Maternal body condition and plasma hormones affect offspring sex ratio in peafowl T HOM AS W . PIK E & MA RI ON PET RIE Evolution and Behaviour Research Group, University of Newcastle upon Tyne (Received 3 August 2004; initial acceptance 15 September 2004; final acceptance 16 December 2004; published online 29 August 2005; MS. number: 8233R) In theory, females that can afford to do so may increase their fitness by investing in offspring of the sex with the greater probability of attaining high reproductive success. This has been observed in a wide variety of mammalian, reptilian and avian species although the proximate mechanism remains a mystery. Using a captive population of peafowl, Pavo cristatus , we investigated the relation between maternal quality, offspring sex ratio and plasma concentrations of the reproductive hormones testosterone and 17 b - oestradiol and the principal avian stress hormone corticosterone. Each peacock was paired with three peahens that differed with respect to their relative body condition, creating a condition and dominance hierarchy within each pen and thus a situation in which we would predict investment in offspring to vary between hens. We found significant intercorrelations between maternal body condition (but not dominance rank or clutch size), maternal plasma levels of corticosterone and testosterone (but not 17 b - oestradiol) and clutch sex ratio, such that good maternal condition, low plasma corticosterone and high levels of testosterone were associated with male biases in the sex ratio and increased investment in male eggs. The observed biases were probably present at laying, and thus add to the growing number of studies showing primary sex ratio adjustment in response to maternal body condition, and furthermore may indicate a role for corticosterone and testosterone in the avian sex manipulation process. Ó 2005 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. In sexually reproducing organisms, frequency-dependent selection generally leads to an evolutionarily stable strat- egy of equal energy expenditure by parents on sons and daughters (Fisher 1930). However, the relative investment of resources should be sensitive to the potential fitness gains through either sex (Charnov 1982). For example, in polygynous species, where the reproductive success of males is more variable than that of females, a male-biased offspring sex ratio is predicted for high-quality parents (Trivers & Willard 1973), assuming that the quality of parent and offspring is correlated. Alternatively, maternal investment could be distributed along a scale in relation to parental quality: mothers in poor condition, for instance, might maximize their fitness by producing a single daughter, whereas females in progressively better condi- tion might achieve maximum fitness by producing a single son, two daughters, a daughter and a son, and so on (Williams 1979). As a result, the investment strategy might depend on the total number of offspring produced in a single breeding attempt. However, it is also possible that sex-biased provisioning could be used to offset a higher mortality of one sex, resulting in the production of an even sex ratio despite biases in investment (Maynard Smith 1980). A number of recent studies on birds have documented sex-biased investment in response to both maternal condition, in captivity (Bradbury & Blakey 1998; Kilner 1998) and in the wild (Nager et al. 1999; Whittingham & Dunn 2000; Kalmbach et al. 2001; Clout et al. 2002), and maternal social status (Leonard & Weatherhead 1996; Nishiumi 1998), although few studies have explicitly addressed the underlying proximate mechanisms (e.g. Geiringer 1961; Petrie et al. 2001; Veiga et al. 2004). One possibility is the involvement of hormones, either circu- lating in the breeding female or deposited in the eggs she produces. Hormone concentrations are remarkably labile, fluctuating rapidly in response to a number of social and environmental factors (Schwabl et al. 1997; Gil et al. 1999; Whittingham & Schwabl 2002). It is this plasticity combined with their close association with reproductive Correspondence and present address: T. W. Pike, Division of Environmental and Evolutionary Biology, Graham Kerr Building, University of Glasgow, Glasgow G12 8QQ, U.K. (email: t.pike@ bio.gla.ac.uk). M. Petrie is at the Evolution and Behaviour Group, School of Biology, University of Newcastle upon Tyne, Newcastle upon Tyne, NE2 4HH, U.K. 745 0003–3472/04/$30.00/0 Ó 2005 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. ANIMAL BEHAVIOUR, 2005, 70, 745–751 doi:10.1016/j.anbehav.2004.12.020 physiology that has led to the suggestion that hormones may form an integral part of the mechanism allowing a breeding female to bias her investment in sons and daughters (Krackow 1999; Pike & Petrie 2003). For in- stance, Veiga et al. (2004) have shown that artificially elevating circulating testosterone levels in breeding female spotless starlings, Sturnus unicolor , induce them to produce a male-biased sex ratio. However, this was most probably mediated by an effect of testosterone on the acquisition and maintenance of a high social dominance rank, which in turn may have affected circulating levels of hormones other than testosterone (Pike & Petrie 2003). Our primary aim in this exploratory study was to create a situation in which investment in male and female offspring would be predicted to vary between breeding hens by varying female quality (in terms of body condi- tion and social dominance rank) while controlling for sire and other environmental effects. As our study species, we used a captive population of peafowl, Pavo cristatus , a highly sexually selected, polygynous species in which males (the larger sex) display to females at leks and contribute no direct resources for reproduction (other than sperm) to either females or their offspring. Under such a system, one male can monopolize the reproduction of many females with the obvious corollary that many males will fail to reproduce. Since investment in the eggs a female produces is constrained by her quality (e.g. Ricklefs & Marks 1983; Bancroft 1984; Bolton et al. 1992), we would expect a female to bias her level of investment according to the sex of the egg. Specifically, we predicted that high-quality peahens would produce a male-biased offspring sex ratio and/or allocate more resources to male eggs. To try to address the proximate mechanisms involved, we related sex ratio biases to plasma concentrations of three steroid hormones: the reproductive hormones tes- tosterone and 17 b -oestradiol and the principal avian stress hormone, corticosterone. These hormones have been implicated in sex ratio deviations (Geiringer 1961; James 1996; Petrie et al. 2001; Veiga et al. 2004), and are also known to vary markedly in response to maternal condi- tion (e.g. Wingfield et al. 1994; Schoech et al. 1997; Kitaysky et al. 1999) and dominance status (Wingfield et al. 1991; Whittingham & Schwabl 2002); thus they may allow the indirect hormonal mediation of sex ratio varia- tion in response to maternal quality. METHODS General Methods We did the study during the summer of 2002 at a peafowl farm in Norfolk, U.K. The birds used were a random sample from the farm population, all at least 3 years old and sexually experienced. They were housed in groups of four (one male with three females) in large outdoor pens (measuring approximately 4 ! 3 m and 3 m high) and supplied with food (pheasant breeder feed) and water ad libitum. At the time of capture (May 2002, just before laying began) we weighed females ( G 1 g) on hanging scales and measured tarsus length ( G 1 mm) with a wing rule. From these measurements we calculated a body condition index as the standardized residual from a linear regression of body weight on tarsus length. There was a highly significant correlation between maternal tarsus length and body weight ( r 2 Z 0.96, F 1,16 Z 342.03, P ! 0.001), and hence this index describes the body weight of a given individual as a deviation from predicted body weight for a bird having a given tarsus length. This index, also known as residual mass, is widely used in bird studies and is strongly positively correlated with the size of an individual’s subcutaneous fat reserves (Merila ̈ & Svensson 1995). Each male ( N Z 6) was randomly assigned to a pen and housed with three females. When allocating these females to pens, we used the large degree of natural intrapopulation variation that exists for body weight and size (mean body weight: 3.63 kg, range 2.80–4.55 kg; mean tarsus length: 196 mm, range 182–214 mm) so that each male was housed with one relatively high-condition female, one relatively low-condition female and a third female whose condition level fell between that of the other two females (mid condition). There was thus a distinct hierarchy of female condition levels within each pen and although no attempt was made to impose condition level boundaries for high-, mid- and low-condition females, mean body condi- tion differed significantly between females of different groups (high condition: 0.74 G 0.17; mid condition: ÿ 0.47 G 0.15; low condition: ÿ 0.99 G 0.15; one-way ANOVA: F 2,15 Z 33.68, P ! 