The Auk 113(1):10-22, 1996

PHYLOGENETIC ANALYSIS OF THE CUCULIDAE (AVES, CUCULIFORMES) USING BEHAVIORAL AND ECOLOGICAL CHARACTERS

JANICEM. HUGHES Departmentof Zoology,University of Toronto,Toronto, Ontario M5S 3G5, Canada

AI3STRACT.--Acladistic analysis of 21 generaof (Aves, Cuculidae)using 28 behav- ioral and ecologicalcharacters produces one shortest-lengthtree (L = 80 steps,CI = 0.52, RI = 0.79, RC = 0.42) that differs from traditional classificationsof the Cuculidae. My results suggestthat two cuculidsubfamilies, the terrestrialNeomorphinae and the Phaenicophaeinae, are polyphyletic. The obligate Tapera() and the facultative brood parasiteCoccyzus (Phaenicophaeinae) are removedfrom otherwisenonparasitic sub- familiesand placedamong the Old World obligateparasites in the Cuculinae.This suggests that: (1) brood parasitismarose only once in the Cuculidae rather than three times as pre- viously thought; and (2) that terrestrialhabitat use in Taperahas evolved secondarily.The placementof Coccyzusamong the obligateparasites implies that the immediateancestor of this genuswas an obligatebrood parasite.Therefore, the facultativebehavior of representsa lossof obligateparasitism, rather than the developmentof facultativeparasitism from a nonparasiticancestor. In life, Coccyzusshares a number of life-historytraits with the obligatebrood parasites that supportthis hypothesis.Based on my analysis,I proposechanges to the classificationof the cuckoosthat are consistentwith the opinions of many early systematists,and the resultsof an unpublishedphylogeny of the cuckoosbased on postcranial osteologicalcharacters. In addition,my findingssuggest that the Hoatzin (Opisthocomushoat- zin)is a cuckoomost closely related to the communallybreeding anis (Crotophaginae). Received 12 May 1994, accepted27 January1995.

THE CUCKOOSare best known for obligately ship with their hosts--that servesto minimize parasiticbreeding habits, whereby a femalewill detectionand destructionof cuckooeggs (Ham- lay eggsin the nestsof hostspecies and, hence, ilton and Orians 1965, Payne 1977, Davies and relinquish the responsibilitiesof parenthood. Brooke 1989). However, recent studies by However, this diverse family of , com- Brooker and Brooker (1989, 1990) have sug- prisedof 129species in 38 genera(Morony 1975), gestedthat theseadaptations to parasitismmay containsat least 4 speciesof facultative brood have evolved due to intraspecificcompetition parasites(Coccyzus spp.; Nolan and Thompson between parasiticfemales that remove an 1975,Ralph 1975,Sick 1993) and 74 nonparasitic from the host nest just prior to the deposition ,4 of which are communal breeders of their own egg.In addition,most chicks (Wyllie 1981). The family is global in distri- will ejectthe host eggsor young from the nest bution and occupiesnearly all temperateand within a few days of hatching. The nestling tropicalbiomes with the exceptionof somere- (Tapera naevia) uses mandibular mote oceanic islands. A few cuckoosare pre- hooks to kill host chicks in a manner similar to dominantly terrestrial,foraging and nestingon that of the parasitichoneyguides of the or near the ground; however, most speciesare Indicator(Piciformes; Morton and Farabaugh arborealand many are long-distancemigrants 1979). Theseadaptations ensure that the para- (Rowan1983). Food habits range through vary- site chick is the sole occupantof the host nest, ing degreesof herbivory and camivory with thereby improving its chancesof fledging suc- many speciesrelying almostentirely on toxic cessfully. aposomaticcaterpillars. Egg color, clutch size, The diversity of the Cuculidae may be the and incubationperiods show little consistency result of a long evolutionary history. Although throughoutthe family (Wyllie 1981).Further- the earliest known cuculid fossil dates from the more,some parasitic genera, such as Cuculus and Eo-Oligocene of France (Weigel 1963), some Chrysococcyx,exhibit egg polymorphism, egg workerssuggest that the cuckoosdiverged from mimicry and egg crypsis--traditionallyattrib- ancestralstock during the Late (Sib- uted to an extended coevolutionary relation- ley and Ahlquist 1990). Fundamental differ- 10 January1996] CuckooPhylogeny Based on Behavior ences in external and internal morphology TABLE 1. Classification of the Cuculiformes sensu within the group have perplexed systematists Peters(1940). Obligately parasiticgenera indicated with asterisk(*). Facultatively parasitic genus in- for decadesand, althoughproblematic, the most dicatedwith diamond(0). acceptedclassification of the cuckoosis that of Peters(1940), which is basedpredominantly on Order Cuculiformes breeding habits and geographic distribution Family Musophagidae (Table 1). Several alternate classificationsof the Family Cuculidae Subfamily Cuculinae Cuculidaehave been proposedin past decades Genus Clamator* followingthe anatomicalstudies of Berger(1952, GenusPachycoccyx* 1954, 1960) and Verheyen (1956a), and more Genus Cuculus* recentlyby Sibleyand Monroe (1990)based on Genus Cercococcyx* Genus Penthoceryx* DNA-DNA hybridization of Sibley and Ahlqu- Genus Cacomantis* ist (1990). However, these studies have not Genus Rhamphomantis* gained wide acceptanceand, as a result, most Genus Misocalius* current classifications still adhere to the se- Genus Chrysococcyx* Genus Chalcites* quencein Peters(1940). Phylogeneticsystem- Genus Caliechthrus* atics had not been used to constructa hypoth- Genus Surniculus* esis of evolutionary relationship among the Genus Microdynamis* cuckoosuntil Seibel (1988) addressedthe family GenusEudynamys* in a cladisticanalysis of postcranialosteological Genus Urodynamis* characters.Not surprisingly,his resultsdid not Genus Scythrops* Subfamily Phaenicophaeinae entirely supporttraditional classifications of the GenusCoccyzus• group. Regrettably, this work remains in dis- Genus Hyetornis sertationform only and, hence,has not received Genus Genus Saurothera criticalattention from the scientificcommunity. Genus Ceuthmochares The considerationof behavior and ecology Genus Rhopodytes has often been used to evaluate the evolution- Genus Taccocua ary relationshipsamong birds (e.g. Whitman Genus Rhinortha 1899,Heinroth 1911,Davis 1942,Mayr and Bond Genus Zanclostomus Genus Rhamphococcyx 1943, Tinbergen 1959, Strauch 1985, Prum and Genus Phaenicophaeus Johnson 1987, Prum 1990). More specifically, GenusDasylophus Baker (1927), Bannerman (1933), and Delacour Genus Lepidogrammus (1947) suggestedthat the cuckooscould be sub- Subfamily Crotophaginae Genus Crotophaga divided basedon their breeding habits. In ad- Genus Guira dition, Delacour and Mayr (1946) noted that Subfamily Neomorphinae habitat use may be a valid systematiccharacter Genus Tapera* for determining some degree of relationship Genus Morococcyx within the family. In the present study, ! use GenusDromococcyx* 28 behavioral and ecologicalcharacteristics to Genus Geococcyx Genus reconstructthe evolutionaryrelationships of the GenusCarpococcyx Cuculidaeby phylogeneticsystematics. The re- Subfamily Couinae suiting topologiesdiffer significantly from tra- Genus ditional classifications of the cuckoos in both Subfamily Centropodinae Genus Centropus placement and membership of subfamilies. Basedmy resultsand other evidence,! propose a new hypothesisfor the evolution of brood parasitismin this taxon. broadlyas those representing movement of all or part of the external anatomyof the and, therefore, encompassa numberof functionalcategories such as METHODS socialinteraction, courtship, nest building, egg lay- ing, and incubation. Several ecologicalcharacters Data.--Twenty-eight behavioral and ecological comprisingdiet and habitatuse alsowere included. characterswere used (seeAppendix 1 for character Data were collected for 38 genera of cuckoos,the descriptions).As suggestedby De Queiroz and Wim- turacos(Musophagidae), and the Hoatzin (Opistho- berger (1993), behavioral characterswere defined comushoatzin) through an extensiveliterature search 12 JANICEM. HUGHES [Auk,Vol. 113

TABLE2. Data matrix of behavioral and ecologicalcharacters for Musophagidae,Opisthocomus hoatzin, and 21 generaof Cuculidae(see Appendix 2 for characterdefinitions). Missing data indicatedby "". Genera markedwith asterisks(*) areobligate parasites; genus marked with diamond(¸) is facultativelyparasitic.

