Behavioral Responses of Free-ranging Rhesus Monkeys to Food Shortage JAMES LOYl Northwestern University, Department of Anthropology, Evanston, Illinois 60201 ABSTRACT During early July 1968, a severe food shortage occurred on Cayo Santiago, a n island colony of free-ranging rhesus monkeys (AMacaca muEatta). The food shortage produced striking changes i n the behavior of the monkeys. Within the one social group intensively studied, the total frequencies of grooming, play and fights decreased significantly; the frequency of matings also dropped; body contact, displacements by other groups, and non-displacement movements decreased, but not in statistically significant amounts. Changes in the percentage of total grooming attributable to related and unrelated monkeys reflected the stability of the rhesus matriline. Comparisons made between Cay0 Santiago and other primate groups under analogous situations reveal similar responses to food shortage. Cay0 Santiago lies just off the southeast corner of Puerto Rico (fig. 1 ) . %At the time of the present study i t was inhabited by over 700 free-ranging rhesus monkeys (Macaca inzilattn) , whose ancestors were transported to the island from India in 1938 (Carpenter, ’42). Besides the monkeys, the island supports several species of crabs and lizards, black rats, and a species of warbler. The island is covered with a heavy growth of vegetation. While the monkeys supplement their diet by foraging on the vegetation and eating a n occasional insect, their dietary staple is Purina Monkey Chow. The chow is distributed at several feeding stations around the island (fig. 1). A delayed shipment of monkey chow caused the July 1968 food shortage and the behavioral changes reported i n this paper. METHODS Subjects On August 4, 1967, the author began a 13 month behavioral study of one of the seven Cay0 Santiago social groups, group F. By June 1, 1968, 1454 hours of observation had been made on the group. An additional 321 hours of observation were made during June and July 1968 (table 3 ) . Observations were tyDically made from 0730-1145 and 1530-1800 Monday through Friday, and from 1500-1800 on Saturday and Sunday. During June and July 1968, observations were made on all days except June AM. J. PHYS.ANTHROP.,33. 263-272 30th. The monkeys were habituated to the author’s presence, and allowed his close approach. The Cayo Santiago monkeys are all tattooed for identification of individuals. Censuses of the population begun in the late 1950’s provide the matrilineal relatives of all but the oldest monkeys. Patrilineal relationships are not known. Prior to the food shartage, group F contained 69 monkeys (table 1 ). Although the fourth largest group on the island in numbers, group F was a t least fifth, and probably sixth, in group dominance, and was repeatedly forced away from feeding stations and rest areas. Although occasionally able to control a feeder for several minutes to a n hour. group F usually fed by making “raids” on temporarily unoccupied feede r ~The . ~ monkeys would quickly fill mouth and cheek pouches, and then sprint off into the undergrowth with a more dominant group at their heels. The group F monkeys spent some part of every day foraging, even when they had already eaten a good deal of monkey chow. Food shortage Provisioned food during June and July 1968 is shown in table 2. The period with1 Present address: 2154 Kennington Road, Knoxville Tennessee 37917. % T h eC a y 0 Santiago colony is administered by the Laboratory of Perinatal Physiology, National Institutes of Health. 3During the entire study, group F was never seen feeding at F-4.The Small Cay was usually occupied by grnup A, which contained over 200 monkeys and was dominant to group F. 263 264 JAMES LOY out provisioned food of any kind was only one week. However, from June 24th-29th the regular provision of 300 pounds of monkey chow per day was cut to 240 pounds per day. Between July 1st-5th over 2000 mangoes were brought to the island, and on July 2nd a h a 1 300 pounds of monkey chow were distributed. Mangoes were not provided after July 5th. The next food given to the monkeys was 800 pounds of whole kernal corn between July 12th14th. On July 15th, monkey chow arrived at Cayo Santiago, and 300 pounds were put out. Subsequently, the