0.001). Behavioural Observations We investigated the female dominance hierarchy within each pen by weekly 10-min observations for the duration of the breeding season. All observations were conducted at approximately the same time of day (1300–1500 hours), combined with the feeding of a favourite food (e.g. hardboiled egg). The order in which the cages were observed was randomized each week. During feeding all agonistic interactions between the females were recorded, and the dominance hierarchy was based on the pro- portion of winning or losing interactions with other females. A female lost an interaction if she retreated when pecked, chased or threatened by another female, and was considered dominant to another if she won more inter- actions than she lost with that female. We ignored agonistic interactions involving the male. There was rarely any physical contact between individuals during agonistic behaviour and interactions, although frequent, were no more severe than those occurring at other times. Sub- ordinate individuals could always escape and, although human intervention was planned if the level of aggression became too high, this never needed to be implemented. We ranked the females from 1 (highest rank) to 3 (lowest rank) according to which individuals they dominated. Hierarchies in all pens were linear and dominance ranks did not change over the breeding season. ANIMAL BEHAVIOUR, 70, 4 746 Egg Collection Observations took place between 18 May and 12 July, during which we watched pens daily between 1700 and 1900 hours (when the majority of eggs are laid). Only three eggs were laid outside this time, and we could accurately assign these to an individual female by know- ing the maternity of other eggs from the same pen. On laying, all eggs were individually marked with a nontoxic marker (date, laying female) and weighed (to G 0.01 g) on digital scales. Each week, all the eggs laid were incubated until hatching in a commercial incubator in which each egg had its own individual compartment so a chick could be reliably assigned to a particular egg. When the chick hatched, we took a small blood sample (2–15 m l), under licence from the Home Office, by brachial vein puncture, a portion of which we immediately transferred to a piece of filter paper. We also removed a piece of embryonic tissue from all fertile, unhatched eggs in which a visible embryo had developed. Blood and tissue samples were immediately stored in absolute ethanol at ÿ 20 C for up to 3 months until sex analysis. Blood Sampling Blood samples (0.5 ml), taken under Home Office licence, were obtained from all females approximately half way through the breeding season (on day 30 of 56). We took blood via a small puncture made to a wing vein with a 26 gauge needle within 2 min of capture to ensure that plasma corticosterone levels reflected baseline levels rather than stress levels. All samples were taken at ap- proximately the same time of day (1900–2000 hours) to control for circadian fluctuations in circulating hormone levels. Collected samples were separated by centrifuge (13 000 rpm for 5 min) and stored at ÿ 20 C for up to 3 months before being assayed for testosterone, 17 b -oestra- diol and corticosterone using commercially available enzyme immunoassay (EIA) kits (Immunodiagnostic Sys- tems Ltd., Bolden, Tyne & Wear, U.K.). All assays were performed precisely according to the manufacturer’s protocol. We ran samples in duplicate and compared hormone concentrations to a standard curve. Mean in- tra-assay coefficients of variation for testosterone, 17 b - oestradiol and corticosterone were 2.9, 4.6 and 5.7%, respectively, and assay sensitivities were 6, 4.6 and 230 pg/ml, respectively (manufacturer’s insert). All