Behavioral Ecology Advance Access originally published online on January 12, 2005
Behavioral Ecology 2005 16(2):442-449; doi:10.1093/beheco/ari018
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sex-biased environmental sensitivity: natural and experimental evidence from a bird species with larger females
Institute of Biomedical and Life Sciences, Graham Kerr Building, University of Glasgow, Glasgow G12 8QQ, UK
Address correspondence to E. Kalmbach, who is now at Animal Ecology Group, University of Groningen, Kerklaan 30, 9755NN Haren, The Netherlands. E-mail: e.kalmbach{at}biol.rug.nl.
Received 27 May 2004; revised 17 September 2004; accepted 18 October 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The larger sex is often more vulnerable, in terms of development and survival, to poor conditions during early life. Differential vulnerability has implications for parental investment strategies such as sex ratio theory. When males are larger, it is not possible to separate the effects of larger size per se and other aspects of the male phenotype on vulnerability. Furthermore, offspring competition might favor the larger sex and thereby mask intrinsic, size-related effects. We studied sex-specific mortality in a bird species with reversed size dimorphism, the great skua Stercorarius skua, under natural and experimentally created poor conditions. Small eggs from extended laying sequences were used to create poor early conditions for the offspring, which were raised as singletons. Daughters had a lower survival in all treatment groups. Survival in natural broods was additionally affected by hatch date and position. Hatch weight was not different for sons and daughters but was lower in experimental than in natural nests. In natural nests, daughters fledged 10% heavier than sons, but in experimental nests, they did not reach a higher mass. The average survival difference between sons and daughters was not increased in experimental broods. However, hatch weight had a strong sex-specific effect. Very light females never survived, and survival probability of daughters increased with increasing hatch weight. By contrast, survival of sons over the same range of hatch weights was not related to weight. These findings support the hypothesis that larger (final) size per se is related to sex-specific offspring vulnerability during early life.
Key words: egg quality, environmental sensitivity, lesser black-backed gull, sex-biased mortality, sex ratio.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Conditions during early development often have differential effects on survival and performance of male and female offspring in birds and mammals (De Kogel, 1997
; Kruuk et al., 1999; Müller et al., 2003; Nager et al., 1999; Tschirren et al., 2003; Velando, 2002). Such sex-biased environmental sensitivity can affect the relative reproductive value of the sexes and is therefore expected to influence parental investment strategies (Clutton-Brock, 1991; Trivers and Willard, 1973). In the majority of studied species, males experience higher mortality rates at some stage between conception and sexual maturity (Cooch et al., 1997; Griffiths, 1992; Røskaft and Slagsvold, 1985). Male-biased mortality is correlated with the degree of sexual size dimorphism exhibited by adults, which suggests a causal relationship between relative final size and vulnerability (Clutton-Brock et al., 1985). The influence of size on sensitivity is also indicated by some studies of species where females are larger (Dijkstra et al., 1998; Torres and Drummond, 1997) and by the absence of sex-linked differences in the young of monomorphic species (Nisbet and Hatch, 1999; Sheldon et al., 1998). It is commonly assumed that the larger sex needs more energy during growth to reach its final size and is therefore more vulnerable to a shortage of resources (Anderson et al., 1993b; Krijgsveld et al., 1998). However, mechanisms independent of size have also been suggested, which could cause male-specific vulnerability. The one that has received most attention is the male phenotype hypothesis. This stipulates that male-specific characteristics, in particular the high levels of testosterone needed for male sexual differentiation, are detrimental to other aspects of development, such as immunocompetence (Fargallo et al., 2002; Müller et al., 2003; Olsen and Kovacs, 1996; Uller and Olsson, 2003).
A general problem in identifying causes of sex-biased mortality is presented by the overlapping predictions these two hypotheses make when males are the larger sex. Recently, the first experimental studies of sex-biased vulnerability in avian species with reversed size dimorphism, that is, larger females, have given new insights. Velando's (2002) findings, that handicapping flight ability of mothers during chick rearing had negative effects on growth of daughters but not sons in the blue-footed booby Sula nebouxi, support the size hypothesis. By contrast, the Fargallo et al. (2002) study on Eurasian kestrels Falco tinnunculus found reduced immune response in male nestlings but not females in food-stressed nests, supporting the male phenotype hypothesis. As in some other studies, which found a higher vulnerability of the smaller sex, sibling competition might have led to the observed disadvantage, as the larger sex might dominate food resources and thereby food-deprive and stress members of the smaller sex (Anderson et al., 1993a; Bortolotti, 1986; Hipkiss et al., 2002; Oddie, 2000).
