Evolutionary biology
rsbl.royalsocietypublishing.org
Male-biased sex allocation in ageing parents; a longitudinal study in a long-lived seabird
Research
Oscar Vedder1, Sandra Bouwhuis1, Marı´a M. Benito2 and Peter H. Becker1 1
Cite this article: Vedder O, Bouwhuis S, Benito MM, Becker PH. 2016 Male-biased sex allocation in ageing parents; a longitudinal study in a long-lived seabird. Biol. Lett. 12: 20160260. http://dx.doi.org/10.1098/rsbl.2016.0260
Received: 30 March 2016 Accepted: 8 July 2016
Subject Areas: evolution, behaviour
2
Institute of Avian Research ‘Vogelwarte Helgoland’, An der Vogelwarte 21, 26386, Wilhelmshaven, Germany tier3 solutions GmbH, Kolberger Strasse 61, 51381 Leverkusen, Germany OV, 0000-0003-4689-8568 Optimal sex allocation is frequency-dependent, but senescence may cause behaviour at old age to be suboptimal. We investigated whether sex allocation changes with parental age, using 16 years of data comprising more than 2500 molecularly sexed offspring of more than 600 known-age parents in common terns (Sterna hirundo), slightly sexually size-dimorphic seabirds. We decomposed parental age effects into within-individual change and sex allocation-associated selective (dis)appearance. Individual parents did not differ consistently in sex allocation, but offspring sex ratios at fledging changed from female- to male-biased as parents aged. Sex ratios at hatching were not related to parental age, suggesting sons to outperform daughters after hatching in broods of old parents. Our results call for the integration of sex allocation theory with theory on ageing and demography, as a change in sex allocation with age per se will cause the age structure of a population to affect the frequency-dependent benefits and the age-specific strength of selection on sex allocation.
Keywords: parental age, senescence, sex ratio, sex-biased mortality, sexual size dimorphism
1. Introduction Author for correspondence: Oscar Vedder e-mail: [email protected]
Electronic supplementary material is available at http://dx.doi.org/10.1098/rsbl.2016.0260 or via http://rsbl.royalsocietypublishing.org.
The optimal allocation of resources to the production of male and female offspring depends on the sex-specific offspring fitness returns per unit of parental investment, which are frequency-dependent and may vary between parents or environments [1]. Given the many factors that potentially affect the costs and benefits of producing each offspring sex, a priori quantitative prediction of the optimal allocation pattern is difficult [1,2]. Several studies have tested for cross-sectional associations between offspring sex ratio and parental age (e.g. [3 –6]), since age often underlies quality differences between individuals [7]. Because sex allocation can be associated with parental lifespan, for example when long-lived (high-quality) individuals invest more in one sex [6,8], or when sex-specific rearing costs are translated into parental survival [9], cross-sectional patterns of sex allocation with parental age may, however, not represent an effect of parental age per se. To establish an effect of age per se, it is required to study individuals longitudinally, and correct for potential overrepresentation of parents that produced a biased sex ratio among older age classes, as well as for confounding variables that have changed over time [10]. We are not aware of any study that meets all requirements to conclude that sex allocation changed with parental age per se. In this study, we test for an effect of parental age per se on offspring sex ratio at hatching and fledging, in a long-lived seabird with slight sexual size dimorphism, the common tern (Sterna hirundo). Our long-term individualbased study allows us to uniquely separate within-individual effects of parental
& 2016 The Author(s) Published by the Royal Society. All rights reserved.