assays were performed blind to the sex ratio data. Molecular Sexing We extracted genomic DNA from approximately 0.1 g of embryonic tissue or 5 m l of blood using a proteinase K digestion followed by sodium chloride extraction and ethanol precipitation (Bruford et al. 1998). We then used the polymerase chain reaction (PCR) to amplify part of the W-linked avian CHD gene ( CHD-W ) in females and its non-W-linked homologue ( CHD-Z ) in both sexes using primers 2718R and 2550F (Fridolfsson & Ellegren 1999). PCR products were separated on 2% agarose gels and visualized with ethidium bromide. Birds were sexed according to the presence of the PCR products of CHD-Z (about 600 base pairs; both sexes) and CHD-W (about 400 base pairs; females only). All eggs that developed a visible embryo were successfully sexed. In all cases, DNA ex- tracted from known-sex individuals was used as standards and samples were sexed blind and in a random order. Statistical Analyses We analysed sex ratios (calculated as the number of males/total number of sexed eggs) at the individual clutch level by fitting a generalized linear mixed model (GLMM, Littell et al. 1996) using the GLIMMIX macro (binomial errors, logit link function) in SAS v.8 (SAS institute, Cary, NC, U.S.A.). Thus instead of using the proportion of male eggs per se, we modelled the response variable as the number of male offspring, with the total number of sexed eggs as the binomial denominator so that the analysis would be sensitive to the number of eggs from which the proportion was estimated. The GLIMMIX macro automat- ically adjusts for overdispersion (Littell et al. 1996). Where necessary, we included pen as a random factor to control for the nonindependence of multiple clutches from a single pen. Means are presented G SE, and all statistics are two tailed with the significance level set at 5%. Where multiple comparisons were made the presented P values have been Bonferroni adjusted with a Z 0.10 (Rice 1989). RESULTS Sex Ratio and Hormones All the females laid a single clutch of 5–20 eggs (mean clutch size 14.17 G 0.94). Altogether this comprised 249 eggs of which 187 developed a visible embryo (and hence could be successfully sexed) and 138 hatched. Six eggs could not be included in the analysis because they were cracked on collection and hence could not be incubated and subsequently sexed (see Arnold et al. 2003). Overall, the population sex ratio did not differ significantly from parity (0.48; binomial test: P Z 0.661), although there was a large range of clutch sex ratios produced by individual females (0.13–0.75). This variation in clutch sex ratio was significantly related to a laying female’s body condition, plasma corticosterone and plasma testosterone levels (Fig. 1), but not to her dominance rank ( F 2,10 Z 5.15, P Z 0.087), plasma 17 b -oestradiol levels ( F 1,11 Z 2.12, P Z 0.173) or clutch size ( F 1,11 Z 0.52, P Z 0.487). How- ever, although maternal body condition was significantly correlated with both plasma corticosterone ( r Z 0.79, F 1,11 Z 15.69, P Z 0.002) and testosterone ( r Z 0.93, F 1,11 Z 34.31, P ! 0.001), partial correlations between sex ratio and each of these three variables in turn, controlling for the remaining two, were not significant (condition: r Z 0.04, F 1,9 Z 0.57, P Z 0.468; corticoste- rone: r Z ÿ 0.09, F 1,9 Z 0.66, P Z 0.436; testosterone: r Z 0.10, F 1,9 Z 1.66, P Z 0.229), suggesting that all three variables explain a large proportion of the same variation in clutch sex ratio. PIKE & PETRIE: HORMONES AND SEX RATIO IN PEAFOWL 747 Timing of Sex Ratio Manipulation The observed sex ratio biases could have occurred as a result of either a difference in the primary sex ratio of eggs laid or differential embryo mortality. However, even though it was impossible to sex infertile eggs or eggs where a visible embryo failed to develop, our data support the former possibility. Although very early differential mortality cannot be ruled out as a potential cause of the observed biases (in many cases such eggs may have ap- peared infertile), if this were the case we would expect the proportion of infertile eggs to be greater at the extremes of sex ratio (see Fig. 2). Expected levels of infertility, calcu- lated by assuming that the probability of a given level of infertility for a given sex ratio follows a Gaussian distri- bution, do not fit the levels of infertility observed for each clutch ( c 2 17 Z 5 : 13, P Z 0.997; Fig. 2), suggesting that the observed sex ratio biases were present at laying and thus provide a close approximation of the primary sex ratio. This does not, however, rule out the possibility of post- laying sex ratio adjustment through sex-differential em- bryo mortality, although the highly significant positive correlation between ‘primary’ (i.e. all fertile eggs) and hatching (i.e. all hatched eggs) sex ratios ( r 2 Z 0.77, F 1,16 Z 54.02, P ! 