SEARCH

SEARCH BY CITATION

Keywords:

  • fecundity;
  • life history;
  • mammals;
  • seals

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

1. The study tested the hypothesis that the occurrence of pregnancy in mammals that are capital breeders will be most closely related to state variables that are indices of food stores. Income breeders would not be expected to have the same relationship.

2. The study examined the relationships between mass, time of year, age and body length with the occurrence of pregnancy in three species of pinniped. This included two capital breeders (Crabeater Seal, Lobodon carcinophagus Hombron & Jacquinot, and Grey Seal, Halichoerus grypus Fabricius) and one income breeder (Antarctic Fur Seal, Arctocephalus gazella Peters). Multiple logistic regression analysis was used to examine the relationships and the interactions between the different state variables.

3. In both the capital breeders the state variables used in the study explained approximately twice the amount of variability (55% compared with 25%) in the occurrence of pregnancy than in the income breeder. Mass was the dominant state variable among both the capital breeders whereas in the income breeder mass was less important both relative to other state variables and in absolute terms. The results support the conclusion that the occurrence of pregnancy in capital breeders is more sensitive to body reserves than it is in income breeders.

4. The results also support the conclusion that life histories are the end result of a variety of functional responses to different state variables that have differing degrees of influence. In particular a size-structured approach to studies of population dynamics in pinnipeds may be a more informative way of predicting population responses to environmental variability than the more traditional age-structured approach.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The relationship between resource availability and allocation to reproduction is critical to understanding the evolution of life-history patterns in organisms (McNamara & Houston 1996). It is a paradox of life-history studies that they are, by definition, time-based approaches to examining variation in the fitness between individuals when time (usually expressed as the age of an individual) probably has less biological importance than other factors (Caswell 1989). For example, among iteroparous species, the age at which an animal becomes sexually mature is likely to be a function of its ability to sustain reproduction while minimizing the risk to its own survival. The growth rate then becomes critical to the age at which individuals begin to breed. Thus, adjustments in the fertility of individuals, including apparent delayed puberty or intermittent infertility, probably depend to a large extent upon an optimal allocation of resources that maximize lifetime fitness.

Pinnipeds have characteristics that make them suitable subjects to study as examples of life-history evolution in a large mammal. In particular, the problem of allocation is simplified because almost all species are annual breeders and they produce a single offspring at each reproductive attempt, despite an order of magnitude variation in adult body size among species.

There is also evidence from pinnipeds that life histories are responsive to changes in environmental conditions. Age at sexual maturity in pinnipeds can be expressed as a decreasing function of growth rate (Laws 1956). Expressed at the level of populations, this has been interpreted to mean that individuals within a population whose size is well below the environmental carrying capacity would experience higher growth rates and would therefore become sexually mature at an earlier age (Bengston & Laws 1985). This is evidence of close interaction between mass and pinniped life histories. Mass, in this case, is likely to be a convenient index of a more specific measure of body condition which may have a threshold level, set by natural selection, that determines whether an individual will breed.

The degree to which body mass contributes to fertility will depend upon the importance of energy reserves for fitness. In some species in which parents have a high rate of energy turnover in order to feed offspring, body mass may be relatively less important than foraging ability or territory quality. In this study, the empirical functional relationship between fertility and body mass among the females of three species of pinnipeds is estimated. In two of these species, Crabeater Seal (Lobodon carcinophagus Hombron & Jacquinot) and Grey Seal (Halichoerus grypus Fabricius), mothers rely entirely upon stored energy reserves to feed their offspring and are extreme examples of capital breeders (Jonsson 1997; Festa-Bianchet, Gaillard & Jorgenson 1998). In the other species, the Antarctic Fur Seal (Arctocephalus gazella Peters), mothers rely mainly on energy that they are able to obtain by foraging during lactation (Boness & Bowen 1996; Boyd 1998) and, in comparison, these are income breeders. The hypothesis being tested is that, in the species that rely upon prebreeding investment in energy reserves, there should be a stronger effect of body mass upon fertility than in those that rely more on current foraging success. It is also predicted that body mass would tend to have a stronger influence than other state variables that were a poorer reflection of the level of energy reserves carried by individuals.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Data sets for Crabeater and Grey Seals were collected at a time when it was deemed acceptable to cull animals for scientific sampling. The data for Antarctic Fur Seals came from the late 1980s when other animal capture and restraint methods were available.

