The timing of exposure to the steroid hormone, testosterone, produces activational and organizational effects in vertebrates. These activational and organizational effects are hypothesized to relate with the number of female mating partners and reproductive success in males. We tested this hypothesis by examining 151 wild degu (Octodon degus) males across a 10-year study. We quantified the association between adult serum testosterone levels (i.e., an indirect index of adult activational effects) and anogenital distance (AGD) length (i.e., a direct index of fetal organizational effects), and their interaction on the number of female mating partners and reproductive success. We found no evidence of an association between adult male serum testosterone levels and the number of female mating partners, or between adult male serum testosterone levels and reproductive success. However, male AGD was positively associated with reproductive success, but not so with the number of female mating partners. Additionally, the positive association between male AGD and male reproductive success was mediated by the number of mates. Our findings do not support major roles of activational or organizational effects of testosterone on the number of female mating partners and its consequences on male reproductive success. Instead, our results suggest that compared with individual male attributes, the female social environment plays a more important role in driving male reproductive success.

Results 3.1

To test the prediction (i) about the potential effects of male serum testosterone levels, male anogenital distance (AGD), and the number of females within the social group, and their interactions, on the total number of female mating partners, run model model 1. This model run with n = 151 replicates, representing all males sampled (i.e., this examination included males with zero effective mates). The response variable is Total number of female mating partners.

To test prediction (ii) about the potential effects of male serum testosterone levels, male anogenital distance (AGD), and the number of females within the social group, and their interactions, on the number of female mating partners within the social group, run model model 2. This model run with n = 89 replicates, representing all males that sired some offspring and whose social group included at least one female. The response variable is Number of female mating partners within the social group.

To test prediction (iii) about the potential effects of male serum testosterone levels, male anogenital distance (AGD), and the number of females within the social group, and their interactions, on the number of female mating partners outside the social group, run model model 3. This model run with n = 107 replicates, representing all males that bred effectively. The response variable is Number of female mating partners outside the social group.

Results 3.2.

To test prediction (iv) about the potential effects of male serum testosterone levels, male anogenital distance (AGD), and the number of females mating partners, and their interactions, on the number of offspring sired by males at weaning, run model model 4. This model run with n = 107 replicates, representing all males that bred effectively. The response variable is Number of offspring sired at weaning.

In all models, year and social group ID (SGID), were random factors. All other factors were considered fixed factors. Serum testosterone does not distribute normally, so it must be included transformed with Log10.

Model fits were assessed with quantile residual dispersion. Best models were chosen by their AICc values and average model weight.