvíce ochrany si zaslouží tropické oblasti, si první dvě místa žebříčku vybojovaly biomy mírného pásu: travinná společenstva (stepi, prérie, pampy) a mediteránní společenstva (např. středomořská macchie). Na dalším místě se pak objevují tropické lesy, ale ty suché (o nichž oproti oslavovaným deštným lesům většina lidí nikdy ani neslyšela). A zjištění, že nejohroženější biomy jsou chráněny stejně nebo ještě méně než tundra a tajga (které jsou zachovány v původní rozloze téměř beze změny a nejsou nijak ohrožené), pak nelze označit jinak než jako naprosté selhání ochrany přírody v globálním měřítku. Jak vůbec mohlo k tak katastrofální situaci dojít? Svůj podíl má na tom nepochybně fakt, že diverzita pozitivně koreluje s hustotou lidského osídlení – kde se dobře bydlí všemožným živáčkům, tam se dobře bydlí i nám. Mnohdy tak na rezervace zbudou plochy, které nikomu nepatří nebo nejsou osídleny (ani lidmi, ani ochranářsky významnými druhy). Vinu však možná nese i způsob, jakým je ochrana přírody propagována.

����

�������������������� �������������������������

���

����

�������������������������

���

����

����������������������� ����������

���

����

������������������������ ������������

���

����

����������������������� ���������������

���

����

����������������������� ����������

����

����

��

��

��

�������������������������������� ���� �����������������������

����

������������������

����

���

����������������������

���

����

���������������� �����������������������

����

����

����������������������������

����

���

���������������������

���

���

������

����

�

�

����������������

na úrovni biomů největší ptačí endemizmus v suchých lesích (90 %) a savanách (80 %). Tato prostředí jsou nesrovnatelně ohroženější O čem se v souvislosti s ochranou přírody než deštné lesy (jistěže místy najdeme deštné nejvíce mluví? Podívejme se třeba na inter- lesy, například v západním Ekvádoru, které netové stránky Greenpeace, celosvětově nej- jsou poškozeny katastrofálně, ale to na uvedeném obecném závěru nemění nic). Diverzita vlivnější ochranářské organizace. Rozšířený vyhledávač mi vyplivl (22. 9. 2005) 9120 odka- je zde k jejich smůle nižší než v deštných lesích. Jedním z neblahých důsledků přístupu zů na „biodiversity“ oproti 109 odkazům na „endemic“ nebo „endemism“. Následující vý- „druhová pestrost je svatá“ je zaměření ochrasledky byly podobně výmluvné: heslo „forest“ nářské propagandy na amazonské deštné lesy 30 400 odkazů, „savannah“ ubohých 106. (z nichž je přes osm desetin zachováno) a značné přehlížení těch nejohroženějších jihoameDalší čísla ze stránky Greenpeace dokreslují obrázek: na heslo „Amazon“ našel vyhledá- rických biotopů – tedy například savan, vysokohorských polylepisových lesů, suchých vač 12 600 odkazů, na „Andes“ ubohých 102 odkazů (Andy mají vyšší biodiverzitu, ende- sezonních lesů či atlantských deštných lesů mizmus i ohroženost než Amazonie a je to na (z nichž je devět desetin zničeno). Není divu, že v základní publikaci o ohrocelém širém světě jediný hotspot v práci D. Ormeho a kol., kde se všechna tři kritéria sho- žení jihoamerických ptáků D. F. Stotze a kol. Neotropical birds. Ecology and conservation z r. 1996 dují). Pak jsem se ovšem konečně přestal divit to- se amazonské deštné lesy objevují v žebříčku mu, že se skoro na všech ochranářských le- míst, která by měla být ochranářskými prioritami v budoucnu, až na jednom z posledních tácích a materiálech týkajících se tropických oblastí, které se mi dostaly do rukou, stan- míst. V separátních analýzách se Amazonie dardně skví fotografie vypáleného či vyká- řadí v míře endemizmu velmi nízko a v kateceného kusu amazonského lesa. Polonahý gorii ohrožených druhů se už neobjevuje vůchlapík s motorovou pilou zajíždějící do ná- bec. Situace je stejně pitoreskní, jako kdyby na jednotce rychlé lékařské pomoci poskytli běhového kořene pralesního velikána zůstane ochranářskou ikonou jistě ještě dlouho. Rol- první pomoc pacientovi s naraženou kostrčí ník s pluhem, ryjící brázdu v panenské stepi, a silně krvácejícího nešťastníka poslali k obvedle něj zřejmě působí příliš kultivovaně až vodnímu lékaři. idylicky. Bohužel. Proč bohužel? Podívejme se na ptactvo – je- Kácení a fragmentace: dinou skupinu zvířat, o níž máme dostateč- hlavní problémy deštných lesů? ně detailní informace na to, abychom se na Na nejobecnější rovině tkví problém pochopitelně v obecně sdílené představě „příroda závěry jejich analýz mohli alespoň trošku je tehdy, když je les“. Ta má svůj podíl na tom, spolehnout. V Jižní Americe např. najdeme Svátost biodiverzity a ochranářská propaganda

��

��

��

���������������� 9. Žebříček ohroženosti hlavních zemských biomů. Bohužel je důvod předpokládat, že zrychlujícím tempem jsou ničeny právě ty biomy, kterých zbývá nejméně (např. travinná společenstva mírného pásu, která mají nejvyšší hustotu zalidnění).

http://www.vesmir.cz | Vesmír 85, březen 2006

145

10. Skrtiče, z nichž nejznámější je rod Ficus (který není na snímku), často vytvářejí esteticky zajímavé struktury. Národní park Tikal, Guatemala (viz Živu 50, 238, 2002/5).

146

že nejméně se v rámci ochrany přírody mluví o těch prostředích, která ji potřebují nejvíce, tedy o ne-lesních biomech. Handikepem těchto nedostatečně „úžasných a životodárných prostředí“ je i to, že v nich nenajdeme žádné okouzlující druhy, které zaujmou každého laika (těžko si představit hnědošedé hrnčiříky či matné tyrany jako vlajkové druhy). Proč jsou ale deštné lesy mnohem méně zasaženy lidskou činností a nejsou tedy – navzdory všem tvrzením ochranářů – v žádném případě ochranářskou prioritou číslo jedna? Vykácet a vypálit les a udělat z něj pastviny není žádná procházka růžovou zahradou. Na rozdíl od savany, která už jaksi pastvinou je. Takže není divu, že lesy se za účelem pastevního hospodaření likvidují méně než savany. Taktéž sucholes lze přeměnit na obyvatelné území snadněji než deštný les, což vysvětluje, proč jsou vlhké lesy ohroženy méně než ty suché. Hlavním globálním faktorem ohrožujícím deštné lesy nejsou japonské, americké či jiné

Vesmír 85, březen 2006 | http://www.vesmir.cz

zahraniční těžařské společnosti, jak ochranáři občas populisticky vytrubují do světa. Místo těchto snadno napadnutelných obětních beránků způsobují likvidaci deštných lesů v první řadě domorodci, kteří lesy vypalují na pastviny či pole, a dřevo vůbec netěží. Kromě toho hraje obrovskou roli dřevo na podpal: například v Demokratické republice Kongo (bývalém Zairu), která má na svém území největší část afrických deštných lesů, je objem dřeva vysbíraného na otop dvěstěkrát větší než objem dřeva vytěženého komerčně. Komerční těžba dřeva má celosvětově na svědomí pouhou pětinu poškozených tropických lesů. Záměrně nepíšu „vykácených“ – většina deštných lesů je kácena selektivně, tzn. z hektaru lesa jsou odvezeny jen ty největší stromy, zatímco ty zbylé, které jsou buď neekonomicky malé, nebo komerčně nevyužitelné, jsou ponechány svému osudu. Většina laiků by podle detailní fotografie z ptačí perspektivy ani nepoznala, že les na ní je uváděn v kolonce „vytěžený“, a to nejen v Latinské Americe, kde se typicky odebírají 1–2 stromy z hektaru, ale i v jihovýchodní Asii, která má ke své smůle poměrně homogenní druhové složení a stromy lze snadno rozčlenit do uživatelských tříd. Není divu, že nejohroženější deštné lesy na světě jsou v jihovýchodní Asii a ne v Amazonii. Kácení ostatně často nepozná ani les sám: ke každému odbornému článku, který zjistil negativní vliv výběrového kácení, lze přiložit jiný odborný článek, který ukazuje, že výběrová těžba buď jen krátkodobě mění strukturu společenstva (běžné druhy se stávají ještě běžnějšími a vzácné ještě vzácnějšími), nebo nemá žádný zjistitelný vliv například na netopýry, motýly, žáby, mravence, ptáky ani další faunu lesa v Americe, Africe či Asii. Práce sledující kácené lesy v delším časovém horizontu také většinou ukazují rychlý návrat do původního stavu (během několika málo let až desetiletí). To by nemělo nikoho překvapovat – pád starých stromů a vytváření světlin je nezbytnou součástí fungování deštného lesa a selektivní těžba tuto dynamiku v podstatě kopíruje. Nejen proto je zavádějící tvrdit, že fragmentace lesů je jedním z hlavních faktorů jejich ohrožení. Např. atlantské deštné lesy byly vykáceny z 95 % a není znám jediný druh ptáka, který by tam vyhynul. Je třeba si uvědomit, že atlantské druhy ptáků jednoduše nemají potřebu v důsledku fragmentace „svých“ lesů vyhynout, poněvadž jejich lesy byly vždy fragmentovány cyklony a ptačí druhy tu vždy existovaly v malých izolovaných populacích. Fragmentace rovněž neznamená úplnou izolaci. Většina ploch bývalých atlantských deštných lesů jsou dnes kávovníkové plantáže. Kávovníku se dobře daří jen při zastínění, a tak plantáže připomínají spíše ochuzený dvouetážový les: nahoře stínící stromy, dole keřové patro kávovníků. Přes takové prostředí se lesní ptáci většinou neštítí šířit do jiných lesních fragmentů. Plantáže kakaovníků ve Venezuele jsou navíc významnou tahovou zastávkou ptáků, a některé vzácné druhy jim dokonce dávají přednost.

