Search

In Antarctic fur seals, Arctocephalus gazella, mothers must identify their own young among hundreds or even thousands of pups, if they are to invest in their own offspring and avoid misdirecting their parental care. When returning to their breeding colony from a foraging trip of several days at sea, mothers have to find and identify their young before suckling can occur. There appears to be little confusion about which pup belongs to a mother, and adoption is absent or rare. Using behavioral observations, we investigated the means by which female Antarctic fur seals identified their pups in a breeding colony of about 750 mother-pup pairs on Kerguelen Island. We evaluated the importance of vision, scent communication, vocalizations, and rendezvous locations as possible explanations of how mothers find their pups. Every pup that a mother examined, whether her own or not, exchanged naso-nasal inspection with her, suggesting a strong role for olfactory communication in individual recognition. Both mothers and pups called to each other, and mothers that searched for pups over a longer period gave more calls and encountered more pups. Thus, vocalizations may have been used to attract pups that might be offspring. Nursing usually occurred in the same place from the end of one maternal visit to the colony and the arrival at the beginning of the next visit, suggesting that nursing locations may serve as a meeting place, or rendezvous, for mothers and pups. These results suggest that finding pups is a two-stage process for females, in which pups for sampling are attracted by calls or examined at the previous nursing location, and then individual identification is made by olfactory cues.

Albatrosses exhibit extremely low reproductive rates, each pair brooding only one egg and subsequent chick at a time. Furthermore, in several of the species, the majority of successful pairs breed only once every second year (termed 'biennial' breeding). Thus, on average, these latter species have an annual fecundity of about half an offspring per year, while other albatrosses produce an egg and chick every year. Using our 40-year bank of demographic data, we compared 12 species of albatrosses according to these two breeding strategies to examine potential causes of biennial breeding. Biennial breeding could be due to physiological constraints, larger animals breeding more slowly, or ecological constraints, more distant pelagic feeding trips being energetically costly, or both. We tested these hypotheses by looking for predicted associations between the duration of the rearing period, the distance to the oceanic feeding zone and breeding frequency. We also looked for associations of these variables with other life-history traits. Body size had a strong influence on the duration of the rearing period, but not on the distance that birds travelled to the feeding zone. Both the duration of the rearing period and distance to the feeding zone appeared to have direct influences on breeding frequency, as revealed by a path analysis, and thus both hypotheses to explain biennial breeding were supported. Finally, breeding frequency exhibited a strong trade-off with adult survival and age at maturity, indicating that slower breeders live through more breeding seasons, perhaps mitigating their lower annual reproductive output.

Environmental factors influence the dynamics and regulation of biological populations through their influences on demographic variables, but demographic mechanisms of population regulation have received little attention. We investigated the demographic basis of regulation of Columbian ground squirrel (Spermophilus columbianus) populations under natural and experimentally food-supplemented conditions. Food supplementation caused substantial increases in population density, and population densities returned to pretreatment levels when the supplementation ended. Control (untreated) populations remained relatively stable throughout the study period (1981-1986). Because food resources regulated the size of the ground squirrel populations, we used life-table response experiment (LTRE) analyses to examine the demographic basis of changes in population growth rate and thus also demographic influences on population regulation. LTRE analyses of two food-manipulated populations revealed that changes in age at maturity and fertility rate of females generally made the largest contributions to observed changes in population growth rate. Thus, our results suggested that abundance of food resources regulated the size of our study populations through the effects of food resources on age at maturity and fertility rates. Our results also indicated that different demographic mechanisms can underlie population regulation under different environmental conditions, because lower juvenile survival substantially contributed to population decline, but in only one of the populations. Demographic analyses of experimental data, such as those presented here, offer a rigorous and unambiguous means to elucidate the demographic basis of population regulation and to help identify environmental factors that underlie dynamics and regulation of biological populations.

Evolutionary theory suggests that mating systems should have substantial effects on gene dynamics of local populations. In polygynous species, local 'breeding groups' may produce significant genetic structure, due to genetic differences among groups, and rate of loss of genetic variation from such populations may be considerably slowed. We examined possible influences of the variable mating system and family group structure on genetic properties of a population of plateau pikas (Ochotona curzoniae). Pika gene dynamics were examined via F-statistics and effective population sizes (N(e)), calculated from genetic correlations within and among individuals and families. Genetic correlations were estimated from mating patterns, population demography, and dispersal patterns. Substantial genetic structure within the population was indicated by a strongly positive F(LS). Genetic influence of natal dispersal out of pika families was indicated by a strongly negative inbreeding statistic (F(IL)=-0.34). Effective size of the population was not greatly different from the census population, whereas a traditional estimate of effective size of the population was much lower, indicating that the family structure of the pikas results in a slowed loss of genetic variation over time. Thus, even though mating patterns of plateau pikas were variable, family structure had a strong influence on pika gene dynamics.