0.001; Fig. 3) suggests that, in this species, embryo mortality after laying is not a major factor influencing hatching (secondary) sex ratios. Egg Weight and Sex There was a significant, negative relation between an egg’s weight and the date on which it was laid ( r 2 Z 0.18; F 1,247 Z 53.30, P ! 0.001). When we controlled for this, an ANCOVA (with laying date as a covariate, pen and female identity as random factors and condition group [high, mid, low] and egg sex as fixed factors) revealed a significant effect of condition group ( F 2,158 Z 3.86, P Z 0.023) and the interaction between egg sex and condition group ( F 2,158 Z 7.71, P Z 0.001), but not egg Maternal body condition (a) (b) (c) 0.8 0.4 0 0.6 0.2 −1.6 1.6 0 −0.8 0.8 Sex ratio (proportion of males) Plasma corticosterone (ng/ml) 0.8 0.4 0 0.6 0.2 0 2.4 1.2 0.6 0.8 Plasma testosterone (ng/ml) 0.8 0.4 0 0.6 0.2 0 1.5 0.5 1 Figure 1. The relation between clutch sex ratio and (a) maternal body condition ( r 2 Z 0.38, F 1,11 Z 12.26, P Z 0.015), (b) maternal plasma levels of corticosterone ( r 2 Z 0.34, F 1,11 Z 8.56, P Z 0.041) and (c) plasma testosterone levels ( r 2 Z 0.26, F 1,11 Z 8.25, P Z 0.046). Point sizes are proportional to the number of successfully sexed eggs per clutch (range 4–17 eggs). Proportion of infertile eggs Sex ratio (proportion of males) 0.5 0.2 0 0.3 0.1 0 1 0.4 0.2 0.6 0.8 Figure 2. Predicted (dashed line) and observed (points) clutch levels of infertility. Predicted levels of infertility were calculated by assuming that the probability of a given level of infertility for a given sex ratio follows a Gaussian distribution. ANIMAL BEHAVIOUR, 70, 4 748 sex alone ( F 1,158 Z 1.39, P Z 0.241) on egg weight. Spe- cifically, high-condition females laid significantly heavier male eggs than female eggs (male egg weight: 109.25 G 0.83 g; female egg weight: 100.06 G 1.02 g; F 1,59 Z 25.07, Bonferroni adjusted P ! 0.001). DISCUSSION Sex Ratio and Hormones Our results suggest that, in peafowl, there is a positive causal relation between a female’s physical condition during egg production and her clutch sex ratio. Similar findings linking maternal condition and sex ratio have been reported for a wide variety of avian species (Bradbury & Blakey 1998; Kilner 1998; Nager et al. 1999; Whitting- ham & Dunn 2000; Kalmbach et al. 2001; Clout et al. 2002), although none have explicitly addressed the question of what proximate mechanism drives the re- corded sex ratio deviations. We investigated hormonal correlates of the sex ratio biases observed in this study and found significant interrelation between a female’s condi- tion and her plasma levels of the hormones testosterone and corticosterone, both of which acted as significant predictors of offspring sex ratio themselves. In general, females in poor condition, which had elevated circulating levels of corticosterone and lowered levels of testosterone, tended to produce proportionally more daughters, where- as the reverse held true for high-condition females. Since all three predictors (condition, corticosterone and testos- terone) appeared to explain much of the same variance in clutch sex ratio, it is possible that testosterone and/or corticosterone are acting as the proximate mechanism through which body condition influences the sex ratio of offspring produced in this species. Corticosterone is the major avian glucocorticoid re- leased in response to stress and is commonly found to be raised in the plasma of birds in poor body condition (e.g. Schoech et al. 1997; Kitaysky et al. 1999). It also has an inhibitory effect on reproductive processes, including testosterone production (e.g. Wingfield et al. 1994), which may explain why females with elevated corticosterone levels tended to