In the present study, we investigate sex differences in environmental sensitivity of great skuas Stercorarius skua, in natural broods and in experimental broods without sibling competition. Great skuas are seabirds with reversed size dimorphism, where adult females are 10% heavier than adult males (Catry et al., 1999; Furness, 1987). Great skuas bias the primary sex ratio in favor of the smaller sex when mothers are in poor condition (Kalmbach et al., 2001). Unfavorable circumstances often highlight the disadvantage of the larger sex (Daunt et al., 2001; Nager et al., 2000; Røskaft and Slagsvold, 1985). We predict that great skua daughters, the larger sex, experience higher mortality than sons and that this difference is exacerbated under poor conditions.
We followed sex-specific offspring survival in unmanipulated broods, as well as investigating effects of experimentally created poor conditions on mortality of sons and daughters. Great skuas normally lay clutches of two eggs, resulting in broods of one and two chicks in unmanipulated nests. To separate effects of egg and parental quality on sex-linked offspring performance, we conducted an egg removal and cross-fostering experiment. Through egg removal, we induced females to lay six instead of the normal two eggs. Female body condition is greatly reduced after the production of six eggs (Kalmbach et al., 2004). We created two groups of experimental chicks, both of which hatched from poor eggs. However, one of the groups was raised by parents in good condition, while the other had the additional disadvantage of poor condition parents. To exclude effects of sibling competition on sex-specific vulnerability, experimental chicks were raised as singletons. If female offspring in the great skua are more vulnerable than males, we expect to see higher female mortality in the experimental treatments than in natural broods. Depending on the relative importance of parental condition or egg quality, sex-linked differences should be equal in both groups (mainly egg effects) or stronger in the second group (additional parental effects).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Study population and treatment groups
The present study was carried out between May and August 2000 on Foula, Shetland (60° 08' N, 2° 05' W), which supports a colony of approximately 2500 breeding pairs of great skuas. Altogether, we used 169 pairs of great skuas for our study, divided over three different treatment groups (Figure 1). From the "unmanipulated" group, background data on breeding parameters and sex-dependent offspring survival in natural broods were collected. From the "removal" pairs, eggs were removed continuously during laying, some of which were then incubated and raised by the "foster" group. Fifty-four pairs were studied in the unmanipulated group, 55 pairs belonged to the removal group, and 60 clutches were originally prepared as foster nests, of which 49 were eventually needed. Chicks from the foster and removal groups are jointly referred to as experimental chicks, in contrast to the unmanipulated chicks from natural broods (Figure 1). In the experimental group we carried out daily territory checks prior to the onset of and during laying, unmanipulated pairs were checked every 2 days. Eggs from the unmanipulated group were marked and measured on finding. During incubation, nest checks were kept to a minimum until the expected time of hatching. From hatching until fledging, unmanipulated as well as experimental chicks were checked every 2 to 3 days. During the peak of hatching, nest checks were carried out more frequently to be able to assign hatching order.
|
Experimental procedure
To minimize variability of adult quality among foster parents, we chose pairs that had laid their first eggs within a time period of 8 days (10–17 May) in the first half of the laying season. Similarly, for the group of removal pairs, we only chose pairs that had initiated laying within a 10-day period (10–19 May). The laying period of the unmanipulated nests fell between 9 May and 5 June (median 16 May) in the study year. Eggs of foster parents were vigorously shaken and painted with clear varnish after clutch completion, in order to prevent development of embryos. For cross-fostering, one of the original eggs was replaced by an experimental egg, so that foster parents were still left with two eggs for incubation but only one of which was developing.
In the removal group, daily checks were carried out and any newly laid egg was removed on finding. Fourth eggs laid by these females were put into nests of foster parents. We attempted to keep egg quality of fourth eggs as similar as possible by only cross-fostering those fourth eggs that had been laid between 20 and 31 of May (12.0 ± 2.17 days after the first eggs were laid). With 46 pairs, we continued egg removal until a sixth egg was laid, which we left in the nests of removal parents. Within 2 days after laying, we added a dummy egg to the nest, to discourage females from laying a further egg. However, 15 females produced a seventh egg despite this. We assume that females who laid a seventh egg were in comparatively better condition at the time when they produced the sixth egg than those females who did not produce a seventh egg. We therefore believe that rather than increasing the variability of parental condition within the removal group, the production of these seventh eggs might have reduced variability between females. In only four of these nests both eggs hatched, in three neither hatched, and of the remaining eight, one chick came from a seventh egg, while the others hatched from the sixth. Because of our aim to prevent competition within broods, we carried out the survival analyses for experimental chicks without the inclusion of the four nests that had hatched two chicks. Females in the removal group were in significantly poorer condition after egg laying than unmanipulated females (body mass unmanipulated: 1441 ± 10.7 g [55], removal: 1404 ± 11.6 g [40]; t = 2.36, p = .02; hematocrit unmanipulated: 0.46 ± 0.004 [55], removal: 0.42 ± 0.005 [40]; t = 6.13, p < .001; see Kalmbach et al., 2004).
