Intraspecific genetic admixture and the morphological diversification of an estuarine fish population complex.

The North-east American Rainbow smelt (Osmerus mordax) is composed of two glacial races first identified through the spatial distribution of two distinct mtDNA lineages. Contemporary breeding populations of smelt in the St. Lawrence estuary comprise contrasting mixtures of both lineages, suggesting that the two races came into secondary contact in this estuary. The overall objective of this study was to assess the role of intraspecific genetic admixture in the morphological diversification of the estuarine rainbow smelt population complex. The morphology of mixed-ancestry populations varied as a function of the relative contribution of the two races to estuarine populations, supporting the hypothesis of genetic admixture. Populations comprising both ancestral mtDNA races did not exhibit intermediate morphologies relative to pure populations but rather exhibited many traits that exceeded the parental trait values, consistent with the hypothesis of transgressive segregation. Evidence for genetic admixture at the level of the nuclear gene pool, however, provided only partial support for this hypothesis. Variation at nuclear AFLP markers revealed clear evidence of the two corresponding mtDNA glacial races. The admixture of the two races at the nuclear level is only pronounced in mixed-ancestry populations dominated by one of the mtDNA lineages, the same populations showing the greatest degree of morphological diversification and population structure. In contrast, mixed-ancestry populations dominated by the alternate mtDNA lineage showed little evidence of introgression of the nuclear genome, little morphological diversification and little contemporary population genetic structure. These results only partially support the hypothesis of transgressive segregation and may be the result of the differential effects of natural selection acting on admixed genomes from different sources.

Mean fecal glucocorticoid metabolites are associated with vigilance, whereas immediate cortisol levels better reflect acute anti-predator responses in meerkats.

Adrenal hormones likely affect anti-predator behavior in animals. With experimental field studies, we first investigated associations between mean fecal glucocorticoid metabolite (fGC) excretion and vigilance and with behavioral responses to alarm call playbacks in free-ranging meerkats (Suricata suricatta). We then tested how vigilance and behavioral responses to alarm call playbacks were affected in individuals administered exogenous cortisol. We found a positive association between mean fGC concentrations and vigilance behavior, but no relationship with the intensity of behavioral responses to alarm calls. However, in response to alarm call playbacks, individuals administered cortisol took slightly longer to resume foraging than control individuals treated with saline solution. Vigilance behavior, which occurs in the presence and absence of dangerous stimuli, serves to detect and avoid potential dangers, whereas responses to alarm calls serve to avoid immediate predation. Our data show that mean fGC excretion in meerkats was associated with vigilance, as a re-occurring anti-predator behavior over long time periods, and experimentally induced elevations of plasma cortisol affected the response to immediate threats. Together, our results indicate an association between the two types of anti-predator behavior and glucocorticoids, but that the underlying mechanisms may differ. Our study emphasizes the need to consider appropriate measures of adrenal activity specific to different contexts when assessing links between stress physiology and different anti-predator behaviors.

Reproductive rate, not dominance status, affects fecal glucocorticoid levels in breeding female meerkats.

Glucocorticoid hormones (GCs) have been studied intensively to understand the associations between physiological stress and reproductive skew in animal societies. However, we have little appreciation of the range of either natural levels within and among individuals, or the associations among dominance status, reproductive rate and GCs levels during breeding. To address these shortcomings, we examined variation in fecal glucocorticoid metabolites (fGC) during breeding periods in free-ranging female meerkats (Suricata suricatta) over 11 years. The vast majority of variation in fGC levels was found within breeding events by the same female (~87%), with the remaining variation arising among breeding events and among females. Concentrations of fGC generally tripled as pregnancy progressed. However, females with a high reproductive rate, defined as those conceiving within a month following parturition (mean = 9 days postpartum), showed significant reductions in fGC in the final 2 weeks before parturition. Despite these reductions, females with a high reproductive rate had higher fGC levels at conception of the following litter than those breeding at a low rate. After controlling for the higher reproductive rate of dominants, we found no association between levels of fGC and either age or dominance status. Our results suggest that one should be cautious about interpreting associations between dominance status, reproductive skew and GCs levels, without knowledge of the natural variation in GCs levels within and among females.