(b) Data and analyses Offspring sex was analysed at the level of the individual chick using generalized linear mixed models with a binomial error distribution and a logit link function and a Markov chain Monte Carlo estimation algorithm with 100 000 iterations. Individual offspring sex was represented by 0 (daughter) and 1 (son), with 1 as the denominator. To account for non-independence of offspring sex within years, parents and broods, ‘year’, ‘parent identity’ and ‘brood identity’ were included as random effects. Sex ratios, as presented in the figures, represent the proportion of males. To maximize sample size, we analysed parental age effects on sex ratio for all offspring for which at least one parent was transponder-marked and of known age. To avoid pseudo-replication, we randomly selected one of the parents for offspring with two transponder-marked parents (n ¼ 959 hatchlings and 794 fledglings), such that individual offspring are represented once in the analyses. We included interactions between within- and betweenindividual parental age terms with parental sex to test for sex differences in parental age effects. We used male age for 1586 and 1405 offspring, and female age for 1413 and 1213 offspring, for analysis of hatchling and fledgling sex, respectively. We included ‘year’ as a continuous fixed effect to ascertain that parental age effects were not confounded by temporal changes in sex allocation over the study period. Between- and within-individual parental age effects were separated by using (i) the average age of each parent (with each brood weighted equally) and (ii) the annual age deviation from this average, as two fixed continuous variables. The average age term represents the between-individual age effect, and corrects for potential selective (dis)appearance of parents
within-individual pattern between-individual pattern
0.8 0.6 0.4 0.2 0 2
4
6
8
2
4
6
8
10
12
14
16
18
20
22
24
10 12 14 16 18 parental age (years)
20
22
24
(b) 1.0 0.8 0.6 0.4 0.2 0
Figure 1. Common tern hatchling (a) and fledgling (b) sex ratio in relation to parental age (effects do not differ between fathers and mothers, see table 1). Circles with error bars represent the cross-sectional average +s.e. per age category. Solid lines represent within-individual model predictions, while dashed lines represent between-individual model predictions. The dotted lines indicate a sex ratio of 0.50. producing a biased sex ratio, while the deviation term represents the within-individual age effect [11]. Significance of these terms was tested with both terms in the model. Because within-individual age effects may not be additive with real age, we also tested for the interaction between the within-individual deviation and average age terms. Full models were simplified by backward stepwise removal of non-significant terms. Significance ( p , 0.05, twotailed) was assessed using the Wald statistic. Analyses were performed in MLwiN 2.22 [14]. We present full models with standardized effect sizes in the electronic supplementary material.
3. Results At hatching, both within- and between-individual parental age effects on offspring sex were non-significant (figure 1a and table 1a). The sex ratio among sexed hatchlings did not significantly change over the study period and the absence of significant random effects of ‘year’, ‘parent identity’ and ‘brood identity’ suggests that the sex of individual hatchlings does not differ more between years, parents and broods than randomly expected (table 1a). Within individual parents, regardless of parental sex, the proportion of males among fledged offspring, however, increased with age (figure 1b and table 1b). Parents that were on average older in our dataset had not produced a significantly more male-biased offspring sex ratio (figure 1b and
Biol. Lett. 12: 20160260
Common terns were studied in Wilhelmshaven on the German North Sea coast (538360 N, 088060 E). Chicks hatched in the study colony have been ringed since 1980, and the oldest bird has so far reached 26 years of age. Common terns are socially and genetically monogamous birds that typically lay a clutch of three eggs per breeding attempt, and generally do not have more than one successful attempt per year [12]. Males feed females during courtship and both pair members share incubation and chick provisioning [12]. Common terns appear sexually monomorphic, but males are slightly larger and reach 3% greater fledging mass [13]. Since 1992, all fledglings, and 101 adult breeders, from the colony have been subcutaneously implanted with transponders (TROVAN ID 100; TROVAN, Ko¨ln, Germany). This allows identification of philopatric breeders by placing an antenna around each nest during incubation. Between 1992 and 2013, the number of breeding pairs fluctuated between 90 and 530, with about 60% of breeders being transponder-marked and of known age. Chicks are ringed within 3 days of hatching. Ringed chicks that die before fledging are collected and stored at 2208C. Between 1998 and 2013, 35.2% of dead chicks (n ¼ 7470) and all fledglings have been molecularly sexed using DNA obtained from muscle tissue (freshly dead chicks) or feathers (fledglings; for details, see [13]). For analyses of hatchling sex, we used only broods from which all hatchlings were sexed (either as dead chick or fledgling), including broods for which all hatchlings fledged. Although this causes a slight bias towards successful broods, dead molecularly sexed chicks composed 44.3% of the dataset on hatchling sex ratio (n ¼ 2999 hatchlings), providing sufficient scope to observe qualitative differences between the hatchling and fledgling sex ratios.
hatchling sex ratio (prop. males)
(a) Study species and data collection
2 1.0
rsbl.royalsocietypublishing.org
2. Material and methods
(a)
fledgling sex ratio (prop. males)
age on offspring sex ratio from between-individual and environmental effects in a single analytical framework [11].