Chick weights and survival
In all treatment groups, chicks were measured and marked by painting different combinations of toenails on the day of hatching or the following day. A 50-µl blood sample was taken when chicks were marked. The blood samples were later used for molecular sexing, following standard methods (Griffiths et al., 1998), with the alterations given in Kalmbach et al. (2001). The reported hatching sex ratios include those chicks that had died at hatch. At around 10 days of age, chicks were ringed with individually numbered British Trust for Ornithology rings. Survival of all chicks was followed by nest and territory checks every 2–3 days. Great skuas defend their nests and chicks against potential predators, including humans, through repeated aerial attacks, which increase in intensity with the proximity of the predator to the nest or chick. Therefore, parental behavior, that is, incidence and strength of attack on investigator, was included as an indicator of death or disappearance of nestlings when chicks had not been found during several checks. Chicks were classified as "fledglings" when they reached an age of 38 days, as chicks start to fly at around 38–44 days old, and survival from 38–44 days is generally high. Chicks' weights were measured again as close to fledging age as possible. Older chicks are very mobile and could sometimes not be found for several days. Fledging weights therefore include weights between 36 and 45 days of age, always using the weight on the day closest to 38–41. The average age at which fledging weight was measured was not significantly different for experimental and unmanipulated chicks (experimentals: 38.63 ± 0.28 days; unmanipulated: 39.59 ± 0.32 days, t73 = 1.901, p = .061).
Data analysis
Due to the egg removal procedure, the mean hatch dates of unmanipulated, foster, and removal chicks were necessarily different (mean hatch dates: unmanipulated 19 June ± 6 days; foster 25 June ± 3 days; removal 5 July ± 6 days). To correct for these differences, we included hatch date as a main factor in all models for comparing chick performance between treatment groups.
Because of nonindependence of unmanipulated chicks from the same nest, as well as experimental chicks with the same biological mother, we used statistical models that allow the inclusion of a random factor. As random factor, we entered the identity of the laying female in our analyses, which in the unmanipulated group corresponds to nest identity. Hatching and fledging weights were analyzed with linear mixed models (LME), which can allow for temporal and/or spatial pseudoreplications and nested designs (Crawley, 2002). Chick survival was analyzed by using generalized estimating equations (GEE) to fit population average models to the data (Hosmer and Lemeshow, 2000). This procedure is similar to generalized linear models but particularly designed for binary data, and it allows the inclusion of a random factor. Originally, we included all two-way interactions of the respective main factors of the full model and then followed backward elimination of factors that contributed least in explaining the observed variance, starting with the interactions. Three-way interactions were only included where visual data inspection suggested an effect. In the final models, only significant interactions are included, plus the corresponding and significant main factors.
Due to reduced hatching success in the experimental group, the final sample sizes for some parameters of foster and removal chicks were relatively small. We therefore firstly compared the two experimental groups with each other. When we found no significant differences between foster and removal chicks, we pooled them as experimental chicks in models where we compare them to the chicks from unmanipulated nests.
Sex ratios are given as proportion of males. The LME and GEE models were fitted using the "lme" and "gee" functions from the open source software R (Ihaka and Gentleman, 1996; Venables and Ripley, 2002). All other statistical analyses were performed using SPSS for Windows, version 11.0. Data are presented as means ± SE with n denoting sample size. All statistical tests are two tailed.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Unmanipulated broods
Overall hatching success of unmanipulated eggs was 75% (n = 108 eggs in 55 nests). Thirty nests hatched two chicks, 21 hatched one, and 4 failed to hatch any eggs. The sex ratio at hatching was 0.52 (n = 84), which was not different from a 1:1 binomial distribution (Fisher's exact p = .744; Figure 2). There was no sex bias in the hatching sequence (first chicks 18
:12
, p = .362; second chicks 17:13, p = .585; single chicks 8:13, p = .383; all Fisher's exact tests), and all combinations of offspring sexes in two-chick broods occurred as frequently as could be expected from a .5 probability for male and female offspring (MF 8; MM 10; FM 7; FF 5; 
2 = 0.023, df = 1, p = .880). Hatch weight was not different for males and females (males 64.6 ± 0.9 g, females 66.8 ± 1.4 g, t67 = 1.31, p = .195; Figure 2), but females were 10% heavier at the time of fledging (males 1062 ± 19.2 g, females 1173 ± 27.9 g, t49 = 3.364, p = .001; Figure 2).
|
Of the hatchlings, 65% survived to fledging age. Chick survival was mainly determined by three factors: offspring sex, hatch date, and position within the brood, as revealed by the results of a GEE model (Table 1). Sex had the strongest effect, as great skua daughters, the larger sex, showed a 20% lower survival than sons (Figure 2; Table 1). This resulted in a fledging sex ratio that appeared slightly biased in favor of males with 0.61 ± 0.07 (n = 53) but was not statistically different from unity (Fisher's exact p = .169). Hatch date also strongly affected survival, with a decreasing survival probability as the season progressed. Hatching second in a brood of two was the least successful position, while survival of single and first chicks was not different (Table 1). There was no interaction effect of position and sex, thus hatching second was equally bad for males and females (position x sex p = .399). Although hatch weight had no overall effect on survival, it interacted with chick position. When hatching in the disadvantaged second position, survival probability increased with increasing hatch weight (Table 1). This indicates that hatch weight might be an important factor for survival under unfavorable circumstances. Considering the existing effect of brood position on survival, it is not surprising that brood size had no additional effect (p = .6738). Neither did the sex of a chick's sibling affect its survival (p = .623).