Antarctic fur seal

Female Antarctic Fur Seals were sampled at random from a breeding site at Bird Island, South Georgia (54 °S, 38 °W) in 1988. This was the location of a long-term mark–recapture study and a method had been developed to capture individuals from randomized positions on the beach, causing little disturbance to the colony in the process and which allowed the individual to be returned to the site of origin after having been measured (Boyd et al. 1995). The reproductive status of individuals was determined by observing the presence or absence of a pup during twice daily visits to the breeding site. Since seals could be viewed from above because of a scaffolding walkway over the beach, it was also possible to determine if a female was pregnant by the degree of distention of her abdomen when she arrived at the pupping grounds. In the case of pregnant females, mass was measured within 24 h after birth.

Each seal was weighed (nearest 0·5 kg) and standard length was measured (nearest cm). A tooth was obtained for determining age using the methods of Arnbom et al. (1992).

Crabeater seal

Data for Crabeater Seals were obtained from a cull that took place from 1967 to 1977 to supply food for sledging dogs at an Antarctic station. All seals were obtained from Margurite Bay (68 °S, 68 °W) between January and April when pregnant females have a well-developed foetus. Animals were shot from an ice-strengthened vessel with a 0·303 calibre rifle when they were hauled out on ice floes. Although it is impossible to sample at random from animals in pack ice, all animals that were available during the period when the ship was deployed for culling were shot so it is assumed that the sample was representative of the population of animals in the region.

After capture, animals were returned to the ship where they were weighed (nearest 1 kg) and standard length was measured (nearest cm) (Bonner & Laws 1993). The reproductive tract was dissected and inspected for the presence of a foetus.

Grey seal

A similar method was used to obtain population samples of Grey Seals. Animals were shot at close range with a 0·222 calibre rifle while out of the water on rocks at the Farne Islands, UK (55°40′ N, 1°35′ W). All samples were obtained during March to October in 1980 and 1981 (Boyd 1985). There was no intentional selection of different classes of individuals although, as for the Crabeater Seal, it was not possible to ensure random sampling from the population. Individuals were weighed (nearest kg) and standard length was measured (nearest cm) and the reproductive tract was examined for the presence of a foetus.

In all species, age was determined from annual growth lines in teeth to the nearest year. In Grey Seals, cementum in the upper canine teeth was used, whereas in Crabeater Seals and Antarctic Fur Seals the cementum of postcanines was used (Laws 1962; Arnbom et al. 1992). In the Fur Seals, animals were captured during the birth season so age was expressed as a whole number of years. For the other two species, age was expressed as the age from the teeth in whole numbers of years plus the number of months after the breeding season given by the date of capture.

Statistics

The values of the state variables examined in the present study are summarized in Table 1. Since pregnancy was a binary response variable, logistic regression analysis (SAS, CATMOD procedure, SAS release 6·11, SAS Institute Inc. Cary, NC) (Brown & Rothery 1993) was used to examine the effect of age, time of year, length and mass on the presence or absence of pregnancy. Models were run for all three species of pinniped with all combinations of independent variables to assess the degree to which each explained the deviance in the presence or absence of pregnancy. This was achieved through tabulation and inspection of the results of the models as shown by Brown & Rothery (1993). These comparisons were made using the chi-square statistic. Those variables that showed no significant deviation from expected were successively excluded from the model.

Table 1.  Mean, median and range of each of the state variables examined for the three species included in the present study
SpeciesState variableMean (±SD)MedianRangen
Atlantic Fur SealAge  7·3 ± 2·9  7  2–15140
 Julien date334·8 ± 9·6 29319–353140
 Length124·6 ± 7·4126 97–140140
 Mass 41·0 ± 7·1 41·25 22–58135
Crabeater SealAge  9·4 ± 6·3  8·5  0–28 90
 Julien date 46·3 ± 15·0 35 33–65 99
 Length230·2 ± 18·4235175–259 99
 Mass191·7 ± 35·8200 70–260 99
Grey SealAge 13·3 ± 9·4 10  1–36 64
 Julien date175·9 ± 65·4165 68–287 65
 Length172·7 ± 13·6175129–201 65
 Mass141·2 ± 33·8143 57–196 65

