Male rank, not paternity, predicts male–immature relationships in mountain gorillas, Gorilla beringei beringei

• We assessed whether dominance rank or paternity predicts male–immature relationships in gorillas.
• Male dominance rank predicted relationship strength much better than paternity did.
• We found little evidence for paternity discrimination in multimale mountain gorilla groups.
• Group size and reproductive skew variance were also associated with relationship differences.

Kin discrimination mechanisms are expected to evolve when they provide fitness benefits. To date, evidence for kin discrimination is mixed across taxa and mating systems even when it would apparently be beneficial. In animals with promiscuous mating systems, males were long believed to abstain from parenting behaviours partly because the costs of offspring misidentification outweighed the benefits of dual parenting. Conversely, males in monogamous systems could parent because of high paternity certainty. However, recent work has shown that in some species males parent despite high false paternity rates, and males in some promiscuous systems discriminate between their own and other males' offspring. Here we evaluate the impact of male dominance rank, paternity and age on male–immature relationships in wild mountain gorillas. Mountain gorillas provide an interesting context for assessing paternal kin discrimination because (1) male–immature relationships are strong, and (2) while their morphological characteristics suggest an evolutionary history of single-male groups, a substantial fraction contain multiple adult males. In our sample of 21 males and 49 genotyped immatures living in multimale groups monitored by the Dian Fossey Gorilla Fund's Karisoke Research Center, we found that male rank was the primary predictor of male–immature relationship strength. There was little evidence that paternity or age were related to relationship patterns. Male–immature dyads were closer social partners in 2011–12 when groups were smaller and reproductive skew lower, than comparable dyads in 2003–04 when groups were larger and skew higher. Gorillas' lack of paternal kin discrimination provides further behavioural evidence that the species' multimale social structure is evolutionarily novel. However, patterning of male–immature relationships and genetic paternity suggest a persistent minority of two-male groups throughout G. beringei's evolutionary history. This may help explain their ability to live in multimale, multifemale social units despite possessing morphological characteristics typical of harem systems.