|
Experimental eggs
Egg sizes declined sharply over the extended laying sequence from egg one to egg six (rmANOVA F5,46 = 93.689, p < .001). Both fourth and sixth eggs were significantly smaller than first eggs (fourth 10.1% smaller, sixth 13.4%, rmANOVA, 4 versus 1: F1,50 = 278.1, p < .001; 6 versus 1: F1,50 = 279.7, p < .001), and also second eggs (fourth 8.0% smaller, sixth 11.4%, rmANOVA, 4 versus 2: F1,50 = 276.3, p < .001; 6 versus 2: F1,50 = 305.4, p < .001). The size difference between the experimental eggs from fourth and sixth positions (3.7%, rmANOVA, 4 versus 6: F1,50 = 22.6, p < .001) was a lot smaller than the difference between either of the experimental eggs (fourth and sixth) and the first or second egg.
Within experimental group comparison (foster versus removal)
In the experimental group, 27 chicks (55%) hatched successfully in foster nests and 29 (48%) in removal nests. Hatch weight was not different between foster and removal chicks, and as in the unmanipulated group, there was no sex difference in hatch weight (Figure 2; Table 2a-I.). Sex ratio at hatching was 0.5 ± 0.09 in the foster group and 0.68 ± 0.09 in the removal group and was not statistically different between the two groups (Figure 2; Fisher's exact p = .2031). However, the sample sizes in the two groups are relatively small, and when tested directly against a binomial distribution, the hatching sex ratio in the removal group appeared male biased (Fisher's exact p = .0679). Fledging weights were not different between foster and removal chicks, and females were not heavier at fledging than males (Figure 2; Table 2b-I).
|
Unlike in unmanipulated broods, hatch date did not affect the survival of experimental chicks (Table 2c-I). As in unmanipulated broods, sex had an overall effect on survival, with females showing higher mortality (Table 2c-I). Contrary to predictions, the sex difference in survival was not larger in the removal than in the foster group (Figure 2; Table 2c-I; no effect of treatment [foster or removal] nor of the interaction of sex x treatment p = .115). Thus, when hatching from small eggs, being raised by parents in poorer condition did not appear to be an additional survival disadvantage and did not affect daughters more than sons. Equal to the unmanipulated chicks, hatch weight did not have an overall effect on survival. However, there was a significant interaction effect of hatch weight and sex on chick survival (Table 2c-I). Very light females had a much lower survival probability than heavier ones (
hatch weight coefficient, b = 0.723 ± 0.325, p = .026; Figure 3a), while male survival was equally poor across the whole range of hatch weights ( hatch weight coefficient, b = 0.045 ± 0.071, p = .528; Figure 3b).
|
= 0.648). Note that each data point represents all chicks of a given hatch weight, which varies between one and four per point.Experimental versus unmanipulated conditions
Hatching success was significantly lower in the experimental group than for unmanipulated eggs (52% versus 75%; Fisher's exact p < .001). Equally, hatch weight was lower for experimental chicks than in the unmanipulated group, even when correcting for the naturally occurring decline with hatch date (Figure 2; Table 2a-II). Fledging weights, by contrast, were on average not lower for experimental chicks and did not decline with hatch date (Figure 2; Table 2b-II.). However, there was a significant interaction effect of treatment and sex on fledging weight (Table 2b-II). Despite the lower hatch weights in the experimental groups, sons fledged equally heavy from unmanipulated and experimental nests. By contrast, daughters in unmanipulated nests fledged heavier than sons, but in the experimental group, they fledged at the same weight as sons, and thereby lighter than unmanipulated females (Figure 2). These results were the same when excluding second-hatched chicks, whose inclusion might have decreased the average for the unmanipulated group.
Survival to fledging was analyzed with a model that originally included the main factors hatch date, treatment (unmanipulated or experimental), sex, and hatch weight, as well as brood size and position. As seen for the unmanipulated chicks alone, later-hatched chicks survived less well, while there was no additional effect of treatment on survival (Table 2c-II). The three-way interaction of sex, treatment, and hatch weight was significant, as was the interaction of sex and treatment, while none of the other main factors or other two-way interactions were (Table 2c-II). The significant three-way interaction mean that the effect of hatch weight on sex-specific survival differed between the unmanipulated and experimental group. This supports the results we had found in the separate analyses of unmanipulated and experimental group: among the light-hatched experimental chicks, there was a clear sex-specific effect of hatch weight on survival, while such an effect was absent for unmanipulated chicks.