Character

Taxon 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Musophagi- dae 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Opisthocomus0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 Clamator* 0 1 1 1 1 1 1 1 1 1 0 3 3 1 0 1 1 1 3 1 0 0 1 0 0 0 1 0 Pachgcoccgx* 0 1 1 0 1 0 3 1 0 1 0 3 3 0 1 1 1 1 3 1 1 1 1 0 0 0 1 0 Cuculus* 0 1 1 1 1 0 2 1 0 0 1 3 3 0 1 1 3 2 3 1 1 1 0 0 0 1 1 0 Cacomantis* 0 1 1 1 0 0 0 1 0 ? ? 3 3 0 1 1 2 2 3 1 1 1 0 0 0 1 1 0 Chrgsococcgx*0 1 1 1 1 0 1 1 0 1 1 3 3 0 1 1 1 2 3 1 1 1 1 0 0 1 1 0 Misocalius* 0 1 1 1 0 0 1 1 0 ? ? 3 3 0 1 1 2 1 3 1 1 1 0 1 0 0 1 0 Chalcites* 0 1 1 1 1 0 1 1 0 1 1 3 3 0 1 1 3 2 3 0 1 1 0 0 0 1 1 0 Eydynamys* 0 1 0 1 0 0 0 1 ? 0 0 3 3 1 0 1 1 1 3 1 1 0 0 0 0 0 1 0 Scythrops* 0 1 0 1 0 0 0 1 0 ? ? 3 3 0 0 1 1 1 3 1 1 0 0 0 0 0 1 0 Cocc!/zus• 0 1 1 1 1 0 0 0 0 0 1 0 1 0 0 1 2 1 0 1 0 0 0 0 0 0 1 0 Piaya 0 1 2 0 0 1 0 0 0 0 ? 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 Saurothera 1 0 3 0 0 0 0 0 ? 0 ? 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Crotophaga 0 0 2 0 0 2 0 0 1 0 0 1 0 0 0 1 0 1 0 0 1 0 0 0 0 0 0 0 Guira 0 0 2 0 1 2 0 0 1 0 0 1 0 0 0 1 0 1 0 0 1 0 0 0 0 0 0 0 Tapera* 1 0 2 0 0 0 0 ? ? 1 1 3 3 0 0 1 2 2 3 0 0 0 0 1 0 0 1 0 Morococcyx 1 0 2 0 0 0 0 ? ? 0 ? 1, 2 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 Neomorphus 1 0 3 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 Geococcyx 1 0 3 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 1 0 3 0 0 0 0 0 ? 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Coua 1 0 3 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Centropus 1 0 3 0 1 1 2 0 1 0 1 2 1 0 0 0 0 0 2 0 0 0 1 0 0 0 0 1

(Appendix 2). Many cuculid speciesdemonstrate a priate characterswere coded in one of two ways ac- high degreeof endemismand often are secretivein cording to Maddison (1993). Binary charactersinap- nature.As a result,they may be poorly studied,and propriate to nonparasiticgenera, such as laying mi- the available literature is lacking in accuratedescrip- metic (character 20), were assignedthe pleiso- tionsof their behaviorand naturalhistory. Therefore, morphicstate to restrainthe algorithmfrom selecting as suggestedby Prum (1990),I have deletedtaxa for a stateapplicable only to a parasiticgenus. Multistate which there is insufficientinformation (considering charactersthat were not applicableto both breeding 20%or more missingdata adequate grounds for omis- strategies,such as host specialization(character 17) sion). This resultedin the exclusionof the following and nestarchitecture (character 12) were assignedan 17 genera; Caliechthrus,Cercococcyx, Ceuthmochares, additional state (e.g. no host usage,does not build Dasylophus,, Lepidogrammus, Microdynamis, nest), that would allow the inclusion of all taxa. Penthoceryx,Phaenicophaeus, Rhamphococcyx, Rhampho- Outgroups.--Characterswere polarized into primi- mantis,Rhinortha, Rhodopytes, Surniculus, Taccocua, Uro- tive and derivedstates through outgroup comparison dynamis,and Zanclostomus.Despite the exclusionof with the turacos(Musophagidae) and the Hoatzin. thesetaxa, all six cuculid subfamiliesrecognized by The tufacostraditionally have been consideredthe Peters(1940) are represented in the analysisby at least sistertaxon to the Cuculidae(e.g. Pycraft1903, Steg- three genera,except those which are monotypic(e.g. mann 1978, Cracraft 1981) and are commonlyclassi- Centropodinaeand Couinae)or ditypic (Crotophag- fied with the cuckoosin the Cuculiformes(e.g. Peters inae). 1940,Howard and Moore 1991). However, DNA-DNA The analysisincluded 20 binary charactersand 8 hybridization evidenceof Sibley and Ahlquist (1990) nonadditive multistate characters (Table 2). Un- suggeststhat thesetwo taxa may not be one another's known charactersfor particular taxa were coded as closestliving relatives. Furthermore,they propose missing (?). Coding inappropriate characterswas that the Hoatzin, often classified in the Galliformes problematic.Because many of the behaviorsconsid- (e.g. Stresemann1934, Verheyen 1961,Cracraft 1981), ered in my study are linked to breeding strategy, is a cuckoothat is most closely related to the com- severalwere not applicableto either nonparasiticor munally breeding anis (Crotophaginae;Sibley and parasiticgenera. Wherever possible,characters were Ahlquist 1972, 1973, 1990). This conclusionhas been defined to avoid exclusion of genera solely due to questionedby Bock(1992), who indicated that fun- breedingstrategy (e.g. character15 and 16). Inappro- damentaldifferences in foot morphologyshould ex- January1996] CuckooPhylogeny Based on Behavior 13 clude the Hoatzin from the cuckoos.Although not branch swapping,and simple stepwiseaddition. Fol- unequivocal,the DNA sequencingresults of Avise et lowing a visual comparisonof the resulting trees,the al. (1994) suggestthat the Hoatzin and the cuckoos behavioraland ecologicaldata were constraintedon may indeed be sister taxa, but do not support the the optimal osteologicaltrees to determine the num- inclusion of the Hoatzin within the small clade of ber of additional stepsrequired to reproducethe to- seven cuckoosexamined in their study. Despite the pologies.A total-evidenceanalysis (heuristic, ACCT- controversy, the alliance of the Hoatzin with the RAN, TBR, simple stepwiseaddition) was performed cuckoosin some capacityjustifies its selectionas an by combining the osteological,behavioral, and eco- outgroup. logical data (total = 76 characters).Since there were Phylogeneticanalyses.--All phylogenetic analyses nearly twice as many osteologicalcharacters as there were performed using PAUP version 3.1 (Phyloge- were behavioral and ecological charactersincluded netic Analysis Using Parsimony;Swofford 1993) on in this data set, the latter were given a weight of 2 an Apple MacintoshQuadra 660AV. An initial branch- to equalizethe influenceeach type of datawould have and-boundsearch proved too time consuming. There- on the resulting topologies. fore, I employedthe heuristic-searchoption, as sug- gestedby Swofford (1993) for usewith data setslarger RESULTS than 20 taxaand with moderatelevels of homoplasy. To find the optimal tree, 36 heuristic searcheswere The phylogenetic analysisof behavioral and performed exhaustingall possiblecombinations of ecologicalcharacters yielded one shortest-length the following searchoptions: (1, optimization) accel- tree of 80 steps(CI = 0.52, RI = 0.79, RC = 0.42; eratedtransformation (ACCTRAN), delayedtransfor- Fig. 1). This tree resulted from 24 heuristic mation (DELTRAN), and minimum F-value (MINF); searchesusing all optimization and stepwise (2, branch swapping) tree bisection-reconnection addition options with the TBR and SPR algo- (TBR),subtree pruning-regrafting (SPR), and nearest- rithms. Only four charactersshowed no ho- neighbor interchanges(NNI), and (3, stepwisead- dition sequence)simple, closest,as-is, and random moplasy:characters 8 (sunbathing),16 (>1 fe- with 10 replicalions. The MULPARS option was in male), 27 (incubation period), and 28 (vocal effect. MAXTREES was set to 1,000 trees with auto- duetting). Four charactershad high levels of matic increase.Zero-length brancheswere collapsed homoplasy,with consistencyindices equal or to yield polytomies.The consistencyindex (CI), re- tention index (RI), and rescaled consistencyindex Musophagidae (RC) were calculated.Suboptimal trees were calcu- Piaya lated by "keeping" all trees a specified number of $aurothera stepslonger than the minimum length tree. Coua Due to the distinct behavioral dimorphism of the Carpococcyx Cuculidae (parasiticvs. nonparasiticgenera), it was Centropus difficult to selectand code enough charactersinde- Morococcyx pendentof brood parasitismto producewell-resolved trees.Despite my efforts to avoid nonindependence, Neomorphus 12 charactersremained in the initial analysisthat were Geococcyx partially or fully dependent on parasitic behavior. Opisthocomus Therefore,to verify that the resulting cladeswere not Crotophaga merely due to clustering of nonindependent char- Guira acters,a secondanalysis (heuristic, ACCTRAN opti- Tapera mization, TBR branch swapping, and simple stepwise Coccyzus addition sequence)was performed including only Eudynamys thosecharacters that were fully independentof brood Scythrops parasitism (characters1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 23, Misocalius

