NEST PARASITISM

Scott K. Robinson* and Stephen I. Rothstein† *University of Illinois at Urbana-Champaign and †University of California at Santa Barbara

I. Natural History ized nests (usually with renesting), or a new cup II. Coevolution between Brood Parasites and Their built over a parasitized clutch. Hosts evolutionary equilibrium hypothesis The hypothesis III. Modeling Host–Parasite Coevolution that frequencies of acceptance of brood parasitism IV. Impacts on Host Population Dynamics reflect an equilibrium between costs and benefits of V. Ecology and Social Behavior of Brood Parasites host defenses against parasitism. VI. Conspecific Brood Parasitism evolutionary lag hypothesis The hypothesis that hosts VII. Conclusions and Research Needs lack defenses against brood parasites because the defenses have not yet had time to evolve. generalist brood parasites Species that parasitize many (up to more than 200) host species. GLOSSARY gentes (singular gens) Lineages of in which individual females specialize on a single host and lay mimetic eggs. acceptance Hosts treat brood parasitic eggs as if they host The individual that is parasitized. were their own eggs. interspecific brood parasites Brood parasites that lay coevolution Cycles of adaptation and counteradapta- their eggs in the nests of other species. tion that occur between interacting lineages. mafia effect Interspecific brood parasites that may de- coevolutionary arms race The continuing bouts of co- stroy clutches from which parasitic eggs have been evolving defenses and counterdefenses that occur ejected by the host. between hosts and parasites. mimicry Brood parasitic eggs or nestlings that closely conspecific brood parasites Brood parasites that lay match those of the hosts. their eggs in the nests of other individuals of the specialist brood parasites Species that parasitize only same species. one or a few host species. American blackbirds in the genus Molothrus, which contains important brood parasites. cuckoos A family of of which approximately half the species (61 of 125) are obligate interspecific brood parasites. BROOD PARASITISM, also called social parasitism, is egg rejection Host responses to parasitism that include the exploitation by one individual (the ) ejection of parasitic eggs, abandonment of parasit- of the parental care of another (the host). Brood para-

Encyclopedia of Biodiversity, Volume 4 Copyright  2001 by Academic Press. All rights of reproduction in any form reserved. 365 366 NEST PARASITISM sites can deposit eggs in the nests or broods of another well-studied common , the species as a whole individual of the same (conspecific brood parasitism) is a generalist, but each individual female is a specialist or of a different (interspecific brood parasitism) species. member of a ‘‘gens.’’ Any one region has only one to Hosts often raise young of the brood parasite, typically several coexisting gentes but different arrays of gentes at the expense of their own young. This article presents occur in different regions. a comprehensive overview of the natural history, evolu- Brood parasitic nestlings also have a variety of mech- tion, and consequences of brood parasitism, with a spe- anisms that enhance their ability to compete with host cial focus on birds, the taxon in which it has been nestlings. Cuckoo nestlings have concave backs that best studied. they use to push out host eggs and nestlings, whereas several other brood parasites have specialized bill hooks that they use to kill nestmates. Some brood parasites apparently increase the amount of food hosts bring to I. NATURAL HISTORY them by mimicking the juvenile plumages or complex mouth markings (the ‘‘gapes’’) of host nestlings or by Brood parasites lay their eggs in the nests of other having large mouths and intense begging behavior. individuals, which then raise the parasitic young at Because brood parasites do not have to engage in the expense of part or all of their own brood. Brood costly breeding activities such as incubation and nest- parasitism was noted by Aristotle and even earlier (ap- ling feeding, they often have more time and energy proximately 2000 B.C.) in India. Brood parasitism is to devote to egg production. Some generalist brood best known in birds, approximately 1% of which are parasites lay 40 or more eggs in a season; estimates for obligate interspecific brood parasites that only lay their one tropical brood parasite suggest that more than 100 eggs in the nests of other species. Interspecific brood eggs/year are typical. Other brood parasites, however, parasitism, however, has also been documented in in- may be only slightly more fecund than their hosts. sects and fish. Conspecific brood parasitism, in which individuals facultatively lay eggs in nests of conspecific individuals, is more widespread. B. A Survey of Avian Brood Parasites Brood parasites have generated intense interest in There are 90–95 species of obligate avian brood para- the public and scientific communities. Brood parasites sites in five unrelated families (Table I). tend to be vilified in the media because of a human tendency to moralize about such things as killing baby 1. Cuculidae birds (which parasitic birds often do) and because at The cuckoos are a diverse family that contains both least some may pose a conservation threat to some parasitic and nonparasitic species. The Cuculinae has of their hosts. Among scientists, brood parasites have traditionally been recognized as an Old World subfam- generated intense debate about the coevolutionary pro- ily, all of whose approximately 50 species are obligate cesses that may be responsible for the seemingly mal- adaptive acceptance of parasitic eggs by hosts.