Table 1. Results from models testing parental age effects on common tern sex ratio, among hatchlings (a) and fledglings (b). Statistics of removed terms are presented as when added to the final model. estimate (s.e.)
x2
Dd.f.
p-value
0.018 (0.018)
0.97
1
0.325
parent identity (random) brood identity (random)
0.019 (0.019) 0.010 (0.011)
0.95 0.90
1 1
0.330 0.343
0.011 (0.012) 0.003 (0.012)
0.95 0.06
1 1
0.330 0.807
20.003 (0.017) 0.029 (0.076)
0.03 0.14
1 1
0.862 0.708
parental sex between-individual age parental sex within-individual age
0.010 (0.025) 0.015 (0.031)
0.16 0.24
1 1
0.689 0.624
between- within-individual age
0.001 (0.006)
0.01
1
0.920
year (random) parent identity (random)
0.009 (0.010) 0.009 (0.013)
0.71 0.54
1 1
0.399 0.462
brood identity (random)
0.010 (0.012)
0.67
1
0.413
between-individual age within-individual age
0.005 (0.013) 0.043 (0.017)
0.16 6.92
1 1
0.689 0.009
removed terms year
0.006 (0.012)
0.28
1
0.597
parental sex parental sex between-individual age
0.058 (0.080) 0.024 (0.028)
0.53 0.72
1 1
0.467 0.396
parental sex within-individual age
0.004 (0.032)
0.01
1
0.920
between- within-individual age (n ¼ 2636 fledglings, 1904 broods, 669 parents)
0.000 (0.007)
,0.01
1
0.975
removed terms year between-individual age within-individual age parental sex
(n ¼ 2999 hatchlings, 1382 broods, 621 parents) (b) dependent variable: fledgling sex
table 1b). There was no significant change in fledgling sex ratio over the study period and no significant random effect of ‘year’, ‘parent identity’ or ‘brood identity’ on fledgling sex (table 1b).
4. Discussion We found that the proportion of male offspring among fledglings increased as parents aged, which could be explained neither by other factors changing over time, nor by parents that are more likely to produce males being overrepresented among older age classes. As such, we are, to our knowledge, the first to have unambiguously shown that sex allocation changed with parental age per se. We observed this pattern only among fledglings, not hatchlings, suggesting that it is the post-hatching care of old parents that favours sons over daughters. We cannot rule out that this, for unknown reasons, represents an adaptive phenomenon. However, despite an increase in the number of fledglings with parental age, the quality of fledglings declines with parental age in our study colony [15], and adult reproductive value (a combination of parental survival and
fledgling production) is known to decline after the age of 8 [16]. The bias towards sons with old age may thus more likely be suboptimal and an epiphenomenon of senescence. Perhaps the slightly larger size of sons provides them with a competitive advantage over their female siblings when the provisioning rate of old parents declines. Such a scenario could have evolved, because selection to ameliorate the fate of daughters at high parental age will be weak, because only few parents reach old age. In general, the strength of selection declines with age causing senescence to occur [17,18], but the possibility of senescence in sex allocation has received little attention. Directly showing senescence in sex allocation may, however, be challenging, as it would require manipulation of the sex ratio that old parents produce, independent of other traits, and testing whether a different sex ratio would provide these parents with greater fitness benefits. Regardless of the ultimate explanation, our observation of a within-individual change in sex allocation with age calls for a better integration of sex allocation theory with theory on age-specific performance and demography, because the age structure of a population will affect the age-specific strength of selection on optimal sex allocation and its frequency-dependent benefits.
Biol. Lett. 12: 20160260
(a) dependent variable: hatchling sex year (random)
rsbl.royalsocietypublishing.org
final model
3
Competing interests. We declare we have no competing interests. Funding. The study was supported by the Deutsche Forschungsge-
4
meinschaft (BE 916/9-1 and 2). O.V. was supported by an Alexander von Humboldt Foundation Research Fellowship. Acknowledgements. We thank the numerous fieldworkers that have contributed to the dataset, and Hedi Sauer-Gu¨rth and Go¨tz Wagenknecht for molecularly sexing all offspring. We thank Kate Lessells and three anonymous reviewers for helpful comments on a previous manuscript version.
rsbl.royalsocietypublishing.org
Ethics. The study was performed under license of the Bezirksregierung Weser-Ems and Wilhelmshaven (63-04/03). Data accessibility. Data are deposited in Dryad: http://dx.doi.org/10. 5061/dryad.tj247 [19]. Authors’ contributions. O.V., P.H.B., S.B. and M.M.B. designed the study; P.H.B., M.M.B. and S.B. collected the data. O.V. and S.B. analysed the data. O.V. wrote the manuscript, with contributions by S.B., P.H.B. and M.M.B. All authors gave final approval for publication and agreed to be accountable for all aspects of the content therein.
