Predicting trait values and measuring selection in complex life histories: reproductive allocation decisions in Soay sheep.

Accurate prediction of life history phenomena and characterisation of selection in free-living animal populations are fundamental goals in evolutionary ecology. In density regulated, structured populations, where individual state influences fate, simple and widely used approaches based on individual lifetime measures of fitness are difficult to justify. We combine recently developed structured population modelling tools with ideas from modern evolutionary game theory (adaptive dynamics) to understand selection on allocation of female reproductive effort to singletons or twins in a size-structured population of feral sheep. In marked contrast to the classical selection analyses, our model-based approach predicts that the female allocation strategy is under negligible directional selection. These differences arise because classical selection analysis ignores components of offspring fitness and fails to consider selection over the complete life cycle.

Patterns of parental relatedness and pup survival in the grey seal (Halichoerus grypus).

Levels of parental relatedness can affect offspring survival and susceptibility to disease. We investigated parental relatedness of live and dead Halichoerus grypus pups between and within island populations and between possible causes of mortality. Nine microsatellites were used to calculate internal relatedness (IR) and standardized mean d2. We find that pups with higher than average levels of IR have significantly lower survival and that this varied between island populations and that certain loci contributed to the effect more than others. Although, there were no significant differences between causes of mortality, peritonitis, infection and stillborn had the highest levels of IR. These results provide evidence that parental relatedness is an important determinant of pre-weaning pup survival in the grey seal and that this may vary with cause of mortality given a larger sample size.

Sex differences in emigration and mortality affect optimal management of deer populations.

Populations of red deer that are limited by food, like those of many other ungulates, commonly include more females than males. We assessed the contribution of variation in sex- and age-specific rates of mortality and emigration to density-dependent changes in the adult sex ratio, using long-term observations and demographic experiments involving the red deer population on Rum, Scotland. We incorporated these effects in a stochastic model of local populations under different management regimes to show here that, when female numbers are allowed to increase to more than 60% of the ecological carrying capacity, the sustainable annual harvest of males from local deer populations will fall. Because males are typically culled by fee-paying hunters and generate more income than females, income will decrease as the male harvest falls. Because numbers of female deer throughout much of the Highlands probably exceed the threshold at which male density starts to be affected, many managers might be able to raise income from local deer populations by reducing female numbers, with potential benefits to the vegetation of Scottish Highland environments.

Sex determination in malaria parasites.

A century ago, W. G. MacCallum identified distinct male and female forms in malaria parasites of both birds and humans. Since then, scientists have been puzzled by the high female-to-male ratios of parasites in Plasmodium infections and by the mechanism of sex determination. The sex ratio of malaria parasites was shown to become progressively more male as conditions that allow motility and subsequent fertilization by the male parasites become adverse. This resulted from an increased immune response against male gametes, which coincides with intense host erythropoietic activity. Natural and artificial induction of erythropoiesis in vertebrate hosts provoked a shift toward male parasite production. This change in parasite sex ratio led to reduced reproductive success in the parasite, which suggests that sex determination is adaptive and is regulated by the hematologic state of the host.

Microsatellites reveal heterosis in red deer.

The fitness consequences of inbreeding and outbreeding are poorly understood in natural populations. We explore two microsatellite-based variables, individual heterozygosity (likely to correlate with recent inbreeding) and a new individual-specific internal distance measure, mean d2 (focusing on events deeper in the pedigree), in relation to two measures of fitness expressed early in life, birth weight and neonatal survival, in 670 red deer calves (Cervus elaphus) born on the Isle of Rum between 1982 and 1996. For comparison, we also analyse inbreeding coefficients derived from pedigrees in which paternity was inferred by molecular methods. Only 14 out of 231 calves (6.1%) had non-zero inbreeding coefficients, and neither inbreeding coefficient nor individual heterozygosity was consistently related to birth weight or neonatal survival. However, mean d2 was consistently related to both fitness measures. Low mean d2 was associated with low birth weight, especially following cold Aprils, in which foetal growth is reduced. Low mean d2 was also associated with low neonatal survival, but this effect was probably mediated by birth weight because fitting birth weight to the neonatal survival model displaced mean d2 as an explanatory variable. We conclude that in the deer population fitness measures expressed early in life do not show evidence of inbreeding depression, but they do show evidence of heterosis, possibly as a result of population mixing. We also demonstrate the practical problems of estimating inbreeding via pedigrees compared with a direct marker-based estimate of individual heterozygosity. We suggest that, together, individual heterozygosity and mean d2, estimated using microsatellites, are useful tools for exploring inbreeding and outbreeding in natural population.

The maintenance of genetic polymorphism in small island populations: large mammals in the Hebrides.

Conventionally, small populations living on islands are expected to lose genetic variation by drift. Fluctuations in population size, combined with polygynous mating systems, are expected to contribute to the process by increasing sampling effects on genetic variation. However, in individually monitored populations of Red deer on Rum and Soay sheep on St. Kilda, which experience fluctuations in population size, two processes have been identified which mitigate loss of genetic variation. First, in a number of examples, population reductions are associated with selection. Selection may be in favour of heterozygotes, or, as we have documented in several cases, it may fluctuate in direction temporally. Second, in Soay sheep, in which mortality over population crashes is male-biased, ostensibly leading to low effective numbers of males, molecular studies show that there are systematic changes in the reproductive success of young males, and in variance in male success, that broaden genetic representation compared with expectation.