No evidence for female kin association, indications for extragroup paternity, and sex-biased dispersal patterns in wild western gorillas

Abstract

Characterizing animal dispersal patterns and the rational behind individuals' transfer choices is a long-standing question of interest in evolutionary biology. In wild western gorillas (Gorilla gorilla), a one-male polygynous species, previous genetic findings suggested that, when dispersing, females might favor groups with female kin to promote cooperation, resulting in higher-than-expected within-group female relatedness. The extent of male dispersal remains unclear with studies showing conflicting results. To investigate male and female dispersal patterns and extragroup paternity, we analyzed long-term field observations, including female spatial proximity data, together with genetic data (10 autosomal microsatellites) on individuals from a unique set of four habituated western gorilla groups, and four additional extragroup males (49 individuals in total). The majority of offspring (25 of 27) were sired by the group male. For two offspring, evidence for extragroup paternity was found. Contrarily to previous findings, adult females were not significantly more related within groups than across groups. Consistently, adult female relatedness within groups did not correlate with their spatial proximity inferred from behavioral data. Adult females were similarly related to adult males from their group than from other groups. Using R _ST statistics, we found significant genetic structure and a pattern of isolation by distance, indicating limited dispersal in this species. Comparing relatedness among females and among males revealed that males disperse farer than females, as expected in a polygamous species. Our study on habituated western gorillas shed light on the dispersal dynamics and reproductive behavior of this polygynous species and challenge some of the previous results based on unhabituated groups.

Keywords: dispersal; great apes; kin association; paternity; polygynous species; western gorillas.

FIGURE 1

Spatial distribution of the study groups, with the group compositions of sampled individuals. CAR1/CAR2 are groups sampled at two field sites in Central African Republic (CAR), and RC1/RC2 are groups sampled at one field site in Republic of Congo (RC). The number of individuals sampled in each group is written in parenthesis (note that one individual from RC1 did not produce enough genetic data so was removed from the dataset for the genetic analyses). Pop, Population; SB, Silverbacks; AF, Adult Females; IM, Immatures, this includes all offspring from each study group, thus young silverbacks, blackbacks, subadults, juveniles and infants; Other SB, corresponds to two solitary silverbacks and two silverbacks from semi‐habituated groups

FIGURE 2

Histogram of the expected average relatedness values within each group obtained by 10,000 random permutations of individuals among groups. The red arrow indicates where the observed value lies. The p‐value was computed as the percentage of permutations where the expected values were greater than or equal to the observed value. All computations were performed with the function grouprel of the R package related (see Methods), and the graphs were also displayed using this function

FIGURE 3

Pairwise values of the population differentiation index R _ST/(1‐R _ST) plotted against the logarithm of the geographical distance using all adults (left) or only female adults (right)

FIGURE 4

Histogram of the expected average relatedness values within each country obtained by 10,000 random permutations of individuals among countries, for females (upper) and for males (lower). The red arrow indicates where the observed value lies. The p‐value was computed as the percentage of permutations where the expected values were greater than or equal to the observed value. All computations were performed with the function grouprel of the R package related (see methods), and the graphs were also displayed using this function