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Antarctic fur seal

Multiple logistic regression analysis of the Antarctic Fur Seal data showed there was no significant effect of mass alone upon the occurrence of pregnancy and there was also no effect of either date or age (Table 2). In contrast, length had a significant influence upon the occurrence of pregnancy. Since all the Antarctic Fur Seal samples were collected within a narrow time window of 30 days the lack of any effect of date was to be expected. When present in combination within the regression model, mass and length explained 25% of the variation in the occurrence of pregnancy. In the absence of length, mass also explained a significant amount of variation in the occurrence of pregnancy when it was present in the model with age and date but the level of variability explained was substantially less than when length was also present in the model. Comparing Model 3 with Model 6 (Table 2) shows that, when age was present in the model with mass then mass gained greater explanatory power. This interaction between mass and age became most apparent in Model 11 (Table 2) where these were introduced as an explicit interaction. Not only did the interaction explain a significant amount of the variability in the occurrence of pregnancy but so did age and, to a lesser extent, mass. However, when present in models together with length, age did not contribute significantly to the explanatory power (Models 10 and 12, Table 2). Therefore in Antarctic Fur Seals, this analysis shows that, of the state variables measured, length was the most important although mass also had an effect that is independent of the one associated with length. These relationships are illustrated in Fig. 1.

Table 2.  Results of a multiple logistic regression analysis of data from Antarctic Fur Seals (n = 140) with pregnancy as the binary response variable. Variables used in each model are shown by the value of the deviance accounted for by the presence of the variable in the model. The significance of the contribution of each variable towards the overall fit of the logistic regression model was assessed using a chi-square test with 1 degree of freedom. The results of this test are shown as superscripts beside each deviance. Mass × date and mass × length interactions were not significant
  Variables in the model 
ModelnAgeDateMassLengthMass × AgeResidual deviance
  • –= variable not used.

  • NS, not significant (P > 0·05).* = significance at 0·01 < P < 0·05;** = significance at 0·001 < P < 0·01;*** significance at P < 0·001.

 00171·0
 11 2·7 NS168·3
 210·2 NS170·0
 31 1·5 NS164·8
 41 9·7**152·8
 52 2·4 NS0·3 NS168·0
 62 4·3* 3·4 NS160·4
 7212·5***17·0***126·5
 821·9 NS 1·7 NS162·9
 93 3·2 NS0·8 NS 3·9*159·6
103 0·2 NS12·7***17·0***126·3
11412·3***0·8 NS 4·5*11·0***146·9
124 0·3 NS0·1 NS13·5***17·9***126·2
image

Figure 1. The relationships among Antarctic Fur Seals for pregnancy with mass (a) or length (b) shown for individual measurements (●) and as means (○). In the case of length, the relationship was significant (Table 2) and the line of the logistic least squares regression is shown together with the equation for the line where P is the probability of pregnancy. There was no significant relationship for mass (Table 2).

Download figure to PowerPoint

Crabeater seal

In contrast, the data from the Crabeater Seal showed that all the state variables measured in the study were important and explained significant amounts of variation in the occurrence of pregnancy (Table 3). The effect of age was greater than the other variables when entered into the model alone (Model 1, Table 3) but this was less than the effect of mass when other variables were present in models together with mass. Length was the least important variable and its effect was not significant when present in a model with mass. When present in models, mass was consistently the most important of the state variables from Crabeater Seals. Overall, the state variables investigated in the present study accounted for 55% of variation in the occurrence of pregnancy. The relationship between mass and the occurrence of pregnancy is illustrated in Fig. 2.

Table 3.  Results of a multiple logistic regression analysis of data from Crabeater Seals (n = 99) with pregnancy as the binary response variable. Variables used in each model are shown by the value of the deviance accounted for by the presence of the variable in the model. The significance of the contribution of each variable towards the overall fit of the logistic regression model was assessed using a chi-square test with 1 degree of freedom. The results of this test are shown as superscripts beside each deviance. Mass × date and mass × age interactions were not significant
  Variables in the model 
ModelnAgeDateMassLengthDeviance
  • –= variable not used.