Argumentovat atlantskými deštnými lesy v obecné rovině by však bylo zavádějící. Lesů pravidelně a přirozeně fragmentovaných cyklony je samozřejmě menšina. Jak je to s těmi ostatními? Např. v Singapuru bylo za poslední dvě století vykáceno 99,7 % primárních deštných lesů. Diverzita ale klesla jen o 28 % (pokud počítáme s nedoloženými extinkcemi, vyhynulo lokálně odhadem asi 50 % druhů). Jako biolog bych tipoval, že nemohlo přežít téměř nic – vždyť pozůstalá plocha primárního lesa měří 1,2 km2! Negativní vliv fragmentace na biodiverzitu je oproti dřívějším očekáváním minimální i na jiných místech. Navíc fragmentace ve větším měřítku biodiverzitu dokonce zvyšuje. Řada druhů je totiž závislá na raných sukcesních stadiích a v primárním lese žít nemůže. Empirická data versus bizarní závěry

Neochota zapomenout teoretické učebnicové pravdy založené na dávno překonaných představách o biodiverzitě a stabilitě, které mají, mírně řečeno, velmi omezenou empirickou podporu, se netýká jen autorů populárněvědeckých knížek, ale i vědců samotných (Vesmír 84, 37, 2005/1). Jako perličku na závěr povídání o fragmentaci cituji výmluvné tvrzení z knihy Biodiversity: An ecological perspective: „Tropické odlesňování a následná fragmentace prostředí jsou hlavními faktory ohrožujícími světovou terestrickou biodiverzitu, ačkoli nezvratné vědecké důkazy pro lokální vyhynutí způsobené fragmentací jsou překvapivě vzácné.“ Těžko říct, jak mohlo tak schizofrenní tvrzení projít recenzním řízením. Buď děláme vědu, a potom musíme svoje oblíbené hypotézy odmítnout, nejsou-li v souladu z důkazy, nebo děláme ideologickou kampaň a mlžíme, pak nás ale žádné důkazy nemají zajímat. Skutečné ochranářské priority

Kácení (deštných) lesů tedy není černobílý problém se zlovolnými těžaři na straně jedné a desítkami vymírajících druhů denně na straně druhé, jak by vyplývalo z televizních a jiných zpráv. I v dalších ohledech je rozdíl mezi praxí a mediálně prezentovaným obrazem priorit ochrany přírody propastný. Na závěr lze těžko říct něco obecného. Diverzita různých skupin někdy koreluje pozitivně, jindy negativně, některé endemity a ohrožené druhy najdeme v místech druhově chudých („coldspots“), jiné na lokalitách biodiverzitou oplývajících, indikátorové druhy často indikují jen samy sebe, kácení a fragmentace lesů některé druhy poškozuje, jiným vyhovuje a dalším je to úplně jedno. Každá oblast a druh jsou – bohužel – unikátní a napasovat na ně nějaká obecná pravidla lze těžko. To málo, co snad lze obecně k prioritám ochrany přírody říci, lze shrnout do „trojatera“ přikázání: 1) nenecháš se poblouznit vějičkou biodiverzity, ale k ochraně vybereš míst, kde druhy endemické a ohrožené dosud přebývají, 2) hýčkati budeš nikoli hvozdy hluboké, ale savany, stepi a jiná místa nelesní a 3) na prvním místě jednání tvého v místech za-

11. Nahoře: Mírné vlhké zimy a horká suchá léta typická pro oblasti kolem 40 stupňů severní a jižní šířky vedou ke vzniku typické vegetace – tvrdolistých křovinatých až lesnatých biomů mediteránního typu. Patří mezi ně nejen kalifornský chaparral, chilský matorral, jihoafrický fynbos a australské mallee, ale především středomořská macchie (na snímku ze severního Peloponésu, Řecko). Ta se vyznačuje fenomenálním endemizmem rostlin – 52 % z 22 500 místních druhů se nevyskytuje nikde jinde na světě. 12. Dole: Uprostřed pulzující aglomerace Singapuru se tyčí do závratné (z místního pohledu) výšky 164 m kopec Bukit Timah se stejnojmennou rezervací. Každý návštěvník je před vstupem do parku upozorněn informační tabulí, že jde o zaručeně primární (tj. nikdy nekácený) les. Na stejné tabuli se ovšem o několik řádků níže doví poněkud zarážející skutečnost, že zde před 60 lety proběhla lítá „bitva o Bukit Timah“, která staví panenskost tohoto území do trochu jiného světla (podstatně férovější by tedy bylo na tabuli uvést „Primární prales – již od roku 1942“). Díky své přístupnosti je Bukit Timah dobře znám botanikům, neboť z této lokality byla popsána řada nových druhů rostlin typických pro Malajský poloostrov. Pozorný návštěvník může při „velké troše“ štěstí zahlédnout i letuchu či luskouna, avšak ani ten nejnepozornější návštěvník nepřehlédne makaky žebrající o potravu na blízkém parkovišti.

lesněných buďtež lesy listnaté v krajích mírných, lesy suché v krajích tropických a ikonu pohanských ochranářů – modloslužebné lesy amazonské – zanech na jednom z posledních míst v seznamu priorit svých. Ö

+

více na www.vesmir.cz

WWW

http://www.vesmir.cz | Vesmír 85, březen 2006

147

34. Grim T. & Jirků, M. 2003: Jak fungují „pekelné konve“. Adaptace masožravých láčkovek. 34. Vesmír 82(10): 559–563. Grim T. & Jirků, M. 2003: Jak fungují „pekelné konve“. Adaptace masožravých láčkovek. Vesmír 82(10): 559–563.

Ý V N O G N SI E D

přírodovědecký časopis

ročník 82 (133) • první číslo vyšlo roku 1871 cena 79 Kč (pro předplatitele 56 Kč)

2003

PEKELNÉ KONVE

BOBŘI V ČECHÁCH LABE MEZI STŘEKOVEM A HŘENSKEM ČESKÉ PIVO TADEÁŠ HAENKE 9 770042 454062

10

BOTANIKA

Sopka Gunung Singgalang (2877 m) převyšuje okolní vrcholy pohoří Barisan. Právě poblíž vrcholu Gunung Singgalang sbíral Odoardo Beccari v 19. století typové exempláře Nepenthes singalana, jedné z prvních horských láčkovek Sumatry objevených Evropany. Snímek © Tomáš Grim