25, 26, 27, 28). Cacomantis Finally, the behavioral and ecologicaldata were Cuculus comparedto the postcranialosteological data of Seibel Chalcites (1988). Taxa not included in both analyseswere de- leted from the data setsin order to facilitate compar- Chrysococcyx ison. Nine generawere removedfrom the osteolog- Clamator ical data matrix: Ceuthmochares,, Dromococcyx, • Pachycoccyx Hyetornis,Penthoceryx, Phaenicophaeus, Rhinortha, Sur- Fig. 1. Optimaltree based on behavioraland eco- niculus,and Urodynamis.Opisthocomus hoatzin was de- logical data (L = 80 steps,CI = 0.52, RI = 0.79, RC = leted from the behavioraland ecologicaldata set.Op- 0.42). Internal nodes are numbered. See Appendix 3 timal trees for the osteologicaldata were estimated for character-changelist correspondingto internal usinga heuristicsearch, ACCTRAN optimization,TBR nodes. 14 Jamc•M. HUGHES [Auk, Vol. 113 less than 0.33: characters5 (pair bond), 9 (al- itself. The position of the facultative parasite lopreening), 11 (song posts),and 21 (egg re- Coccyzuswithin a cladeof obligateparasites im- moval). SeeAppendix 3 for a character-change plies that this genusis exhibitinga lossof ob- list, as well as characterconsistency and reten- ligate parasitism.Three additionalsteps would tion indices.The heuristicNNI algorithmfound be requiredto reversethe positionsof Coccyzus the optimal tree (MIN-F, random), as well as and Tapera;a topologythat would supportthe eight treesof 81 steps(CI = 0.52, RI = 0.79, RC hypothesis that facultative parasitism repre- = 0.41) that differed only in the arrangement sentsan intermediatestage in the evolution of of taxawithin the major cladeof parasiticcuck- obligate brood parasitismin the cuckoos. oos (Clamator,Pachycoccyx, Cuculus, Cacomantis, The resultsof my secondanalysis, including Chrysococcyx,Chalcites, and Misocalius). only those charactersindependent of brood The treesbased on behavioraland ecological parasitism,upheld the strength of the Tapera- charactersdiffer significantly from traditional Coccyzus-Cuculinaeclade. As in the first anal- classificationsof the cuckoos.Most importantly, ysis,Coccyzus clustered among the obligatepar- two cuculid subfamilies sensuPeters (1940) were asiteswhen 12 characterspartially or fully de- severed.The Neomorphinae(previously com- pendent on brood parasitismwere deletedfrom prised of Geococcyx,Neomorphus, Carpococcyx, the data set (16 trees of 42 steps;CI = 0.50, RI Morococcyx,Dromococcyx, and Tapera)was di- = 0.78, RC = 0.39). Taperaoccupied a position vided into two groups. Morococcyx,Geococcyx, at the baseof the neomorphine-Saurothera-Coua and Neomorphusform a cladewith Centropusand cladeon all shortest-lengthtrees, but could be Carpococcyx.However, the neomorphine obli- returned to its basalposition in the Cuculinae gatebrood parasite Tapera is positionedbasally with the addition of only one step.By deleting in a cladecomprised of the obligatelyparasitic character1 (generalizedhabitat use), the pri- Cuculinae and the facultative-parasiticgenus mary characteristictraditionally usedto define Coccyzus.Five additional steps would be re- the neomorphine cuckoos,Tapera resumed its quired to unite Tapera with the nonparasitic position in the Cuculinae(156 treesof 40 steps; neomorphineclade. This suggeststhat terres- CI = 0.50, RI = 0.77, RC = 0.39) as depicted on trial habitat use by Taperahas evolved second- the optimal tree. arily. On the shortest-lengthtree (L = 80 steps),the The Phaenicophaeinae,represented in my Hoatzin is the sistertaxon to the crotophagine analysisby Saurothera,Coccyzus, and Piaya,has cuckoos,Crotophaga and Guira,despite its selec- been partitioned into three parts. Saurothera tion asan outgroup.This positionis maintained formsa cladewith Couathat is the sistergroup on 17 of 22 suboptimaltrees with a length of to the nonparasiticneomorphine cuckoos and 81 steps. The remaining five trees place the Centropus.Coccyzus is clusteredamong the par- Hoatzin assister taxon to a cladecontaining the asitic Cuculinae.Piaya occupiesthe basalposi- crotophagineand cuculine cuckoos.Two ad- tion on the trees.Seven and 10 additional steps ditional steps are required to constrain the are needed to join Coccyzuswith the other Hoatzin to the outgroup.Deleting the Hoatzin phaenicophaeinecuckoos, Piaya and Saurothera,from the analysis,using only the Musophagidae respectively.If these topologiesare accepted, as the outgroup taxon, does not change the the Neomorphinaeand Phaenicophaeinaesen- overalltopology of optimaltrees (three trees of su Peters(1940) are not monophyletic. 79 steps;CI = 0.53, RI = 0.78, RC = 0.42). How- In addition, my results suggestthat brood ever, there is a minimal loss of resolution with- parasitismevolved only once in the cuckoos, in neomorphinecuckoos. Using the Hoatzin as rather than three times as advocatedby tradi- the sole outgroup doesnot alter relationships tional classifications.In my analysis,the Tapera- within the majorclades, but merely causesthe Coccyzus-Cuculinaeclade is the bestsupported crotophaginecuckoos to becomethe basalclade clade,being united by sixcharacter-state changes on the shortest-lengthtopologies (12 trees of on the optimal tree. All suboptimaltrees ex- 79 steps;CI = 0.53, RI = 0.79, RC = 0.42). amined-one steplonger (81 steps;n = 22) and The heuristic reanalysisof Seibel's(1988) 48 two stepslonger (82 steps;n = 359) than the postcranialosteology characters resulted in 16 shortest tree--supported the position of Coc- equallyparsimonius trees that varied primarily cyzusand Taperaamong the Cuculinae,despite in the arrangementof genera within the Cu- the loss of resolution within the cuculine clade culinae (L = 66, CI = 0.86, RI = 0.94, RC = 0.81). January1996] CuckooPhylogeny Based on Behavior 15

Musophagidae Musophagidae Crotophaga Piaya Guira Saurothera Centropus Ooua Carpococcyx Carpococcyx Ooua Centropus • Morococcyx • Morococcyx Neomorphus Neomorphus Geococcyx Geococcyx Crotophaga i PiayaSaurothera c Guira Eudynamys Tapera Scythrops Coccyzus Tapera Eudynarnys Clamator Scythrops Coccyzus Clamator • Pachycoccyx Pachycoccyx • Cuculus Chrysococcyx

• Cacomantis Cuculus

• Chrysococcyx Chalcites Misocalius Cacomantis

Chalcites Misocalius

Fig. 2. Strict-consensustree of 16 equally parsi- Fig. 3. Strict-consensustree of three equally par- monius trees basedon postcranialosteological char- simoniustrees basedon combinedosteological, be- acters (after Seibel 1988; L = 66, CI = 0.86, RI = 0.94, havioral and ecologicaldata (CI = 0.63, RI = 0.82, RC RC = 0.81). = 0.52).

Figure 2 illustratesa strict-consensustree of the behavioraland ecologicalcharacters to Seibel's resultingtopologies. Perhaps the moststriking data provided more resolutionwithin the cu- similarity between Seibel'strees and those de- culine and neomorphineclades. A strict-con- rived from the behavioral and ecologicaldata sensus tree for the combined data is illustrated is the placementof Coccyzusand Taperawithin in Figure 3. the Cuculinae.Also of importanceis the inclu- DISCUSSION sion of Centropusand Couain the clade contain- ing the neomorphine cuckoos.In traditional The classification of the Cuculidae has re- classifications,both Centropusand Coua form mained virtually unchanged for decades,with monotypic subfamilies(see Table 1). Forcing taxonomiesbased on Peters (1940) still being the behavioral and ecological charactersonto widely accepted.However, the advent of phy- the topologyof Seibel'soptimal trees requires logenetic systematicshas allowed the recon- 19 additional steps. However, 12 of these ad- structionof phylogeniesusing a robustmeth- ditional stepsserve merely to provide resolu- odologyunavailable to systematistslike Peters tion within the cuculine and neomorphine and others.Therefore, it is imperative that it be cladesthat may not be presentin the behavioral usednot only in the teevaluationof traditional data if suboptimaltrees (L = 81 steps)are ex- classifications,but also in the proposalof new amined. Combining the behavioraland ecolog- hypothesesof evolution(Brooks and McLennan ical data with the osteologicaldata resulted in 1991, Harvey and Pagel 1991). My reconstruc- three equally parsimoniustrees (CI = 0.63, RI tion of the phylogeny of the cuckoosdiffers = 0.82, RC = 0.52). Although, the CI, RI and RC from many traditional classificationsin two for the combined data was lower than that of fundamentalways--the polyphyly of the Neo- the osteologicaldata alone, the addition of the morphinaeand of the Phaenicophaeinae. 16 JANICEM. HUGHES [Auk,Vol. 113