  • NS, not significant (P > 0·05).* = significance at 0·01 < P < 0·05;** = significance at 0·001 < P < 0·01;*** significance at P < 0·001.

 00129·8
 1132·6*** 82·4
 21 4·6*123·3
 3123·7*** 85·7
 419·3**112·2
 5227·6***10·7** 73·9
 6210·2***25·1*** 66·1
 72 5·1*23·0** 78·8
 8237·0***4·5* 81·0
 9237·8***1·1 NS 85·3
103 9·9**11·6***20·7*** 58·9
11417·2***17·9***20·4***0·5 NS 58·8
image

Figure 2. The relationship in Crabeater Seals between pregnancy and mass shown for individual measurements (●) and as means (○). The relationship was significant (Table 3) and the line of the logistic least squares regression is shown together with the equation for the line where P is the probability of pregnancy.

Download figure to PowerPoint

Grey seal

In the Grey Seal, age, date, length and mass also all explained a significant amount of variation in the occurrence of pregnancy when entered into the logistic regression models on their own (Models 1–4, Table 4). However, the effects of age and date disappeared when they were present in models with either mass or length or with both. As an individual state variable, mass contributed more explanatory power to variability in the occurrence of pregnancy than any of the other variables that were measured. This effect of mass was enhanced by the inclusion of length in the model (Model 3 compared with Model 9, Table 4). Overall, the state variables investigated here accounted for 49% of variation in the occurrence of pregnancy. The relationship between mass and the occurrence of pregnancy is illustrated in Fig. 3.

Table 4.  Results of a multiple logistic regression analysis of data from Grey Seals (n = 65) with pregnancy as the binary response variable. Variables used in each model are shown by the value of the deviance accounted for by the presence of the variable in the model. The significance of the contribution of each variable towards the overall fit of the logistic regression model was assessed using a chi-square test with 1 degree of freedom. The results of this test are shown as superscripts beside each deviance. Mass × date and mass × age interactions were not significant
  Variables in the model 
ModelnAgeDateMassLengthDeviance
  • –= variable not used.

  • NS, not significant (P > 0·05).* = significance at 0·01 < P < 0·05;** = significance at 0·001 < P < 0·01;*** significance at P < 0·001.

 0062·2
 11   4·5*55·4
 214·1*58·0
 3116·2***34·6
 415·3*52·8
 52   4·3*4·4*52·1
 62   1·8 NS16·4***32·1
 721·3 NS16·2***33·6
 82<0·1 NS7·7**47·8
 9221·3***7·8**31·7
103   1·3 NS0·2 NS17·5***32·0
114   0·3 NS0·1 NS25·0***5·5*32·0
image

Figure 3. The relationship in Grey Seals between pregnancy and mass shown for individual measurements (●) and as means (○). The relationship was significant (Table 4) and the line of the logistic least squares regression is shown together with the equation for the line where P is the probability of pregnancy.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The power that the suite of state variables used in this study had to explain the occurrence of pregnancy varied from 25% of the total variation explained in the Antarctic Fur Seal to 55% in the Crabeater Seal. In the case of Fur Seals, most of the explained variability came from body length whereas in the Crabeater and Grey Seals length contributed relatively little to the explanatory power. In these cases the most important variable was mass which is a useful indication of the level of body energy reserves in pinnipeds (Arnould 1995; Fedak, Arnbom & Boyd 1996).