Jak fungují

TOMÁŠ GRIM MILOSLAV JIRKŮ

„pekelné konve“

Adaptace masožravých láčkovek

Ačkoli můžete jedny z nejzajímavějších masožravých rostlin – láčkovky (r. Nepenthes) – vidět ve sklenících téměř každé botanické zahrady, vědci do jejich života pronikají až v posledních letech. Celkem 86 druhů1 se vyskytuje od Madagaskaru po Novou Kaledonii a od severovýchodní Indie a jižní Číny po severní Austrálii (centrem výskytu jsou Borneo a Sumatra). Většina oblastí tohoto rozsáhlého areálu zůstává prozkoumána jen chabě (např. Sumatra), což je škoda – život „pekelných konví“ je fascinující. Láčkovky jsou dvoudomé liány dosahující délky až kolem 20 m. Kořenují v zemi, jen několik málo druhů, převážně ve vyšších nadmořských výškách, je epifytických. Rostou na stanovištích s dostatkem slunečního svitu a vysokou vlhkostí vzduchu. Samotné lapací orgány – láčky – vznikly během evoluce přeměnou listové čepele, kdežto dnešní „čepel“ je druhotně modifikovaný řapík. Svinutí a srůst listu do kornoutu je nejspíš dost jednoduše řízený děj, protože se u běžných rostlin s plochými listy občas přihodí jako „chyba v programu“, známe to třeba u našich lip. Svou kořist láká láčkovka jednak svým vzhledem (některé láčky jsou pestře zbarvené, s různou kresbou), jednak chemicky. Vábnou vůni vydávají především sekrety pachových

a nektarových žlázek, které se vyskytují na víčku a kolem obústí, v malém množství i na listech a stoncích. Druhotně se jako atranktant uplatňuje i pach kořisti rozkládající se v konvici. Předpokládá se, že také samotná aktivita hmyzu vábeného láčkami může přilákat další kořist, která by jinak neměla důvod láčky navštěvovat, například pavouky, kudlanky, zákeřnice (Reduviidae), nebo dokonce žáby, ještěry a drobné savce (viz rámeček na této straně). Nachytaný hmyz (hlavně mravenci), pavouci a další kořist poskytují rostlině živiny, kterých je v neúživném tropickém deštném lese v důsledku rychlého vymývání a příjmu ostatními rostlinami nedostatek. Dolní baňkovité láčky ležící na substrátu jsou rozpoznatelné podle nápadných zubatých „křídel“. Jejich funkce bývala tradičně vysvětlována tím, že dovedou hmyz k příčně žebrovanému

Tomáš Grim, Ph.D., viz Vesmír 82, 437, 2003/8. Mgr. Miloslav Jirků (1979) vystudoval Přírodovědeckou fakultu UP v Olomouci. V Ústavu parazitologie na Veterinární a farmaceutické univerzitě v Brně se v rámci postgraduálního studia zabývá zejména biologií kokcidií.

1) Současný stav k dnešnímu dni – nové taxony jsou objevovány bezmála každý rok.

Velikostní rekord kořisti je pták zvíci holuba, který byl kdysi nalezen v jedné láčce (a láčka právě odumírala, neboť se přejedla). Články o láčkovkách jeden od druhého opisují, že to byl přímo holub. To by snad, aspoň na úrovni čeledi nebo řádu měkkozobých, bylo biogeograficky možné, ale čert těm cestovatelům věř! Jak to asi reálně v terénu vypadá? Dokážeme si představit, že příslušný objevitel nahne konvici, z níž stoupá puch obzvlášť ohavný, a z ní vyhrkne něco úplně příšerného, slizký žvanec peří. Opatrně nakopnut špičkou boty: „No… holub. Byl to holub, dejme tomu.“ Jiří Sádlo

http://www.vesmir.cz | Vesmír 82, říjen 2003

559

Snímky na této dvoustraně © Tomáš Grim

Horský mlžný les ve výšce přibližně 2500 m nad mořem na západním úbočí hory Gunung Talang, na nějž je vázána většina horských láčkovek. Jeho rozšíření je závislé na horizontálních srážkách z oblaků, které ve výškách 2000 až 3000 m po celý rok zahalují hory západní Sumatry. Výskyt terestrických láčkovek (většina druhů) je zpravidla omezen na malé osluněné plošky vzniklé sesuvem půdy či pádem stromu.

hladkému obústí. Tradovalo se také, že křídla horních láček jsou podstatně redukována kvůli tomu, že k horním konvicím hmyz přilétá. Výsledky experimentu však ukázaly, že odstranění „naváděcích“ křídel nemá na zachycení kořisti významný vliv. Skutečný důvod, proč se křídla vyvinula, zůstává záhadou (viz rámeček na této straně). Ať už je funkce křídel jakákoli, osud hmyzích obětí v horních láčkách je stejný jako v láčkách dolních. Po hladkém obústí kořist uklouzne a spadne do tekutiny vyplňující až polovinu konvice. Pokud se hmyz rovnou neutopí a začne lézt po stěně konvice nahoru, obalí se mu přísavky na chodidle jemnými voskovými šupinkami, což mu – spolu s převislým okrajem obústí – zabrání v dalším pokusu o útěk. Láčkovka pak stojí před problémem jak využít živiny obsažené v kořisti. Funkci če-

JAK TO S TĚMI KŘÍDLY LÁČKOVEK VLASTNĚ JE? Našince nad tím napadají čtyři věci: l Že něco „zůstává záhadou“? Inu, to je běžné a podezřelé klišé. Takhle si většinou dramatizujeme realitu, abychom pokus, který se nám nepovedl, aspoň nějak publikovali. Zůstává to záhadou, tj. zajisté se statisíce vědců třesou na objasnění zásadního problému, proč že má láčkovka křídla. l Stejnou záhadou je ovšem tvar listů rostlin vůbec. Leccos se o tom už ví (o tom až jindy), ale přesto pohoříme, budeme-li chtít funkčně vysvětlit třeba evidentní tvarový rozdíl v listech našich javorů a dubů. Vlastně právě láčkovky a podobné masožravky jsou jedny z mála rostlin, u nichž dokážeme tvar listu aspoň nějak funkčně vysvětlit. l Za testovatelnou úvahu by možná stála stabilizační funkce křídel. Visuté láčky se mohou aktivními pohyby úponků narovnat, jsou-li převrženy, kdežto láčky ležící na zemi už o tuto možnost přicházejí. l Podobná křídla mají na lodyhách např. četné naše bodláky a pcháče, o ostropsu nemluvě (rody Carduus, Cirsium, Onopordon). Různě dlouhá, široká, v různém počtu, různě moc trnitá. Obstojí vůbec jakékoli funkční vysvětlení? Jiří Sádlo

560

Vesmír 82, říjen 2003 | http://www.vesmir.cz

listí, zubů a žvýkacích svalů jí obstarají symbiotické organizmy. Nejprve nastoupí larvy komárů a jiných dvoukřídlých, které kořist „naporcují“. Poté se do trávení zapojí mikroorganizmy, a produkty jejich trávení pak již hostitelská rostlina vstřebá sama. Tekutina v láčce je vysoce viskózní a kyselá (pH 1,9 – 5,9). Vysoká kyselost i proteolytické enzymy, které rostlina produkuje, slouží pouze k trávení kořisti, nikoli k jejímu usmrcení – kořist se utopí, vyhladoví nebo zahyne vyčerpáním po marných pokusech uniknout. Na rozdíl od jiných stojatých vodních zdrojů tekutina v láčce nehnije, a navíc je vysoce prokysličená. Čím je to způsobeno? Chloroplasty jsou rozloženy nerovnoměrně, buňky na vnitřní straně láčky mají podstatně vyšší obsah chloroplastů než na straně vnější. Kyslík jakožto odpad fotosyntetických procesů pak uniká přednostně do láčky a ne do okolního prostředí. Láčky nejsou zrovna efektivní pastí – např. z mravenčích návštěvníků láčkovky jich skončí v trávicí tekutině jen 0,3–1,3 %. Paradoxně ale právě neefektivnost pasti je základem úspěchu masožravých láčkovek – kdyby byla úspěšnost lapání mravenců stoprocentní, skončili by v láčce všichni průzkumníci, kteří dorazí jako první, ale nebylo by čím přilákat další mravence (průzkumníci vyrábějí pachovou stezku mezi zdrojem potravy a mraveništěm). Vzhledem k tomu, že průzkumníků je poměrně málo oproti počtu dělníků, kteří mohou láčku navštívit, jestliže alespoň někteří průzkumníci přežijí, je vlastně malá efektivnost lovu v konečném důsledku výhodou. Přestože většina láčkovek získává živiny převážně z „ulovených“ bezobratlých, vznik-

la přinejmenším u některých druhů neobvyklá alternativní strategie získávání dusíku a dalších živin (viz rámeček na s. 562). V láčkách bornejské N. lowii, rostoucí v horách, nalezneme jen mizivé množství hmyzu, zato dostatek ptačího trusu. Ten pochází převážně od strdimilů rodu Aethopyga, kteří navštěvují nektarové žlázky na obústí. Bizarní specia-

lizace N. lowii pravděpodobně souvisí s tím, že s nadmořskou výškou klesá početnost mravenců (hlavní potravy většiny láčkovek). Další neobvyklá strategie získávání živin se vyvinula u nížinné N. ampullaria, rozšířené od jižního Thajska po Novou Guineu, která na zemi vytváří husté koberce soudečkovitých láček se značně redukovaným víčkem. Do nich

