Do peahens not prefer peacocks with more elaborate trains?

Available online at www.sciencedirect.com Do peahens not prefer peacocks with more elaborate trains? AD ELIN E L OYA U , MA RION PE TR IE †, M ICH E L S AIN T J AL M & G AB RI E L E SORC I§ *Station d’Ecologie Expe´rimentale du CNRS a` Moulis, CNRS UMR 2936 yEvolutionary Biology Group, Newcastle University zMuse´um National d’Histoire Naturelle (MNHN), UMR 5173 xBioge´osciences, CNRS UMR 5561, Universite´ de Bourgogne (Received 17 June 2008; initial acceptance 9 July 2008; final acceptance 23 July 2008; published online 7 September 2008; MS. number: D-08-00401) Keywords: female mate choice; interpopulation variability; ornament; Pavo cristatus; peafowl Ever since , the peacock’s train has been population can be misleading if generalized to the whole cited as the icon of an extravagant conspicuous second- ary sexual trait that has evolved through female mate The peacock’s train is a complex structure that cannot be summarized with only three morphological traits lenged this idea. They monitored female mate choice (number of eyespots, train symmetry and train length).
during 7 years in a feral peafowl, Pavo cristatus, popula- Two previous studies showed that the density and the tion in Japan and found no correlation between male coloration and iridescence of eyespots in the train have mating success and three morphological train traits.
the potential to be involved in mate choice They concluded that ‘combined with previous results, our findings indicate that the peacock’s train is not cur- cannot discard the possibility that they did not measure rently the universal target of female choice’ and pro- elements of the signal most relevant for female mate posed ‘that the peacock’s train is an obsolete signal for choice. Their main concern was the absence of correla- which female preference has already been lost or weak- tions between the number of eyespots and mating suc- cess, consistent over the 7 years of their study. One their conclusions are far too strong, particularly since explanation for this absence of correlation could be three independent studies have found a relationship be- that, in the Japanese population, the train contains tween train features and mating success ( a trait preferred by the females that is not always posi- tively correlated with the number of eyespots. Hence, it would not be possible to detect whether the train fore to draw attention to alternative explanations and contains a signal under sexual selection. On the basis conclusions that are essential for the understanding of the complexity of mate choice. We first suggest some that train symmetry and train length were not compo- possible nonadaptive and adaptive explanations for the nents of the signal received by the females. However, reported differences in female preferences in the peafowl.
does that mean that no signal exists? More recently, We then show that plasticity in mate choice is a wide- a correlational study suggested that females may use spread phenomenon across a large spectrum of species.
Therefore, we suggest that findings based on a single than the number of eyespots, perhaps because thisfeature is more quickly assessed. This finding is notinconsistent with the fact that experimentally reducingthe number of eyespots in the train decreased mating Correspondence: A. Loyau, Station d’Ecologie Expe´rimentale du CNRS a` Moulis, CNRS UMR 2936, 09200 Saint Girons, France (email: ). M. Petrie is at the School of Clinical Medical Sciences, William Leech Building, Newcastle University, Newcastle- tions between eyespot number, train length and mating upon-Tyne NE1 4HH, U.K. M. Saint Jalme is at the Muse´um National success, although the relationship between eyespot d’Histoire Naturelle (MNHN), UMR 5173, MNHN-CNRS-Paris 6,France. G. Sorci is at Bioge´osciences, CNRS UMR 5561, Universite´ de number and train length was negative in a sample of culled birds from one lek, all shot on the same day Ó 2008 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
density was the trait with the highest coefficient of vari- Takahashi and colleagues may be able to test for an ation (10.3%) while the coefficient of variation of the effect of eyespot density, since they measured the number of eyespots and the train length of 24 pea- It is unfortunate that Takahashi et al. did not provide cocks for 3 years. This would be a valuable addition the pattern of variation of the eyespot density in their to our understanding of how the various components population. Low variation in train morphology could of the signal in the male’s train affect female mate arise if genetic variability is low. Consideration of varia- all the studies of peahen mate choice were carried between studies and suggested that missed observations out in an unnatural environment. Apart from the obvi- of mating and small sample sizes of previous studies ous problem that feral populations of peafowl do not un- may play a part, claiming that their study had ‘the lon- dergo the same selection pressures as wild populations, gest observation with the largest sample size among pea- these populations were established outside the original distribution range of the species several centuries ago, that the number of copulations observed is critical to usually using a small number of individuals which could a meaningful analysis of variance in mating success.