The interspecific comparison presented in this study involving data collected in similar ways from three species of pinnipeds has shown that the two species (Crabeater and Grey Seals) which are recognized as those that depend upon stored body reserves to rear their offspring are also those for which body mass was the most important single state variable. In contrast, Antarctic Fur Seals rely upon foraging during lactation to deliver food to their pups (Boyd 1999). State variables that reflect foraging ability are less easily measured than those that reflect current energy stores which is probably one reason why the suite of variables measured in this study explained such a low proportion of the variability in the occurrence of pregnancy in Fur Seals. State variables such as mass and length were also found to explain a small amount of the variation in pregnancy rate among South African Fur Seals (Arctocephalus pusillus, Schreber) (Guinet et al. 1998). Length was the variable that gave the greatest explanatory power in Antarctic Fur Seals. Since swimming energetics is, to a degree, a function of length (Williams 1987), length may therefore have a loose association with foraging performance. However, it is also possible that the decision about whether to reproduce is made in some pinnipeds at the time of implantation, which is approximately 8 months before birth (Boyd 1991). Data from Antarctic Fur Seals tend to confirm that the state of individuals at the time of implantation influences the subsequent birth rate (Lunn & Boyd 1993). The lack of influence of state variables that reflect body energy reserves at the time of birth in fur seals, as found in the present study, may reflect greater selection for the influence of state variables at the time of implantation in fur seals.

The state-dependent approach to life-history analysis assumes that decisions about the allocation of resources to reproduction are optimal with respect to lifetime fitness. In the present analysis, rather than examine this intraspecifically (Marrow et al. 1996; Morris 1996) which requires detailed information about the relative success of different decisions by individuals, the comparison is made across species with widely varying reproductive investment strategies. It remains possible that the samples obtained in this study were not completely representative of the populations being studied. For example, the Antarctic Fur Seal samples all came from a single year and it is possible that some of these relationships vary between years in which there is contrasting food abundance. Nevertheless, it seems most probable that the functional relationships, as opposed to the population mean response in any year, between the occurrence of pregnancy and state variables are unaffected by such environmental variability because they represent optimal responses that have arisen due to natural selection. Although it is not possible to test directly for optimality in the allocation decisions made by each species in relation to the different state variables, it is logical that those species that rely mainly upon food stores for success of reproduction should make the decision about whether to reproduce based upon the level of those food stores. The observation that the species for which food stores are less important to the success of reproduction (in this case Fur Seals) did not exhibit this dependency supports the view that life histories are the end result of a variety of different functional responses to a set of state variables with different degrees of influence.

Life-history variables, in addition to the birth/pregnancy rate examined in the present study, that are likely to be subject to such functional relationships with state variables include survival rates and weaning success. Antarctic Fur Seals incur a substantial survival and reproductive cost, in terms of reduced future survival and reproduction, with each reproductive attempt (Boyd et al. 1995) suggesting that, as income breeders, they may be less well buffered against environmental variability than the capital breeders. There are insufficient data for Antarctic Fur Seals, or either of the other species of pinniped examined in the present study, to allow the quantification of the relationship between the survival rate and the type of range of state variables illustrated in the present study. Only in Northern Fur Seals (Callorhinus ursinus) has a relationship between weaning mass and subsequent survival rate been illustrated (Baker & Fowler 1992). Examination of pinniped population dynamics in general (Eberhardt & Siniff 1977; Harwood & Prime 1978) suggests that populations are relatively insensitive to changes in the birth rate but they are sensitive to changes in survival rate. Therefore, the functional relationship illustrated in the present study is likely to have less power to explain life-history variation than the equivalent type of relationship with survival as the dependent variable.

The results of the present study suggest that the classical age-dependent approach to life-history analysis, which has been applied widely to pinnipeds (Eberhardt & Siniff 1977; Eberhardt 1990; Sydeman et al. 1991; Boyd et al. 1995) as well as many other long-lived iteroparous vertebrates, contains fewer insights into the dynamics of life histories than a state-dependent approach (McNamara & Houston 1996). Life-history analyses in the three species examined in the present study (Harwood & Prime 1978; Boyd et al. 1995; Boveng 1993) have all been age-based even though only in the case of the Crabeater Seal does the explanatory power of age approach that of mass for the occurrence of pregnancy. More recent studies, most especially in Grey, Harbour (Phoca vitulina) and Southern Elephant (Mirounga leonina Linnaeus) Seals (Bowen et al. 1994; Fedak et al. 1996; Arnbom, Fedak & Boyd 1997; Mellish, Iverson & Bowen 1999; Pomeroy et al. 1999), have begun to demonstrate the importance of considering body condition, rather than age, as a factor in the way that investment decisions are made by individuals. Similar effects of state variables upon the survival of pinnipeds have also been observed (Reiter & Le Boeuf 1991; Lunn, Boyd & Croxall 1994; Boyd et al. 1995; Hastings & Testa 1998). The contribution made by this study has been to demonstrate empirical functional relationships that could eventually be incorporated formally into state-dependent life-history analyses of pinnipeds.

