(2) Relatedness-based mate choice
To test if titis avoid mating with related individuals, we compared relatedness between real and randomly generated mating partners using the pairwise relatedness estimator implemented in STORM (Frasier, 2008). First, relatedness of real mating pairs (n = 10) was calculated and averaged. The expected relatedness of random mating pairs was then calculated by generating mating pairs from female and male breeding pools over 1000 simulations and compared to the observed relatedness values. The breeding pool of 12 females and 12 males included all sampled adults and the oldest subadult male from Group 6 who had dispersed from his natal group in the beginning of the study period and was thus available as a potential mating partner by the end of the sampling period.