We studied mate choice in the wandering albatross, Diomedea exulans, using data from 32 years of banding returns in the population of the Crozet Islands. We studied mating choices in a single year, when the Crozet Islands population was male biased (8:5, males:females). Thus, we expected that females might show great flexibility of choice of partners. Because age and experience might influence mate choice, we tested the expectation that females would choose the oldest and most experienced males for pair bonding. Pair bonds usually last until one member of the pair dies (0.3% of the birds 'divorce'), so mate choice should be especially important. We found that the ages of males and females in both displaying and bonded (breeding) pairs were significantly correlated. These age-associated pairings were not a passive phenomenon, but appeared to be due to an active process of selection of mates of similar age. First-time breeders sought mates of similar age, but preferred those with the most experience. Remating, experienced birds whose mates had died did not pair with individuals of significantly similar age, but predominantly paired with other widowed birds that, on average, were also relatively old. Mate fidelity in wandering albatrosses may be due to the cost of finding and bonding with a new mate. Pair bonds, and thus breeding, took an average of 3.2 and 2.3 years to establish, for males and females, respectively. Thus, remating exerts a potential average reproductive cost of about 15% of lifetime reproductive success. Copyright 1999 The Association for the Study of Animal Behaviour.

Considerable controversy surrounds the importance of inbreeding in natural populations. The rate of natural inbreeding and the influences of behavioral mechanisms that serve to promote or minimize inbreeding (e.g., philopatry vs. dispersal) are poorly understood. We studied inbreeding and social structuring of a population of black-tailed prairie dogs (Cynomys ludovicianus) to assess the influence of dispersal and mating behavior on patterns of genetic variation. We examined 15 years of data on prairie dogs, including survival and reproduction, social behavior, pedigrees, and allozyme alleles. Pedigrees revealed mean inbreeding coefficients (F) of 1-2%. A breeding-group model that incorporated details of prairie dog behavior and demography was used to estimate values of fixation indices (F-statistics). Model predictions were consistent with the minimization of inbreeding within breeding groups ("coteries," asymptotic F IL = -0.18) and random mating within the subpopulation ("colony," asymptotic F IS = 0.00). Estimates from pedigrees (mean F IL = -0.23, mean F IS = 0.00) and allozyme data (mean F IL = -0.21, mean F IS = -0.01) were consistent with predictions of the model. The breeding-group model, pedigrees, and allozyme data showed remarkably congruent results, and indicated strong genetic structuring within the colony (F LS = 0.16, 0.19, and 0.17, respectively). We concluded that although inbreeding occurred in the colony, the rate of inbreeding was strongly minimized at the level of breeding groups, but not at the subpopulation level. The behavioral mechanisms most important to the minimization of inbreeding appeared to be patterns of male-biased dispersal of both subadults and adults, associated with strong philopatry of females. Incest avoidance also occurred, associated with recognition of close kin via direct social learning within the breeding groups., (© 1997 The Society for the Study of Evolution.)

Sexual size dimorphism might be influenced by environmental constraints on sexual selection or by intraspecific competition between males and females. We studied bobcats (Lynx rufus) in collections of museum specimens from western North America to examine these hypotheses. Structural body size was estimated from several measurements of the skull, ln-transformed and indexed through principal components analysis. Sexual dimorphism in body size was estimated from the difference in size index of males and females, and compared to geographic and climatic variables associated with biotic provinces (ecoregions). Of several climatic variables that were associated with bobcat body size, only seasonality of climate was associated with sexual dimorphism. Sexual size dimorphism, longitude, elevation, and seasonality were intercorrelated. As longitude decreased (moving inland from west-coastal ecoregions), sexual dimorphism decreased with the increased elevation and seasonality of continental climates of the Rocky Mountains. We suggest that increased seasonality and the need for fasting endurance by females may place constraints on the degree of sexual dimorphism in bobcats. Sexual dimorphism of body size and sexual size dimorphism of trophic structures (teeth) exhibited a strong positive association over geography, thus indirectly supporting the hypothesis that intrasexual competition for prey could account for the geographic variation in sexual size dimorphism. Thus, both environmental constraints on sexual selection of body size and intersexual competition were supported as possible explanations of the degree of sexual size dimorphism that occurs in populations of bobcats.

