We conducted observations of three bonobo groups at Lola Ya Bonobo Sanctuary, Kinshasa, DR Congo, between September and November 2008 and between August and November 2009. The daily routine at the sanctuary remained the same throughout observation periods (electronic supplementary material). In 2008, we observed individuals in one of the two largest enclosures, henceforth ‘group 1a’. In 2009, we collected data from two groups housed in the same and the adjacent enclosure, henceforth ‘group 1b’ and ‘group 2’. Group composition in the first enclosure changed between the two study periods (group 1a: n = 9 females, n = 9 males, n = 4 infants; group 1b: n = 7 females, n = 9 males, n = 4 infants; group 2: n = 5 females, n = 11 males; n = 3 infants; electronic supplementary material, table S1). We pooled the data across the three groups and combined data for dyads that met again in the second year (n = 9 female–female dyads; n = 19 male–female dyads).
In bonobos, sexual behaviour can take a variety of forms, such as heterosexual copulations (with pelvic thrusts and intromission), mountings, homosexual genital contacts and genital stimulation using an object or body part. Here, we recorded behaviours and vocalizations of females engaging in copulations with males or genital contacts with females. Copulation calls were acoustically distinct and never observed in contexts other than sexual interactions. We conducted observations using all-day focal and ad libitum sampling methods (approx. n
= 1093 h), balanced across individuals. We recorded female vocalizations at distances of 3–20 m using a Sennheiser MKH816T directional microphone and Marantz PMD660 solid-state recorder (sampling rate of 44.1 kHz, 16 bits accuracy). Bonobo copulation calls typically consist of a single or succession of high-frequency squeaks and screams that usually begin during the copulation [9
] (electronic supplementary material, figure S1). We also recorded female reproductive states following veterinary assessments and collected daily records of swelling sizes, using Furuichi's [15
] 4-point scale based on degree of wrinkling.
To assess the influence of dominance, we used hierarchies calculated for another study on the same study groups and period [14
], based on the outcome of dyadic agonistic interactions. We used ‘fleeing upon aggression’ as a behavioural marker for subordinance [16
]. Dominance relationships and linearity were calculated with the Matman matrix analysis program (Noldus
v. 1.1). We calculated and tested the adjusted linearity index h
′, corrected for unknown relationships [16
], as well as the directional consistency index (electronic supplementary material, table S2). For significantly linear hierarchies, we calculated individual cardinal ranks, using normalized David's scores corrected for chance (electronic supplementary material; tables S1 and S2) [16
]. Using regression plots of the rank scores, we divided the females into either high or low ranks, based on their position in the hierarchies (high: n
= 6; low: n
= 8). For the males, the absence of significant linearity prevented calculations of cardinal ranks. We thus assigned high-rank status to any male who dominated (i.e. elicited fleeing behaviour) at least 50 per cent of the other males in the group (high: n
= 7; low: n
= 17, electronic supplementary material, tables S1–S2).