have depressed levels of testosterone, and vice versa. However, until recently (Pike & Petrie 2003; Hayward & Wingfield 2004; but see Geiringer 1961) corticosterone has received little attention regarding its potentially important role in avian sex allocation, al- though high circulating levels of corticosterone have been implicated in inducing female-biased litters in rats, Rattus norvegicus (Geiringer 1961). Instead, interest has focused almost exclusively on the roles of the reproductive hor- mones testosterone and oestradiol (e.g. Bowden et al. 2000; Petrie et al. 2001). For example, Veiga et al. (2004) showed that artificial elevation of circulating testosterone levels in breeding females could induce male biases in the sex ratio. Our results are consistent with this finding, but we cannot tell whether the observed relation between hormones and sex ratio is one of cause or effect, or whether hormone concentrations are related to some other, as yet unidentified variable. Consequently, while this finding suggests a possible hormonal mechanism of avian sex ratio manipulation, further studies are clearly needed to address this in species where we know sex ratio adjustment is taking place. Timing of Sex Ratio Manipulation With the exception of studies demonstrating postlaying manipulation of offspring sex, for example through sex- differential nestling mortality (Kilner 1998; Williams 1999; Nager et al. 2000), only one other study (Komdeur et al. 2002) has explicitly considered the timing of sex ratio adjustment. Here we have provided evidence for the prelaying manipulation of sex by showing that infertile eggs were unlikely to explain the observed sex ratio biases. Although the sex ratios used in these analyses were not true primary sex ratios because we were unable to sex infertile eggs and those eggs in which a visible embryo failed to develop, the data suggest that the bias must have occurred by the time of laying, because there was no evidence that hatchability was lower in the clutches of females producing either an excess of sons or daughters. We can thus be confident that the sex ratios we used were a close approximation of the sex ratio present at laying. In birds females are the heterogametic sex (producing Z- and W-bearing ova), so mechanisms of sex ratio adjust- ment could potentially be under their control (Oddie 1998). It has been postulated that maternal hormone levels around the time of sex determination might cause pre-ovulation control to occur through segregation distor- tion of sex chromosomes during meiosis (e.g. Petrie et al. 2001; but see Krackow 1999 for a critique), differential provisioning of ova of different sexes to influence the order in which they are ovulated or through the selective resorption of male and female follicles prior to ovulation (Pike & Petrie 2003). Egg Weight and Sex A number of studies have reported sexual dimorphism in avian eggs (Howe 1977; Fiala 1981; Ankney 1982; Mead Hatching sex ratio Laying sex ratio 0.9 0.5 0.1 0.7 0.3 0.1 0.8 0.5 0.4 0.3 0.6 0.2 0.7 Figure 3. The relation between the sex ratio used in this paper (assumed to be a close approximation of the primary sex ratio) and the hatching sex ratio. PIKE & PETRIE: HORMONES AND SEX RATIO IN PEAFOWL 749 et al. 1987; Cordero et al. 2001; Petrie et al. 2001), but the adaptive significance of this is not always immediately obvious. In this study, we found that females in the best condition produced significantly heavier male eggs than female eggs (see also Petrie et al. 2001), but this effect was not present for females in poorer condition. Since there is no evidence that the size or quality of the egg per se can affect the sex of the chick (Rutkowska & Cichon ́ 2002), or that male embryos suffered higher mortality because they were starved of resources inside a smaller egg, it seems likely that females with enough resources to do so were diverting more to male eggs in order to improve their fitness. What is not known, however, is whether this increased egg weight is due to a larger quantity of albumin or yolk. We assume that in peafowl, as in chickens, Gallus gallus domesticus , eggs are ovulated and fertilized sequen- tially, and yolk size is determined before fertilization whereas the albumin and shell are deposited around the egg as it passes down the