have created a strong genetic bottleneck. Isolated small populations are expected to have reduced genetic vari- tions in 7 years, which amounts to ca. 38 copulations ability because of a combination of founder effects, in- per annum from 20 to 37 territorial males observed in a number of instances where the conspicuous coloration 116 copulations in 1 year (1989) from observations of of mainland birds has been lost on small island popula- 30 territorial males (almost three times as many). We sus- tions and genetic drift is often cited as a cause pect the difference in the number of copulations ob- and references therein). Another possibility is served may be caused by a difference in population size that, in some captive populations, not all founders and in particular the number of reproductively active have come from the same place in the native range of hens observed. The Japanese study reported a total pop- the species, resulting in a higher genetic diversity. There- fore, isolation combined with relaxed or modified selec- stated is male biased, whereas the Whipsnade population tion pressures and/or mixed origins of founders could was estimated at 179 birds. Unfortunately, watching the explain why Takahashi et al. found different results to same small population for a number of years does not overcome the problem of error associated with small Reasons for the discrepancy between Takahashi et al.’s sample size, as the dependent variable is the number of results and previous studies may not only reflect differ- copulations per male per annum; the possibility there- ences in methodology. There are several other explana- fore remains that there may be insufficient observations tions for what might be called plastic female choice.
of copulations to detect an effect in the Japanese study.
Indeed, divergence in behaviour among populations There are other possible explanations for the discor- of a given species is widespread ; see also dance between studies, including uncontrolled variation as a result of small methodological differences. For example, train length was measured ‘in early spring’ in In particular, there is a growing body of evidence demon- beginning to the peak of the mating season in Japan.
strating a significant variation in male traits and female This could confound the relationship, as there is marked variation in the timing of the start and length of time a wide range of taxa including insects, amphibians, fishes over which feathers drop. The number of feathers lost by the end of the season could relate to performance during the mating season, if expenditure on reproductive guppies, Poecilia reticulata, among 11 populations in Tri- effort or mating success leads to an earlier or compressed nidad and showed that mate preferences varied in inten- sity, direction and the number of traits used. In this species, females also differ in the relative importance male mating success. The most successful male obtained that they place on male orange coloration or iridescence only 14.9e31.4% of copulations per year whereas in ear- lier studies these values were consistently higher (36.4% Regardless of methodological differences (as suggested by ) or small sample sizes (as suggested by by Takahashi et al. there appeared to be a low consensus tions can arise by several mechanisms. Populations might among females about whom to mate with, an unusual differ in the amount of genetic variation for the trait of finding in a lekking species. Could that be a result of interest because of a founder effect and/or a genetic drift.
low variation in train morphology which does not allow Geographical variation in selection pressures can also pro- peahens to distinguish between the different potential duce genetic structuring among populations (e.g. mates? In the French population, we found that eyespot genetic differences may account for the differences of pat- is a galliform bird, the train is likely to develop in the cost of trait expression may vary with environmental con- absence of oestrogen and consequently is unlikely to be ditions such as climate, predation risk, pathogen preva- lence, conspecific density and sex ratio (e.g. Therefore, they expected females to discard this trait in A given trait within one environment may not reliably indicate mate quality in another environment male preference for the male’s train is supposed to have been ‘lost or weakened’. If so, it implies that this prefer- in female mate choice is thought to result from spatial ence existed at some point and thus may still exist. Eye- and temporal variation in environmental conditions spots have independently evolved in several taxa (birds and butterflies) and it is likely that such a trait originally evolved through the exploitation of a sensory bias ( ). A spectacular illustration of the concept of ) which would also account for the ‘hypnotic effect’ adaptive plasticity in female mate choice has been re- of the train on peahens described by .
The coexistence of the heritability of female preference ting, Calamospiza melanocorys, sexual selection on male (through the sensory bias) and the heritability of the traits can vary dramatically across years and even show re- versals in the direction of a single trait. Importantly, in support for the idea that the train evolved through Fish- a given year, these changes usually favour male traits er’s runaway process (). At some point, as the that correlate with the highest reproductive success train became an extreme trait, it presumably crossed a threshold and started to impose costs on the bearer, that a seemingly inconsistent female mate choice across with only the better quality males being able to grow years in fact targets the male phenotype that within and display the more ornamented trains. Some results al- each year provides the best reproductive output. Even though different selection mechanisms may arise for temporal and spatial variation, this example illustrates has evolved as a Fisherian trait and is maintained as that, if temporal differences can be found within a given a good genes indicator, whatever the hormonal control.
