Reply from R.C. Connor et al.

Abstract

Marino correctly points out that we overlooked her work relating brain and group size in odontocetes. This is an important subject, and worthy of study. However, there are problems with her analysis that prevent us from embracing her finding of a `significant positive correlation between pod size and encephalization quotient among 21 odontocete species' as being equivalent to Dunbar's conclusions for primates[1xDunbar, R.I.M. J. Hum. Evol. 1992; 20: 469–493Crossref | Scopus (660)See all References[1].The hypothesis in question holds that the size of the brain (or, more specifically, the neocortex) places a limit on the number of social relationships that an individual can handle simultaneously[1xDunbar, R.I.M. J. Hum. Evol. 1992; 20: 469–493Crossref | Scopus (660)See all References[1]. Individuals in Dunbar's primate `groups' have their primary social relationships with each other and not individuals of other groups. Thus his `group size' is very closely related to the mean number of social relationships of an individual. In contrast, the `pod size' reported most often for odontocetes, and apparently used by Marino, is simply the number of individuals that are usually observed together at a given point in time. This may be very different from the number of social relationships of an individual for several reasons.First, coastal bottlenose dolphins (Tursiops sp.), and probably many other cetaceans, live in fission–fusion societies in which the typical number of individuals found together ( 100) (Refs [2xWells, R.S., Scott, M.D., and Irvine, A.B. Curr. Mammal. 1987; 1: 247–305CrossrefSee all References[2],[3xSmolker, R.A. et al. Behaviour. 1992; 123: 38–69Crossref | Scopus (250)See all References[3]). If small-brained odontocetes such as Inia, Platanista and Pontoporia live in similar fission–fusion societies, then available `pod size' data will significantly underestimate the number of social relationships individuals maintain.Second, large groups of large-brained pelagic delphinids (e.g. Lagenorhynchus) might reflect nonsocial assemblages of smaller social units attracted to food sources or minimizing predation risk. Until these species are studied we simply do not know.We conclude that while the number of social relationships maintained by individuals in a few large-brained, well studied odontocetes clearly rival or exceed nonhuman primates, a correlation between the number of social relationships individuals maintain and brain size among odontocetes has not been established.