1.
2.
3.
4.
5.
6.
Frank SA. 1990 Sex allocation theory for birds and mammals. Annu. Rev. Ecol. Syst. 21, 13 –55. (doi:10.1146/annurev.es.21.110190.000305) Cockburn A, Legge S, Double MC. 2002 Sex ratios in birds and mammals: can the hypotheses be disentangled? In Sex ratios: concepts and research methods (ed. ICW Hardy), pp. 266–286. Cambridge, UK: Cambridge University Press. Blank JL, Nolan V. 1983 Offspring sex ratio in redwinged blackbirds is dependent on maternal age. Proc. Natl Acad. Sci. USA 80, 6141 –6145. (doi:10. 1073/pnas.80.19.6141) Clutton-Brock TH, Iason GR. 1986 Sex ratio variation in mammals. Q. Rev. Biol. 61, 339 –374. (doi:10. 1086/415033) 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. (doi:10.2307/6021) Weimerskirch H, Lallemand J, Martin J. 2005 Population sex ratio variation in a monogamous long-lived bird, the wandering albatross. J. Anim. Ecol. 74, 285–291. (doi:10.1111/j.1365-2656.2005. 00922.x)
7.
8.
9.
10.
11.
12.
13.
Forslund P, Pa¨rt T. 1995 Age and reproduction in birds—hypotheses and tests. Trends Ecol. Evol. 10, 374 –378. (doi:10.1016/S0169-5347(00)89141-7) Svensson E, Nilsson JA. 1996 Mate quality affects offspring sex ratio in blue tits. Proc. R. Soc. Lond. B 263, 357 –361. (doi:10.1098/rspb.1996.0055) Helle S, Lummaa V, Jokela J. 2002 Sons reduced maternal longevity in preindustrial humans. Science 296, 1085. (doi:10.1126/science.1070106) van de Pol M, Verhulst S. 2006 Age-dependent traits: a new statistical model to separate withinand between-individual effects. Am. Nat. 167, 766 –773. (doi:10.1086/503331) van de Pol M, Wright J. 2009 A simple method for distinguishing within- versus between-subject effects using mixed models. Anim. Behav. 77, 753 –758. (doi:10.1016/j.anbehav.2008.11.006) Becker PH, Ludwigs JD. 2004 Sterna hirundo common tern. In Birds of the Western Palearctic, update 6 (ed. D Parkin), pp. 91 –137. Oxford, UK: Oxford University Press. Becker PH, Wink M. 2003 Influences of sex, sex composition of brood and hatching order on mass growth in common terns Sterna hirundo. Behav.
14.
15.
16.
17. 18.
19.
Ecol. Sociobiol. 54, 136–146. (doi:10.1007/s00265003-0605-4) Rasbash JF, Steele F, Browne W, Prosser B. 2004 A user‘s guide to MLwiN, version 2.0. London, UK: Centre for Multilevel Modelling. Bouwhuis S, Vedder O, Becker PH. 2015 Sex-specific pathways of parental age effects on offspring lifetime reproductive success in a long-lived seabird. Evolution 69, 1760– 1771. (doi:10.1111/evo.12692) Zhang H, Rebke M, Becker PH, Bouwhuis S. 2015 Fitness prospects: effects of age, sex and recruitment age on reproductive value in a longlived seabird. J. Anim. Ecol. 84, 199– 207. (doi:10. 1111/1365-2656.12259) Medawar PB. 1952 An unsolved problem in biology. London, UK: HK Lewis. Hamilton WD. 1966 The moulding of senescence by natural selection. J. Theor. Biol. 12, 12 –45. (doi:10. 1016/0022-5193(66)90184-6) Vedder O, Bouwhuis S, Benito MM, Becker PH. 2016 Data from: Male-biased sex allocation in ageing parents; a longitudinal study in a long-lived seabird. Dryad Digital Repository. See http://dx.doi. org/10.5061/dryad.tj247.