In a previous analysis of the Grey Seal data used in this study (Boyd 1984), it was suggested that mass was an important covariate in determining the occurrence of pregnancy but the effect of natural seasonal changes in mass could not be distinguished from those of mass itself upon pregnancy. In this analysis, the use of multiple logistic regression showed that the seasonal effect, defined by the date variable, was insignificant compared with mass (Model 7, Table 4). Therefore, even though the samples from Grey Seals were collected over most of the period of active foetal growth (7–8 months), the effects of mass on the occurrence of pregnancy were independent of this time span.

The results of the present study suggest that, in future, size- rather than age-structured models would provide a more informative way of approaching studies of population dynamics, at least in pinnipeds, if not for large mammals in general. It also points the way to a requirement for a greater understanding of growth patterns and dynamics.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

I am grateful to all those who have helped to collect the data used in this study. I am particularly grateful to Dr R. M. Laws who ensured that information was collected from Crabeater Seals. J. H. Prime and Dr N. J. Lunn were important members of the team that collected the data from Grey and Antarctic Fur Seals, respectively.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Arnbom, T., Lunn, N.J., Boyd, I.L. & Barton, T. (1992) Aging live Antarctic fur seals and southern elephant seals. Marine Mammal Science 8, 3743.
  • Arnbom, T., Fedak. M.A. & Boyd, I.L. (1997) Factors affecting maternal expenditure in southern elephant seals during lactation. Ecology 78, 471483.
  • Arnould, J.P.Y. (1995) Indices of body condition and body composition in female Antarctic fur seals (Arctocephalus gazella). Marine Mammal Science 11, 301313.
  • Baker, J.D. & Fowler, C.W. (1992) Pup weight and survival of northern fur seals, Callorhinus ursinus. Journal of Zoology, London 227, 231237.
  • Bengston, J.L. & Laws, R.M. (1985) Trends in crabeater seal age at sexual maturity: an insight into Antarctic marine interactions. Antarctic Nutrient Cycles and Food Webs (eds W. R.Siegfried, P. R.Condy & R. M.Laws), pp. 667675. Springer, Berlin.
  • Boness, D.J. & Bowen, W.D. (1996) The evolution of maternal care in pinnipeds. Bio Science 46, 645654.
  • Bonner, W.N. & Laws, R.M. (1993) Morphometrics, specimen collection and preservation. Antarctic Seals: Research Methods and Techniques (ed. R. M.Laws), pp. 161171, Cambridge University Press, Cambridge.
  • Boveng, P.L. (1993) Variability in a crabeater seal population and the marine ecosystem near the Antarctic Peninsula. PhD Thesis, Montana State University, Bozeman, MT.
  • Bowen, W.D., Oftedal, O.T., Boness, D.J. & Iverson, S.J. (1994) The effect of maternal age and other factors on birth mass in the harbour seal. Canadian Journal of Zoology 72, 814.
  • Boyd, I.L. (1984) The relationship between body condition and the timing of implantation in pregnant grey seals (Halichoerus grypus). Journal of Zoology, London 203, 113123.
  • Boyd, I.L. (1985) Pregnancy and ovulation rates in grey seals (Halichoerus grypus) on the British coast. Journal of Zoology, London 205, 265272.
  • Boyd, I.L. (1991) Environmental and physiological factors controlling the reproductive cycles of seals. Canadian Journal of Zoology 69, 11351148.
  • Boyd, I.L. (1998) Time and energy constraints in pinniped lactation. American Naturalist 152, 717728.
  • Boyd, I.L. (1999) Foraging and provisioning in Antarctic fur seals: interannual variability in time–energy budgets. Behavioral Ecology 10, 198208.
  • Boyd, I.L., Croxall, J.P., Lunn, N.J. & Reid, K. (1995) Population demography of Antarctic fur seals: the costs of reproduction and implications for life-histories. Journal of Animal Ecology 64, 505518.