monkeys received the regular 300 pounds of Purina chow every day (except Sunday, when no food was provided). In total, 22 days elapsed between periods of uninterrupted D provisioning with the full complement of chow-. Typical behavior. during the transition period between the birth and mating seasons The food shortage occurred during late June-early July. It thus fell at the end of the birth season (January-June) and the beginning of the mating season (JulyDecember) (Koford, '65). At this time there is much play among infants and juveniles. Although fertile matings typically do not begin until August, an occasional male-to-female series-mounting is seen in June and July (possibly with the motor patterns characteristic of ejaculation). When the animals are well-fed and relaxed, there is normally a lot of grooming, sitting close together and sitting in body contact. The frequency of agonistic interactions is very high. Intergroup and intragroup fights are common, especially around the feeders. RESULTS Group F mortality Six group F animals died during the food shortage. They ranged in age from just over one year to seven years. Five of the six dead were females (not significantly different from expected figures; Binomial test; p = 0.18). One of the dead females CAY0 SANTIAGO - 100 Meters TABLE 2 Provisioned food during June and July 1968 Date Provisioned food 22 June and prior 300 pounds Purina Monkey Chow every day (execpt Sunday) 240 pounds of monkey chow every day 800 mangoes 300 mangoes; 300 pounds monkey chow 200 mangoes None 800 mangoes None 800 pounds of whole kernel corn distributed over the three days 300 pounds Purina Monkey Chow None 300 pounds of monkey chow every day (except Sunday) South Point Fig. 1 Map of Cay0 Santiago. Modified from Altmann ('62). F-1 is no longer used and therefore is not shown on the map. TABLE 1 1 July 2 July Age-sex profile of Cayo Santiago group F: July 1st, 1968 Ace Male Female 6 and older 5 11 5 1 3 4 5 5 3 5 5 7 3 4 7 29 40 2 Yearlings Infants 5 3 2.4-29 June 3 July 4 July 5 July G 1 1 July 12-14 July 15 July 16 July 17 July and subsequently BEHAVIORAL RESPONSES TO FOOD SHORTAGE suffered from Diabetes menitus, which may have been exacerbated by the food shortage. General behavior during the food shortage The general activity level of the monkeys appeared to be depressed during the famine. There was a great increase in the amount of time spent foraging from the vegetation cover. Same monkeys, especially the young, appeared to lose weight. A few of' the monkeys that subsequently died appeared lethargic immediately before they were missed from their group. These same monkeys foraged constantly up to their death or disappearance. Among the plants eaten by the group F monkeys during the food shortage were Roble flowers (Tabeuia heterophylla), needles of the Australian pine ( Casuarina equisetifolia), Corcho leaves and twigs (Torrubia fraguns), Sotacaballo leaves (Randia aculeta), Coconut palm fronds (COCOS nucifera), and the fruit and leaves of the Sea Grape (CoccoEoba uvifera). Unfortunately, as emphasis was placed on social behavior, precise, quantitative data with regard to identity and amount of all plant foods eaten was not recorded. The diabetic adult female in group F w7as observed to eat two lizards on July 5th (probably Anolis spp.; Heatwole, Sade and Hildreth, '63). The lizards, each four to six inches long, were consumed in toto. Earlier in the spring of 1968, the same female was seen eating a single lizard of about the same size. This lizard u7as also entirely consumed, head first. Such carnivorous behavior had never before been observed among the Cay0 Santiago monkeys. It is not known whether this female's diabetes was connected with her consumption of animal protein. Quantitative changes in behavior During the focd shortage all forms o€ behavior were depressed with the single exception of foraging (figs. 2-8). The food shortage data was compared to averages for June 1968, June 16th-30th 1968, and the combined averages for April, May and June 1968 (tables 3, 4 ) . Social grooming (self grooming not included) dropped from the average of 17.7 k F4 a B m 22 2 .n Qa 265 266 JAMES LOY episodes/hour in the latter half of June 1968 to 8.4 episodes /hour during the food shortage (July 1st--1l