To account for the potential bias through different brood sizes in experimental and unmanipulated nests, we also analyzed survival with an alternative model. We excluded second-hatched chicks, whose survival was significantly lower than that of first and single chicks. The remaining unmanipulated chicks thereby represent a group that is more comparable to the singly raised experimental chicks. In this alternative model of survival, the hatch date effect disappeared (p = .071), while treatment effects became significant, with a lower survival probability for experimental chicks (treatment –15.38 ± 6.51, 2 = 5.57, p = .018). Offspring sex was also a significant predictor of survival (sex –1.34 ± 0.46, 2 = 8.37, p = .004), as was the interaction of treatment and hatch weight (treatment x hatch weight –0.23 ± 0.11, 2 = 4.56, p = .033). This, again, supports the result that hatch weight had a dissimilar effect under unmanipulated and experimental conditions.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We found that in the reversed sexually size-dimorphic great skua, offspring mortality during the nestling period was female biased. This pattern existed in natural broods, where a sibling was often present, as well as in experimental broods, where chicks were raised as singletons. Contrary to our predictions, the sex difference in mortality was not stronger under poor circumstances, that is, in the groups with poorer egg and parental condition. However, in the groups with reduced egg size, daughters showed increased sensitivity to low mass at hatching, while survival of sons was independent of hatch weight. Additionally, daughters that fledged under experimental conditions did so with reduced mass compared to unmanipulated conditions, while fledging weight of sons was not affected. These results support the hypothesis that size per se makes the sex, which grows to a larger final size, more vulnerable (Clutton-Brock et al., 1985
). This conclusion is strengthened by the comparison of our study with the closely related lesser black-backed gull Larus fuscus, in which males are the larger sex and the ones that exhibit higher nestling mortality (Griffiths, 1992; Nager et al., 2000).
Sibling competition or intrinsic vulnerability?
In contrast to some studies of reversed size-dimorphic species, which have found a competitive disadvantage and higher mortality of the smaller sex within broods (Bortolotti, 1986; Hipkiss et al., 2002), our results from natural broods gave no indication of such a pattern. Hatching second affected survival of both sexes negatively, while the sex of a sibling did not have an effect. However, our sample size of two-chick broods is too small to allow a thorough test of the interaction effect of brood position and sibling sex on survival, as there are four possible sex combinations in broods with two chicks. Also, we found no evidence that the presence of a sibling as such had a negative effect on survival. Even if competition would differentially affect survival of the smaller sex when there is a sibling present, our results point toward the existence of an intrinsic mechanism, which causes a high vulnerability of the finally larger, that is, female, chicks. Such a mechanism could be based on sex-specific growth rates, if female chicks had to achieve a higher initial growth rate in order to develop normally throughout the nestling phase (Gebhardt-Henrich and Richner, 1998). Female great skua chicks did not hatch heavier than male chicks, but at the end of the nestling phase, they fledged at a higher mass than males. This means that their development pattern must be different from that of males. Mortality occurred mainly in the first 10 days after hatching but not exclusively. Beside the immediate effect of early conditions on mortality, it is also possible that processes in the early days of life influence later survival (Lindström, 1999). The important role early conditions can play for differential development is underlined by our results from the experimental chicks.
Effects of egg and parental quality
Survival of experimental chicks was 15% to 20% lower than for chicks from natural broods, but this difference was only statistically significant when excluding the disadvantaged second-hatched chicks. This is a somewhat unexpected result, considering that conditions in the experimental groups were almost certainly poorer, also than for second-hatched chicks in natural broods. In addition to the small egg sizes, females in the removal group were in poorer body condition after laying than females who had laid two eggs and did not recover condition fully during incubation (Kalmbach et al., 2004). Male partners might compensate in their parenting effort for poor body condition and reduced effort of their females. However, it is unlikely that extra male investment would have fully compensated for the differences in female condition and thus the rearing capacity between the two groups, considering that the production of one extra egg already had a measurable effect on parenting ability in the lesser black-backed gull (Monaghan et al., 1998). Being raised in a natural two-chick brood had a skewed effect on the survival of the two brood members. Second-hatched chicks suffered reduced survival, while first chicks did equally well as singletons in natural broods. A general sex effect on survival was only found when correcting for this biased mortality (including brood position in survival model of unmanipulated chicks) or when excluding second-hatched chicks (alternative survival model experimental versus unmanipulated chicks). These results support findings from other studies that within-brood asymmetries, such as in size, age, or competitive ability, can mask intrinsic effects of sex-specific vulnerability (Bortolotti, 1986; Oddie, 2000).
