Social grooming in the common vampire bat, Desmodus rotundus

Grooming patterns among 65 common vampire bats in hollow tree day roosts were studied by behavioural sampling techniques during a 15-month period. Self-grooming occurred more than social grooming in response to ectoparasites since the proportion of time spent self-grooming and the amount of ectoparasite infestation covaried positively among tree roosts, while the time spent grooming others was independent of roost and ectoparasite level. Rates of social grooming were not independent of the sex of participants due to infrequent social grooming by adult males. Two variables, the level of relatedness and a measure of roosting association, which previously were shown to predict food sharing by regurgitation, correlated positively with the rate of social grooming. Since social grooming occurred more often than expected before a regurgitation and correlated with regurgitation frequency, it is suggested that this behaviour facilitates identification of food sharing partners by enabling a grooming bat to monitor other animals' potential for giving or receiving blood. The functional significance of social grooming (licking, nibbling or picking at the fur ofa conspecitic) has yet to be unequivocally demonstrated in any species. If one assumes that social grooming costs a donor time that could have been allocated to self-maintenance or feeding but benefits a recipient either by removing ectoparasites (Freeland 1976), by helping to heal wounds (Simonds 1965), or by increasing the likelihood of support in an aggressive dispute (Seyfarth 1977, 1980, 1983; de Waal 1982), then either kin selection (Kurland 1977; Silk 1982) or reciprocity (Seyfarth & Cheney 1984) must be invoked to explain the occurrence of this behaviour. However, Dunbar & Sharman (1984) reviewed 40 baboon studies which measured the amount of time spent feeding, resting and social grooming and failed to find the expected negative relationship between time spent feeding and time spent social grooming. Instead, they discovered that the time allotted to ~social grooming was conserved. During periods of food shortage, resting time decreased to permit extended feeding periods. Consequently, at present there is no evidence that social grooming costs a donor by decreasing its reproductive success or survival. Furthermore, although some primates groom conspecifics in locations that are hard to reach (Hutchins & Barash 1976) and around recent wounds (Simonds 1965), there is little evidence to demonstrate that social grooming directly benefits a recipient by enhancing its survival or reproduc* Present address: Department of Zoology, University of Maryland, College Park, Maryland 20742, U.S.A. tion. Grooming might indirectly affect an animal's fitness, however, by influencing a partner's behaviour in future, more costly, situations. For example, Seyfarth (1977) has suggested that one female may groom a higher-ranking female to gain support from her in a subsequent aggressive alliance. If females form a linear dominance hierarchy with relatives in adjacent ranks and all of the females attempt to groom higher-ranking animals, then high-ranking females and close relatives will be groomed most often (Seyfarth 1977). This interpretation is supported by observations of kin-biased grooming distributions in several primate species (Seyfarth 1980, 1983). In a recent study, however, Seyfarth & Cheney (1984) present results from a playback experiment on vervet monkeys which suggest that monkey A will support an unrelated animal B in an aggressive interaction if B recently groomed A. This apparent reciprocal altruism did not occur among closely related monkeys. Of course, such an explanation does not preclude the operation of kin selection since aggressive alliances sometimes form more frequently among kin (Silk 1982) and can be independent of dominance relations (Dunbar 1980). This brief review of primate grooming studies suggests that social grooming may serve purposes other than ectoparasite control in species with complex social behaviour. Unfortunately, very few non-primate grooming studies have been reported for comparison. Here, therefore, I present data on the frequency and distribution of social grooming in the vampire bat, Des'modus rotundus. This ani-