The Neomorphinae.--Traditionally,the Neo- era to be unlike that of Geococcyx,Morococcyx, morphinaeare comprisedof five generaof New and Carpococcyx.In addition, Seibel (1988) found World cuckoosand one genus of Old World that Taperaand Dromococcyxdiffered from the cuckoos(Peters 1940; see Table 1), and contains nonparasiticneomorphine genera in 21 of 48 both parasitic and nonparasitic species. Al- postcranialosteological characters. though this diverse subfamily differs signifi- Upon examining nonparasitic terrestrial cantlyin life historyand externalmorphology, cuckoos,Shufeldt (1886) placed Geococcyxand in the pastthey have been groupeddue to their Centropusin the samesubfamily based on ana- distribution and their terrestrial habits (Berger tomicalcharacters. Berger (1960) noted similar- 1960). Basedon my results,I recommendthe ities in the internal anatomyof Morococcyxand partitioning of the Neomorphinae into two Coua.Although Couatraditionally has occupied groups.The subfamily Neomorphinaeshould a monotypicsub family, Berger(1960) found lit- be comprised of the traditional nonparasitic tle reasonfor this, suggestingsuch a designa- neomorphine genera Carpococcyx,Geococcyx, tion likely wasdue to the restrictionof the nine- Morococcyx,Neomorphus plus Centropus.The po- speciesgenus to Madagascar.Seibel's (1988) re- sition of Coua is less clear. Paired with Sauroth- vised classificationof the Cuculidae presented era,it mayrepresent the sistertaxon to the Neo- a new clade, BranchCentropodes, comprised of morphinae (Fig. 1). Alternatively, it could be Centropus,Coua and the nonparasiticneomor- includedwithin the Neomorphinae(Figs. 2 and phine cuckoos.Shelley (1891) placedCoua and 3). The obligateparasite Tapera should be moved Saurotherain the samesubfamily. to the Cuculinae, the large subfamily that in- The Phaenicophaeinae.--Peters(1940) was dis- cludes all 47 speciesof parasitic Old World satisfied with his classification of the Phaeni- cuckoos.Dromococcyx, another genus of para- cophaeninae,calling it a "catch-all"(Berger 1960: sitic New World cuckoos,traditionally seen as 94) for genera that could not be allocated into the sistergenus of Tapera(e.g. Sclater and Salvin other subfamilies.This is evidencedby the high 1873,Shelley 1891, Verheyen 1956a), also should morphological,behavioral, and distributional be included in the Cuculinae basedon the post- diversity among this group of Old World and cranialosteological characters it shareswith that New World cuckoos.In a study of nine phaen- group(Seibel 1988). This removalof Taperaand icophaeinegenera, Berger (1960) found signif- Dromococcyxfrom the Neomorphinaesuggests icant differencesin pterylosis,appendicular that obligate brood parasitismarose only once muscular,and skeletal elements.Based on my amongthe Cuculidaerather than twice as pro- analysis,I suggestthe subdividingof Peters' posedby Peters(1940). In addition, the poly- Phaenicophaeinaeinto at least two groups by phyly of the Neomorphinaedemonstrates that transferring Coccyzusto the Cuculinae. terrestrial habitat use is a derived state among In the past, Coccyzuswas classified in the the cuckoos that has evolved at least twice in Phaenicophaeinaeprimarily on the basisof ex- the family. ternal morphology,distribution, and nonobli- These modifications to the classification of gate parasitic nesting habits. However, many the neomorphinecuckoos are supportedin part systematistshave questionedthis association. by the literature. Prior to Peters(1940), many Beddard(1885) placedboth Coccyzusand Piaya systematistshad suggestedthe severanceof the in the Cuculinaebased on similaritiesin syrinx, parasiticneomorphine cuckoos from the re- musculature,and pterylosis.Shufeldt (1886)and mainder of the group (Linnaeus 1766, Sclater Shelley (1891) also supported the inclusion of and Salvin 1873, Beddard 1885, Shelley 1891, Coccyzusin the Cuculinae. Pycraft (1903) Gadow and Selenka 1891). Beddard (1885) and groupedCoccyzus with the obligatelyparasitic Gadow and Selenka (1891) classified Taperaas Cuculus,Eudynamys, and Scythropsbased on his being within the Cuculinae based on muscu- study of sternum configurationand appendic- lature, pterylosis,and configurationof the syr- ular musculature.In reviewing the musculature inx. More recently, Verheyen (1956a) recog- of Coccyzus,Berger (1952) concluded that this nized a number of anatomical differences be- genus should not be included in the Phaeni- tween parasiticand nonparasiticneomorphine cophaeinae.Verheyen (1956a) erected a new cuckoos and, as a result, erected a new subfam- subfamilyfor Coccyzusand two genera of ob- ily to contain Taperaand Dromococcyx.Berger ligatelyparasitic cuckoos (Clamator and Pachy- (1960) found the appendicularmuscles of Tap- coccyx),raising the parasiticcuckoos to the rank January1996] CuckooPhylogeny Based on Behavior 17 of suborder based on similarities in their anat- 1943,Hamilton and Hamilton 1965). This gen- omy. Seibel(1988) found that Coccyzusdiffered erally is consideredto be an adaptationfor suc- from other phaenicophaeinaecuckoos by as cessfulparasitism since it allows the parasitic many as 11 osteologicalcharacters. chick to hatch first and, therefore, gain a size Theevolution of broodparasitism in thecuckoos.- advantageover the host young (Hamilton and The placementof Coccyzusand Taperawithin a Orians 1965,Payne 1977).Consequently, para- clade of obligateparasites suggests a new hy- sitic cuckooshave incubation periodsbetween pothesisfor the evolution of brood parasitism 10 and 15 days,with nonparasiticcuckoos hav- in the Cuculidae. In traditional classifications, ing incubation periods greater than 15 days. parasitismwas thought to have evolved three Also,Coccyzus chicks have a very shortnestling times:obligate parasitism in the Neomorphinae period of only sevento nine days.In contrast, and Cuculinae,and facultativeparasitism in the the nestlings of many nonparasitic cuckoos Phaenicophaeinae.This in itself is counterin- fledgeafter about18 to 20 daysin the nest(Wyl- tuitive sinceit is unlikely that sorare a behavior lie 1981).Although early fledginglikely would would originatemany timesin one family. Only reducesurvival in a nonparasiticspecies, it could about 1% of all avian speciesare obligate par- increasesurvival in a nestlingthat is beingcared asites (Payne 1977). If the conclusionsof my for by host parents,since it lessensthe time analysis are accepted,the facultative parasite during which recognitionand rejectionof the Coccyzusshared an ancestorwith the Cuculinae foreign offspring may occur.In addition, many that wasan obligateparasite and, therefore,must parasiticspecies breed later in the seasonthan be demonstratinga lossof obligatelyparasitic nonparasitespresumably to ensurea supplyof habitsrather than de novo developmentof par- clutchesin varyingstages of incubationin which asitism from a nonparasitic ancestor. Several to depositan egg without detection(Hamilton studieshave illustratedthe propensityof Coc- and Orians 1965). Late breeding may also add cyzusfor occasionalintraspeciflc and interspe- stabilityto