oviduct. Since the sex-determin- ing meiotic division occurs only hours before ovulation (Romanoff & Romanoff 1949), if heavier eggs are due to increased albumin deposition it suggests that peafowl have the ability to discriminate internally between male and female ova and adjust the quantity of albumin deposited accordingly. On the other hand, larger quanti- ties of yolk would suggest that sex is predetermined. Further studies are needed in species where this phenom- enon is known to occur to resolve this issue. Acknowledgments We are grateful to BBSRC and NERC for financial support, Quinton Spratt for allowing us to work at his farm and to Marie Hale and two anonymous referees for valuable comments on the manuscript. References Ankney, C. D. 1982. Sex ratio varies with egg sequence in lesser snow geese. Auk , 99 , 662–666. Arnold, K. E., Orr, K. J. & Griffiths, R. 2003. Primary sex ratios in birds: problems with molecular sex identification of undeveloped eggs. Molecular Ecology , 12 , 3451–3458. Bancroft, G. T. 1984. Patterns of variation in size of boat-tailed grackle Quiscallus major eggs. Ibis , 126 , 496–509. Bolton, M., Houston, D. & Monaghan, P. 1992. Nutritional constraints on egg formation in the lesser black-backed gull: an experimental study. Journal of Animal Ecology , 61 , 521–532. Bowden, R. M., Ewert, M. A. & Nelson, C. E. 2000. Environmental sex determination in a reptile varies seasonally and with yolk hormones. Proceedings of the Royal Society of London, Series B , 267 , 1745–1749. Bradbury, R. B. & Blakey, J. K. 1998. Diet, maternal condition, and offspring sex ratio in the zebra finch, Poephila guttata Proceedings of the Royal Society of London, Series B , 265 , 895–899. Bruford, M. W., Hanotte, O., Brookfield, J. F. Y. & Burke, T. 1998. Multilocus and single-locus DNA fingerprinting. In: Molecular Genetic Analysis of Populations: A Practical Approach , 2nd edn. (Ed. by A. R. Hoelzel), pp. 287–336. Oxford: IRL Press. Charnov, E. L. 1982. Parent–offspring conflict over reproductive effort. American Naturalist , 119 , 736–737. Clout, M. N., Elliott, G. P. & Robertson, B. C. 2002. Effects of supplementary feeding on the offspring sex ratio of kakapo: a dilemma for the conservation of a polygynous parrot. Biological Conservation , 107 , 13–18. Cordero, P. J., Vinuela, J., Aparicio, J. M. & Veiga, J. P. 2001. Seasonal variation in sex ratio and sexual egg dimorphism favouring daughters in first clutches of the spotless starling. Journal of Evolutionary Biology , 14 , 829–834. Fiala, K. L. 1981. Sex ratio constancy in the red-winged blackbird. Evolution , 35 , 898–910. Fisher, R. A. 1930. The Genetical Theory of Natural Selection . Oxford: Clarendon Press. Fridolfsson, A. K. & Ellegren, H. 1999. A simple and universal method for molecular sexing of non-ratite birds. Journal of Avian Biology , 30 , 116–121. Geiringer, E. 1961. Effect of ACTH on sex ratio in the albino rat. Proceedings of the Society for Experimental Biology and Medicine , 106 , 752–754. Gil, D., Graves, J., Hazon, N. & Wells, A. 1999. Male attractiveness and differential testosterone investment in zebra finch eggs. Science , 286 , 126–128. Hayward, L. S. & Wingfield, J. C. 2004. Laying Japanese quail transfer corticosterone to egg yolk. American Zoologist , 41 , 1468– 1469. Howe, H. F. 1977. Sex ratio adjustment in the common grackle. Science , 198 , 744–745. James, W. H. 1996. Evidence that mammalian sex ratios at birth are partially controlled by parental hormone levels at the time of conception. Journal of Theoretical Biology , 180 , 271–286. Kalmbach, E., Nager, R. G., Griffiths, R. & Furness, R. W. 2001. Increased reproductive effort results in male-biased offspring sex ratio: an experimental study in a species with reversed sexual size dimorphism. Proceedings of the Royal Society of London, Series B , 268 , 2175–2179. Kilner, R. 1998. Primary and secondary sex ratio manipulation by zebra finches. Animal Behaviour , 56 , 155–164. Kitaysky, A. S., Wingfield, J. C. & Piatt, J. F. 1999. Dynamics of food availability, body condition and physiological response in breeding black-legged kittiwakes. Functional Ecology , 13 , 577– 585. Komdeur, J., Magrath, M. J. L. & Krackow, S. 2002. Pre-ovulation control of hatchling sex ratio in the Seychelles warbler. Proceedings of the Royal Society