A. Adaptations of Obligate Avian TABLE I Brood Parasites Major Groups of Avia Brood Parasites Brood parasites search for host nests, synchronize their Taxon (family/subfamily) No. of species laying with that of the host, and often remove a host Old World cuckoos (Cuculinae)a 50 egg from nests they parasitize. Brood parasitic eggs often New World cuckoos (Neomorphinae)3 have unusually thick eggshells, are usually small rela- (Africa and Asia) (Indicatoridae) 17? tive to the size of the parasite (but large relative to the Viduine finches (Africa) (Ploceidae)14 size of the host), often mimic the coloration of the Cuckoo-finch (Africa) (Ploceidae)1 hosts’ eggs, and have rapidly developing embryos. All Cowbirds (New World) (Icteridae)5 these traits increase fitness of the parasite by reducing Black-headed duck (South America) (Anatidae)1 competition with host nestmates and making parasitic eggs more difficult to detect and remove. Egg mimicry a Recent work indicates that at least one genus in this subfamily is most pronounced in specialist brood parasites. In the belongs in another cuckoo subfamily (see text). NEST PARASITISM 367 brood parasites. Most species are host specialists that warblers in grasslands. Recently reported DNA evi- are relatively uncommon and only a few have been well dence suggests that parasitism evolved once in a single studied (the and the great spotted lineage that then gave rise to both Anomalospiza and cuckoo). Another traditionally recognized subfamily, the Viduines. the Neomorphinae or New World ground cuckoos, has 11 species, 3 of which are relatively rare obligate para- 5. Icterinae sites. The latter tend to parasitize hosts that build domed nests and are poorly known overall. Nestlings The five species of New World brood parasitic cowbirds of one species, Tapera naevia, have pincer-like bills that include the most generalized of all brood parasites, they use to kill host nestlings, a case of covergence on the brown-headed and shiny cowbirds, both of which honeyguides (Indicatorinae). Recently published DNA parasitize more than 200 host species. Another species, sequence data indicate that at least 1 genus (Clamator) the tropical giant , parasitizes almost exclu- within the Cuculinae is more closely related to the sively colonial American blackbirds. Of the two re- Phaenicophaeinae, a subfamily of New World cuckoos. maining species, the is a host general- Ͼ None of the currently recognized members of the Phae- ist ( 80 host species), whereas the nicophaeinae are obligate parasites but some parasitize is one of the most specialized of all brood parasites. It conspecifics and occasionally other species. These new mainly parasitizes a single species of blackbird, which DNA data place obligate parasitism within three groups was formerly thought to be a species of cowbird. Unlike of cuckoos, two of which have both parasitic and non- most brood parasites, cowbirds are often more abundant parasitic species, and indicate that parasitism evolved than many of their hosts and can pose a significant separately up to three times in cuckoos or that some threat to populations of rare, localized host species. parasitic cuckoo lineages reverted to parental behavior. Most cowbirds benefit greatly from changes in the land- scape caused by humans, which has enabled several 2. Indicatoridae species to expand their geographic ranges and increase The honeyguides, a family named for the habit of one their population sizes. As a result, they are coming into species that guides people and the African honey badger contact with new hosts that have not had recent contacts or ratel to beehives, has 17 obligately parasitic species. with brood parasitism. Some of these new hosts suffer The species of Indicator and two related genera parasit- extremely high levels of parasitism. Cowbirds appear ize cavity nesters, whereas 3 species of Protodiscus para- to be extremely fecund; the may lay more sitize open-cup nesters. Most species are poorly known than 100 eggs a year. The invasion of North America and the eggs of many species have never been described. by this cowbird may pose an additional threat to species Some honeyguides have raptorial hooks on their bills, that are not currently being parasitized by the brown- which they use to kill host nestlings. headed cowbird. The brown-headed cowbird may be the most intensively studied North American . 3. Viduine Finches The approximately 16 species of obligate brood parasitic 6. Black-Headed Duck (Heteronetta viduines occur in Africa and provide some of the best atricapilla) examples of nestling mimicry, particularly of the intri- This species is the only obligately parasitic species with cate gape patterns of their hosts. Compared with other precocial young that feed themselves when they hatch. brood parasites, viduines exert relatively low costs on For this reason, this brood parasite has little effect on their hosts, which can usually raise mixed broods. Most the nesting success of its hosts, which only have to Vidua species specialize on single species of grassfinches incubate an extra egg or two. Because incubation is so (Estrildidae), to which they may be related. Unlike most similar among all bird species, Heteronetta successfully brood parasites, which are insectivorous, hosts of vidu- parasitizes a wide range of birds, such as gulls and ines are granivorous. Some species of viduines incorpo- ibises, as well as other ducks. rate the songs of their hosts into their own songs, which are then used by females to choose males that have been raised by the same species. 7. Conspecific Brood Parasitism Recent studies have shown widespread conspecific 4. Cuckoo-Finch (Anomalospiza imberbis) brood parasitism, especially in waterfowl, gallinaceous This African species is closely related to Viduine birds, and in a small number of songbirds (generally finches and parasitizes approximately 11 species of species with rare nest sites or that breed colonially). 368 NEST PARASITISM