the parasite-hostrelationship by al- cific parasitism(Nolan and Thompson 1975, lowing the host to successfullyraise their first Ralph 1975, Fleischer et al. 1985, Sick 1993). broodunaffected by parasitism(May and Rob- Although their most common hosts are con- inson 1985).Although the North Americanspe- specificor congeneric,the Yellow-billed Cuck- cies of Coccyzuswinter in the tropics, their oo (Coccyzusamericanus) and the Black-billed breedingseason does not generallybegin until Cuckoo(C. erythropthalmus)have parasitizedat June and extends into September(Nolan and least 13 speciesof North American birds, in- Thompson1975, Cadman et al. 1987),long after cluding the American Robin (Turdusmigrato- many migratory specieshave departed the rius), Northern Cardinal (Cardinaluscardinalus), breedinggrounds. In addition,the obligatepar- Gray Catbird (Dumetellacarolinensis), and Wood asites exhibit a constant readiness to breed Thrush (Hylocichlamustelina; Darwin 1859, At- (within season)to exogeneousstimuli, such as twater 1892, McIlwraith 1894, Bendire 1895, hostavailability. Similarly, the onsetof breed- Herrick 1910, Forbush 1927, Bent 1940, Sprunt ing in Coccyzusappears to be regulatedexter- and Chamberlain 1949, Nolan and Thompson nally by resourceavailability rather than by en- 1975). dogeneouscues (Hamilton and Hamilton 1965, Unlike nonparasiticcuckoos, Coccyzus shares Ralph 1975).In parasiticcuckoos, egg mimicry many life-history traits with the obligate par- is a commonadaptation to hostdefenses. When asitesof the Cuculinae that are adaptive to a parasitizinginterspecifically, Coccyzus spp. may parasiticlifestyle. Kendeigh (1952), after Her- lay mimetic eggs(unpubl. data).Since it is un- rick (1910), noted that some cuckoosexhibited likely that egg mimicry would evolve in a gen- a characteristic"disassociation" of egg laying erally nonparasiticspecies that only rarely lays from the "normal" nesting sequenceof court- its eggsin otherbirds' nests (Davies and Brooke ship, nest building, egg laying, incubation,and 1988, 1989, Rothstein 1990), egg mimicry by care of young. In Coccyzus,this disassociation Coccyzuscould be an artifactof an intense,and is manifestedin the laying of eggsat irregular perhapsobligate, relationship that onceexisted intervals,or before the nestis completed(Spen- betweenthese parasites and their hosts.There- cer 1943,Ralph 1975,Potter 1980). Like the ob- fore, theselife-history traitsthat Coccyzusshares ligate parasites,Coccyzus has an extremelyshort with the Cuculinae are not preadaptationsto a incubation period of 10 or 11 days (Spencer future obligately parasiticlife style (Hamilton 18 JANICEM. HUGHF• [Auk, Vol. 113 and Orians 1965), but the selectively neutral AMBROSE,S.J. 1987. Adult Fan-tailed Cuckoo (Cu- artifactsof an ancestralbreeding strategy. culuspyrrhophanus) feeds fledgling. Emu 87:69. The Hoatzin.--The taxonomic position of the ATTW^TER, H. P. 1892. List of birds observed in the Hoatzin has perplexed systematistsfor more vicinity of San Antonio, Bexar County, Texas. Auk 9:229-238. than two centuries. In an excellent review of its classification,Sibley and Ahlquist (1990) not- AvISE, J. C., W. S. NELSOIq,AND C. G. SIBLE¾. 1994. Why one-kilobasesequences from mitochondrial ed that the systematistshad allied the Hoatzin DNA fail to solvethe Hoatzin phylogeneticenig- with the Galliformes in 17, the tufacos in 4, and ma. Mol. Phylogen. Evol. 3:175-184. the cuckoos in 8 classifications. In addition, the BAKER,E. C. S. 1927. The fauna of British India, Hoatzin has been placed in a monotypic order including Ceylon and Burma.Birds, vol. 4. Taylor 12 times.They suggestedthat many studiesclas- and Francis, London. sifying the Hoatzin with the Galliformeshave BANNERMAN,D.A. 1933. The birds of tropical West been biased by the original description of the Africa, with specialreference to thoseof Gambia, species(Milllet 1776) and the blind adherence SierraLeone, the Gold Coastand Nigeria. Crown to tradition by the systematiststhat followed Agentsfor the ColoniesPublishers, London. (e.g. Huxley 1867, Gadow 1893, Peters 1934, BEDDARD, F. E. 1885. On the structural characters Howard 1950, Cracraft 1981). However, in the and classification of the cuckoos. Proc. Zool. Soc. Lond. 1885:168-187. past 40 years, several molecular and morpho- logical studieshave placedthe Hoatzin outside B•E, C. W. 1909. A contributionto the ecologyof of the Galliformes(Verheyen 1956b,Hudson et the adult Hoatzin. Zoologica 1:45-66. BELL,H. L. 1986. The participation by cuckoosin al. 1959, Streseman1965, Sibley and Ahlquist mixed-speciesflocks of insectivorousbirds in 1972, 1973, 1990, De Queiroz and Good 1988). south-eastern Australia. Emu 86:249-253. Like Sibley and Ahlquist (1972, 1973, 1990),my BEh•DIP•,C.E. 1895. Life histories of North American results include the Hoatzin in a clade with the birds. U.S. Natl. Mus. Spec.Bull. 3. communally breeding crotophagine cuckoos BEIqT,A. C. 1940. Life histories of North American (Crotophagaand Guira).Beebe (1909) and others cuckoos,goatsuckers, hummingbirds, and their have noted the behavioral similarities between allies. Bull. U.S. Natl. Mus. 176. the Hoatzin and the Crotophaginae.Although BERGER,A.J. 1952. The comparativefunctional mor- my results support the Hoatzin's inclusion in phologyof the pelvicappendage in threegenera the Cuculidae, it is possiblethat the characters of Cuculidae. Am. Midl. Natl. 47:513-605. aligning the Hoatzin with the cuckoosmay be BERGER,A.J. 1954. The royologyof the pectoralap- convergentadaptations to communalbreeding pendage of three genera of American cuckoos Univ. Mich. Zool. Misc. Publ. 85:1-35. rather than synapomorphiesof the clade. Fur- ther molecularand morphologicalanalyses are BERGER,A. J. 1960. Some anatomical characters of the Cuculidae and the Musophagidae. Wilson neededto resolvethis enigma. Bull. 72:60-104. Boot, W.J. 1992. Methodology in avian systematics. ACKNOWLEDGMENTS Bull. Br. Ornithol. Club Centenary Supp1.112A: 53-72. I am grateful to D. A. McLennan and M. E. Siddall BOh•D,J. 1960. Birdsof the West Indes. Collins Pub- for their helpful commentson earlier drafts of this lishers, London. manuscript.I also thank A. De Queiroz, P. Houde, S. M. Lanyon, and G. D. Schnell for their constructive BROOm, L. C., AND M. G. BROO•CER.1990. Why are criticism, and J. C. Barlow for his editorial assistance. cuckooshost specific? Oikos 57:301-309. This work was supportedby a fellowship to me from BROOm, M. G., AND L. C. BROOm. 1989. The com- the Natural Sciences and Engineering Research parative breeding behavior of two syrupattic Council.In addition,some aspects of this studywere cuckoos,Horsfield's Bronze-Cuckoo Chrysococcyx funded by NSERC grant A3472 to J.C.B.I dedicate basalisand the Shining Bronze-CuckooC. lucidus, this manuscriptto the memory of my father, Arthur in Western Australia: A new model for the evo- Edward Hughes, the man who always knew what it lution of egg morphologyand hostspecificity in took me 10 years to discover. arian brood parasities.Ibis 131:528-547. BROOm, M. G., AND L. C. BROOm. 1992. Evidence