of London, Series B , 269 , 1067–1072. Krackow, S. 1999. Avian sex ratio distortions: the myth of maternal control. In: 22nd International Ornithological Congress (Ed. by N. Adams & R. Slotow), pp. 425–433. Johannesburg: BirdLife South Africa. Leonard, M. L. & Weatherhead, P. J. 1996. Dominance rank and offspring sex ratios in domestic fowl. Animal Behaviour , 51 , 725– 731. Littell, R. C., Milliken, G. A., Stroup, W. W. & Wolfinger, R. S. 1996. SAS System for Mixed Models . Carey, North Carolina: SAS Institute. Maynard Smith, J. 1980. A new born theory of sexual investment. Behavioral Ecology and Sociobiology , 7 , 247–251. Mead, P. S., Morton, M. L. & Fish, B. E. 1987. Sexual dimorphism in egg size and implications regarding facultative manipulation of sex in mountain white-crowned sparrows. Condor , 89 , 798– 803. Merila ̈, J. & Svensson, E. 1995. Fat reserves and health state in migrant goldcrest Regulus regulus Functional Ecology , 9 , 842–848. Nager, R. G., Monaghan, P., Griffiths, R., Houston, D. C. & Dawson, R. 1999. Experimental demonstration that offspring sex ratio varies with maternal condition. Proceedings of the National Academy of Sciences, U.S.A. , 96 , 570–573. ANIMAL BEHAVIOUR, 70, 4 750 Nager, R. G., Monaghan, P., Houston, D. C. & Genovart, M. 2000. Parental condition, brood sex ratio and differential young survival: an experimental study in gulls ( Larus fuscus ). Behavioral Ecology and Sociobiology , 48 , 452–457. Nishiumi, I. 1998. Brood sex ratio is dependent on female mating status in polygynous great reed warblers. Behavioral Ecology and Sociobiology , 44 , 9–14. Oddie, K. 1998. Sex discrimination before birth. Trends in Ecology and Evolution , 13 , 130–131. Petrie, M., Schwabl, H., Brande-Lavridsen, N. & Burke, T. 2001. Sex differences in avian yolk hormone levels. Nature , 412 , 489. Pike, T. W. & Petrie, M. 2003. Potential mechanisms of avian sex manipulation. Biological Reviews , 78 , 553–574. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution , 43 , 223–225. Ricklefs, R. & Marks, H. L. 1983. Egg characteristics of lines of Japanese quail selected for four-week body mass. Poultry Science , 62 , 1330–1332. Romanoff, A. L. & Romanoff, A. J. 1949. The Avian Egg . New York: J. Wiley. Rutkowska, J. & Cichon ́ , M. 2002. Maternal investment during egg laying and offspring sex: an experimental study of zebra finches. Animal Behaviour , 64 , 817–822. Schoech, S. J., Mumme, R. L. & Wingfield, J. C. 1997. Corticosterone, reproductive status, and body mass in a coopera- tive breeder, the Florida scrub-jay ( Aphelocoma coerulescens ). Physiological Zoology , 70 , 68–73. Schwabl, H., Mock, D. W. & Gieg, J. A. 1997. A hormonal mechanism for parental favouritism. Nature , 386 , 231. Trivers, R. L. & Willard, D. E. 1973. Natural selection of parental ability to vary the sex ratio of offspring. Science , 179 , 90–91. Veiga, J. P., Vinn ̃uela, J., Cordero, P. J., Aparicio, J. M. & Polo, V. 2004. Experimentally increased testosterone affects social rank and primary sex ratio in the spotless starling. Hormones and Behavior , 46 , 47–53. Whittingham, L. A. & Dunn, P. O. 2000. Offspring sex ratios in tree swallows: females in better condition produce more sons. Molecular Ecology , 9 , 1123–1129. Whittingham, L. A. & Schwabl, H. 2002. Maternal testosterone in tree swallow eggs varies with female aggression. Animal Behaviour , 63 , 63–67. Williams, G. C. 1979. The question of adaptive sex ratio in outcrossed vertebrates. Proceedings of the National Academy of Sciences, U.S.A. , 205 , 567–580. Williams, T. D. 1999. Parental and first generation effects of exogeneous 17 b -estradiol on reproductive performance of female zebra finches ( Taeniopygia guttata ). Hormones and Behavior , 35 , 135–143. Wingfield, J. C., Hegner, R. E. & Lewis, D. M. 1991. Circulating levels of luteinizing hormone and steroid hormones in relation to social status in the cooperatively breeding white-browed sparrow weaver, Plocepasser mahali Journal of Zoology , 225 , 43–58. Wingfield, J. C., Suydam, R. & Hunt, K. 1994. The adrenocortical responses to stress in snow buntings ( Plectrophenax nivalis ) and Lapland longspurs ( Calcarius lapponicus ) at Barrow, Alaska. Comparative Biochemistry and Physiology, Part B , 108 , 299–306. PIKE & PETRIE: HORMONES AND SEX RATIO IN PEAFOWL 751