t h for calculation purposes) (fig. 2 ) . This was a drop of 52.5%. The same type of drop in frequency of grooming and other social activities is often seen on weekends even during normal provisioning (see Sunday, July 21st, on figs. 2, 3 , 5 , 6). When the chow provided on Friday and Saturday is exhausted, the monkeys turn almost exclusively to foraging. The reverse of this situation was seen on July 2nd. After a two day break in provisioning, distribution of monkey chow resulted in a dramatic rise i n fights, body contact, and grooming (figs. 2, 3 , 5, 6 ) . Grooming episodes between relatives rose from 55.3% of the total June 1968 episodes to 73.9% of the total during the food shortage, while grooming episodes between non-relatives dropped from 44.7% to 26.1% of the respective totals for the same periods. Body contact per hour (the instances of 2 animals sitting touching each other) decreased slightly, from 6.9 episodes/hour for June 16th-30th to 6.4 episodes/hour f o r July 1st-11th (fig. 3 ) . This was a 7.2% drop. Body contact between relatives rose from 70.5% of the total during June to 73.9% during the food shortage, while body contact between non-relatives dropped TABLE 4 Per cent decreases i n behavior: group F . Expressed as u percentage drop during the period o f food shortage ( J u l y 1 - 1 1 ) f r o m the time period shown. Example: the average numher of grooming episodes/hour dropped 52.5% during J u l y 1st-11th from the m e a n of June 7630, 1968. Grooming/ hour Body Contact, hour Play/ hour Fights/ hour Non-feeder fights/hour Displace- mc::z/ Non-displ. movements/ hour % 5% 5% % % 7.2 5% 90.n 7% 52.5 52.5 32.3 15.6 33.3 Ave, for April, May and June 1968 45.8 7.2 90.4 48.1 35.4 30.3 45.4 June 16-30 (L 30- -30 3 0 I a w a :20- -2 0 D 0 v, a w m z 5 -1 0 10- 0 a W -0 0JUNE 16 JULY I WG. I Fig. 2 Grooming episodes in group F per hour of observation during June 16-30 and July 1-31, 1968. 267 BEHAVIORAL RESPONSES TO FOOL) SHORTAGE from 29.5% to 26.1% of the respective totals for the two periods. The number of play episodes decreased dramatically from 2.51 episodes/hour during June 16th-30th to 0.15 episodes/hour during July 1st-11th (a drop of 90.0% ) (fig. 4 ) . Fights, including both inter- and intragroup fights at and away from feeders, €ell from 6.8 fights/hour in the last half of June to 3.2 fightsihour during the food shortage (fig. 5). This was a drop of 52.5%. Non-feeder fights (those inter- and intragroup fights that occurred away from the feeding stations) also showed a drop, fallClZ 73 0 NO FOOD FOOD I IH ClZ 111 a 07 W ing from 4.2 fights/hour during the latter part of June to 2.8 fights/hour during the food shortage (fig. 6 ) . This was a decrease of 32.3%. Displacements of group F by other groups showed little change during the food shortage. Between June 16th-30th there were 1.1displacements of group F per hour; between July 1st-llth, 0.9 clisplacements/hour (drop of 15.6%) (fig. 7). Neither did the number of times that group F moved without obviously being displaced by another group (non-displacement movements) show much change. Group F moved on its own on the average of 0.9 v///m FOOD 20- -20 10- - 10 0 0 5!? a W I- 2 lz 0 0 > a 0 m 0--t 20 JUNE 16 10 -0 20 AUG. I JULY 1 Fig. 3 Body contact episodes i n group F per hour of observation during June 16-30 July 1-31, 1968. NO FOOD FOOD P 3 1 6- FOOD 1 -6 -5 -4 -3 -2 - I -0 JUNE IS- JULY I AUG. I Fig. 4 Play episodes in group F per hour of observation during June 16-30 and July 1-31, 1968. 268 JAMES LOY times/hour during the latter half of June, total fights ( p < 0.001), and non-feeder and 0.6 times/hour during the food short- fights ( p < 0.025) for the two periods. The drops in frequency of body contact ( p > age (fig. 8). This was a drop of 33.3%. Even though the mating season was ap- 0.25), displacements of group F by other proaching, series-mountings of females by groups ( p > O . l O ) , and non-displacement males decreased from ten in June 1968 to movements ( p > O . O 5 ) were €ound to be non-significant . one in July 1968. The significant decrease in number of DISCUSSION AND COMPARISONS fights per hour is no doubt connected to Several statistical tests of significance the fact that the amount of food provided were run on the data for June 16th-30th in the feeders went from almost a normal versus the food shortage data. The “t” test level to no food at all very rapidly. Thererevealed significant decreases between the fore chow was not an incentive for intermean frequencies of grooming episodes or intragroup fights. Also, the monkeys ( p < 0.001), play episodes ( p < O . O O l ) , were widely scattered as they foraged, and -I5 N O FOOD FOOD -to -5 0--I .) 