population, it is not surprising to find differences In other galliform species the development of the train between geographically distinct populations. Finally, may never have reached the threshold, leading to variation in male traits and female preferences can a loss of ornamental traits and associated display account for learning and cultural transmission. Indeed, male bowerbirds may copy peers when building bowers not see why the form of hormonal control of train and this would explain interpopulation differences in expression negates previous findings and know of no evidence to suggest why it should. The peahens may brown-headed cowbirds, Molothrus ater, are a conse- ‘appear to be quite exceptional if they indeed select quence of complex interactions between male learning and population differences in female preferences (re- page 1215). However, the existence of secondary sexual ornaments and display behaviours as complex and influences such as sexual imprinting and mate choice extravagant as the peacock’s train and display is itself copying can induce variation in female preferences male song can be influenced by social experience in it is important to publish negative results, and hopefully female zebra finches, Taeniopygia guttata, and canaries, further such studies will be published so that a more meaningful meta-analysis can be carried out. However, Taken together, these examples highlight the fact that the failure to detect evidence of mate choice in one great caution is required when generalizing results from study based on a limited array of traits does not mean a single population to the whole set of populations of that females do not prefer males with more elaborate a given species, as already pointed out by trains. Only a very strict experimental study across several captive and wild populations could demonstrate by three independent groups on three independent pop- that. To date, only one study on peafowl mate choice and, unfortunately, the number of eyespots was not re- elaboration in mate choice, it is thus rather surprising corded. Further studies of wild populations with natural levels of genetic variation will be particularly useful proposed that ‘the peacock’s train is an obsolete sig- in extending our understanding of peahen mating nal for which female preference has already been lost or We thank Dirk S. Schmeller, Alexis S. Chaine, L. Morris the peacock’s train might be an indicator of good genes, Gosling and Jean Clobert who provided valuable com- despite strong evidence supporting that hypothesis ments and discussion and two anonymous referees who Kwiatkowski, M. A. & Sullivan, B. K. 2002. Geographic variation in sexual selection among populations of an iguanid lizard, Sauroma- Baird, T. A., Fox, S. F. & McCoy, J. K. 1997. Population differences lus obesus (¼ ater). Evolution, 56, 2039e2051.
in the roles of size and coloration in intra- and intersexual selection Loyau, A., Saint Jalme, M. & Sorci, G. 2005a. Intra and inter-sexual in the collared lizard, Crotaphytus collaris: influence of habitat and selection for multiple traits in the peacock (Pavo cristatus). Ethol- social organization. Behavioral Ecology, 8, 506e517.
Brooks, R. 2002. Variation in female mate choice within guppy pop- Loyau, A., Saint Jalme, M., Cagniant, C. & Sorci, G. 2005b. Multiple ulations: population divergence, multiple ornaments and the sexual advertisements honestly reflect health status in peacocks maintenance of polymorphism. Genetica, 116, 343e358.
(Pavo cristatus). Behavioral Ecology and Sociobiology, 58, 552e557.
Chaine, A. S. & Lyon, B. E. 2008. Adaptive plasticity in female mate Loyau, A., Gomez, D., Moureau, B., The´ry, M., Hart, N. S., Saint choice dampens sexual selection on male ornaments in the lark Jalme, M., Bennett, A. T. D. & Sorci, G. 2007a. Iridescent struc- turally-based coloration of eyespots correlates with mating success Coleman, S. W., Patricelli, G. L. & Borgia, G. 2004. Variable female in the peacock. Behavioral Ecology, 18, 1123e1131.
preferences drive complex male displays. Nature, 428, 742e745.
Loyau, A., Saint Jalme, M., Mauget, R. & Sorci, G. 2007b. Male Dale, S. T., Slagsvold, T., Lampe, H. M. & Sætre, G. P. 1999. Population sexual attractiveness affects the investment of maternal resources divergence in sexual ornaments: the white forehead patch of Norwe- into the eggs in peafowl (Pavo cristatus). Behavioral Ecology and gian pied flycatchers is small and unsexy. Evolution, 53, 1235e1246.
Danchin, E., Giraldeau, L.-A., Valone, T. J. & Wagner, R. H. 2004.
Madden, J. R. 2006. Interpopulation differences exhibited by spot- Public information: from nosy neighbors to cultural evolution.
ted bowerbirds Chlamydera maculata across a suite of male traits and female preferences. Ibis, 148, 425e435.
Darwin, C. 1871. The Descent of Man and Selection in Relation to Sex.
Ma´rquez, R. & Bosch, J. 1997. Male advertisement call and female Princeton, New Jersey: Princeton University Press.
preference in sympatric and allopatric midwife toads. Animal Endler, J. A. & Houde, A. E. 1995. Geographic variation in female pref- erences for male traits in Poecilia reticulata. Evolution, 49, 456e468.
Martins, E. P., Bissell, A. N. & Morgan, K. K. 1998. Population differ- Espmark, Y. E. & Lampe, H. M. 1993. Variations in the song of the ences in a lizard communicative display: evidence for rapid change pied flycatcher within and between breeding seasons. Bioacoustics, in structure and function. Animal Behaviour, 56, 1113e1119.
Møller, A. P. & Petrie, M. 2002. Condition dependence, multiple Fisher, R. A. 1930. The Genetical Theory of Natural Selection. Oxford: sexual signals, and immunocompetence in peacocks. Behavioral Forstmeier, W., Hasselquist, D., Bensch, S. & Leisler, B. 2006.