8. Other Brood Parasite Systems regions where several cuckoo species coexist, they tend a. to show little or no overlap in host use, which reduces competition for hosts. Most cuckoos are much less An African catfish is the only fish species known to be an abundant than their hosts and probably have little effect obligate brood parasite. It parasitizes mouth-breeding on host population dynamics because they affect only cichlids and apparently consumes all of its host’s young a small proportion of their host populations. while in the mouth of its foster parent. Conspecific Some cuckoo–host systems may have reached an brood parasitism and occasional parasitism of other evolutionary equilibrium, whereas others show strong species occur in many nest-building fish. Some host evidence of ongoing coevolution and provide some of defenses such as increased nest guarding may have de- the strongest examples in vertebrates of microevolution veloped in response to this parasitism but egg recogni- that has occurred during ongoing research. tion has not been reported. b. A range of insects practice brood parasitism, especially B. Adaptations of Cuckoos members of the order Hymenoptera (bees, , and 1. Behavior ants). Some of these species are obligate brood parasites, Cuckoos typically approach nests stealthily and may such as certain ants that kill the queens of nonparasitic quickly drop their thick-shelled eggs onto host eggs to species and use the workforce of the entire colony to increase chances of breaking host eggs. Males of some rear their own young. Other species are facultative para- cuckoos resemble hawks. This resemblance or simply sites that victimize conspecifics or raise their own host recognition of cuckoos incites host aggression and young. To our perception, some of these parasitic in- males in some species use this aggression to draw hosts sects look remarkably different from their hosts so their away from nests. This male distraction is coordinated acceptance by hosts seems inexplicable. However, olfac- with and facilitates nest entry by the stealthy and drab- tion is critical to these insects and it is likely that para- colored females in several cuckoo species. Nestlings of sitic species have evolved chemical cues that mimic some cuckoos mimic the begging calls or plumage of those of their hosts. Because both humans and birds host young. There is evidence that great spotted cuck- perceive the world largely via vision and audition, and oos revisit nests they have parasitized and destroy host have relatively poor olfactory abilities, it is not surpris- eggs or nestlings if the parasitic egg has been removed ing that parasitic birds have been subjects of much more by the host. Such a protection racket or ‘‘mafia effect’’ research than have parasitic insects, even though the could select for acceptance of host eggs. latter undoubtedly have many equally interesting exam- Cuckoos find nests by watching hosts from hidden ples of coevolved adaptations. perches and laying is synchronized with that of the hosts. Most cuckoos lay one egg per nest, except in a few species in which cuckoo nestlings do not kill those II. COEVOLUTION BETWEEN BROOD of the host. Cuckoos defend territories against other PARASITES AND THEIR HOSTS cuckoos and usually remove a host egg before laying one of their own. A. Cuckoo–Host Systems 2. Fecundity Cuckoo–host systems are much more highly coevolved Cuckoos lay 8–25 eggs per season, which is generally than the other well-studied systems, those of cowbirds, more than their hosts lay. Eggs are laid every other day. and DNA divergence data show that they are also much older. Most cuckoo species use only one or a few host species. Some species; such as the Eurasian common 3. Egg Adaptations cuckoo, might be called generalists when viewed over Cuckoo eggs usually mimic the coloration of their hosts’ their entire ranges because they parasitize 50–100 or eggs. Cuckoos that parasitize many host species have more host species. However, even these cuckoos are individually specialized females, with each group of more properly viewed as specialists because they use females or gens (plural gentes) mimicking the eggs of only one to several host species in each region and a single host species or a group of hosts with similar individual females, or members of a gens, specialize on eggs. Egg mimicry is usually assumed to have arisen in a single host species or several similar host species. In response to host egg recognition. An alternative hypoth- NEST PARASITISM 369 esis argues that egg mimicry evolved to keep a second eggs and is therefore costly or because they are simply cuckoo from identifying and removing a cuckoo egg too small to eject cuckoo eggs. Egg rejection is most from an already parasitized nest and then laying its own prevalent in hosts that have a long coevolutionary his- egg in the nest. However, it is not clear that more than tory with cuckoos. These hosts, which often show fine one cuckoo attempts to parasitize the same host nest abilities to discriminate among egg types, are species often enough for this to be a significant selective pres- that have intrinsic characteristics that make them suit- sure. Most cuckoos parasitize hosts that are much able for parasitism, such as food appropriate smaller than themselves and have correspondingly for the nestling parasites and nests accessible to adult small eggs for their body sizes. However, cuckoo eggs cuckoos. Species unsuitable as hosts, such as ones that are usually slightly larger than those of the hosts. The have specialized nestling diets (e.g., seeds in some finch few species that parasitize hosts larger than themselves species), have presumably never been parasitized in- do not have smaller eggs. The thick shells of cuckoo tensely in the past and generally lack egg recognition eggs may be adaptive because they reduce chances of altogether. Among suitable hosts, egg recognition is breakage of cuckoo eggs during rapid laying, facilitate more prevalent in species that are currently rarely para- breakage of host eggs (thereby reducing competition sitized than in species that are currently common hosts. with hosts), or make cuckoo eggs resistant to pecking This trend suggests that cuckoos shift away from using by hosts. A counteradaptation to egg mimicry that some suitable hosts with well-developed egg recognition. It hosts have developed is variable egg types. In response, is possible that cuckoos have a dynamic system of host some cuckoo species have evolved an equivalent range usage in which they repeatedly switch from hosts with of egg types; but hosts still seem to benefit because well-developed defenses only to switch back to these individual female cuckoos lay only one type and may hosts after their defenses have declined and defenses not always parasitize a host female who lays the same in other hosts have increased. Such a system could type. The highly variable host eggs may also function result in never-ending cycles of host switches and co- in the context of conspecific parasitism. evolution, but it is also possible that hosts retain high levels of egg recognition for long periods in the absence 4. Nestling and Fledgling Adaptations of parasitism, which would force cuckoos to become increasingly more specialized. Long-term retention is Nestlings in most cuckoo species have concave backs indicated by high levels of egg recognition in New that they use to push host nestlings or eggs out of nests. World magpies and shrikes, which are currently not Cuckoos