20 JUNE 16 10 3M -0 30t AUG. I 20 JULY I Fig. 5 Total fights in group F per hour of observation during June 16-30 and July 1-31, 1968. NO FOOD FOOD -9 0 -0 U -7 - 6 -5 -4 -3 -2 -1 -0 30) lo 20 30 9 JULY I AUG. I Fig. 6 Non-feeder fights in group F per hour of observation during June 16-30 and July 1-31, 1968. 0- 1. 20 JUNE 16 269 BEHAVIORAL RESPONSES TO FOOD SHORTAGE there were few squabbles over choice items of vegetation. If a small amount of chow had continued to be provided, a rise in the number of fights per hour might have been expected to occur. The relative stability of the frequency of body contact is related to the stability of intragenealogy activities. During normal provisioning, the percentages of both grooming episodes and body contact episodes are consistently higher between relatives than non-relatives (Sade, '65, '66; personal observations). Intragenealogy grooming typically comprises 60-70% of all grooming, and intragenealogy body contact 70-80 % 20 301 of all episodes of body contact. During the food shortage, total intergenealogy interactions decreased much more than total intragenealogy interactions. This is seen from the changes in the percentage of grooming and body contact between relatives and non-relatives. The small decrease in intragenealogy activities was not enough to significantly affect the mean number of body contact episodes per hour. These data reflect the importance of the matriline to rhesus monkey social organization and behavior (see Sade, '65. '66). Even though there was no food to prompt group interactions around the feeders, the 10 JULY I Fig. 7 Displacements of group F per hour of observation during June 16-30 and July 1-31, 1968. u) t- z w E y 2- -2 I- --I 0 x I- z W r w c) a -I a E I 0 z -0 0- JULY I Fig. 8 Non-displacement movements by group 16-30 and July 1-31, 1968. JUNE 16 AUG. I F per hour of observation during June 270 JAMES LOY frequency of displacements of group F did various stages of estrus were fairly comnot decrease significantly. This is prob- mon among mainland populations in the ably explained by ( 1 ) the crowded condi- same area; ( 4 ) no mating, mounting, or tions on Cay0 Santiago which made it presenting was observed; (5) many of the practically impossible for any group to animals had the appearance of lethargy; move without contact with anothcr group and ( 6 ) the number of intragroup squab(and in the case of group F, contact almost bles was reduced. The above observations closely parallel always led to displacement), and ( 2 ) the possibility that during the food shortage those made on group F during the food the other groups were attracted to the au- shortage. Group F, however, did not split thor, expecting him to distribute chow. In into foraging sub-groups, probably due to approaching the author, the other groups the crowded conditions on Cay0 Santiago would almost certainly have displaced and the need for group unity in intergroup encounters. The baboon group-split and the group F. In general, the group F monkeys were reduction in fighting resembles the situa( 1 ) foraging from the vegetation cover tion on Desecheo Island. The intragenealmuch more than usual during the food ogy unity observed among the rhesus monshortage, ( 2 ) grooming, playing, fighting keys (in grooming and body contact data and mating much less, and (3) moving on from Cay0 Santiago, and in the compositheir own, being displaced by other groups, tion of some of the sub-groups on Deand sitting in body contact slightly less secheo) suggests that perhaps the small than usual. Social interactions between foraging parties observed by Hall were non-related monkeys showed a much composed of related individuals. Rosenblum, Kaufman and Stynes ('69) greater decrease in frequency than those reported on the behavioral effects of brief between matrilineal relatives. On Desecheo Island, another