The hypothesis that food resources regulate population size was tested in Columbian ground squirrels, Spermophilus columbianus, from 1981 to 1986 in southwestern Alberta, Canada. Two replicate populations received supplemental food resources from 1981 to 1983, and were subsequently monitored until 1986. Two reference (unmanipulated) populations were monitored throughout the 6 years. During the experiment, dramatic increases in population size of about 500% occurred. After supplementation, spring populations declined by about 20% per year, under conditions that produced stressful shortages of food (as evidenced by significant decreases in body mass of ground squirrels). The demographic process that contributed most to increasing and decreasing experimental populations was production of yearlings, through changes in reproductive success and survival of young. Changes in migration contributed secondarily to changes in population size, and changes in adult survival were least important. While similarities of demographic processes were evident between replicates during population increases, a dramatic decrease in survival of young contributed to population decline at one replicate, and "normal" levels of reproduction and survival were sufficient to produce a decline in the other replicate. Demographic mechanisms may vary, but population regulation of mammalian populations appears strongly dependent on food resources.

From analyses of published data and a review of the literature, I studied indirect and direct measures of gene flow among populations of Columbian ground squirrels, Spermophilus columbianus. New analyses were used to examine an allozyme data set (seven polymorphic loci) that had been collected by Zammuto and Millar (1985a) from six populations of ground squirrels that were spread over 183 km. G-tests indicated significant variation in allele frequencies among populations, but F-statistics revealed relatively little population differentiation (average F ST =0.026). F ST values were used to estimate rates of gene flow indirectly and indicated fairly high rates of gene flow (average N e m=13.5). Recorded dispersal distances of individual ground squirrels were fairly short (most<4 km, maximum recorded distance was 8.5 km), and the minimum distance between populations used to create the allozyme data set was about 25 km. Thus, direct dispersal among the populations in the allozyme data set was highly unlikely. Small genetically effective populations may have experienced high rates of migration over short distances (about 43% of adults in local populations were immigrants), however, resulting in homogeneous allele frequencies over the geographic range. This explanation provides an alternative to invoking gene flow in the recent past to explain discrepancies between dispersal distances in the field and homogenization of allele frequencies over large ranges, Mammalian species that have virtually complete dispersal of subadult males from the natal area might be expected to exhibit relatively high rates of gene flow, regardless of actual dispersal distances. Genetically effective populations may be much smaller than more extensive ecological populations and experience higher rates of gene flow.

Columbian ground squirrels (Spermophilus columbianus) exhibit an elevational cline in several life-history traits and in body mass. Body mass is a trait that might interact with the elevational changes in life histories, but it could reflect either or both of two underlying factors: structural size and physiological condition. I studied these factors in several populations of ground squirrels. Structural size was estimated from first principal component scores in an analysis of postcranial bone lengths. Physiological condition was inferred from differences in body mass that were independent of structural size. Field body mass and structural size of yearlings decreased with increasing elevation, which indicates slower growth at higher elevations. For adults, physiological condition was not measured directly, but evidence suggested that it was reflected by changes in body mass: body mass declined with increasing elevation but structural size did not, residuals of body mass on structural size showed the same patterns of change as body mass, most of the variation in body mass was orthogonal to a "general size factor" in principal components analyses, and significant changes in body mass of individuals could be stimulated by experimental supplementation of food. Life-history traits and physiological condition covaried along the elevational cline and were phenotypically plastic, but the structural size of adult ground squirrels was not an extremely plastic trait.

The cost of reproduction for female Columbian ground squirrels was assessed using seven years of data from three populations in southwestern Alberta, Canada. Short-term costs, measured as reduced survival or lowered subsequent fecundity, were not directly associated with whether a female reproduced or not, nor with the females' litter size. In one of the populations, females that first reproduced as yearlings had slightly shorter lifespans, but almost 1.5 times the lifetime reproductive success, compared to those that first reproduced when older. Although short-term reproductive costs were not apparent from our data, we cannot conclude that reproduction was not costly. Such costs might exist but be masked by females' adjusting reproductive investment to their body condition. A weak association of spring (minimum) body weight of females and subsequent reproduction supported this possibility, as did correlated increases in body weight and litter size in a population given supplemental food.