that parasitize larger hosts do not show this parasitized by any brood parasites but are descended eviction behavior but can compete successfully with from Old World ancestors that are cuckoo hosts. Popu- host nestlings because they beg more loudly and inces- lations of some suitable hosts that have only recently santly, hatch earlier, and develop more rapidly. Mimicry come into contact with cuckoos have apparently under- of host begging calls occurs mainly, but not exclusively, gone rapid increases in egg ejection and discrimination, in species that do not evict host nestlings. apparently because they possessed some recognition even in the absence of parasitism. This possession could be due to retention of recognition from past bouts of C. Host Defenses Against Cuckoos parasitism or to gene flow from parasitized populations 1. Defenses against Adults of the same species. Hosts often respond aggressively to cuckoos, but this 3. Nestling Discrimination may actually provide a cue to cuckoos about their stage Despite there being many species that have some degree of the nest cycle and the proximity of nests. Mobbing of mimicry of host nestling appearance or begging calls, cuckoos has both genetic and learned components. One there are no known cases in which cuckoo hosts show possible advantage of mobbing cuckoos is that it may outright rejection (i.e., removal) of nonmimetic nest- trigger ejection of parasitic eggs. lings. This lack of outright rejection appears to be true in all other systems of avian parasitism that show equiv- 2. Egg Rejection alent or even better mimicry of host nestlings. There- Hosts reject cuckoo eggs by abandoning parasitized fore, the selective value of nestling mimicry may relate nests or ejecting eggs directly. Small hosts are more to ensuring that parasitic nestlings receive high-quality likely to abandon (and renest), perhaps because ejection care from their hosts rather than avoiding outright re- results in the incidental breakage of some of their own jection. 370 NEST PARASITISM

D. Cowbird–Host Systems before the host eggs or after hosts have finished laying occur more commonly than in cuckoos. Although gen- The five brood parasitic cowbirds include one of the eralist cowbird species avoid parasitizing some species most specialized of all brood parasites and the two most that are clearly unsuitable as hosts, such as doves, there generalized of all brood parasites. We have chosen to is little evidence that they select the best host species highlight cowbird–host interactions in this article be- in a community. Many cowbird eggs are seemingly cause there is a vast literature on cowbird–host systems wasted in nests of hosts that feed their nestlings inap- and because these systems are quite different from propriate diets of seeds or fruit or that eject cowbird cuckoo–host systems. eggs. However, parasitism of the latter species may sometimes be adaptive for two reasons. First, some 1. Cowbird Adaptations species that exercise egg recognition and ejection have a. Egg Adaptations low nest predation rates. Second, these species learn As described previously, some cowbirds are extremely the appearance of their own eggs by imprinting on fecund, an apparent reflection of a trade-off between the first egg or eggs that they lay during their lives. egg production and parental care. In captivity, brown- Accordingly, naive hosts parasitized about the time they headed cowbirds can lay up to 77 eggs in the short begin laying may come to learn cowbird eggs as their temperate breeding season and field estimates of shiny own eggs and may provide cowbirds with nests that cowbird productivity exceed 100 eggs during the much have a relatively low likelihood of failing due to nest longer tropical breeding season. As in cuckoos, the eggs predation. Multiple parasitism (two or more cowbird of the smaller cowbirds are small relative to the size of eggs in a nest) occurs in approximately one-third of all the cowbird but are larger, thicker shelled, rounder, nests parasitized by cowbirds, especially with some and faster to develop than those of most of their hosts. large hosts. Several female cowbirds sometimes lay in Only the giant cowbird, which parasitizes hosts as large the same nest even though multiple parasitism usually as itself, has normally proportioned eggs. The other reduces the success rate of cowbird eggs and many host smaller species usually parasitize smaller hosts. Egg nests receive no cowbird eggs. Female cowbirds are mimicry has not been firmly established in any cow- usually highly aggressive toward one another in habitats birds, but cowbird egg shape and patterns have evolved, in which hosts occur (but not in which cowbirds feed), perhaps in response to diffuse selection by dominant but there is still substantial overlap among home ranges. hosts or groups of hosts. The two most generalized Cowbirds sometimes, but not always, remove host eggs cowbirds, the shiny and brown-headed, have general- by puncturing them. Egg removal is rarely done during ized spotted egg colorations that are similar to those of the same nest visits on which females lay but may occur many . Although cowbird egg coloration is the afternoon before or later in the morning on which not finely tuned to host egg coloration, as in cuckoos, cowbird eggs are laid. There are many observations of it has undergone some shifts, presumably in response cowbirds depredating unparasitized host nests and to host use, during the evolution of the cowbird lineage. there is evidence that cowbirds do this regularly to Different cowbird species have generalized spotted eggs stimulate renesting, which will increase the future avail- or immaculate unspotted ones; however, two species, ability of host nests to be parasitized. However, some the giant and shiny cowbirds, have both kinds of eggs. studies have found no evidence for this depredation- renesting hypothesis and the behavior may occur only b. Behavioral Adaptations of Adults in special circumstances, if at all. Similar to cuckoos, cowbirds search for nests primarily by observing the behavior of hosts. Cowbirds approach c. Nestling and Fledging Adaptations nests stealthily and lay eggs very early in the morning In the specialized screaming cowbird, there is almost before most hosts appear at their nests during the egg- perfect mimicry of host plumage and vocalizations of laying period. In contrast, giant cowbirds appear to its usual host species, the bay-winged cowbird (which circumvent the defenses of colonial hosts by visiting is not closely related to the parasitic cowbirds despite colonies as a group in which males appear to distract its name). This mimicry does not seem to be essential hosts while females stealthily enter nests. Screaming for