Petrie, M. 1992. Peacocks with low mating success are more likely Does song reflect age and viability? A comparison between two to suffer predation. Animal Behaviour, 44, 585e586.
populations of the great reed warbler Acrocephalus arundinaceus.
Petrie, M. & Halliday, T. 1994. Experimental and natural changes in Behavioral Ecology and Sociobiology, 59, 634e643.
the peacock’s (Pavo cristatus) train can affect mating success.
Foster, S. A. 1999. The geography of behaviour: an evolutionary Behavioral Ecology and Sociobiology, 35, 213e217.
perspective. Trends in Ecology & Evolution, 14, 190e195.
Petrie, M. & Kempenaers, B. 1998. Extra-pair paternity in birds: Freeberg, T. M. 2000. Culture and courtship in vertebrates: a review explaining variation between species and populations. Trends in of social learning and transmission of courtship systems and mating patterns. Behavioural Processes, 51, 177e192.
Petrie, M., Halliday, T. & Sanders, C. 1991. Peahens prefer peacocks Gil, D., Cobb, J. L. S. & Slater, P. J. B. 2001. Song characteristics are with more elaborate trains. Animal Behaviour, 41, 323e331.
age-dependent in the willow warbler, Phylloscopus trochilus.
Petrie, M., Cotgreave, P. & Stewart, I. 1996. Variation in the train morphology of peacocks (Pavo cristatus). Journal of Zoology, 238, Griffith, S. C., Stewart, I. R. K., Dawson, D. A., Owens, I. P. F. & Burke, T. 1999. Contrasting levels of extra-pair paternity in mainland and is- Petrie, M., Cotgreave, P. & Pike, T. W. In press. Variation in the land populations of the house sparrow (Passer domesticus): is there an peacock’s train shows a genetic component. Genetica.
‘island effect’? Biological Journal of the Linnean Society, 68, 303e316.
Pfennig, K. S. 2000. Female spadefoot toads compromise on mate Hagelin, J. C. & Ligon, J. D. 2001. Female quail prefer testosterone- quality to ensure conspecific matings. Behavioral Ecology, 11, mediated traits, rather than the ornate plumage of males. Animal Ptacek, M. B. & Travis, J. 1996. Inter-population variation in male Hill, G. E. 1994. Geographic variation in male ornamentation and mating behaviours in the sailfin mollie, Poecilia latipinna. Animal female mate preference in the house finch: a comparative test of models of sexual selection. Behavioral Ecology, 5, 64e73.
Ridley, M. W., Lelliott, A. D. & Rands, M. R. W. 1984. The court- Jaenike, J. R. 1973. A steady state model of genetic polymorphism ship display of feral peafowl. World Pheasant Association Journal, 9, on islands. American Naturalist, 107, 793e795.
Jennions, M. D. & Petrie, M. 1997. Variation in mate choice and Rodd, F. H., Hughes, K. A., Grether, G. F. & Baril, C. T. 2002. A mating preferences: a review of causes and consequences. Biolog- possible non-sexual origin of mate preference: are male guppies mimicking fruit? Proceedings of the Royal Society of London, Series Kimball, R. T., Braun, E. L., Ligon, J. D., Lucchini, V. & Randi, E. 2001. A molecular phylogeny of the peacock-pheasants (Galliformes: Polyplec- Ryan, M. J. 1990. Sexual selection, sensory systems and sensory tron spp.) indicates loss and reduction of ornamental traits and display exploitation. Oxford Surveys in Evolutionary Biology, 7, 157e195.
behaviours. Biological Journal of the Linnean Society, 73, 187e198.
Safran, R. J. & McGraw, K. J. 2004. Plumage coloration, not length Klappert, K., Mazzi, D., Hoikkala, A. & Ritchie, M. G. 2007. Male or symmetry of tail-streamers, is a sexually selected trait in North courtship song and female preference variation between American barn swallows. Behavioral Ecology, 15, 455e461.
phylogeographically distinct populations of Drosophila montana.
Shine, R., Webb, J. K., Lane, A. & Mason, R. T. 2006. Flexible mate choice: a male snake’s preference for larger females is modified by the sizes of females encountered. Animal Behaviour, 71, 203e Wiens, J. J. 2001. Widespread loss of sexually selected traits: how the peacock lost its spots. Trends in Ecology & Evolution, 16, 517e Takahashi, M., Arita, H., Hiraiwa-Hasegawa, M. & Hasegawa, T.
2008. Peahens do not prefer peacocks with more elaborate trains.
Yasmin, S. & Yahya, H. S. A. 1996. Correlates of mating success in

Source: http://www.adeline-loyau.net/publications/Loyau_etal_AnimBehav2008.pdf