Comparison of chick survival between the two experimental groups gives an indication about the importance of egg and parental quality for chick performance. Despite small size differences, in both treatments, chicks hatched from eggs that were considerably smaller than unmanipulated eggs and subsequently hatched at low weights. However, the two groups of chicks were raised by parents who differed in condition. Nonetheless, survival did not differ between the two experimental groups. This indicates that poor parental condition had no additional negative effect on offspring survival or, vice versa, good condition did not increase chick survival. At first sight it seems an unlikely result that parental condition did not affect offspring performance (Monaghan et al., 1998; Nager et al., 2000; Velando, 2002), and relatively small final sample sizes might be one reason for this. On the other hand, the existence of an egg quality or size threshold could explain that below a certain limit, egg effects on survival become so strong that they can override any parental effects (Nager et al., 1999). A weight threshold, rather than a continuous decline, has also been found to affect postfledging survival of birds (e.g., Van der Jeugd and Larsson, 1998). Whether this is indeed the case for egg size or hatching weight needs to be investigated with more detailed experiments.
The size disadvantage
Although poor early conditions did not increase survival differences between daughters and sons, fledging weights of daughters but not sons were reduced. Additionally, there was a distinct sex-linked effect of very low hatch weight on female survival. This pattern mirrors the situation in humans, where the increased mortality of boys among very low birth weight infants is known as the "male disadvantage" (Stevenson et al., 2000). Weight at hatching is tightly correlated to egg mass and reflects to a certain extent the amount of resources a chick possesses during the first days of life (Carey, 1996; Furness, 1983). Different metabolic requirements as well as differences in resource allocation of male and female chicks during the initial growth phase could explain sex-specific sensitivities to low hatch weight. Other studies of size-dimorphic species have found that the larger sex is more restricted in its growth pattern (Lindén, 1981). That is, the larger sex has not got an equal scope to vary growth or development in response to resource availability. An inability to vary growth might also be related to the process known as fetal programming (Lucas, 1991). It assumes that during very early development, the embryo is organized according to the currently available resources. This organization might fix, for example, the number of cells or metabolic pathways (Lucas et al., 1996; Robinson et al., 1999). Even when conditions improve after the initial poor phase, that is, after hatching when fed plentifully by good parents, the fixed body organization cannot be reversed. A sudden change in conditions might in fact be detrimental (Metcalfe and Monaghan, 2001). In the larger sex, this might have a stronger effect, as the aim of growing to a larger final size might increase the need for catch-up growth already during the first days of life. Despite hatching equally heavy as sons, daughters in unmanipulated nests reached a 10% higher weight at fledging. By contrast, females of the light-hatched group were not able to reach a higher fledging mass than males, irrespective of parental condition. It appears that particularly females in the great skua are not able to catch-up with the "normal" growth pattern after a bad start, even under good conditions. These results support the suggestion that vulnerability of the larger sex is not merely a consequence of higher energy requirements during posthatch growth (Torres and Drummond, 1999).
Implications for adaptive sex ratios
Sex-specific offspring performance has implications for many aspects of life history, including the theory of adaptive sex allocation. According to the hypothesis by Trivers and Willard (1973), parents should skew their offspring sex ratio in favor of the more beneficial sex. For the present study, that would mean sons are more beneficial in circumstances when the hatch weight is likely to be below a threshold weight. In a previous study, we found that great skuas skewed the primary sex ratio in favor of sons at the end of an experimentally extended laying sequence, when eggs are considerably smaller than in normal clutches (Kalmbach et al., 2001). We found the same sex ratio skew in last eggs replicated in the present study (when calculating the sex ratio in the same way as in the previous study, that is, taking the last viable egg of each female, the sex ratio was significantly male biased: sex ratio = 0.73, n = 26, binomial test p = .029). In combination, these results suggest a possible adaptive nature of the observed male-biased sex ratio in last eggs as a response to the female-biased vulnerability to very low hatch weight.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Our study in the reversed size-dimorphic great skua adds support to the theory that intrinsic vulnerability of the larger sex is linked to size per se and not specific to the male sex (Torres and Drummond, 1997
; Velando, 2002). Vulnerability was related to conditions before the peak demand of energy during nestling growth, suggesting differences in resource allocation prior to the appearance of size dimorphism. This does not exclude the possibility that in the same species, in broods larger than one, the smaller sex might exhibit higher mortality due to sibling competition (Bortolotti, 1986; Hipkiss et al., 2002; Oddie, 2000). Both mechanisms could act at the same time and need to be taken into account when making predictions about sex-specific offspring costs and benefits for parents. The fact that hatch weight played a differential role for sons and daughters suggests that circumstances during early development are crucial for these costs and that good parental care cannot necessarily offset early negative effects. |
ACKNOWLEDGEMENTS |
|---|
We thank the Holborn family for permission to work on Foula. Darren Evans and Kate Orr were a great asset during fieldwork. Egg removal was licensed by Scottish Natural Heritage, and blood was taken under license from the British Home Office. Lukas Keller helped with the statistical analysis, and he, Jan Komdeur, Alberto Velando, and an anonymous referee gave helpful comments on the manuscript. E.K. was funded by a University of Glasgow, Institute of Biomedical and Life Sciences scholarship.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Anderson DJ, Budde C, Apanius V, Gomez JEM, Bird DM, Weathers WW, 1993a. Prey size influences female competitive dominance in nestling American kestrels (Falco sparverius). Ecology 74:367–376.[CrossRef]
Anderson DJ, Reeve J, Gomez JEM, Weathers WW, Hutson S, Cunningham HV, Bird DM, 1993b. Sexual size dimorphism and food requirements of nestling birds. Can J Zool 71:2541–2545.