Biol. Lett. 12: 20160260
References
rsbl.royalsocietypublishing.org
Male-biased sex allocation in ageing parents; a longitudinal study in a long-lived seabird
Research
Oscar Vedder1, Sandra Bouwhuis1, Marı´a M. Benito2 and Peter H. Becker1 1
Cite this article: Vedder O, Bouwhuis S, Benito MM, Becker PH. 2016 Male-biased sex allocation in ageing parents; a longitudinal study in a long-lived seabird. Biol. Lett. 12: 20160260. http://dx.doi.org/10.1098/rsbl.2016.0260
Received: 30 March 2016 Accepted: 8 July 2016
Subject Areas: evolution, behaviour
2
Institute of Avian Research ‘Vogelwarte Helgoland’, An der Vogelwarte 21, 26386, Wilhelmshaven, Germany tier3 solutions GmbH, Kolberger Strasse 61, 51381 Leverkusen, Germany OV, 0000-0003-4689-8568 Optimal sex allocation is frequency-dependent, but senescence may cause behaviour at old age to be suboptimal. We investigated whether sex allocation changes with parental age, using 16 years of data comprising more than 2500 molecularly sexed offspring of more than 600 known-age parents in common terns (Sterna hirundo), slightly sexually size-dimorphic seabirds. We decomposed parental age effects into within-individual change and sex allocation-associated selective (dis)appearance. Individual parents did not differ consistently in sex allocation, but offspring sex ratios at fledging changed from female- to male-biased as parents aged. Sex ratios at hatching were not related to parental age, suggesting sons to outperform daughters after hatching in broods of old parents. Our results call for the integration of sex allocation theory with theory on ageing and demography, as a change in sex allocation with age per se will cause the age structure of a population to affect the frequency-dependent benefits and the age-specific strength of selection on sex allocation.
Keywords: parental age, senescence, sex ratio, sex-biased mortality, sexual size dimorphism
1. Introduction Author for correspondence: Oscar Vedder e-mail: [email protected]
Electronic supplementary material is available at http://dx.doi.org/10.1098/rsbl.2016.0260 or via http://rsbl.royalsocietypublishing.org.
The optimal allocation of resources to the production of male and female offspring depends on the sex-specific offspring fitness returns per unit of parental investment, which are frequency-dependent and may vary between parents or environments [1]. Given the many factors that potentially affect the costs and benefits of producing each offspring sex, a priori quantitative prediction of the optimal allocation pattern is difficult [1,2]. Several studies have tested for cross-sectional associations between offspring sex ratio and parental age (e.g. [3 –6]), since age often underlies quality differences between individuals [7]. Because sex allocation can be associated with parental lifespan, for example when long-lived (high-quality) individuals invest more in one sex [6,8], or when sex-specific rearing costs are translated into parental survival [9], cross-sectional patterns of sex allocation with parental age may, however, not represent an effect of parental age per se. To establish an effect of age per se, it is required to study individuals longitudinally, and correct for potential overrepresentation of parents that produced a biased sex ratio among older age classes, as well as for confounding variables that have changed over time [10]. We are not aware of any study that meets all requirements to conclude that sex allocation changed with parental age per se. In this study, we test for an effect of parental age per se on offspring sex ratio at hatching and fledging, in a long-lived seabird with slight sexual size dimorphism, the common tern (Sterna hirundo). Our long-term individualbased study allows us to uniquely separate within-individual effects of parental
& 2016 The Author(s) Published by the Royal Society. All rights reserved.
