Marc D. Hauser Peter Marler Animal Communication Laboratory, Department of Zoology, University of California, Davis, CA 95616-8761, USA Address reprint requests to M. D. Hauser at the Depanment of Biological Anthropology, Haf\'ard University, Cam- bridge, MA 02138. 26 1991 Revised 13 April 1992 Second revision 12 May 1992 Accepted 10 June 1992 1 045-2249/93!$5.00 <tl 1993 International Society for Behavioral Ecology Food-associated calls in rhesus macaques (Macaca mulatta): I. Socioecological factors Upon discovering food, free-living rhesus macaques (Macaca mulaUa) on the island of Cayo Santiago, Puerto Rico, produce a complex of vocal signals consisting of five acoustically distinguishable calls. This report examines the socioecological factors eliciting call production and the information potentially con- veyed to others. The primary contexts for three vocalizations ("warbles," "harmonic arches," and "chirps") are encounters with rare and highly preferred foods (e.g., coconut). Two other vocalizations ("coos" and "grunts") are produced both in food (primarily provisioned chow) and in nonfood contexts, such as during mother-infant separation and grooming interactions. Grunts given upon encountering food are acous- tically distinct from those given in nonfood contexts. In contrast, coos associated with food are statistically indistinguishable from coos given in other contexts. When conspecifics hear these food-associated calls, they typically approach the caller. Coos are less likely to lead to approach than other food-associated calls. Results from ali-day follows on adult males and adult females reveal that changes in hunger level influence call rate but not call type; the different call types are produced throughout the day. We infer that the structure of food-associated calls provides information about the quality of the food discovered, whereas call rate conveys information about the relative hunger level of the caller. In this population, adult males give fewer food-associated calls than adult females. In addition, females within large matrilines call more than females within smaller matrilines, and males who are resident in a group are more vocal than peripheral males. Kry words: food-associated calls, foraging behavior, hunger, Macaca mulaUa, referential signalling, rhesus monkeys, socioecology. [&hav Ecol 4:194-205 (1993)] A n important goal of foraging studies is to un- derstand the kinds of rules underlying or guiding observed patterns of feeding (Stephens and Krebs, 1986). Central to such investigations is the role of information about available food, including knowledge that individuals may derive both from individual experience and from the experiences of companions. Regarding the latter, a number of studies have shown that individuals often vocalize when they encounter food (see reviews in Elowson et al., 1991; Heinrich and Marzluff, 1991). Such food-associated calls provide an important poten- tial source of information to group members about the location of food and should influence the for- aging patterns of others in hearing range. The so- cioecological conditions under which food-associ- ated calls are given and the kinds of information potentially conveyed to group members thus have direct ecological relevance. Most studies of food-associated calls (see Table 1) have concentrated either on the causes or the consequences of calling (e.g., Brown et al., 1991; Chapman and Lefebvre, 1990; Elgar, 1986; Good- all, 1986; Hauser and Wrangham, 1987; Hauser et al., in press; Heinrich, 1988; Wrangharn, 1977). In addition, some have attempted to specify the in- formation conveyed and received (Dittus, 1984, 1988; Elowson et al., 1991; Gyger and Marler, 1988; Marler et al., 1986a). In these studies, results in- dicate that call production is influenced by the quantity, quality, and divisibility of food discovered and that conspecifics hearing such calls tend to move in the direction of the signaler. Food-asso- dated calls thus cause an increase in group size at the food source and may consequently lead to ben- efits in terms of predator defense and costs in terms of feeding competition. Concerning call meaning, studies reveal that a majority of calls are restricted to the context of food and may provide information about food in general, its quality, and the caller's affective or emotional relation to the food encoun- tered (e.g., its food preference). With the exception of Dittus's (1984, 1988) re- search on toque macaques (Macaca sinica), none of the studies reviewed above have combined obser- vations of the causes and consequences of produc- ing food-associated calls with detailed analyses of the information potentially conveyed by the signal. However, understanding the kind of information profferred may be critical to understanding both the factors influencing call production and the po- tential benefits obtained from hearing such calls. For example, calls that convey information about the discovery of food, and perhaps about its quality or quantity, can provide other group members with specific information about an alternative food patch, permitting the fine tuning of giving-up time as well as other foraging parameters (Stephens and Krebs, 1986). Information offered by specific food-asso- ciated calls is likely to reduce the signaler's food intake as a result of recruiting others to the food source, thereby imposing a Calls that have a less specific relationship to food and that also occur in other circumstances may be designed as general recruitment signals, serving to increase group size. Consequently, such calls are less informative to for- 194 Behavioral Ecology Vol. 4 No.3