MATERNAL AGGRESSION IN THE SOUTH-AMERICAN FUR-SEAL IN PERU

Maternal aggression was examined with regard to its role in maternal defense of offspring in the South American fur seal (Arctocephalus australis) at Punta San Juan, Peru (15-degrees-22'S, 75-degrees-12'W). Female fur seals' responses to conspecifics and to predatory southern sea lions (Otaria byronia) were examined pre- and post-partum. Although the defense of offspring from predators may be expected to be a major component of maternal care, mothers rarely defended their offspring from raiding sea lions, possibly due to the high risks involved in attacking a large and potentially very dangerous predator. The costs of defending the pup from sea lions appeared to outweigh the potential benefit of increased survival of the pup, due to the already high pup mortality at the site. Maternal defense from conspecifics may also be a possible function of female aggression. Females with newborn pups were more aggressive towards other females than were females without pups, particularly during the perinatal attendance period. They threatened a higher proportion of approaching females and won more aggressive encounters postpartum. However, mothers of surviving pups did not appear to be any more aggressive, their pups did not receive any fewer threats from unrelated females, nor did they reside in areas of less aggression, than mothers whose pups died. Aggression towards male conspecifics appeared to serve a further purpose, as females threatened approaching adult and subadult males regardless of whether the females had a pup. Unlike colonially breeding phocid pinnipeds, there seems to be no correlation between increased maternal-aggression and neonate survival in the South American fur seal. This may be because females have to leave their offspring to forage when the offspring are still vulnerable to attack by unrelated conspecifics. Alternatively, maternal aggression may have been subjected to such intense selection that although it is responsible for increased pup mortality at the site, there is insufficient variability to measure differential consequences.