(b) Data and analyses Offspring sex was analysed at the level of the individual chick using generalized linear mixed models with a binomial error distribution and a logit link function and a Markov chain Monte Carlo estimation algorithm with 100 000 iterations. Individual offspring sex was represented by 0 (daughter) and 1 (son), with 1 as the denominator. To account for non-independence of offspring sex within years, parents and broods, ‘year’, ‘parent identity’ and ‘brood identity’ were included as random effects. Sex ratios, as presented in the figures, represent the proportion of males. To maximize sample size, we analysed parental age effects on sex ratio for all offspring for which at least one parent was transponder-marked and of known age. To avoid pseudo-replication, we randomly selected one of the parents for offspring with two transponder-marked parents (n ¼ 959 hatchlings and 794 fledglings), such that individual offspring are represented once in the analyses. We included interactions between within- and betweenindividual parental age terms with parental sex to test for sex differences in parental age effects. We used male age for 1586 and 1405 offspring, and female age for 1413 and 1213 offspring, for analysis of hatchling and fledgling sex, respectively. We included ‘year’ as a continuous fixed effect to ascertain that parental age effects were not confounded by temporal changes in sex allocation over the study period. Between- and within-individual parental age effects were separated by using (i) the average age of each parent (with each brood weighted equally) and (ii) the annual age deviation from this average, as two fixed continuous variables. The average age term represents the between-individual age effect, and corrects for potential selective (dis)appearance of parents
within-individual pattern between-individual pattern
0.8 0.6 0.4 0.2 0 2
4
6
8
2
4
6
8
10
12
14
16
18
20
22
24
10 12 14 16 18 parental age (years)
20
22
24
(b) 1.0 0.8 0.6 0.4 0.2 0
Figure 1. Common tern hatchling (a) and fledgling (b) sex ratio in relation to parental age (effects do not differ between fathers and mothers, see table 1). Circles with error bars represent the cross-sectional average +s.e. per age category. Solid lines represent within-individual model predictions, while dashed lines represent between-individual model predictions. The dotted lines indicate a sex ratio of 0.50. producing a biased sex ratio, while the deviation term represents the within-individual age effect [11]. Significance of these terms was tested with both terms in the model. Because within-individual age effects may not be additive with real age, we also tested for the interaction between the within-individual deviation and average age terms. Full models were simplified by backward stepwise removal of non-significant terms. Significance ( p , 0.05, twotailed) was assessed using the Wald statistic. Analyses were performed in MLwiN 2.22 [14]. We present full models with standardized effect sizes in the electronic supplementary material.
3. Results At hatching, both within- and between-individual parental age effects on offspring sex were non-significant (figure 1a and table 1a). The sex ratio among sexed hatchlings did not significantly change over the study period and the absence of significant random effects of ‘year’, ‘parent identity’ and ‘brood identity’ suggests that the sex of individual hatchlings does not differ more between years, parents and broods than randomly expected (table 1a). Within individual parents, regardless of parental sex, the proportion of males among fledged offspring, however, increased with age (figure 1b and table 1b). Parents that were on average older in our dataset had not produced a significantly more male-biased offspring sex ratio (figure 1b and
Biol. Lett. 12: 20160260
Common terns were studied in Wilhelmshaven on the German North Sea coast (538360 N, 088060 E). Chicks hatched in the study colony have been ringed since 1980, and the oldest bird has so far reached 26 years of age. Common terns are socially and genetically monogamous birds that typically lay a clutch of three eggs per breeding attempt, and generally do not have more than one successful attempt per year [12]. Males feed females during courtship and both pair members share incubation and chick provisioning [12]. Common terns appear sexually monomorphic, but males are slightly larger and reach 3% greater fledging mass [13]. Since 1992, all fledglings, and 101 adult breeders, from the colony have been subcutaneously implanted with transponders (TROVAN ID 100; TROVAN, Ko¨ln, Germany). This allows identification of philopatric breeders by placing an antenna around each nest during incubation. Between 1992 and 2013, the number of breeding pairs fluctuated between 90 and 530, with about 60% of breeders being transponder-marked and of known age. Chicks are ringed within 3 days of hatching. Ringed chicks that die before fledging are collected and stored at 2208C. Between 1998 and 2013, 35.2% of dead chicks (n ¼ 7470) and all fledglings have been molecularly sexed using DNA obtained from muscle tissue (freshly dead chicks) or feathers (fledglings; for details, see [13]). For analyses of hatchling sex, we used only broods from which all hatchlings were sexed (either as dead chick or fledgling), including broods for which all hatchlings fledged. Although this causes a slight bias towards successful broods, dead molecularly sexed chicks composed 44.3% of the dataset on hatchling sex ratio (n ¼ 2999 hatchlings), providing sufficient scope to observe qualitative differences between the hatchling and fledgling sex ratios.
hatchling sex ratio (prop. males)
(a) Study species and data collection
2 1.0
rsbl.royalsocietypublishing.org
2. Material and methods
(a)
fledgling sex ratio (prop. males)
age on offspring sex ratio from between-individual and environmental effects in a single analytical framework [11].