nestling care but instead seems to be an absolute cowbirds also visit their communally breeding hosts in requirement for receiving host care after the parasites groups, but males do not appear to help distract the fledge from host nests. Such perfect mimicry is lacking hosts. Cowbird eggs are usually synchronized properly in the other, more generalized cowbirds. However, giant with host laying but improperly timed eggs that appear cowbird nestlings have white bills, like those of their NEST PARASITISM 371 hosts. Bills turn the usual cowbird black recognize their own eggs and remove cowbird eggs even after the fledglings become independent. Nestling cow- if these outnumber their own eggs, unless they have birds grow more rapidly than the nestlings of many, misimprinted on cowbird eggs laid in their very first but not all, of their hosts and beg more loudly. They nest. have relatively large mouths and show intraspecific vari- ation in their gape colors, unlike the majority of nonpar- asitic nestlings. This gape variation may relate to host use in some unknown way. There is a recent report of III. MODELING HOST–PARASITE a cowbird ejecting a host nestling, although the ejection COEVOLUTION could have been inadvertent. There is little doubt that the adaptations and counterad- 2. Host Defenses against Cowbirds aptations described in Section II result from coevolution a. Preventing Access to Nests between parasites and their hosts. Parasitic counterde- Many hosts react aggressively to cowbirds and colonial fenses to host defenses must correspond to the charac- nesting may provide protection for some hosts. Aggres- teristics of hosts, as in mimicry of host eggs. Thus, sive responses to cowbirds, however, may reveal nest parasites cannot have effective defenses against more locations to cowbirds and be counterproductive for than a few hosts at any one time because hosts vary in small hosts, which may not be able to drive cowbirds key characteristics such as egg coloration. The tradi- away from nests. Some hosts sit on their nests when tional view therefore hypothesizes that parasites spe- approached by a cowbird, but there are records of cow- cialize on increasingly fewer hosts as a parasite–host birds pulling hosts off their nests. Some hosts reduce system becomes older and more hosts evolve defenses. parasitism by nesting in cavities or in dense vegetation, This view is supported by studies showing that (i) re- but even cavity nests are often parasitized in many cently derived brood parasites tend to be more general- species. ized than older species (e.g., cowbirds versus cuckoos); (ii) the costs of parasitism are often much higher than b. Egg Rejection the costs of potential host defenses, which suggests that Cowbird host species can be divided into acceptors, there is a continuing evolutionary arms race; (iii) some which accept cowbird eggs, and rejecters, which aban- systems in which hosts and parasites have only recently don parasitized nests, build a new nest on top of the old come in contact show either an absence or low levels nest, or, most commonly, eject cowbird eggs. Within of host defenses; (iv) host defenses have evolved since acceptor species, nearly 100% of the individuals accept recent contact with a brood parasite; (v) parasites have cowbird eggs. Rejecter species show nearly 100% rejec- evolved egg mimicry in response to host defenses; and tion and there is little geographic variation in a species’ (vi) cuckoo–host systems are dynamic as evidenced by responses to cowbird eggs. Birds that eject cowbird eggs host shifts in parasitic cuckoos. In accord with these either puncture them or grasp them and usually drop generalizations, most Old World passerines in Africa them at least several meters from the nest. Smaller hosts and Eurasia, which are exposed to numerous species of tend to puncture–eject, whereas larger hosts grasp– an ancient group (the cuckoos), posses egg recognition. eject. Nest abandonment occurs most frequently when However, most New World passerines, which are ex- cowbird eggs are laid too early in a host’s laying cycle posed to a smaller and relatively younger group of para- and when hosts are too small to eject eggs (or to eject sites (the cowbirds) lack egg recognition. without breaking too many of their own eggs). Nest The question then arises, why do so many hosts desertion appears to be triggered by encounters at the accept easily distinguishable parasitic eggs? Two nest between hosts and cowbirds. In nearly all host hypotheses have been proposed: ‘‘evolutionary lag’’ and species, detection of cowbird eggs in nests does not ‘‘evolutionary equilibrium.’’ There is a burgeoning liter- appear to play a role in eliciting desertion, even in hosts ature by proponents of each hypothesis. The evolution- whose eggs are highly divergent from cowbird eggs. ary lag hypothesis is that these acceptors have not yet Some hosts may begin incubating early in the morning had time to evolve antiparasite adaptations. Acceptance to prevent undetected cowbird parasitism. Most species of parasitic eggs by some hosts in heavily parasitized, that reject cowbird eggs occur in open habitats where unproductive populations may also result from gene they may have a long coevolutionary history with cow- flow from populations in areas where cowbirds are ab- birds, all of which forage in open areas in association sent and reproduction is high. The lag hypothesis is with grazing ungulates. Hosts that eject cowbird eggs clearly applicable to situations in which parasites have 372 NEST PARASITISM begun to use host species with no recent history of Nevertheless, brood parasites can pose threats to exposure to parasitism but may even apply to some host populations, especially generalist brood parasites host species with long histories of parasitism because and those that have undergone recent range expansions egg recognition/rejection is a totally new feature that and population increases. In a few cases, cuckoos have may be difficult to evolve. In the evolutionary equilib- expanded their ranges as a result of human-caused in- rium hypothesis, hosts accept parasitic eggs because of creases of early successional habitat. In areas recently the high costs of rejection. In this scenario, parasitism colonized by cuckoos in Japan, levels of parasitism often is costly, but trying to reject parasitism would be even exceed 40%, although in most of these areas the inci- more costly, i.e., acceptors are making the best of a dence of host defenses appears to be increasing rapidly, bad situation. probably because these host populations already had Distinguishing between lag and equilibrium has some level of defense before being parasitized. Such proven to be difficult. Testing either hypothesis