Bortolotti GR, 1986. Influence of sibling competition on nestling sex-ratios of sexually dimorphic birds. Am Nat 127:495–507.[CrossRef]
Carey C, 1996. Female reproductive energetics. In: Avian energetics and nutritional ecology (Carey C, ed). New York: Chapman and Hall; 324–374.
Catry P, Phillips RA, Furness RW, 1999. Evolution of reversed sexual size dimorphism in skuas and jaegers. Auk 116:158–168.
Clutton-Brock TH, 1991. The evolution of parental care. Princeton: Princeton University Press.
Clutton-Brock TH, Albon SD, Guinness FE, 1985. Parental investment and sex differences in juvenile mortality in birds and mammals. Nature 313:131–133.[CrossRef]
Cooch E, Lank D, Robertson R, Cooke F, 1997. Effects of parental age and environmental change on offspring sex ratio in a precocial bird. J Anim Ecol 66:189–202.[CrossRef]
Crawley MJ, 2002. Statistical computing. An introduction to data analysis using S-Plus. Chichester: John Wiley and Sons.
Daunt F, Monaghan P, Wanless S, Harris MP, Griffiths R, 2001. Sons and daughters: age-specific differences in parental rearing capacities. Funct Ecol 15:211–216.[CrossRef]
De Kogel CH, 1997. Long-term effects of brood size manipulation on morphological development and sex-specific mortality of offspring. J Anim Ecol 66:167–178.[CrossRef]
Dijkstra C, Daan S, Pen I, 1998. Fledging sex ratios in relation to brood size in size-dimorphic altricial birds. Behav Ecol 9:287–296.
Fargallo AJ, Laaksonen T, Pöyri V, Korpimaki E, 2002. Inter-sexual differences in the immune response of Eurasian kestrel nestlings under food shortage. Ecol Lett 5:95–101.
Furness RW, 1983. Variations in size and growth of great skua Catharacta skua chicks in relation to adult age, hatching date, egg volume, brood size and hatching sequence. J Zool Lond 199:101–116.
Furness RW, 1987. The skuas. Calton: T & AD Poyser.
Gebhardt-Henrich S, Richner H, 1998. Causes of growth variation and its consequences for fitness. In: Avian growth and development (Starck JM, Ricklefs RE, eds). Oxford: Oxford University Press; 324–339.
Griffiths R, 1992. Sex-biased mortality in the lesser black-backed gull Larus fuscus during the nestling stage. Ibis 134:237–244.
Griffiths R, Double MC, Orr K, Dawson RJG, 1998. A DNA test to sex most birds. Mol Ecol 7:1071–1075.[CrossRef][Medline]
Hipkiss T, Hörnfeldt B, Eklund U, Berlin S, 2002. Year-dependent sex-biased mortality in supplementary-fed Tengmalm's owl nestlings. J Anim Ecol 71:693–699.
Hosmer DW, Lemeshow S, 2000. Applied logistic regression, 2nd ed. New York: John Wiley and Sons.
Ihaka R, Gentleman R, 1996. R: a language for data analysis and graphics. J Comput Graph Stat 5:299–314.[CrossRef]
Kalmbach E, Griffiths R, Crane JE, Furness RW, 2004. Effects of experimentally increased egg production on female body condition and laying dates in the great skua. J Avian Biol 35:501–514.[CrossRef]
Kalmbach E, Nager RG, Griffiths R, Furness RW, 2001. Increased reproductive effort results in male-biased offspring sex ratio: an experimental study in a species with reversed sexual size dimorphism. Proc R Soc Lond B 268:2175–2179.[Medline]
Krijgsveld K, Dijkstra C, Visser HG, Daan S, 1998. Energy requirements for growth in relation to sexual size dimorphism in marsh harrier Circus aeruginosus nestlings. Physiol Zool 71:693–702.[Medline]
Kruuk LEB, Clutton-Brock TH, Rose KE, Guinness FE, 1999. Early determinants of lifetime reproductive success differ between the sexes in red deer. Proc R Soc Lond B 266:1655–1661.[Medline]
Lindén H, 1981. Growth rates and early energy requirements of captive juvenile capercaillie, Tetrao urogallus. Finn Game Res 39:53–67.