Table 1. Results from models testing parental age effects on common tern sex ratio, among hatchlings (a) and fledglings (b). Statistics of removed terms are presented as when added to the final model. estimate (s.e.)
x2
Dd.f.
p-value
0.018 (0.018)
0.97
1
0.325
parent identity (random) brood identity (random)
0.019 (0.019) 0.010 (0.011)
0.95 0.90
1 1
0.330 0.343
0.011 (0.012) 0.003 (0.012)
0.95 0.06
1 1
0.330 0.807
20.003 (0.017) 0.029 (0.076)
0.03 0.14
1 1
0.862 0.708
parental sex between-individual age parental sex within-individual age
0.010 (0.025) 0.015 (0.031)
0.16 0.24
1 1
0.689 0.624
between- within-individual age
0.001 (0.006)
0.01
1
0.920
year (random) parent identity (random)
0.009 (0.010) 0.009 (0.013)
0.71 0.54
1 1
0.399 0.462
brood identity (random)
0.010 (0.012)
0.67
1
0.413
between-individual age within-individual age
0.005 (0.013) 0.043 (0.017)
0.16 6.92
1 1
0.689 0.009
removed terms year
0.006 (0.012)
0.28
1
0.597
parental sex parental sex between-individual age
0.058 (0.080) 0.024 (0.028)
0.53 0.72
1 1
0.467 0.396
parental sex within-individual age
0.004 (0.032)
0.01
1
0.920
between- within-individual age (n ¼ 2636 fledglings, 1904 broods, 669 parents)
0.000 (0.007)
,0.01
1
0.975
removed terms year between-individual age within-individual age parental sex
(n ¼ 2999 hatchlings, 1382 broods, 621 parents) (b) dependent variable: fledgling sex
table 1b). There was no significant change in fledgling sex ratio over the study period and no significant random effect of ‘year’, ‘parent identity’ or ‘brood identity’ on fledgling sex (table 1b).
4. Discussion We found that the proportion of male offspring among fledglings increased as parents aged, which could be explained neither by other factors changing over time, nor by parents that are more likely to produce males being overrepresented among older age classes. As such, we are, to our knowledge, the first to have unambiguously shown that sex allocation changed with parental age per se. We observed this pattern only among fledglings, not hatchlings, suggesting that it is the post-hatching care of old parents that favours sons over daughters. We cannot rule out that this, for unknown reasons, represents an adaptive phenomenon. However, despite an increase in the number of fledglings with parental age, the quality of fledglings declines with parental age in our study colony [15], and adult reproductive value (a combination of parental survival and
fledgling production) is known to decline after the age of 8 [16]. The bias towards sons with old age may thus more likely be suboptimal and an epiphenomenon of senescence. Perhaps the slightly larger size of sons provides them with a competitive advantage over their female siblings when the provisioning rate of old parents declines. Such a scenario could have evolved, because selection to ameliorate the fate of daughters at high parental age will be weak, because only few parents reach old age. In general, the strength of selection declines with age causing senescence to occur [17,18], but the possibility of senescence in sex allocation has received little attention. Directly showing senescence in sex allocation may, however, be challenging, as it would require manipulation of the sex ratio that old parents produce, independent of other traits, and testing whether a different sex ratio would provide these parents with greater fitness benefits. Regardless of the ultimate explanation, our observation of a within-individual change in sex allocation with age calls for a better integration of sex allocation theory with theory on age-specific performance and demography, because the age structure of a population will affect the age-specific strength of selection on optimal sex allocation and its frequency-dependent benefits.
Biol. Lett. 12: 20160260
(a) dependent variable: hatchling sex year (random)
rsbl.royalsocietypublishing.org
final model
3
Competing interests. We declare we have no competing interests. Funding. The study was supported by the Deutsche Forschungsge-
4
meinschaft (BE 916/9-1 and 2). O.V. was supported by an Alexander von Humboldt Foundation Research Fellowship. Acknowledgements. We thank the numerous fieldworkers that have contributed to the dataset, and Hedi Sauer-Gu¨rth and Go¨tz Wagenknecht for molecularly sexing all offspring. We thank Kate Lessells and three anonymous reviewers for helpful comments on a previous manuscript version.
rsbl.royalsocietypublishing.org
Ethics. The study was performed under license of the Bezirksregierung Weser-Ems and Wilhelmshaven (63-04/03). Data accessibility. Data are deposited in Dryad: http://dx.doi.org/10. 5061/dryad.tj247 [19]. Authors’ contributions. O.V., P.H.B., S.B. and M.M.B. designed the study; P.H.B., M.M.B. and S.B. collected the data. O.V. and S.B. analysed the data. O.V. wrote the manuscript, with contributions by S.B., P.H.B. and M.M.B. All authors gave final approval for publication and agreed to be accountable for all aspects of the content therein.