requires rapid development of host defenses, however, appears measurements of the costs of rejection in species that to be absent from hosts of the brown-headed and shiny accept, which has not yet been done effectively. Some- cowbirds. Furthermore, range expansions are much times, likely costs for acceptors can be estimated from more pervasive in these two generalist cowbirds than similar species that eject. For example, the number of in any cuckoo species. Cowbirds have undergone enor- their own eggs that small ejecting hosts break when mous population increases as the result of human activi- they eject cowbird or cuckoo eggs can be extrapolated ties, especially those associated with . As a result, to acceptor hosts of similar size. These ejection costs, many host populations with no recent history of parasit- generally less than 0.5 host eggs per ejection, do not ism are being exposed to massive levels of brood parasit- seem to be sufficiently high to make acceptance a more ism. For some species, especially forest-nesting ones, adaptive option because acceptance usually results in widespread parasitism may be occurring for the first the loss of all the host’s young for cuckoo hosts and time in the history of the species’ lineage. Parasitism for small cowbird hosts. There is evidence that rejecter levels of many forest species nesting in the midwestern hosts mistakenly eject their own eggs on occasion, but United States, for example, exceed 80% with most nests it is not clear if this cost is sufficient to outweigh the receiving multiple cowbird eggs. Most newly exposed demonstrable costs of egg acceptance. Equilibrium may host lineages lack effective defenses. Several recent be most likely in species for which parasitism is rela- studies have linked large-scale population declines of tively infrequent and in which parasitic eggs resemble songbirds with increasing levels of cowbird parasitism. those of the host, which greatly increases the probability Concerns about cowbird parasitism have stimulated an of mistakes. Nevertheless, distinguishing conclusively enormous amount of research since the late 1980s and between lag and equilibrium will require new experi- two major symposia have been published recently on mental and phylogenetic studies. this subject. This increased focus on cowbirds has resulted in pressure to trap and kill cowbirds to reduce their impacts on hosts. However, host species that have IV. IMPACTS ON HOST declined have experienced massive loss of habitat POPULATION DYNAMICS and increased rates of nest predation in addition to increased rates of cowbird parasitism. For example, Brood parasitism is undoubtedly almost always costly most species that breed in riparian habitat for individual hosts, but the extent to which brood in the southwestern United States have declined in parasites threaten host populations is much less clear. the past century but so too has their habitat. Dams, Most brood parasites have little impact on host popula- water diversions, overgrazing, urbanization, and exotic tions, presumably because coevolutionary processes re- plants have resulted in the degradation or loss of duce the frequency of parasitism and because most para- more than 90% of the riparian habitat present a sites are much less abundant than their hosts. In century ago in the Southwest. Even if cowbirds are Europe, for example, less than 5% of most host nests not the primary cause of some or all declines, they are parasitized by the common cuckoo. One brood para- may now be exposing some reduced host populations site, the giant cowbird, has even been hypothesized to to additional stresses that threaten the populations benefit its hosts because its nestlings remove parasitic with extinction. Importantly, a cowbird population botfly eggs from its host nestmates. can be stable or grow even as it pushes a rare host NEST PARASITISM 373 to extinction because individual female cowbirds do may have prevented the extinction of several endan- not specialize on single host species. gered host species and races, and two endangered hosts Nevertheless, it is not clear whether cowbird parasit- have increased in population since cowbird control be- ism threatens more than a few host species. Species gan. However, control programs for two other endan- that are most at risk are those with small geographic gered hosts resulted in no increases in the sizes of host ranges that are wholly included within areas that con- breeding populations, even though all control programs tain abundant cowbird foraging habitat (pastures, feed- have resulted in increases in host reproductive output. lots, mowed grass, and bare soil) and that cannot be Unfortunately, even when host populations increase, rescued by emigrants from more productive popula- cowbird control must be done every year because cow- tions. This list includes several endangered species that bird removal has little or no year-to-year effect on the are brown-headed cowbird hosts (Kirtland’s warbler, numbers of cowbirds that occur in an area due to high black-capped vireo, least Bell’s vireo, and southwestern dispersal by cowbirds. Therefore, most workers view willow flycatcher) and many species restricted to islands cowbird control as a temporary, stop-gap measure, al- in the Caribbean that have recently been invaded by though a U.S. Fish and Wildlife recovery plan for the shiny cowbirds (yellow-shouldered blackbird and least Bell’s vireo in California advocates cowbird control Puerto Rican vireo). in ‘‘perpetuity.’’ Larger scale control programs, such as Most species in North America have very large geo- killing cowbirds by the millions at winter roosts, have graphic breeding ranges that include regions in which been suggested by some workers but do not seem justi- cowbirds are rare and restricted to areas near human fied given that regional breeding season control pro- habitations. These refugia from cowbird parasitism tend grams are effective in eliminating nearly all cowbirds to occur in large, unfragmented habitats that may act from the ranges of endangered hosts. Such large-scale as ‘‘sources’’ of surplus host young that can recolonize killing programs may also raise important ethical issues populations in more fragmented habitats in which levels and some workers have argued that an undue emphasis of parasitization (and nest predation) are often very on cowbirds may detract from more productive and high. In these population ‘‘sinks’’ in fragmented habi- more long-lasting management actions such as habi- tats, levels of reproduction may be too low to compen- tat restoration. sate for adult mortality; that is, such populations can only be sustained by immigration from source habitats. Evidence for this source–sink scenario derives from