Vlevo nahoře: Typicky protáhlá horní láčka Nepenthes spectabilis, která je endemitem severní Sumatry, kde obývá světliny horských lesů ve výškách kolem 2000 m. Gunung Pangulubao, Sumatra. Vpravo nahoře: Jednotlivé druhy láčkovek se v přírodě běžně kříží, na obrázku je horní láčka přírodního hybrida Nepenthes spectabilis a Nepenthes ovata. Gunung Pangulubao, Severní Sumatra. Vlevo dole: Drobná Nepenthes dubia roste v porostech kapradin nad zónou zapojeného horského lesa na svazích jediné hory, její horní láčky mají redukované víčko a jsou velké pouze několik centimetrů. Gunung Talamau, Sumatra. Vpravo dole: Nepenthes tobaica úspěšně obydlela kulturní krajinu po přeměně původních lesů na políčka a rýžoviště. Jižní úbočí Gunung Pangulubao, Sumatra.

http://www.vesmir.cz | Vesmír 82, říjen 2003

561

Snímek © Tomáš Grim

Snímek © Miloslav Jirků

Snímek © Tomáš Grim

Vlevo nahoře: Srovnání horní láčky (vlevo) a dolní láčky Nepenthes talangensis, endemita vrcholku hory Gunung Talang. Vpravo nahoře: Pro horní láčky Nepenthes inermis je charakteristická absence obústí a „křídel“. Tato láčkovka je výjimečná epifytickým růstem. Gunung Talang, Sumatra. Dole: Dolní láčka Nepenthes vieillardii s výraznými „křídly“. Vnitřní povrch láčky je tvořen vrstvičkou šupinek kutikuly, jež kořisti zabraňují v útěku.

„lapá“ nejrůznější organický materiál, který je deštěm splachován z okolní vegetace. Většinou láčky fungují jako gravitační pasti, do nichž kořist pasivně padá. Zajímavá adaptace se vyvinula u N. inermis, která je endemitem několika horských vrcholů střední Sumatry. Pasti mají nálevkovitý tvar, zcela chybí obústí a víčko je značně úzké – nemůže tedy chránit obsah láčky před deštěm. Vnitřní povrch láčky je však pokryt vrstvou vysoce viskózní tekutiny, na niž se chycená kořist přilepí a pomalu sklouzává dolů. Vazká tekutina, Dusík a fosfor? Nejprve – dávno tomu – se myslelo, že masožravým rostlinám jde přímo o bílkoviny kořisti (na způsob Adély večeřící bifteky). Pak se to svedlo na dusík a na fosfor, a toto vysvětlení se od té doby traduje. To by znamenalo, že když dáme masožravým rostlinám bezmasou dietu, začnou brzy chátrat – dusík a fosfor jsou přece pro život rostliny zásadní prvky. Ukázalo se však, že to nejde tak rychle. Teprve po mnoha generacích začnou produkovat semena se sníženou klíčivostí a málo vitální potomstvo. To ukazuje, že těmto rostlinám jde o něco jiného – o stopové prvky, které jsou v malém množství potřebné zejména v semenech. Mravenec tedy položí život pro trošku (třeba) vanadu nebo molybdenu, které jsou v neúživném tropickém deštném pralese zbožím mnohem úzkoprofilovějším než zmíněný dusík a fosfor. Jiří Sádlo

Snímek © Tomáš Grim Vlevo nahoře: Koberec láček Nepenthes ampullaria na povrchu substrátu zachycuje organický materiál padající z okolní vegetace. Vpravo nahoře: Na stoncích Nepenthes ampullaria, jež se ze dna lesa šplhají do korun menších stromů (tam kvetou), se někdy objevují také visuté láčky. Vpravo dole: Nepenthes madagascariensis (na snímku) z Madagaskaru byla první objevenou láčkovkou (popsána 1798).

jež se dešťovou vodou neředí, spolehlivě zabraňuje vyplavení kořisti. Tekutina na stěnách je tak viskózní, že po naklonění láčky vytvoří i několikametrový provazec. N. inermis tak jako jediná z celého rodu do značné míry opustila systém padací pasti a zkombinovala jej se strategií lapání kořisti na lepivý povrch, která je běžná u jiných skupin masožravých rostlin. V láčkách se vyskytuje řada organizmů, z nichž některé se na kořisti láčkovek přiživují, jiné jsou dokonce predátory těchto příživníků. V konvicích se vyvíjejí larvy komárů, pakomárů a much, dále zde nalezneme pavouky, kraby (Geosesarma malayanum), roztoče, housenky motýlů či pulce. Snad nejzajímavějším obyvatelem konvic je běžníkovitý pavouk Misumenops nepenthicola. Jeho nejčastějšími hostiteli jsou N. rafflesiana a N. gracilis. V každé láčce žije jen jeden jedinec (běžník byl nalezen asi ve 20 % konvic). Během svého života se jedinci stěhují z jedné láčky do druhé, protože žijí mnohem déle než samotné láčky. Specializace na život v láčce může být absolutní. Např. mravenec Camponotus schmitzi si staví hnízda pouze ve ztluštělých úponcích bornejské láčkovky N. bicalcarata. Dělnice spolupracují při získávání kořisti z láčky tak, že některé vytvářejí ze svých těl živý žebřík, po němž ostatní vytahují kořist nahoru. Mravenci oblafli všechny lovecké adaptace láčkovek – drze se potápějí za kořistí do trávicí tekutiny. Ö

Snímek © Tomáš Grim

Snímek © Pavel Hošek

35. Šumbera R. & Grim T. 2005: Pantanal − na otevřené scéně. 35. Živa 53(2): 93–96. Šumbera R. & Grim T. 2005: Pantanal − na otevřené scéně. Živa 53(2): 93–96.

Pantanal — na otevfiené scénû

Radim ·umbera, TomበGrim Na hranicích tfií jihoamerick˘ch zemí, hluboko v srdci kontinentu leÏí jedno z nejzajímavûj‰ích a faunisticky nejbohat‰ích míst Nového svûta. Unikátní oblast o rozloze více neÏ 240 000 km2 zasahuje do v˘chodní Bolívie, severní Paraguaye a nejvût‰í ãástí, více neÏ polovinou, leÏí v západní Brazílii. S trochou nadsázky se toto místo známé jako Pantanal (portugalsky moãál, baÏina) dá pfiirovnat ke gigantickému amfiteátru. Nejen proto, Ïe pomûrnû pfiesnû naznaãuje, jak celá oblast geomorfologicky vypadá, ale také v˘stiÏnû charakterizuje místo, kde mÛÏeme b˘t svûdky opravdu spektakulární podívané. Nejvût‰í svûtov˘ mokfiad Pantanal je vskutku rozlehl˘m amfiteátrem s ploch˘m dnem a zv˘‰en˘mi okraji z pfiilehl˘ch v˘‰in. Dno pánve tvofiené aluviálními holocenními náplavy v tlou‰Èce aÏ nûkolika desítek metrÛ se pohybuje v nadmofiské v˘‰ce jen 80–150 m a prÛmûrn˘ v˘‰kov˘ rozdíl v severojiÏním smûru je pouhé 2 cm na 1 km. Pantanalskou pláÀ ze v‰ech svûtov˘ch stran svírají vysoãiny (500–700 m n. m.), jako je Serra de Maracajú na v˘chodû, Serra da Bodoquena na jihu, bolivijské a paraguayské chaco na západû, resp. na jihu a koneãnû Serra do SÇo Geronimo ze severu. Na jihu je bariéra pfieru‰ena a tudy také vytéká mohutná fieka Paraguay, která odvádí vût‰inu vody z Pantanalu. Bûhem období de‰ÈÛ (fiíjen–bfiezen) kotlinu zaplavují obrovské masy vody. SráÏky nad Pantanalem nejsou nijak extrémní (pfiibliÏnû 1 000–1 400 mm roãnû), pfiesto se rozsáhlé oblasti rychle zaplaví, neboÈ k de‰ÈÛm nad planinou se pfiidává voda pfiitékající z vysoãin bohat˘ch na sráÏky. ¤eky v Pantanalu nezvládají mohutn˘ pfiítok kompenzovat, coÏ na nûkter˘ch místech zpÛsobuje vzestup vodní hladiny aÏ o 2–3 m. Bûhem kulminace vody jsou tak nûkde celé oblasti aÏ na v˘‰e poloÏená místa zvaná cordilheiras z velké ãásti zaplavené. Zatopeny b˘vají aÏ dvû tfietiny celkové rozlohy, coÏ je mimo-