LITERATURE CITED for individual female hostspecificity in two Aus- tralian bronze-cuckoos(Chrysococcyx spp.). Aust. ALI, S., AND S. D. RIPLE•. 1987. Compact handbook J. Zool. 40:485-493. of the birds of India and Pakistan. Oxford Univ. BROOKS,D. R., ANDD. A. MCLFJqNAN.1991. Phylog- Press, London. eny, ecology,and behavior:A researchprogram January1996] CuckooPhylogeny Based on Behavior 19

in behavioral biology. Univ. Chicago Press,Chi- ecologyof (Geococcyx californianus) cago. in South Texas. Southwest. Natl. 23:1-28. BROWN,J.L. 1987. Helping and communalbreeding FORBUSH, E. H. 1927. Birds of Massachusetts and in birds: Ecologyand evolution. Princeton Univ. other New England states,vol 2. Massachusetts Press,Princeton, New Jersey. Dept. Agriculture, Boston,Massachusetts. CADMAN,M. D., P. F. J. EAGLES,AND F. M. HELLEINER. FRIEDMANN,H. 1928. The origin of host specificity 1987. Atlas of breeding birds of Ontario. Univ. in the parasitic habit in the Cuculidae. Auk 45: Waterloo Press, Waterloo, Ontario. 33-38. COATES,B.J. 1985. The birds of PapuaNew Guinea. FRIEDMANN,H. 1964. Evolutionary trends in the avi- VoL 1, Nonpasserines.Dove Publications,Ad- an genus Clamator.Smithson. Misc. Coil. 146. dericy, Australia. FRIEDMANN,H. 1968. The evolutionary history of COATES,B.J. 1990. The birds of Papua New Guinea. the avian genusChrysococcyx. Bull. U.S. Natl. Mus. Vol. 2, .Dove Publications,Adderley, 265. Australia. FRY, C. H., S. KEITH, AND E. K. URBAN. 1988. The CRACRAFT,J. 1981. Toward a phylogenetic classifi- birds of Africa, vol. 3. Academic Press, London. cation of the Recent birds of the world (class GADOW,H. 1893. V6gel. II. SystematischerTheil. In Aves). Auk 98:681-714. Bronn'sKlassen und Ordnungen desTheir-Reichs CRAMP,S. 1985. Handbook of the birds of Europe, (H. G. Bronn, Ed.), vol. 6. C. F. Winter, Leipzig. the Middle East and North Africa. The birds of GADOW,H., AND E. SELENKA.1891. V6gel. I. Ana- the West Palearctic.Vol. 4, Terns to woodpeckers. tomischer Theil. In Bronn's Klassen und Ord- Oxford Univ. Press, London. nungeL des Their-Reichs(H. G. Bronn, Ed.), vol. DARWIN,C. 1859. On the origin of speciesby means 6. C. F. Winter, Leipzig. of naturalselection, 1st ed. JohnMurray, London. GILL, B. J. 1983. Brood parasitismby the Shining DAVIES,N. B., AND M. DE L. BROOKE.1988. Cuckoo Cuckoo Chrysococcyxlucidus at Kaikoura, New versusreed warblers: Adaptationsand counter- Zealand. Ibis 125:40-55. adaptations.Anim. Behav.36:262-284. GINN, P. J., W. G. MCILLERON,AND P. LE S. MILSTEIN. DAVIES,N. B., AND M. DE L. BROOKE.1989. An ex- 1989. The completebook of South African birds. perimental study of coevolutionbetween the Struik Winchester Publishers,Cape Town, South cuckoo, Cuculuscanorus, and its hosts.I. Host egg Africa. discrimination. J. Anim. Ecol. 58:207-224. GOS?ER,D. 1964. Observationson the breeding of DAVIS,D.E. 1942. The phylogenyof socialnesting the Koel. Emu 64:39-41. habitsin the Crotophaginae.Q. Rev.Biol. 17:115- GRIMMER,J.L. 1962. Strangelittle world of the Hoat- 134. zin. Natl. Geog. Mag. 122:391-401. DELACOUR,J. 1947. Birdsof Malaysia. Macmillan Co., HAFFER,J. 1977. A systematicreview of the Neo- New York. tropical ground-cuckoos (Aves, Neomorphus). DELACOLIR,J. T., AND P. JABOLIILLE.1931. Les Oiseaux Bonn. Zool. Beitr. 28:48-77. de l'Indochine Francaise,vol 2. Exposition Co- HAMILTON, W. J., III, AND M. E. HAMILTON. 1965. loniale Internationale, Paris. Breeding characteristicsof Yellow-billed Cuck- DELACOLIR,J. T., AND E. MAYR. 1946. Birds of the oos in Arizona. Proc. Calif. Acad. Sci. 32:405-432. Philippines. Macmillan Co., New York. HAMILTON,W. J., AND G. H. ORIANS. 1965. Evolution DE QuEIROZ, A., AND P. H. WIMBERGER. 1993. The of brood parasitismin altricial birds. Condor 67: usefulnessof behaviorfor phylogenyestimation: 361-82. Levelsof homoplasyin behavioraland morpho- HARVEY, P. H., AND M.D. PAGEL. 1991. The com- logical characters.Evolution 47:46-60. parativemethod in evolutionarybiology. Oxford DE QLIEIROZ,K., AND D. A. GOOD. 1988. The scleral Univ. Press, Oxford. ossiclesof Opisthocomusand their phylogenetic HAVE•$CHMIDT,F. 1961. Der Kuckuck Taperanaevia significance.Auk 105:29-35. und seine Wirte in Surinam. J. Ornithol. 102:353- EHRLICH,P. R., D. S. DOBKIN, AND D. WHEYE. 1987. 359. The birder's handbook. Simon and Schuster, Inc., HAVERSCHMIDT, F. 1968. Birds of Surinam. Oliver New York. and Boyd Publishers,Edinburgh. FFRENCH,R. 1973. A guide to the birds of Trinidad HEINROTH,O. 1911. Beltragezur Biologic,Nametlich and Tobago.Livingston Publishing Co. Wynne- Ethologie und Psychologieder Anatiden. Verh. wood, Pennsylvania. Ver. Int. Ornithol. Kongr. (Berlin) 1910:589-702. FLEISCHER,R. C., M. T. MURPHY, AND L. E. HUNT. 1985. HERKLOTS,G. A. C. 1965. Birds of Trinidad and To- Clutch size increaseand intraspecificbrood par- bago. Collins, London. asitism in the Yellow-billed Cuckoo. Wilson Bull. HERRICK,F.H. 1910. Life and behavior of the cuckoo. 97:125-127. J. Exp. Zool. 9:169-233. FOLSE,L. J., AND K. A. ARNOLD. 1978. Population HILTY,S. L., ANDW. L. BROWN.1986. A gu.ide to the 20 JANICEM. HUGHES [Auk, Vol. 113

birds of Columbia. Princeton Univ. Press, Prince- adaptations of the . Condor 75:140- ton, New Jersey. 149. HOWARD, H. 1950. Fossil evidence of avian evolu- OLIVER, W. R. B. 1955. New Zealand birds. A. W. tion. Ibis 92:1-21. Reed Publishers,Wellington. HOWARD,R., AND A. MOORE. 1991. A complete PAYNE,R. B. 1973. Individual laying historiesand checklist of birds of the world, 2nd ed. Academic the clutch size and numbers of eggsof parasitic Press, London. cuckoos. Condor 75:414-438. HUDSON, W. H. 1920. Birds of the La Plata. J. M. PAYNE,R. B. 1977. The ecology of brood parasitism Dent and Sons, New York. in birds. Annu. Rev. Ecol. Syst. 8:1-28. HUDSON, G. E., P. J. LANZILLOTTI,AND G. D. EDWARDS. PETERS,J. L. 1934. Checklist of birds of the world, 1959. Muscles of the pelvic limb in galliform vol. 2. Museum of Comparative Zoology, Cam- birds. Am. Midl. Natl. 61:1-67. bridge, Massachusetts. HUXLEY, T. H. 1867. On the classification of birds; PETERS,J. L. 1940. Checklist of birds of the world, and on the taxonomic value of the modifications vol. 4. Museum of Comparative Zoology, Cam- of certain of the cranial bones observable in that bridge, Massachusetts. class. Proc. Zool. Soc. Lond. 1867:415-472. POTTER,E.F. 1980. Notes on nesting Yellow-billed KENDEIGH,S.C. 1952. Parental care and its evolution Cuckoos. J. Field Ornithol. 51:17-28. in birds. Illinois Biol. Monogr. 22. PRUM,R.O. 1990. Phylogeneticanalysis of the evo- LANGRAND,O. 1990. Guide to the birds of Mada- lution of display behavior in the Neotropical gascar.Yale Univ. Press,New Haven, Connect- manakins(Aves: Pipridae). Ethology84:202-231. icut. PRUM,R. O., AIqDA. E. JOHNSON.1987. Display be- LINNAEUS,C. 1766. Systemanaturae per regna tria havior, foraging ecology,and systematicsof the naturae, 12th ed. Vol. 1. Lipsiae. Golden-winged Manakin (Masius chrysopterus). MACEDO,R.H. 1992. Reproductive patterns and so- Wilson Bull. 99:521-539. cial organizationof the communalGuira Cuckoo PYCRAFT,W. P. 1903. Contributions to the osteology (Guiraguira) in central Brazil.Auk 109:786-799. of birds. Part 6, Cuculiformes. Proc. Zool. Soc. Lond. 1903:258-289. MADDISON,W.P. 1993. Missing data versus missing charactersin phylogeneticanalysis. Syst. Biol. 42: QUINN, J. S., R. MACEDO, AND B. N. WHITE. 1994. 576-581. Genetic relatedness of communally breeding Guira Cuckoos. Anim. Behar. 47:515-529. MAY, R. M., AND S. K. ROBINSON.1985. Population dynamicsof avian brood parasitism.Am. Natl. RALPH,C. P. 1975. Life style of Coccyzuspumilus, a 126:475-494. tropical cuckoo.Condor 77:60-72. ROTH, P. 1981. A nest of the Rufous-vented Ground- MAYa, E., AND J. BOND. 1943. Notes on the classifi- cation of the swallows, Hirundinidae. Ibis 85: Cuckoo(Neomorphus geoffroyi). Condor 83:388. 334-341. ROTHSTEIN,S. I. 1990. A model system for coevo- lution: Arian brood parasitism. Annu. Rev. Ecol. McCLuRE,H. E. 1967. The composition of mixed Syst. 21:481-508. speciesflocks in lowland and submontaneforests RowAN,M. K. 1983. The doves,parrots, louries and of Malaya. Wilson Bull. 79:131-154. cuckoos of South Africa. Croom Helm Ltd., Lon- MclLwRA•TH, T. 1894. The Birds of Ontario, 2nd ed. don. William Briggs,Toronto. SCLATER,P. L., AIqD O. SALVIN. 1873. Nomenclator MORONS, J. J., JR., W. J. BOChe,AND J. FARaAND, JR. avium neotropicalium. London. 1975. Reference list of the birds of the world. SEIBEL,D. E. 1988. A phylogenetic analysis of the AmericanMuseum of Natural History,New York. Cuculiformesand Opisthocomus,based on post- MORTON, E. S., AND S. M. FARABAUGH.1979. Infan- cranial skeletal characters. Ph.D. dissertation. ticideand otheradaptations of the nestlingStriped Univ. Kansas, Lawrence. Cuckoo,Tapera naevia. Ibis 121:212-213. SHELLEY,G.E. 1891. Catalogueof the Picariaein the MOLLER, P. L. S. 1776. Des Ritters Carl von Linne. collection of the British Museum, vol. 19. In Cat- Vollsffindigen Natursystems Supplements und alogueof the Birdsof the BritishMuseum. British Register-bandfiber alle sechsTheile oderClassen Museum, London. des Thiefreichs. G. N. Raspe,Nfirnberg. SHUFELDT,R.W. 1886. Contributionsto the anatomy NOLAN, V., JR., AND C. F. THOMPSON. 1975. The oc- of Geococcyxcalifornianus. Proc. Zool. Soc.Lond. currence and significanceof anomalousrepro- 1886:466-491. ductive activities in two North American cuckoos SIBLEY,C. G., AND J. E. AHLQUIST.1972. A compar- Coccyzusspp. Ibis 117:496-503. ative study of the egg white proteins of non- OATES,E. W. 1903. Catalogue of the collection of passefine birds. Peabody Mus. Natl. Hist. Bull. bird's eggsin the British Museum,vol. 3. Printed 39. by the Order of the Trustees,London. SIBLEY,C. G., AND J. E. AHLQUIST. 1973. The rela- OHMART,R. D. 1973. Observationson the breeding tionshipsof the Hoatzin. Auk 90:1-13. January1996] CuckooPhylogeny Based on Behavior 21