V. ECOLOGY AND SOCIAL BEHAVIOR well-documented sink populations in fragmented mid- western U.S. forests. These populations are nevertheless OF BROOD PARASITES relatively stable, probably due to immigration from pop- ulations in large, unfragmented forests in the region in The lack of parental care in brood parasites potentially which levels of both parasitism and nest predation are sets them apart from most other birds in their foraging very low and reproductive success is sufficiently high ecology, mating systems, spacing behavior, and vocal for populations to act as sources. Such large-scale development. source–sink population dynamics can slow the evolu- tion of host defenses because most young are being produced in areas in which cowbird parasitism is rare. A. Foraging Ecology Most brood parasites have unusual diets and foraging behavior. Honeyguides eat wax, many cuckoos eat hairy A. Cowbird Management and toxic caterpillars, and cowbirds prefer to forage in Many cowbird control programs have been initiated short grass close to ungulates such as cattle, horses, or as a result of concern regarding several endangered bison. In each of these cases, either the diet of the brood songbird species. Controlling cowbirds, which is easily parasite would be difficult for nestlings to digest (wax done because their social nature attracts them into traps and hairy caterpillars) or the foraging habitat may be that contain cowbirds that function as decoys or ‘‘bait,’’ so ephemeral (the proximity of ungulates) that it may has become a multimillion dollar a year business in the not be available for an entire nesting cycle. It is not American Southwest and there are also active programs known, however, if these unusual foraging ecologies elsewhere in North America and in the Caribbean. Local were precursors of brood parasitism or if they were control efforts, in conjunction with habitat restoration, made possible by the evolution of brood parasitism. 374 NEST PARASITISM

B. Mating Systems In areas of home range overlap, some host nests are often parasitized by several female cowbirds. Cuckoos Mating system theory predicts that birds freed from the may occupy areas that are more mutually exclusive and needs of parental care should be promiscuous. Never- thus this is closer to classic territoriality. In areas of low theless, avian brood parasites show a remarkable array abundance, cowbirds may occupy mutually exclusive of mating systems, including monogamy and resource- areas and thus may appear to be territorial. based polygyny. This variation in mating behavior has Cowbirds often breed and feed on a daily basis in been attributed to the diverse ways in which parasites different areas that can be separated by as much as 15 gain access to nests (some of which require cooperation km. After searching for nests in the morning, cowbirds between several individuals), to the great variety of commute to feeding sites in pastures, plowed fields, foraging ecologies of brood parasites, and to such un- and in other habitats with short mowed grass. The usual features of individual host–parasite systems as uncoupling of breeding and feeding areas made possible strongly male-biased sex ratios in cowbirds. In general, by brood parasitism allows cowbirds to select a wide there have been few detailed studies of the mating sys- variety of breeding habitats, even if there are only a few tems of brood parasites. The mating system of the foraging sites in a region. Some cuckoos also appear to brown-headed cowbird has received the most study have very large home ranges. Home range size is related among parasitic birds and has been described as promis- to a bird’s body size and trophic level, with predators cuous, polygynous, or monogamous. Promiscuity has needing especially large areas. Although cowbirds and been proposed because it is a common sight to see cuckoos are not predators of other birds, adult breeding two or more males associating with a single female. birds are a key resource for them, and they have home However, studies of color-marked cowbirds that en- ranges similar in size to those of raptors that feed on abled researchers to determine which males and females adult birds. mated together and associated with one another the most have indicated monogamy. A recent DNA finger- printing study confirmed that monogamy prevails and even found that cowbirds have fewer matings outside D. Vocal Behavior of their pair bonds than do most nonparasitic songbirds. Brood parasites have provided some of the clearest ex- Field observations show that the mate faithfulness in amples of genetically hard-wired vocal behavior, but cowbirds is due both to males guarding females from recent studies have also shown a key role for subsequent the advances of other males and to females being reluc- learning in the modification of vocalizations. Because tant to mate with males other than their usual consort. cowbirds are raised by different species, it is not surpris- Because males outnumber females, many males do not ing that at least one of their songs, the perched song, acquire a mate. A single study demonstrated promiscu- is genetically programmed. Cowbirds, however, learn ity in an area in which extremely high cowbird abun- to modify this song in response to female preferences dances may have made it difficult for the birds to main- and interactions with other males and develop individ- tain pair bonds. ual repertoires of approximately five perched song In one Asian , males defend beehives and types. A second song type, the flight whistle, is almost only allow access to wax if the females mate with them. totally learned. It occurs as discrete spatial dialects gen- In some viduine weavers and cuckoos, males display erally tens of kilometers in diameter, which are so vari- from prominent perches and are chosen by females able that trained observers may fail to recognize as presumably on the basis of mate quality, which may be cowbird vocalizations whistles from dialects they have indicated by song elements or plumage. not yet experienced. These dialects are true examples of culture and show that considerable amounts of bio- diversity within a species can be related to learned/ cultural differences rather than to genetic differences. C. Spacing Behavior Unlike the majority of songbirds that complete their Many brood parasites defend breeding areas that are vocal development by the time they are 1 year old, male rich in hosts. At least some cowbirds and many cuckoos cowbirds in some regions do not master local versions defend home ranges against conspecifics. However, the of whistles and perched songs until they are 2 years old. home ranges are technically not true territories because Vocal learning occurs among one other