Ïiva 2/2005

chodem území vût‰í neÏ âeská republika, a Pantanal je tak jednou z nejvût‰ích sezonnû zaplavovan˘ch oblastí na svûtû. S postupujícím suchem fieky zvládají vodu odvádût, takÏe se voda udrÏuje zejména v prohlubních zvan˘ch baías, které mohou mít v prÛmûru od nûkolika desítek metrÛ aÏ po nûkolik kilometrÛ. PrÛmûrné roãní teploty se pohybují kolem 25 °C. Kromû extrémních letních „ãtyfiicítek“ v‰ak mÛÏe teplota v zimû klesnout pohybem chladného antarktického vzduchu aÏ k nule. V˘par vody je vysok˘, takÏe ta rychle mizí i z rozsáhl˘ch ploch, coÏ v extrémnû such˘ch letech mÛÏe mít za následek vyschnutí i velk˘ch fiek. Období sucha trvá vût‰inou od kvûtna do záfií. Celková situace se tak bûhem nûkolika mûsícÛ úplnû obrátí. Pozemní fauna se bûhem de‰ÈÛ koncentruje na such˘ch místech, kdeÏto v období vrcholícího sucha se stahuje k rychle mizící vodû. A v tomto fenoménu je hlavní kouzlo Pantanalu, zejména pro zoology. KfiiÏovatka jihoamerick˘ch biomÛ Podobu Pantanalu ovlivÀují pfiedev‰ím tfii dÛleÏité ekosystémy JiÏní Ameriky: suché chaco na jihu a západû, savanové cerrado na v˘chodû a amazonsk˘ prales na severu. Objevují se zde i elementy v˘chod-

93

ních atlantsk˘ch de‰tn˘ch lesÛ. Spoleãnû s topografií terénu a sezonním klimatem se tyto ekosystémy podílejí na vytvofiení pestré unikátní mozaiky such˘ch, pravidelnû zaplavovan˘ch nebo vodních biotopÛ. Jaká vegetace kde poroste, záleÏí pfiedev‰ím na tom, zda je místo zaplavováno a jak dlouho se zde voda udrÏí. Velká ãást rostlin je adaptována zároveÀ na vysokou vodní hladinu i na extrémní sucho. S tím souvisí i adaptace rostlin na oheÀ — ãasté poÏáry v období sucha pfiirozenû regulují udrÏování savanového prostfiedí. V Pantanalu najdeme celou ‰kálu biotopÛ od úplnû otevfien˘ch po lesnaté, od such˘ch po vlhké: suché travnaté porosty skoro bez stromÛ vût‰inou ve vy‰‰ích polohách (campo limpo), suché travnaté plánû s kfiovinat˘mi porosty a nízk˘mi stromy (campo cerrado — biotop typick˘ pro centrální Brazílii), cerradÇo — podobn˘ biotop s hustûj‰ím stromov˘m porostem a uzavfien˘m korunov˘m patrem, vÏdyzelené de‰tné lesy na severu, poloopadav˘ les, opadav˘ les a galeriov˘ les. Galeriové lesy na zv˘‰en˘ch bfiezích velk˘ch fiek hrají v Pantanalu velmi dÛleÏitou úlohu, neboÈ slouÏí jako biokoridor, kter˘m se do su‰‰ích oblastí ‰ífií vliv vlhké Amazonie. Dal‰ími typy vegetace jsou sezonnû zaplavované lesy v blízkosti velk˘ch fiek nebo sezonnû zaplavované savanové porosty (campo alagado), kde se voda z velk˘ch fiek po de‰tích rozlévá do rozlehl˘ch oblastí. Campo alagado — vlhká savana — se ãasto nachází na málo propustném substrátu, kde se voda drÏí, dokud se nevypafií. V zaplavovan˘ch oblastech hrají velmi dÛleVlevo oblast Nhecolandia v jiÏním Pantanalu (brazilská ãást), která je známa sv˘mi chud˘mi písãit˘mi pÛdami. Na nezaplavovan˘ch místech mÛÏeme najít ostrÛvky lesa oznaãovan˘ch jako capÇo. Po poÏárech zde nastupují pion˘rské rostliny, jako napfi. terestrická bromelie Bromelia balansae, která roste ãasto v souvisl˘ch porostech (v popfiedí). Typick˘mi rostlinami jsou také kaktus Cereus peruvianus, stromy Callisthene fasciculata (ãel. Vochysiaceae) a Curatella americana (ãel. Dilleniaceae) ãi palma Astrocaryum aculeatum, která je amazonsk˘mi indiány velice cenûna pro vlákna z ní získávaná. Foto T. Grim ♦ V kopcích na jihu je krasová oblast s fiekami, tÛnûmi a jezírky, takÏe voda mífiící do pánve je nasycena vápenat˘mi a hofieãnat˘mi uhliãitany (vpravo)

www.cas.cz/ziva

Ïitou úlohu jakákoli rozsáhlá vyv˘‰ená místa, která slouÏí jako útoãi‰tû pro místní zvífienu. Posledními typy biotopÛ jsou místa s permanentní pfiítomností vody, jako velké fieky, jezírka, mokfiady a baÏiny, jejichÏ okraje b˘vají celoroãnû hustû porostlé vegetací. Na prvních mapách jihoamerického kontinentu z pfielomu 16. a 17. stol. b˘val Pantanal zobrazen jako obrovské jezero Eupana Lacus, dominujícím biotopem Pantanalu je v‰ak savana. Vysoká biodiverzita, nízk˘ endemismus Pestrá nabídka nejrÛznûj‰ích biotopÛ pfiispívá k velmi vysoké biodiverzitû. Porovnáme–li napfi. savãí faunu brazilského Pantanalu a brazilské Amazonie, tak pfiestoÏe je brazilská ãást Amazonie témûfi ‰edesátkrát vût‰í, je zde jen tfiikrát více druhÛ. V Pantanalu Ïije více druhÛ savcÛ neÏ v dal‰ích rozlohou vût‰ích jihoamerick˘ch travinn˘ch ekosystémech, jako jsou napfi. paraguyajské chaco nebo brazilská caatinga. Pantanal je díky své geomorfologii spí‰e biogeografick˘m koridorem neÏ bariérou, takÏe aã je oblastí s vysokou biodiverzitou (uvádí se, Ïe tu lze vidût nejvût‰í koncentraci divok˘ch zvífiat v Novém svûtû), je zároveÀ místem s velmi mal˘m endemismem. PfiestoÏe je z Pantanalu známo skoro 2 000 druhÛ rostlin, nebyl dosud popsán Ïádn˘ endemick˘ druh a zÛstaneme–li u srovnání savãí fauny Pantanalu a Amazonie, savãích endemitÛ je zde necel˘ch 2,5 % oproti 60 % amazonsk˘m. Z opefiencÛ sem zasahuje sv˘m v˘skytem jedin˘ druh endemick˘ pro Brazílii, guan stfiedobrazilsk˘ (Penelope ochrogaster). V Pantanalu je opravdu Ïivo. V této pfiekvapivû málo prozkoumané oblasti bylo zaznamenáno 656 druhÛ ptákÛ (700 i s blízk˘m okolím), pfies 95 druhÛ savcÛ (nûkteré zdroje uvádûjí více neÏ 120), 162 druhÛ plazÛ, 45 druhÛ obojÏivelníkÛ a 263 (celkov˘ poãet se odhaduje na více neÏ 400) druhÛ ryb. Pro vût‰inu tûchto skupin to urãitû nebudou koneãná ãísla. Pro místní zvífienu je charakteristická nízká teritorialita. Ryby, ptáci, savci, ale i nûktefií plazi jednodu‰e následují ustupující hladinu vody. Mimo nápadné stûhování terestrick˘ch obratlovcÛ jsou migrace typicwww.cas.cz/ziva

Vlevo jelenec pampov˘ (Ozotoceros bezoarticus), jehoÏ poãetnost se v Pantanalu i pfies siln˘ loveck˘ tlak odhaduje na 35 000 jedincÛ ♦ Mraveneãníka velkého (Myrmecophaga tridactyla) je moÏné zastihnout v savanû i bûhûm dne, vpravo

se zde setkat i s tejovit˘m je‰tûrem Dracaena paraguayensis, kter˘ je sv˘m v˘skytem omezen na Pantanal a jeho blízké okolí.

