The overarching goal of my thesis is to characterize the relationship between kinship and
social behaviour in a species with a cooperative, multilevel social structure – the sperm
whale. To do so, I use a combination of genetic, behavioural and acoustic data collected
during a longitudinal study of sperm whale social units, in the eastern Caribbean. Social
units are a stable and basal component of sperm whale social structure. Associations
between social units occur within large cultural groups, called vocal clans. To deal with
degraded DNA from non-invasive sampling, I develop a protocol that maximizes
genotyping success with degraded DNA, while quantifying and minimizing error rates.
Using microsatellite loci and mitochondrial DNA haplotypes, I evaluate kinship among
sperm whales, and I examine its relationship to social association, alloparental care and
vocal repertoires. First, I characterize the extent and pattern of kinship in and among
sperm whale social units, and test whether association is predicted by kinship. I
document that social units have a clear matrilineal basis, but do not appear to be strictly
matrilineal. My findings also indicate paternal relatedness between social units. Within
units, I find individuals associate more with their closer relatives, but this is not the case
among units. Second, I investigate calf care in relation to kinship. I demonstrate that
behavioural observations are not always sufficient for assigning maternity, and that
alloparental care is considerable in some cases and correlates positively with maternal
kinship. Exceptions to the general pattern, however, demonstrate that, in addition to kinselection, other factors influence alloparental care, perhaps including reciprocity, group
augmentation or gaining maternal experience. Lastly, I examine acoustic repertoires of
individuals and social units, in the context of kinship and social association. Variation in
vocal repertoires was not explained by close kinship or social bonds. This supports the
prevailing hypothesis that these vocalizations are culturally transmitted, and not
determined genetically. Further, this suggests that vocal learning occurs broadly within
clans, rather than preferentially from close kin or close social associates, or that biases in
vocal learning at lower levels of social structure are diffused by clan-level processes.
Also, by observing an absence of signals of kinship in vocalizations, my results suggest
that a different mechanism, perhaps familiarity, regulates kin-selection among sperm
whales. In conclusion, kinship clearly influences social unit composition, association
preferences and alloparental care among sperm whales. However, I also reveal variability
in social behaviour that is unexplained by kinship, which highlights the complexity of
drivers behind social structure, cooperation and communication in this cultural, highly
social and large-brained species.