SIBLEY,C. G., ANDJ. E. AHLQUIST.1990. Phylogeny using parsimony,version 3.1. Illinois Natural and classification of the birds of the world. Yale History Survey,Champaign, Illinois. Univ. Press, New Haven, Connecticut. TERRES,J. K. 1988. The Audubon Societyencylo- SIBLEY,C. G., AND B. L. MONROE, JR. 1990. Distri- pediaof North Americanbirds. Alfred A. Knopf, butionand taxonomyof birdsof the world. Yale New York. Univ. Press, New Haven, Connecticut. TINnERGEN,N. 1959. Behavior,systematics, and nat- SICK,H. 1993. Birds in Brazil. Princeton Univ. Press, ural selection. Ibis 101:318-330. Princeton, New Jersey. VERHENCAMP,S. L., B.C. BOWEN,AND R. R. KOFORD. SKUTCH,A.F. 1966. Life historynotes on three trop- 1986. Breedingroles and pairingpatterns within ical American cuckoos. Wilson Bull. 78:139-165. communalgroups of Groove-billedAnis. Anita. SOLER,M. 1990. Relationshipsbetween the Great- Behav. 34:347-366. spottedCuckoo, Clamator glandarius, and its corv- VERHEYEN,R. 1956a. Contribution a l'anatomie eta id hostsin a recentlycolonized area. Ornis Scand. la systematiquedes touracos (Musophagi) et des 21:212-223. coucous(Cuculiformes). Bull. Inst. R. Sci. Natl. SPENCER,O. R. 1943. Nesting habits of the Black- Belg. 32:17-28. billed Cuckoo. Wilson Bull. 55:11-22. VERHEYEN,R. 1956b. Note syst•matiquesur Opisth- SPRUNT,A. JR.,AND E. B. CHAMBERLAIN. 1949. South ocomushoatzin (St-M/iller). Bull. Inst. R. Sci. Natl. Carolina birdlife. Contr. Charleston Mus. 11:1- Belg. 32:1-8. 585. VERHEYEN,R. 1961. A new classification for the non- STEGMANN,B.C. 1978. Relationshipsof the super- passefinebirds of the world. Bull. Inst. R. Sci. ordersAlectoromorphae and Charadriomorphae Natl. Belg. 37:1-36. (Aves):A comparativestudy of the avian hand WEIGEL,R. D. 1963. Oligocenebirds from Saskatch- (R. A. Paynter, Jr., Ed.). Nuttall Ornithol. Club ewan. Q. J. Fla. Acad. Sci. 26:257-262. Publ. 17. WETMORE,A. 1968. The birds of the Republicof STILES,F. G., ANDA. F. SKUTCH.1989. A guide to the Panama. Part 2, Columbidae to Picidae. Smith- birds of CostaRica. ComstockPublishing Asso- son. Misc. Coil. 150:1-605. ciates, Ithaca, New York. WHISTLER,H. 1949. Popular handbookof Indian STRAHL,S. D. 1985. The behaviorand socio-ecology birds.Gurney and Jackson,London. of the Hoatzin, Opisthocomushoatzin, in the llanos W•IITMAN,C. O. 1899. behavior. Pages285- of Venezuela. Ph.D. dissertation. State Univ. New 338 In Biological lectures, Wood's Hole, (C. O. York. Albany. Whitman, Ed.). Ginn and Co., Boston. STRAUCH,J. G., JR. 1985. The phylogenyof the A1- WHITSON,M. A. 1971. Field and laboratory inves- cidae. Auk 102:520-539. tigationsof the ethology of courtshipand cop- STRESEMANN,E. 1934. Aves. In Handbuch der Zool- ulation in the (Geococcyxcal- ogie, vol. 7, part 2 (W. K/ikenthal and T. Krum- ifornianus-Aves,Cuculidae). Ph.D. dissertation. bach, Eds.).Waiter de Gruyter, Berlin. Univ. Oklahoma, Norman. STRESEMANN,E. 1965. Die Mauserder H•hnerv/Sgel. WYLLIE, I. 1981. The Cuckoo. Universe Books, New J. Ornithol. 106:58-64. York. SWOFFORD,D.L. 1993. PAUP: Phylogeneticanalysis

APPENDIX1. Referencesconsulted for collectionof behavioraland ecologicaldata.

Cacomantls:Bell 1986,Coates 1985, 1990, Oliver 1955,Rowan 1983, Wyllie 1981.Carpococcyx: Delacour and Jabouille1931, Oates1903. Centropus:Ali and Ripley 1987,Coates 1985, Fry et al. 1988,Ginn et al. 1989,Langrand 1990, Rowan 1983, Wyllie 1981.Chalcites: Brooker and Brooker1989, 1992, Friedmann 1968, Gill 1983,Oliver 1955,McClure 1967,Rowan 1983, Wyllie 1981. Chrysococcyx:Bannerman 1933, Bell 1986,Friedmann 1968,Fry et al. 1988,Ginn et aL 1989,Payne 1973,Rowan 1983, Wyllie 1981.Clamator: All and Ripley 1987,Bannerman 1933, Cramp 1985, Friedmann 1964, 1968, Fry et aL 1988,Rowan 1983, Soler 1990, Wyllie 1981. Coccyzus:Bent 1940, Cadman et al. 1987, Ehrlich et al. 1987, Fleisher et al. 1985, Hamilton and Hamilton 1965,Hilty and Brown 1986,Stiles and Skutch1989, Terres 1988, Wyllie 1981.Coua: Langrand 1990, Rowan 1983.Crotophaga: Bent 1940,Brown 1987,ffrench 1973,Verhencamp et al. 1986,Wetmore 1968, Wyllie 1981. Cuculus:All and Ripley 1987,Ambrose 1987, Bannerman 1933, Bell 1986,Cramp 1985,Friedmann 1968, Fry et aL 1988,Ginn et al. 1989, Rowan 1983,Wyllie 1981.Eudynamys: All and Ripley 1987,Brown 1987,Friedmann 1928, 1964, Gosper 1964, Rowan 1983, Whistler 1949,Wyllie 1981.Geococcyx: Bent 1940,Folse and Arnold 1978,Ohmart 1973,Rowan 1983,Whitson 1971, Wyllie 1981.Gulra: Brown 1987, Hudson 1920, Macedo 1992, Quinn et al. 1994,Sick 1993,Wyllie 1981.Misocallus: Friedmann 1968, Rowan 1983,Wyllie 1981. Morococcyx:Herklots 1965,Hudson 1920,Sick 1993,Skutch 1966, Stiles and Skutch1989. Mu- sophagidae:Fry et al. 1988,Ginn et al. 1989.Neomorphus: Haffer 1977,Haverschmidt 1968, Hilty and Brown 1986,Roth 1981,Wetmore 1968, Wyllie 1981.Opisthocomus: Beebe 1909, Grimmer 1962,Sibley and Ahlquist 1990,Strahl 1985.Pachy- coccyx:Fry et aL 1988,Ginn et al. 1989,Rowan 1983,Wyllie 1981.Piaya: firerich 1973,Haverschmidt 1968, Herklots 1965, Skutch1966, Wetmore 1968, Wyllie 1981.$aurothera: Bond 1960,Wyllie 1981.$cythrops: Coates 1985, Friedmann 1964, Rowan 1983,Wyllie 1981. Taperatfirerich 1973, Friedmann 1933, Flaverschmidt 1961, 1968, Hilty and Brown 1986,Morton and Farabaugh1979, Stiles and Skutch 1989, Wyllie 1981. 22 J^NICEM. HUGHES [Auk, Vol. 113

A??E•DIX2. Behavioraland ecologicalcharacter descriptions.