ké i pro ryby. Sezonní migrace ryb, nûkdy aÏ na stovky kilometrÛ, se naz˘vají piracemas. Bûhem sucha se dospûlé ryby po proudu stahují na místa s dostatkem vody. KdyÏ se bûhem de‰ÈÛ hladina vody zv˘‰í, zaãínají se rozmnoÏovat a spoleãnû s potûrem osidlují zaplavené oblasti. Ty jsou v‰ak ãasto chudé kyslíkem, coÏ ãasto vede k hromadn˘m úhynÛm ryb. V dobû, kdy de‰tû konãí a hladina vody opadává, se pfieÏiv‰í vykrmené ryby zaãínají vracet do fiek. Nejznámûj‰í místní ryby jsou samozfiejmû dravé piranû r. Serrasalmus, ale také Raphiodon vulpinus nebo velká paku Colossoma mitrei (obû z ãel. tetrovit˘ch — Characidae). Vyskytuje se zde i bizarní bahník americk˘ (Lepidosiren paradoxa), pfiíslu‰ník starobylé skupiny dvojdy‰n˘ch (Dipnoi), která je sesterskou skupinou suchozemsk˘ch obratlovcÛ (Tetrapoda). Jako na mnoha místech JiÏní Ameriky i zde mezi rybami dominují fiády trnobfiich˘ch (Characiformes) a sumcovit˘ch (Siluriformes). Pro obû skupiny zde bylo zji‰tûno pfies 100 druhÛ. Fauna pantanalsk˘ch obojÏivelníkÛ je nedostateãnû známa a zatím bylo v oblasti zji‰tûno jen nûco málo pfies ãtyfii desítky druhÛ. Mezi hojnûj‰í patfií rosniãky (Hylidae), napfi. rosniãka kfiehká (Hyla nana) nebo rosniãka ãakobytná (H. raniceps), a dále hvízdalky (Leptodactylidae) jako hvízdalka hranatá (Leptodactylus ocelatus) a hvízdalka klokotavá (L. podicipinus). Na obojÏivelníky jsme nemûli ‰tûstí a narazili jsme zde jen na ropuchu kururu (Bufo paracnemis). Z plazÛ je jednoznaãnû nejtypiãtûj‰ím zástupcem kajman Ïakare (Caiman yacare, viz obr.), kter˘ Ïije u kaÏdé vût‰í vodní plochy. Tento druh, pfiíbuzn˘ severnûji Ïijícímu kajmanu br˘lovému (C. crocodyllus), zde dosahuje nejvût‰í populaãní hustoty zaznamenané pro krokod˘ly kdekoli na svûtû a jeho celková poãetnost v oblasti je odhadována pfies 10 milionÛ jedincÛ. Dal‰ími typick˘mi plazy jsou aÏ pûtimetrová anakonda Ïlutá (Eunectes notaeus) a populární leguán zelen˘ (Iguana iguana). MÛÏeme

Pantanal, kter˘ je mimo jiné i v˘znamn˘m zimovi‰tûm severoamerick˘ch ptákÛ, je v‰eobecnû povaÏován za vÛbec nejlep‰í ornitologickou lokalitu na svûtû. Nikoli proto, Ïe by byl druhovû nejbohat‰í — v tomto smûru vede peruánsk˘ NP Manu s 980 druhy, i kdyÏ se pfiedpokládá, Ïe ‰patnû prozkouman˘ NP Madidi v severozápadní Bolívii ob˘vá aÏ 1 100 druhÛ ptákÛ; v Pantanalu najdeme „pouze“ 656 druhÛ. Pantanal je „nejlep‰í“ proto, Ïe je zde nejsnaz‰í pozorovat velice ‰iroké druhové spektrum v obrovsk˘ch poãetnostech. Jak uvádí N. Wheatley (1994) „mnoÏství ptákÛ, a to zvlá‰tû koncem období sucha v záfií, musí ãlovûk vidût na vlastní oãi, aby tomu vÛbec uvûfiil.“ Symbolem Pantanalu je ãáp jabiru (Jabiru mycteria); je poãetn˘, nápadn˘ a monumentálních rozmûrÛ (viz obr.). Dobfie také symbolizuje hlavní lákadlo Pantanalu — vodní ptactvo. Je samozfiejmû nemoÏné podat v takto krátkém textu reprezentativní obrázek místní avifauny, ale zmiÀme se alespoÀ o tûch zástupcích, které náv‰tûvník mÛÏe snadno uvidût. Nejvût‰ím ptákem cel˘ch neotropÛ je nandu pampov˘ (Rhea americana), kter˘ je v Pantanalu sice bûÏn˘, ale má (alespoÀ pro fotografa) aÏ nepfiíjemnû velkou útûkovou vzdálenost. Asi nejpoãetnûj‰ími obyvateli vodních ploch jsou volavky a kachny. Velká hejna tvofií pfiedev‰ím husiãka vdovka (Dendrocygna viduata), husiãka podzimní (D. autumnalis) a h. dvoubarvá (D. bicolor). Dále se tu bûÏnû setkáme s kachniãkou amazonskou (Amazonetta brasiliensis) nebo ãájou obojkovou (Chauna torquata), která sv˘m vzhledem i „pravûk˘m“ hlasem pÛsobí znaãnû starobyle. Z brodiv˘ch ptákÛ se setkáme s volavkou bûlostnou (Egretta thula), v. modro‰edou (E. caerulea), „na‰í“ v. bílou (E. alba), v. rusohlavou (Bubulcus ibis), v. jihoamerickou (Ardea cocoi) ãi elegantní volavkou hvízdavou (Syrigma sibilatrix). BûÏní jsou i bukaã ãervenav˘ (Tigrisoma lineatum), ibis ‰edokfiídl˘ (Theristicus melanopis), ibis tmav˘ (Phimosus infuscatus), ibis

94

Ptactvo

Ïiva 2/2005

lesní (Mesembrynibis cayennensis) nebo nesyt americk˘ (Mycteria americana). Pfiekvapivû velká je i diverzita drav˘ch ptákÛ — 45 druhÛ. Mezi nejznámûj‰í neotropické dravce jistû patfií lunûc baÏinn˘ (Rostrhamus sociabilis), znám˘ specialista na vodní plÏe, které vytahuje sv˘m ‰tíhl˘m zahnut˘m zobákem z ulit. Kánû ãernohrdlá (Busarellus nigricollis), která preferuje blízkost vodních ploch, se ãasto prozradí sv˘m nepfiíjemn˘m skfiípav˘m hlasem. Typick˘m obyvatelem otevfiené krajiny je pak kánû savanová (Buteogallus meridionalis). Z fiádu krátkokfiídl˘ch se v Pantanalu vyskytuje seriema rudozobá (Cariama cristata), u vody pak kurlan chfiástalovit˘ (Aramus guarauna), kter˘ na první pohled pfiipomíná spí‰e ibise. U bfiehu se na sv˘ch dlouh˘ch nohách procházejí pisily ãáponohé (Himantopus himantopus mexicanus), vodu brázdí prodlouÏen˘mi spodními ãelistmi sv˘ch bizarních zobákÛ zobouni ameriãtí (Rynchops niger). Z Pantanalu je známo 26 druhÛ papou‰kÛ, vãetnû nûkolika druhÛ arÛ a amazoÀanÛ. Spí‰e v severní ãásti Pantanalu lze napfi. spatfiit nejvût‰ího, stále vzácnûj‰ího aru hyacintového (Anodorhynchus hyacinthinus) s celkovou populací v pfiírodû kolem pouh˘ch 3 000 jedincÛ (viz obr.) a dal‰í velké druhy, jako tfieba aru zelenokfiídlého (Ara chloroptera). Jednûmi z nejnápadnûj‰ích ptákÛ otevfiené krajiny jsou kukaãky ani (Crotophaga ani) a kukaãky guiry (Guira guira), známé z uãebnic etologie díky svému kooperativnímu hnízdûní. Mezi nejãastûji sly‰ené, ale nejménû viditelné ptáky patfií kukaãka Ïíhaná (Tapera naevia). Do severní ãásti Pantanalu zasahuje areál bizarního hoacina chocholatého (Opisthocomus hoazin), známého cestovatele ptaãím systémem, kter˘ byl fiazen do rÛzn˘ch zcela nepfiíbuzn˘ch skupin (hrabaví, krátkokfiídlí, mûkkozobí, tinamy, turakové, chfiástali, kukaãky) a jehoÏ fylogenetické postavení zÛstává nevyfie‰eno. Z ptákÛ s noãní aktivitou jmenujme alespoÀ krásnû zbarveného lelka skvrnokfiídlého (Podager nacunda), ãi mistra kryptického zbarvení — potu obecného (Nyctibius griseus). Ménû známou skupinou jsou lenivky (napfi. l. teãkovaná — Nystalus maculatus) a leskovci (v Pantanalu leskovec neotropick˘ Ïiva 2/2005