1.
2.
3.
4.
5.
6.
Frank SA. 1990 Sex allocation theory for birds and mammals. Annu. Rev. Ecol. Syst. 21, 13 –55. (doi:10.1146/annurev.es.21.110190.000305) Cockburn A, Legge S, Double MC. 2002 Sex ratios in birds and mammals: can the hypotheses be disentangled? In Sex ratios: concepts and research methods (ed. ICW Hardy), pp. 266–286. Cambridge, UK: Cambridge University Press. Blank JL, Nolan V. 1983 Offspring sex ratio in redwinged blackbirds is dependent on maternal age. Proc. Natl Acad. Sci. USA 80, 6141 –6145. (doi:10. 1073/pnas.80.19.6141) Clutton-Brock TH, Iason GR. 1986 Sex ratio variation in mammals. Q. Rev. Biol. 61, 339 –374. (doi:10. 1086/415033) 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. (doi:10.2307/6021) Weimerskirch H, Lallemand J, Martin J. 2005 Population sex ratio variation in a monogamous long-lived bird, the wandering albatross. J. Anim. Ecol. 74, 285–291. (doi:10.1111/j.1365-2656.2005. 00922.x)
7.
8.
9.
10.
11.
12.
13.
Forslund P, Pa¨rt T. 1995 Age and reproduction in birds—hypotheses and tests. Trends Ecol. Evol. 10, 374 –378. (doi:10.1016/S0169-5347(00)89141-7) Svensson E, Nilsson JA. 1996 Mate quality affects offspring sex ratio in blue tits. Proc. R. Soc. Lond. B 263, 357 –361. (doi:10.1098/rspb.1996.0055) Helle S, Lummaa V, Jokela J. 2002 Sons reduced maternal longevity in preindustrial humans. Science 296, 1085. (doi:10.1126/science.1070106) van de Pol M, Verhulst S. 2006 Age-dependent traits: a new statistical model to separate withinand between-individual effects. Am. Nat. 167, 766 –773. (doi:10.1086/503331) van de Pol M, Wright J. 2009 A simple method for distinguishing within- versus between-subject effects using mixed models. Anim. Behav. 77, 753 –758. (doi:10.1016/j.anbehav.2008.11.006) Becker PH, Ludwigs JD. 2004 Sterna hirundo common tern. In Birds of the Western Palearctic, update 6 (ed. D Parkin), pp. 91 –137. Oxford, UK: Oxford University Press. Becker PH, Wink M. 2003 Influences of sex, sex composition of brood and hatching order on mass growth in common terns Sterna hirundo. Behav.
14.
15.
16.
17. 18.
19.
Ecol. Sociobiol. 54, 136–146. (doi:10.1007/s00265003-0605-4) Rasbash JF, Steele F, Browne W, Prosser B. 2004 A user‘s guide to MLwiN, version 2.0. London, UK: Centre for Multilevel Modelling. Bouwhuis S, Vedder O, Becker PH. 2015 Sex-specific pathways of parental age effects on offspring lifetime reproductive success in a long-lived seabird. Evolution 69, 1760– 1771. (doi:10.1111/evo.12692) Zhang H, Rebke M, Becker PH, Bouwhuis S. 2015 Fitness prospects: effects of age, sex and recruitment age on reproductive value in a longlived seabird. J. Anim. Ecol. 84, 199– 207. (doi:10. 1111/1365-2656.12259) Medawar PB. 1952 An unsolved problem in biology. London, UK: HK Lewis. Hamilton WD. 1966 The moulding of senescence by natural selection. J. Theor. Biol. 12, 12 –45. (doi:10. 1016/0022-5193(66)90184-6) Vedder O, Bouwhuis S, Benito MM, Becker PH. 2016 Data from: Male-biased sex allocation in ageing parents; a longitudinal study in a long-lived seabird. Dryad Digital Repository. See http://dx.doi. org/10.5061/dryad.tj247.
Biol. Lett. 12: 20160260
References