Lindström J, 1999. Early determinants of fitness in birds and mammals. Trends Ecol Evol 14:343–348.[CrossRef][Medline]
Lucas A, 1991. Programming by early nutrition in man. In: The childhood environment and adult disease (Bock GR, Whelan J, eds). Chichester: John Wiley & Sons; 38–55.
Lucas A, Baker BA, Desai M, Hales CN, 1996. Nutrition in pregnant or lactating rats programs lipid metabolism in the offspring. Br J Nutr 76:605–612.[CrossRef][ISI][Medline]
Metcalfe NB, Monaghan P, 2001. Compensation for a bad start: grow now, pay later? Trends Ecol Evol 16:254–260.[CrossRef][Medline]
Monaghan P, Nager RG, Houston DC, 1998. The price of eggs: increased investment in egg production reduces the offspring rearing capacity of parents. Proc R Soc Lond B 265:1731–1735.[CrossRef]
Müller W, Groothuis TGG, Dijkstra C, 2003. Inter-sexual differences in T-cell-mediated immunity of black-headed gull chicks (Larus ridibundus) depend on the hatching order. Behav Ecol Sociobiol 55:80–86.[CrossRef]
Nager RG, Monaghan P, Griffiths R, Houston DC, Dawson R, 1999. Experimental demonstration that offspring sex ratio varies with maternal condition. Proc Natl Acad Sci USA 96:570–573.
Nager RG, Monaghan P, Houston DC, Genovart M, 2000. Parental condition, brood sex ratio and differential young survival: an experimental study in gulls (Larus fuscus). Behav Ecol Sociobiol 48:452–457.[CrossRef]
Nisbet IDT, Hatch JJ, 1999. Consequences of a female-biased sex-ratio in a socially monogamous bird: female-female pairs in the roseate tern Sterna dougallii. Ibis 141:307–320.
Oddie KR, 2000. Size matters: competition between male and female great tit offspring. J Anim Ecol 69:903–912.[CrossRef]
Olsen NJ, Kovacs WJ, 1996. Gonadal steroids and immunity. Endocr Rev 17:369–384.[CrossRef][ISI][Medline]
Robinson JJ, Sinclair KD, McEvoy TG, 1999. Nutritional effects on foetal growth. Anim Sci 68:315–331.
Røskaft E, Slagsvold T, 1985. Differential mortality of male and female offspring in experimentally manipulated broods of the rook. J Anim Ecol 54:261–266.[CrossRef]
Sheldon BC, Merilä J, Lindgren G, Ellergren H, 1998. Gender and environmental sensitivity in nestling collared flycatchers. Ecology 79:1939–1948.[CrossRef]
Stevenson DK, Verter J, Fanaroff AA, Oh W, Ehrenkranz RA, Shankaran S, Donovan EF, Wright LL, Lemons JA, Tyson JE, Korones SB, Bauer CR, Stoll BJ, Papile LA, 2000. Sex differences in outcomes of very low birthweight infants: the newborn male disadvantage. Arch Dis Child Fetal Neonatal Ed 83:F182–F185.
Torres R, Drummond H, 1997. Female-biased mortality in nestlings of a bird with size dimorphism. J Anim Ecol 66:859–865.[CrossRef]
Torres R, Drummond H, 1999. Does large size make daughters of the blue-footed booby more expensive than sons? J Anim Ecol 68:1133–1141.[CrossRef]
Trivers RL, Willard DE, 1973. Natural selection of parental ability to vary the sex ratio of offspring. Science 179:90–91.
Tschirren B, Fitze PS, Richner H, 2003. Sexual dimorphism in susceptibility to parasites and cell-mediated immunity in great tit nestlings. J Anim Ecol 66:859–865.
Uller T, Olsson M, 2003. Prenatal exposure to testosterone increases ectoparasite susceptibility in the common lizard (Lacerta vivipara). Proc R Soc Lond B 270:1867–1870.[Medline]
Van der Jeugd HP, Larsson K, 1998. Pre-breeding survival of barnacle geese Branta leucopsis in relation to fledging characteristics. J Anim Ecol 67:953–966.[CrossRef]
Velando A, 2002. Experimental manipulation of maternal effort produces differential effects in sons and daughters: implications for adaptive sex ratios in the blue-footed booby. Behav Ecol 13:443–449.
Venables WN, Ripley BD, 2002. Modern applied statistics with S, 4th ed. New York: Springer.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. J. Verspoor, O. P. Love, E. Rowland, E. H. Chin, and T. D. Williams Sex-specific development of avian flight performance under experimentally altered rearing conditions Behav. Ecol., November 1, 2007; 18(6): 967 - 973. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Fargallo, V. Polo, L. de Neve, J. Martin, J. A. Davila, and M. Soler Hatching order and size-dependent mortality in relation to brood sex ratio composition in chinstrap penguins Behav. Ecol., September 1, 2006; 17(5): 772 - 778. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||