— Galbula ruficauda), ktefií byli dfiíve fiazeni mezi ‰plhavce a dnes jsou vyãleÀováni do samostatného fiádu leskovci (Galbuliformes). Jde o skupinu endemickou pro neotropy, pro níÏ je typick˘ zpÛsob lovu — ospalé vysedávání na „pozorovatelnách“, pfieru‰ované bleskov˘mi útoky na prolétající kofiist a následované návratem na pÛvodní „posed“. Ze ‰plhavcÛ pozná kaÏd˘ laik tukana obrovského (Ramphastos toco). Kromû nûj lze ãasto vidût i men‰ího arassari hnûdouchého (Pteroglossus castanotis) ãi datla kampového (Colaptes campestris). Kapitola sama pro sebe jsou pûvci. Pantanal hostí ‰irokou ‰kálu zástupcÛ typick˘ch neotropick˘ch ãeledí: klouzálkÛ (Dendrocolaptidae), hrnãifiíkÛ (Furnariidae), mravenãíkÛ (dnes ve dvou ãeledích: „zemní“ mravenãíci Formicariidae, „praví“ mravenãíci Thamnophilidae) a samozfiejmû asi vÛbec nejtypiãtûj‰í americkou ptaãí ãeleì — tyrany (Tyrannidae). Kromû klasick˘ch v‰udypfiítomn˘ch druhÛ (napfi. tyran bentevi — Pitangus sulphuratus) zde najdeme i skvostné druhy s v˘raznû prodlouÏen˘mi ocasními pery (napfi. bûÏn˘ tyran savanov˘ — Tyrannus savana nebo pomûrnû vzácn˘ tyranovec vidlochvost˘ — Gubernetes yetapa). Na rozdíl od vût‰iny tyranÛ najdeme nûkteré druhy spí‰e na zemi, tfieba tyranovce dlouhonohého (Machetornis rixosus), kter˘ se, podobnû jako v˘‰e zmínûné kukaãky ani, ãasto pfiiÏivuje na hmyzu vypla‰eném pasoucím se dobytkem. Z drobnûj‰ích pûvcÛ se blízko vody vyskytují jednotlivû nebo v mal˘ch hejnech kardinálové ãernohfibetí (Paroaria capitata) a k. ‰edí (P. coronata). Celou fiadou druhÛ jsou zastoupeni strnadovití knûÏíci (Sporophila) a ‰afránky (Sicalis). Díky sv˘m námluvním rituálÛm a podivuhodn˘m hlasÛm patfií mezi nejatraktivnûj‰í americké ptaãí skupiny nepochybnû vlhovci (Icteridae). Kromû druhÛ, které si v koloniích stavûjí zavû‰ená vakovitá hnízda (napfi. vlhovec ãervenohfibet˘ — Cacicus haemorrhous), jsou nûkteré druhy hnízdními parazity. V JiÏní Americe je to pfiedev‰ím vlhovec modroleskl˘ (Molothrus bonariensis), kter˘ klade svá vejce do hnízd pfies 200 druhÛm pûvcÛ — mezi nimi i blízce pfiíbuznému vlhovci ‰edohnûdému (M. badius). Oba najdeme bûÏnû sbírat potravu ve spoleãn˘ch hejnech v otevfiené krajinû. 95

Vlevo typick˘ interiér poloopadavého pantanalského lesa. Foto T. Grim ♦ âáp jabiru (Jabiru mycteria), dÛstojn˘ symbol Pantanalu (vpravo) Savci Pantanal není proslul˘ pouze sv˘m ptactvem nebo mnoÏstvím kajmanÛ. SavcÛ zde Ïije také mnoho a nûkdy se Pantanal dokonce povaÏuje za nejlep‰í místo na pozorování velk˘ch savcÛ v JiÏní Americe. Nejvût‰í ‰elmou je zde jaguár (Panthera onca), jehoÏ poãetnost se v Pantanalu odhaduje na tisíc jedincÛ. Místní exempláfie dosahují mezi jihoamerick˘mi jaguáry nejvût‰ích rozmûrÛ. Vyskytuje se zde i puma (Puma concolor), ocelot velk˘ (Felis pardalis) nebo z celého jihoamerického kontinentu rychle mizející v‰eÏrav˘ vlk hfiivnat˘ (Chrysocyon brachyurus). Tuto ‰elmu místní farmáfii podezfiívají z loupení domácího zvífiectva, a proto ji nemilosrdnû likvidují, pfiestoÏe vlk hfiivnat˘ není dobytku nijak nebezpeãn˘. V období de‰ÈÛ jsou jeho hlavní potravou plody lilkovité rostliny Solanum lycocarpum, zvané fruta de lobo (vlãí ovoce). Pfiedpokládá se, Ïe tento plod mÛÏe slouÏit jako obrana proti hlístu Dioctophyma renale, kter˘ napadá ledviny vlka hfiivnatého i dal‰ích ‰elem. Zajímavou psovitou ‰elmou je i vysoce sociální pes pralesní (Speothos venaticus) z lesnat˘ch oblastí. NejbûÏnûj‰í a snadno pozorovatelnou ‰elmou je medvídkovit˘ nosál ãerven˘ (Nasua nasua). Vodní predátory zastupují nejvût‰í sladkovodní vydra svûta, vydra obrovská (Pteronura brasiliensis) s charakteristickou krémovou kresbou na hrudi a krku, Ïijící ãasto pospolitû ve vût‰ích fiekách, a její men‰í pfiíbuzná vydra jihoamerická (Lutra longicaudis). Nemohou zde chybût ani zástupci starobylé skupiny, tak typické pro tento kontinent — chudozubí (Edentata, novûji Xenarthra). Na pfiítomnost nejvût‰ího pásovce, noãního pásovce velkého (Priodontes maximus) dosahujícího aÏ 60 kg ukazují vût‰inou jen ‰iroké nory v zemi. Vidût tu mÛÏeme spí‰e jeho men‰í pfiíbuzné, jako napfi. pásovce devítipásého (Dasypus novemcinctus), nebo pásovce ‰estipásého (Euphractus sexcinctus). V lesnat˘ch oblastech Pantanalu nechybûjí ani lenochodi. Snadno se zde dají vidût i myrmekofágní zástupci této skupiny, ktefií jako jediní chuwww.cas.cz/ziva

Vlevo nahofie kajman Ïakare (Caiman yacare), dosahující nejvût‰í populaãní hustoty mezi krokod˘ly na svûtû. Snímky R. ·umbery, není–li uvedeno jinak ♦ Ara hyacintov˘ (Anodorhynchus hyacinthinus) je nejvût‰í a kriticky ohroÏen˘ papou‰ek (vpravo). Foto T. Grim ♦ Vfie‰Èan ãern˘ (Alouatta caraya), vlevo dole. Foto T. Grim

dozubí dûlají ãest svému ãeskému názvu, neboÈ nemají v ãelistech ani jedin˘ zub — oproti nûkter˘m pásovcÛm, ktefií mohou mít i pfies sto jednoduch˘ch zubÛ. Men‰í zástupce s chápav˘m ocasem — mraveneãník ãtyfiprst˘ (Tamandua tetradactyla) — má striktnû noãní aktivitu, kdeÏto mraveneãníka velkého (Myrmecophaga tridactyla) je moÏno spatfiit i ve dne (viz obr.), hlavnû poblíÏ termiti‰È, která sv˘mi masivními drápy rozhrabává. Oba druhy mají ‰patn˘ zrak a sluch a orientují se pfiedev‰ím ãichem. Pantanal je také domovem tapíra jihoamerického (Tapirus terrestris), nejvût‰ího jihoamerického jelena — jelence bahenního (Blastocerus dichotomus), jelence pampového (Ozotoceros bezoarticus, viz obr.), pekari páskovaného (Pecari tajacu) nebo pekari bûlobradého (Tayassu pecari), kter˘ se ãasto pohybuje ve stádech o desítkách ãi stovkách kusÛ. Velmi snadno lze spatfiit nejvût‰í hlodavce na svûtû — kapybary (Hydrochaeris hydrochaeris), Ïijící ve skupinách kdekoli poblíÏ vody. Populace kapybar se v Pantanalu odhaduje na milion jedincÛ a tito hlodavci tvofií jednu z hlavních potravních sloÏek velk˘ch predátorÛ. Velk˘ch a nápadn˘ch druhÛ savcÛ je samozfiejmû mnohem ménû neÏ men‰ích druhÛ, ktefií díky skrytému zpÛsobu Ïivota a noãní aktivitû unikají pozornosti. Mezi drobn˘mi savci se slu‰í zmínit o jihoamerick˘ch vaãnatcích, kter˘ch Ïije v Novém svûtû pfiibliÏnû 70 druhÛ. Nejnápadnûj‰ím je nûkolikakilogramová vaãice bûlobfiichá (Didelphis albiventris) se soumraãnou i denní aktivitou. Dal‰ím zajímav˘m druhem je vaãice ãtyfioká (Philander opossum). Tato milovnice vlhãích biotopÛ si ãesk˘ název vyslouÏila párem svûtl˘ch skvrn nad oãima pÛsobících dojmem dal‰ího páru oãí, coÏ má pravdûpodobnû komunikaãní funkci. Naopak spí‰e na su‰‰ích místech a ãasto v blízkosti lidsk˘ch sídel se vyskytuje malá vaãice krysí (Monodelphis domestica). V Pantanalu samozfiejmû Ïije i nûkolik desítek druhÛ letounÛ, zejména zástupci neotropické ãel. listonosovit˘ch (Phyllostomidae). Nejnápadnûj‰ími jsou pfiedev‰ím dva druhy velk˘ch ryboÏrav˘ch netop˘rÛ r. Noctilio, létající veãer nad vodní hladinou. Zajímavou potravní ekolowww.cas.cz/ziva