  • Brown, D. & Rothery, P. (1993) Models in Biology: Mathematics, Statistics and Computing. Wiley, London.
  • Caswell, H. (1989) Matrix Population Models. Sinauer Associates, Inc., Sunderland, MA.
  • Eberhardt, L.L. (1990) A fur seal population model based on age structured data. Canadian Journal of Fisheries and Aquatic Sciences 47, 122127.
  • Eberhardt, L.L. & Siniff, D.B. (1977) Population dynamics and marine mammal management policies. Journal of the Fisheries Research Board of Canada 34, 183190.
  • Fedak, M.A., Arnbom, T. & Boyd, I.L. (1996) The relationship between size of southern elephant seal mothers, the growth of their pups and the use of maternal energy, fat and protein during lactation. Physiological Zoology 69, 887911.
  • Festa-Bianchet, M., Gaillard, J.M. & Jorgenson, J.T. (1998) Mass- and density-dependent reproductive success and reproductive costs in a capital breeder. American Naturalist 152, 367379.
  • Guinet, C., Roux, J.P., Bonnet, M. & Mison, V. (1998) Effect of body size, body mass, and body condition on reproduction of female South African fur seals (Arctocephalus pusillus) in Namibia. Canadian Journal of Zoology 76, 14181424.
  • Harwood, J. & Prime, J.H. (1978) Some factors affecting the size of British grey seal populations. Journal of Applied Ecology 15, 401411.
  • Hastings, K.K. & Testa, J.W. (1998) Maternal and birth colony effects on survival of Weddell seal offspring from McMurdo Sound, Antarctica. Journal of Animal Ecology 67, 722740.
  • Jonsson, K.I. (1997) Capital and income breeding as alternative tactics of resource use in reproduction. Oikos 78, 5766.
  • Laws, R.M. (1956) Growth and sexual maturity in aquatic mammals. Nature 178, 193194.
  • Laws, R.M. (1962) Age determination in pinnipeds with special reference to growth layers in the teeth. Zeitschrift für Säugetierkunde 27, 129146.
  • Lunn, N.J. & Boyd, I.L. (1993) Influence of maternal characteristics and environmental variation on reproduction in Antarctic fur seals. Symposium of the Zoological Society of London 66, 115128.
  • Lunn, N.J., Boyd, I.L. & Croxall, J.P. (1994) Reproductive performance of female Antarctic fur seals: the influence of age, breeding experience, environmental variation and individual quality. Journal of Animal Ecology 63, 827840.
  • Marrow, P., McNamara, J.M., Houston, A.I., Stevenson, I.R. & Clutton-Brock, T.H. (1996) State-dependent life history evolution in soay sheep: dynamic modelling of reproductive scheduling. Philosophical Transactions of the Royal Society of London, Series B 351, 1732.
  • McNamara, J.M. & Houston, A.I. (1996) State-dependent life histories. Nature 380, 215221.
  • Mellish, J.A.E., Iverson, S.J. & Bowen, W.D. (1999) Variation in milk production and lactation performance in grey seals and consequences for pup growth and weaning. Physiological and Biochemical Zoology 72, 677690.
  • Morris, D.W. (1996) State-dependent life history and senescence of white-footed mice. Ecoscience 3, 16.
  • Pomeroy, P.P., Fedak, M.A., Rothery, P. & Anderson, S. (1999) Consequences of maternal size for reproductive expenditure and pupping success of grey seals at North Rona, Scotland. Journal of Animal Ecology 68, 235253.
  • Reiter, J. & Le Boeuf, B.J. (1991) Life-history consequences of variation in age at primiparity in northern elephant seals. Behavioral Ecology and Sociobiology 28, 153160.
  • Sydeman, W.J., Huber, H.R., Emlie, S.D., Ribic, C.A. & Nur, N. (1991) Age-specific weaning success of northern elephant seals in relation to previous breeding experience. Ecology 72, 22042217.
  • Williams, T.M. (1987) Approaches for the study of exercise physiology and hydrodynamics in marine mammals. Marine Mammal Energetics (eds A. C.Huntley, D. P.Costa, G. A. J.Worthy & M. A.Castellini), pp. 127146. Special Publication no. 1. Society for Marine Mammalogy. Allen Press, Lawrence, KA.