parasitic several males and females often occupy the same area. group. Male viduine weavers incorporate elements of NEST PARASITISM 375 their host’s song into their repertoire, which presum- The rejection of parasitic eggs laid before the host has ably enables females to choose a male raised by the begun laying is a common adaptation against CBP but same host species. As such, viduine weavers show a high does not necessarily involve egg recognition. True egg potential for sympatric speciation by cultural learning. recognition, in which a bird discriminates among egg Such assortative mating maintains coevolutionary adap- types, is a rare response to CBP because it is difficult tations such as the gape mimicry that is specific to just to evolve because parasitic and host eggs are similar in one host. Assortative mating is essential because gape appearance, given that they are from the same species. appearance is due to paternal and maternal genes. In Also, CBP is less deleterious to hosts than is interspecific contrast, common cuckoo and cowbird males do not parasitism because it does not involve adaptations for incorporate host songs into their repertoires, perhaps killing host nestlings or asymmetries in size and incuba- because egg appearance, which is essential for accep- tion period that accomplish the same end. However, tance by their hosts, is determined solely by the mater- egg recognition in response to CBP does occur in some nal genotype. birds such as ploceid weaver finches of Africa and Asia, which have extensive variation among the eggs of con- specific females. Indeed, it has been hypothesized that the extreme intraspecific egg variation of these weavers VI. CONSPECIFIC BROOD PARASITISM is an evolutionary response to CBP. Some species with typical amounts of variation in egg appearance may lay Female birds sometimes lay their eggs in the nests of smaller clutches than they could potentially feed to conspecifics. This behavior has been documented in leave room for parasitic eggs. If hosts cannot recognize more than 100 bird species and has also been docu- parasitic eggs, then laying a smaller clutch might avoid mented in some insects and fish. Among birds, conspe- starvation of most or all nestlings when parasitism cific brood parasitism (CBP) is most prevalent among does occur. precocial species, such as ducks and gallinaceous birds, CBP may actually affect population sizes. In species and among altricial species that breed colonially and/ with frequent CBP, fitness may be reduced at high popu- or use specialized nest sites, such as cavities. CBP is lation densities, which may cause populations to be rare or absent in most songbirds, even some that are cyclic. Nests of wood ducks that are placed too close colonial, and it has been well studied in swallows. together are subject to extreme levels of CBP, which In many cases, CBP occurs when individuals lose can reduce population fitness. their nests during the laying period or are unable to obtain a territory or nest site. In these situations, indi- viduals may be making the best of a bad situation. In some species, CBP may be an alternative reproductive strategy that increases reproductive success by spread- VII. CONCLUSIONS AND ing the risk among several nests, reducing competition RESEARCH NEEDS within a female’s own nest, and exploiting the parental care of others. Recent theoretical work suggests that Brood parasitism raises fascinating questions about co- CBP is a necessary precursor to the evolution of inter- evolution and conservation. Studies of brood parasitism specific brood parasitism. Conspecific brood parasitism provide some of the strongest evidence of microevolu- is especially prominent in New World cuckoos of the tion yet documented in vertebrates. The extreme mobil- genus Coccyzus, which tend their own nests and also ity of cowbirds and their potential threats to host popu- parasitize conspecifics and congeners. In at least some lations illustrate the importance of landscape-level species, such as the ostrich and some ducks, individuals processes in conservation biology. Cowbird song devel- may benefit from CBP because extra eggs or young opment has become a model system showing that both dilute losses to predators. genetic and environmental factors are important to the Nest defenses against CBP parallel those used to development of behavior in general. Despite these and combat interspecific brood parasitism. Hosts guard their other lessons learned from studies of brood parasitism, nests, especially at high population densities in barn there are many unanswered questions that are vital to swallows, which may exert a significant cost of nesting our understanding of this subject. colonially. Parasitic females of another swallow species, We still have much to learn about topics such as the cliff swallow, actually carry their eggs between nests. (i) the frequency of recognition errors in hosts with 376 NEST PARASITISM rejection behavior, (ii) reasons for the general lack of See Also the Following Articles parasite nestling recognition by hosts, (iii) the fre- quency of ‘‘mafia’’-like behaviors and nest predation by BIRDS, BIODIVERSITY OF • COEVOLUTION • PARASITISM • • brood parasites, (iv) how brood parasites choose from POPULATION DYNAMICS SPECIES COEXISTECE available hosts, (v) whether or not cowbirds pose a significant threat to populations of widespread host species, and (vi) the genetic mating systems of brood Bibliography parasitism. The recent invasion of the United States by Davies, N. B., Bourne, A. F. G., and de L. Brooke, M. (1989). Cuckoos a new generalist brood parasite, the shiny cowbird, also and parasitic ants: Interspecific brood parasitism as an evolution- offers an excellent opportunity to study its interactions ary arms race. Trends Ecol. Evol. 8, 2–4. Johnsgard, P. A. (1997). The Avian Brood Parasites: Deception at the with other cowbirds and its new hosts. Finally, most Nest. Oxford Univ. Press, New York. host–parasite systems remain very poorly known, espe- Ortega, C. P. (1998). Cowbirds and Other Brood Parasites. Univ. of cially those of the New World cuckoos, many Old Arizona Press, Tucson. World cuckoos, and some old honeyguides. Some mem- Payne, R. B. (1997). Avian brood parasitism. In Host–Parasite Coevolu- bers of the two latter groups are so poorly known that tion: General Principles and Avian Models (D. H.Clayton and J.Moore, Eds.), pp. 338–369. Oxford Univ. Press, New York. there is no direct evidence that they are parasitic. How- Rothstein, S. I., and Robinson, S. K. (1998). Parasitic Birds and Their ever, they are assumed to be so because closely related Hosts: Studies in Coevolution, Oxford Ornithology Series. Oxford species are known to be parasitic. Univ. Press, Oxford.