The content of the present PhD thesis deals with the application of microsatellite analysis to the study of two species of large mammals, referring to some aspects of their social and mating systems. The Italian wolf (Canis lupus) and the fallow deer (Dama dama) were chosen as case studies, since genetic investigations on their populations result, for different reasons, problematic.

The wolf in Italy is a particularly protected species, recovering throughout the peninsula from the effects of a recent bottleneck. Sampling wolves may not rely on capturing or killing them, therefore an alternative, non-invasive, approach was adopted in my study. Scats, shed hairs and blood drops collected on the snow represented the main source of DNA for the analysis. Methodological cares were necessary to obtain reliable wolf genotypes. A set of ten canine microsatellites was employed to achieve unique multilocus genotypes in the population. Fifty-two individuals were typed in the period 1998 - 2003. In some cases, pack composition was determined, confirming that familiar bonds are at the basis of wolf social units. An unpredictable high local differentiation was found among geographic areas. Early dispersal seem to be common in the population, but its effects on the gene flow are not detectable, at least at my study scale. I proposed that most of this dispersal may be unsuccessful or over long distances. The study population, indeed, seem to have reached a high level of local saturation, with clumped pack territories and high reproductive rates, and thus possibly represents a source, from which wolves disperse toward sink areas.

The second study involves an enclosed population of fallow deer. Lekking is seldom observed in mammals, and among them, it is more common in ungulate species. Fallow deer is one of the most studied lekking ungulate and this particular population is object of long-term researches on male mating strategies. Mature bucks in the population join into leks during the breeding season: a costly strategy, which apparently does not guarantee high direct benefits (fitness). In this study, for the first time, I used a molecular approach to verify the existence of a genetic basis to lek formation. The recourse to microsatellites enabled to overcome the strong limitation due to the extremely monomorphism of the species, documented in several studies. Though the low variability even at microsatellite loci, the hypotheses of kin selection (territorial bucks in a lek are relatives) and of heterozygosity advantage (territorial bucks in a lek have an overall high heterozygosity) were tested and resulted not supported by data. Hence, future studies should be addressed towards phenotypic variation and consider in detail physiological and ecological factors, in order to clarify the reasons why lekking takes place in fallow deer.