gii má i velk˘ predátor listonos ÏáboÏrav˘ (Trachops cirrhosus), kter˘ se neÏiví jen Ïábami, ale i drobn˘mi plazy ãi men‰ími druhy netop˘rÛ. V Pantanalu nechybí ani povûstmi opfieden˘ krevsající upír obecn˘ (Desmodus rotundus), kter˘ se v˘znamnû podílí na ‰ífiení vztekliny mezi domácím dobytkem. Pestrou kolekci savcÛ doplÀuje i pût druhÛ primátÛ, z nichÏ mimo malpy hnûdé (Cebus appela) je díky silnému teritoriálnímu fievu a barevnému pohlavnímu dimorfismu nejnápadnûj‰í vfie‰Èan ãern˘ (Alouatta caraya, viz obr.). Pantanal ohroÏen˘ Pantanal je v mnoha aspektech natolik unikátní fenomén, Ïe se slu‰í zmínit se i o vztahu ãlovûka k tomuto místu. Mimo ovlivÀování klimatu, zásobárny a zároveÀ pfiirozené ãistiãky pitné vody má Pantanal pro domorodé obyvatele v˘znam jako zdroj potravy (pfiedev‰ím ryb), dopravní tepna a v neposlední fiadû i zdroj penûz ze stále rostoucího turistického ruchu. PfiestoÏe je Pantanal oproti zbytku Brazílie fiídce osídlen (1,8 obyv./km2), vût‰inu pÛdy si rozparcelovaly farmy (paraguayská ãást je témûfi neobydlená, liduprázdná je i bolívijská ãást). Na dosavadním zachování pfiírody Pantanalu má zásluhu pfiedev‰ím unikátní vodní cyklus, kter˘ jej neumoÏÀuje extenzivnû vyuÏívat pro jiné neÏ pastevní zemûdûlství. Hlavnû díky nûmu zÛstává Pantanal je‰tû jedním z mála míst, kterému se ãlovûk svou aktivitou musí pfiizpÛsobovat. Nejnápadnûj‰í lidskou ãinností se stal chov dobytka. Jihoamerické savany jsou charakteristické nepfiítomností velk˘ch stád kopytníkÛ znám˘ch ze savan africk˘ch. Velcí kopytníci jsou v Pantanalu nahrazeni pfiibliÏnû 10 miliony kusÛ skotu, kter˘ Ïije celoroãnû na pastvû polodivok˘m zpÛsobem a pfiekvapivû v souladu s pfiírodními procesy. Negativnû pÛsobí lovecká aktivita lidí. Z nûkter˘ch ãástí Pantanalu jiÏ vymizeli predátofii jako jaguár, puma a vlk hfiivnat˘. Podobn˘ dopad má i plenûní hnízd velk˘ch papou‰kÛ. Napfi. jen bûhem 80. let 20. stol. bylo podle odhadÛ odebráno z pfiírody kolem 10 000 jedincÛ ary hyacintového, pfiiãemÏ poptávka a následnû i intenzita pytla96

ãení je‰tû stoupla po r. 1987, kdy byl druh zafiazen do CITES I. Nevhodné zásahy se ãlovûku rychle vracejí zpût. Velké sníÏení poãetnosti mraveneãníka velkého (chutné maso) zpÛsobuje pfiemnoÏení mravencÛ a termitÛ, ktefií sv˘mi stavbami znehodnocují pastviny. Hubení kajmanÛ pro kÛÏe pfiineslo nárÛsty poãtu piraní natolik, Ïe ohroÏují poãetnost ostatních ryb. Velmi negativní vliv mají diamantové a zlaté doly (mimo jiné zamofiováním prostfiedí rtutí), ropovody, komunální odpad vypou‰tûn˘ pfiímo do fiek, splachování hnojiv z okolních vysoãin, lidmi zakládané ohnû, extenzivní turismus a nejrÛznûj‰í hydroprojekty. Tyto vlivy jsou bolestné, ale neohroÏují samou podstatu Pantanalu. V poslední dobû se v‰ak objevila hrozba, pfiiná‰ející zkázu celé oblasti. Jde o gigantick˘ projekt Paraguay–Paraná Hydrovia, kter˘ by pomocí nov˘ch pfiehrad a bagrování koryt fiek pfiispûl k vût‰ímu splavnûní obou fiek a vodnímu propojení Bolívie, Brazílie, Paraguaye, Uruguaye a Argentiny. Tfii a pÛl tisíce kilometrov˘ splavn˘ komplex by mûl usnadnit transport vytûÏen˘ch surovin a potravin, usnadnit pfiístup vnitrozemsk˘ch státÛ k mofii a pfiispût k vût‰ímu ekonomickému rozvoji celé oblasti. Cílem je umoÏnit i proplutí lodí s hlubok˘m ponorem (aÏ 2,8 m) a to dokonce i bûhem období sucha. Tento podnik by kromû mnohem vût‰ího zneãi‰tûní, zv˘‰eného rizika povodní a sníÏení rybolovu mûl zásadní vliv na vodní cyklus (roz‰ífiení koryta fieky Paraguay by zpÛsobilo rychlej‰í odtok vody z oblasti) a zapfiíãinil by vysou‰ení rozsáhl˘ch oblastí a likvidaci podstatné ãásti Pantanalu. Propoãty ukazují, Ïe sníÏení hladiny fieky Paraguay o pouh˘ch 25 cm by mûlo za následek sníÏení celkové zaplavované plochy skoro o ãtvrtinu, coÏ je oblast dvakrát vût‰í neÏ rozloha zbyl˘ch floridsk˘ch mokfiadÛ. Právû osud unikátních severoamerick˘ch mokfiadÛ by mohl b˘t mementem naznaãujícím, jak by to mohlo dopadnout v Pantanalu. Na‰tûstí se v‰ak Pantanal stal i pfiedmûtem ochranáfisk˘ch snah. Celkem ãtyfii chránûná území se dostala na seznam Ramsarské úmluvy o mokfiadech: Pantanal Boliviano (3 189 888 ha), brazilsk˘ Pantanal Matogrossense (135 000 ha), taktéÏ brazilská, ale soukromá rezervace Reserva Particular do Patrimonio Natural SESC Pantanal (87 871 ha) a paraguaysk˘ Národní park Río Negro (370 000 ha). Ochrana na papífie je v‰ak ponûkud odtrÏena od reality. Napfi. bezmála 30 000 ha paraguayského chránûného území bylo ilegálnû pfievedeno do soukromého vlastnictví, probíhá nezákonná tûÏba dfieva atd. Takov˘mto katastrofám se snaÏí zabránit napfi. ekologická organizace Ecotrópica. Ta se kromû osvûty mezi místním obyvatelstvem vûnuje v˘kupu pozemkÛ sousedících s národním parkem. Zatím se tímto postupem chránûná brazilská oblast roz‰ífiila o více neÏ 500 km2. Kromû toho bylo v r. 2000 témûfi 190 000 ha (1,3 % brazilského Pantanalu) chránûn˘ch oblastí v Pantanalu zapsáno na seznam svûtového dûdictví UNESCO. Ïiva 2/2005