NINDS ( 2 day) food deprivation among laboratory colony on Puerto Rico's west coast, the groups of pigtail macaques (Macaca rhesus monkeys are unprovisioned and nemestrina) and bonnet macaques ( M a must forage for all of their food among caca radiata). During deprivation the pigplants including grasses, shrubs, herbs, tails showed ( 1 ) marked drops in social and cactus. Morrison (n.d.) made the fol- play, social grooming and autogrooming, lowing observations on the Desechco mon- (2) an increase in the amount of motherkeys : ( 1) the original group of monkeys infant contact (during the daylight phase placed on the island in 1966 (taken from of the experiment), and (3) an increase 1 social group on Cay0 Santiago) has split in foraging through the bedding material. into structured sub-groups which allow a The bonnets failed to show either a signifimaximum scatter for feeding in the severe cant increase in mother-infant contact or environment (some, but not all, of these decrease in social grooming during deprivasub-groups appear to consist of related in- tion. Bonnet autogrooming decreased, and dividuals) ; and (2) grooming, juvenile foraging showed a brief increase. Strangely, play, and total fighting are much less fre- the bonnet infants showed a slight increase in social play during the hunger day. quent than on Cay0 Santiago. Recognizing the limitations of theorizing Hall ( ' 6 3 ) reports on the behavior of chacma baboons (Papio ursinus ) stranded from the small sample presented above, the on an island in Lake Kariba, Southern Rho- behavioral reaction of Old World monkeys desia. These animals were subjected to a to food shortage (with some exceptions as severe shortage of vegetation food items. seen from the data on bonnet macaques) Hall noted the following: ( 1 ) one or more seems generally characterized by decreasof the groups may have split up into small ing frequencies of almost all social activforaging parties of normally onc, two, or ities (with at least some species showing three animals; ( 2 ) almost all of the mon- a greater drop in interactions between nonkeys' activity was directed at food-seeking: related animals than between relatives), ( 3 ) not a single female in full- or partial- and a concomitant increase in food-seeking estrus was observed, although females in activities. BEHAVIORAL RESPONSES TO FOOD SHORTAGE ACKNOWLEDGMENTS The author was supported while on Cay0 Santiago by NSF grant GS-1777 to Dr. Donald Stone Sade, Northwestern University. Thanks go to Dr. Sade €or his critical review of the manuscript; Dr. R. E. Myers, Chief, Laboratory of Perinatal Physiology, San Juan, Puerto Rico. for permission to work on Cay0 Santiago; and Dr. John Morrison for his kind permission to use unpublished material from Desecheo Island. LITERATURE CITED Altmann, S. A. 1962 A field study of the sociobiology of rhesus monkeys, Macaca mzrlatta. Ann. N. Y. Acad. Sci., 102: 338-435. Carpenter, C. R. 1942 Sexual behavior of freeranging rhesus monkeys (Macaca mulatta): Specimens, procedures and behavorial characteristics of estrus. J. Comp. Psych., 33: 113-142. Hall, K. R. L. 1963 Variations in the ecology of the chacma baboon. In: The Primates. J. 27 1 Napier and N. A. Barnicot, eds. The Zoological Society of London, pp. 1-28. Heatwole, H., D. S. Sade and R. W. Hildreth 1963 Herpetogeography of Puerto Rico. I. Herpetofauna of Cay0 Santiago and Cay0 Batata. Carib. J. Sci., 3: 1-5. Koford, C. B. 1965 Population dynamics of rhesus monkeys on Cay0 Santiago. In: Primate Behavior: Field Studies of Monkeys and Apes. I. DeVore, ed. Holt, Rinehart and Winston, New York, pp. 160-174. Morrison, J. (n.d.) Report on present conditions of rhesus monkeys in Group H at Desecheo Island and Cayo Santiago, May 27-31, 1968. Unpub. ms. Rosenblum, L. A., I. C. Kaufman and A. J. Stynes 1969 Interspecific variations in the effects of hunger on diurnally varying behavior elements i n macaques. Brain Behav. Evol., 2: 119-131. Sade, D. S. 1965 Some aspects of parent-offspring and sibling relations i n a group of rhesus monkeys, with a discussion of grooming. Am. J. Phys. Anthrop., 23: 1-1s. 1966 Ontogeny of social relations i n a group of free-ranging rhesus monkeys (Macaca mulatta Zimmerman). Unpub. Ph.D. dissertation. University of California, Berkeley.