1. Generalizedhabitat use: (0) arboreal;(1) terrestrial.2. Flying ability: (0) weak flyer;(1) strongflyer. 3. Primarydiet: (0) fruitsand foliage; (1) primarilycaterpillars; (2) insectsand other arthropods; (3) invertebratesand . 4. Migration: (0) sedentary;(1) migratory.5. Pair bond:(0) monogamous;(1) somepolygamy or promiscuity.6. Degreeof socialinteraction duringbreeding season: (0) solitary;(1) smallflocks; (2) gregarious.7. Territorialityduring breeding season: (0) territory defendedby both;(1) territorydefended predominantly by male;(2) territorydefended predominantly by female;(3) not territorial.8. Adultssunbathe by droppingwings and exposingback to sun:(0) yes;(1) no. 9. Aiiopreeningin matedpair duringcourtship: (0) no; (1) yes.10. Ritualizeddisplay flights by male or femaleduring courtship: (0) no; (1) yes.11. Male or female call from songposts during courtship:(0) no; (1) yes. 12. Nest architecture:(0) poorly constructedsaucerlike platform;(1) opencup; (2) well constructedenclosed dome; (3) doesnot build nest. 13. Nest location:(0) placedin trees above4 m; (1) placedin shrubs,bushes or grassbetween 1-4 m; (2) placedon or near ground;(3) doesnot build nest.14. Parasiticmale lureshost femaleaway from nestprior to parasiticegg depositionby mate:(0) no; (1) yes.15. Individual femaleslay morethan one eggin nest:(0) yes;(1) no. 16. Eggslaid in nestby morethan one female:(0) no; (1) yes.17. Degreeof hostspecialization: (0) no hostusage; (1) specific-hostspecialists; (2) generalist;(3) individual-hostspecialists. 18. Color of egg:(0) monomorphic,immaculate white; (1) monomorphicnonwhite, or white with markings;(2) polymorphic within a species.19. Sexincubating eggs: (0) both;(1) predominantlyfemale; (2) predominantlymale; (3) doesnot incubate. 20. Parasiticfemale lays egg that mimicshost egg: (0) no; (1) yes.21. Removalof egg from nestby adult: (0) no; (1) yes. 22. Removalof eggsand/or young from nest by juvenile: (0) no; (1) yes. 23. Chick excretesfoul-smelling liquid from cloacawhen disturbed:(0) no; (1) yes.24. Adult bird feedsfledglings: (0) yes;(1) no. 25. Adult producesbiii-clacking vocalizationby snappingmandibles together: (0) no; (1) yes.26. Adultsparticipate in mixed-species-flockfeeding: (0) no; (1) yes.27. Incubationperiod: more than 15 days(0); 15 daysor less(1). 28. Mated pair participatesin vocalduetting: (0) no; (1) yes.

APPENDIX3. Character-changelist for behavioraland ecologicalcharacters on shortest-lengthtree (Fig. 1). Double-lined arrowsindicate that changesoccurred on all possiblereconstructions. Single-lined arrows indicate that changesoccurs only under somereconstructions. CI = consistencyindex. RI = retentionindex.

Character1 (CI = 0.50;R/ = 0.86) 0 [node35] • 1 [node Tapera];0 [node42] • 1 [node41]. Character2 (CI = 0.50;R/ = 0.90)0 [node35] • 1 [node34]; 0 [node43] • 1 [Piaya].Character 3 (CI = 0.50;RI = 0.75)0 [node44] --• 2 [node43]; 2 [node 25] • 0 [Opisthocomus];2 [node 35] • 1 [node 34]; 1 [node 33] • 0 [node 32]; 2 [node 42] • 3 [node 41]; 3 [node 38] • 2 [Morococcyx].Character 4 (CI = 0.50;R/ = 0.88)0 [node35] • 1 [node34]; 1 [node26] • 0 [Pachycoccyx].Character 5 (CI = 0.25;RI = 0.57) 0 [node24] • 1 [Guira];0 [node 30] • 1 [node29]; 0 [node34] • 1 [Coccyzus];0 [node 39] • 1 [Centropus]. Character6 (CI = 0.50;RI = 0.60) 1 [node43] • 0 [node42]; 0 [node36] • 2 [node 25]; 0 [node26] • 1 [CIamator];0 [node 39]• 1 [Centropus].Character 7 (CI = 0.60;RI = 0.50)0 [node33] --• 1 [node31]; 1 [node26] • 3 [Pachycoccyx];1 [node 29] --* 2 [Cuculus];1 [node 30] --• 0 [Cacomantis];0 [node 39] • 2 [Centropus].Character 8 (CI = 100;RI = 100) 0 [node 34] • 1 [node33]. Character9 (CI • 0.33;RI = 0.33) 0 [node25] • 1 [node24]; 0 [node26] • 1 [Clamator];0 [node40] --* 1 [node39]. Character10 (CI = 0.50;RI = 0.75) 0 [node29] • 1 [node28]; 0 [node35] • 1 [Tapera].Character 11 (CI = 0.25;RI = 0.57) 0 [node36] • 1 [node35]; 1 [node27] • 0 [node 26]; 1 [node 33] --* 0 [node32]; 0 [node41] • 1 [node40]. Character12 (CI = 0.50;RI = 0.75) 0 [node 26] • 1 [node 24]; 0 [node 36] --* 3 [node 35]; 3 [node 34] --• 0 [Coccyzus];0 [node 41] • 1 [node 37];0 [node38] • 1 [Morococcyx];0 [node 39] • 2 [Centropus].Character 13 (CI = 0.60;RI = 0.80) 0 [node36] • 3 [node 35]; 3 [node34] • 1 [Coccyzus];0 [node 41] • 1 [node40]; 1 [node38] • 2 [Morococcyx];0 [node 43] • 1 [Piaya].Character 14 (CI = 0.50;R1 = 0) 0 [node26] • 1 [CIamator];0 [node 32] • 1 [Eudynamys].Character 15 (CI = 0.50;R1 = 0.80) 0 [node33] • 1 [node31]; 1 [node26] • 0 [CIamator].Character 16 (CI = 100;R/= 100) 0 [node42] • 1 [node 36]. Character17 (CI = 0.75; R/= 0.88) 0 [node 36] • 2 [node 35]; 2 [node 30] • 3 [node29]; 3[node28] • 1 [node 27];2 [node 33] • 1 [node 32]. Character 18 (CI = 0.50;RI = 0.83) 0 [node 42] • 1 [node 36]; 1 [node 31] • 2 [node 30]; 2 [node27] • 1 [node 26]; 1 [node 35] • 2 [Tapera].Character 19 (CI = 0.75;RI = 0.90)0 [node36] --• 3 [node35]; 3 [node34] --• 0 [Coccyzus];0 [node 40] • 1 [node38]; 0 [node39] • 2 [Centropus].Character 20 (CI = 0.50;RI = 0.88)0 [node35] • 1 [node34]; 1 [node28] • 0 [Chalcites].Character 21 (CI = 0.33;RI = 0.78) 0 [node25] • 1 [node24]; 0 [node34] • 1 [node 33]; 1 [node 26] • 0 [Clamatofl.Character 22 (CI = 0.50;RI = 0.80) 0 [node 33] • 1 [node 31]; 1 [node 26] • 0 [CIamator].Character 23 (CI = 0.50;RI = 0.67) 0 [node 28] • 1 [node27]; 0 [node39] • 1 [Centropus].Character 24 (CI = 0.50;RI = 0) 0 [node31] • 1 [Misocalius];0 [node 35] • 1 [Tapera]. Character25 (CI = 0.50;RI = 0.67)0 [node40] • 1 [node38]; 0 [node43] • 1 [Piaya].Character 26 (CI = 0.50;RI = 0.67)0 [node31] • 1 [node30]; 1 [node27] • 0 [node26]. Character27 (CI = 100;RI = 100) 0 [node36] • 1 [node39]. Character 28 (CI = 100; RI = 100) 0 [node 40] • 1 [node 39].