Behavior of an All-juvenile Group of Rhesus Monkeys JAMES LOY AND KENT LOY Caribbean Primate Research Center,' School of Medicine, University of Puerto Rico, Repto. Univ., Culle 1 0 , Sun German, Puerto Rico 00753 K E Y WORDS Macaca mulatta ganization . Matriline. . Juvenility . Behavior . Or- ABSTRACT The behavior and organization of a group of 33 juvenile rhesus monkeys removed from a single freeranging group were found to parallel closely intact rhesus social groups. Dominance ranking among the juveniles was 95% predictable from prior knowledge of mothers' ranks. Grooming, sitting close together, sitting touching, and play were found to occur much more frequently between matrilineal kin than expected from random selection of partner. Some estrous behavior was observed, including two instances of peri-menstrual estrus. One lengthy homosexual interaction between two two-year-old females was observed. The basis of much of the behavior and organization shown by the juveniles was found to be the matrilineal kin bonds brought from their natal group. Among Cercopithecoid monkeys, the developmental period occurring between nutritional weaning and sexual maturity is commonly referred to as the juvenile stage. During this period, individuals have been described as expanding their social relationships, becoming increasingly integrated into the social group, and learning adult behavior patterns (Jay, '65). Although recognized as a period of critical importance to individual development and social organization, relatively little detailed information is available regarding juvenility in monkeys. This paper reports on the behavior of a unique assortment of juvenile rhesus monkeys (Macaca mulatta). Thirty-three yearling and two-year-old monkeys were removed from a single free-ranging social group on Cay0 Santiago and observed in a corral at the La Parguera, Puerto Rico primate colony. Thus, the interactions within a single (though broad) ontogenetic stage and among animals from a natural social unit were observed away from the direct influences of animals of other developmental stages. The value of the study group was greatly increased by the availability of information regarding the behavior and organization of their natal group on Cay0 Santiago. These data were gathered in 1968-1969 by Dr. E. A. Missakian, who kindly provided unpublished information. AM. J. PHYS. ANTHROP.,40: 83-96. The unique history and composition of our study group allowed us to test current theories regarding the predictability of dominance relationships, and to analyze juvenile grooming, play and sexual behavior. METHODS Animals On November 18, 1971, 21 rhesus monkeys (1 1 males and 10 females) were transported to La Parguera from the Cay0 Santiago, Puerto Rico primate colony. All of these monkeys were from Cay0 Santiago Group A, and all had been born during the 1970 birth season. (On Cay0 Santiago, the birth season extends from January through mid-July, with March and April as the peak birth months [Koford, '651 .) These 21 animals were placed in a one-half acre, outdoor corral and observations were begun on November 19, 1971. On December 3, 1971, 12 more monkeys from Cay0 Santiago Group A arrived at La Parguera and were introduced into the corral with the earlier arrivals. Six of the new monkeys were males and six were females, and all were born during the 1969 birth season. The monkeys were individually marked 1 The Caribbean Primate Research Center was supported by contract NIH-71-2003 from the NINDS, PHS, HEW until June 30, 1972. Since July 1, 1972, the Center h a s been supported by contract NIH-DRR-71-2003 from the Division of Research Resources of NIH, PHS, HEW. 83 84 JAMES LOY AND KENT LOY by chest and thigh tattoos, and ear-notches. Matrilineal kin of all the study group animals were known from census records at Cay0 Santiago. Ten of the two-year-olds had siblings in the yearling age-class. Ob- servations on the group of 33 were begun on December 6, 1971. Figure 1 presents the study group genealogies. During the study the monkeys fed ad libitum on either Gravy Train dog chow or YEAR OF BIRTH '52 '63 '64 '6,s '66 '67 '66 '6F '60 HlOH ' ' ' ' 'bl 'b,t '68 '6? '66 '66 'qO!* '6.9 '7.0 I I Y 2 4 a Q sa W z W Q LOW sdss Fig. 1 Study Group Matrilines. This figure shows the genealogies present during Dr. E. A. Missakian's study of Cay0 Santiago Group A in 1968-1969. The relative rankings of the genealogies were presented in Missakian ('72). Only those monkeys born in 1969 and 1970 were in the present study group. (d) and (r) indicate respectively animals that had died or had been removed from Group A prior to the removal of the entire group from Cayo Santiago. BEHAVIOR OF JUVENILE RHESUS MONKEYS Wayne Monkey Diet. Bananas and oranges were occasionally provided. Water was available at all times. OBSERVATIONS The juveniles were typically observed on Monday-Friday mornings. Between November 19th and December 3, 1971, 22.2 hours of observation were made on the group of 21 yearlings. Between December 6, 1971, and January 21, 1972, 95.3 hours were spent watching the entire group of 33. Thus the yearlings were observed for 117.5 hours and the two-year-olds for 95.3 hours. On January 25, 1972, the group was divided into two parts of approximately equal size and composition preparatory to a longterm study of the effects of castration and ovariectomy. (AU animals were intact during the present study.) While initially shy of the observers, the monkeys, already habituated on Cay0 Santiago to human presence, soon allowed close observation. Observations were made within the corral by both authors following a schedule which prevented an overlap of information. Prose notes were taken, and the data were categorized, tabulated and plotted on matrices at the end of each day. Numerous checks revealed a high degree of inter-observer agreement in the description and interpretation of rhesus monkey behavior. Positive identification of individuals was made in all interactions reported in this paper unless specifically noted otherwise. Visibility within the corral was excellent due to a lack of cover and obstructions. The observers continually scanned the study group for social interactions, rather than focusing on individual monkeys for long periods. It is our assumption that the differences in total observation time for individual monkeys were negligible. However, since this assumption was not tested, we will limit ourselves in this paper to informal inferences from the observed behaviors. RESULTS Dominance relationships Both Kawai (’65) and Kawamura (‘65), studying different groups of Japanese macaques (Macaca fuscata), concluded that the position of an individual in a hierarchy of agonistic dominance was strongly 85 influenced by the ranking of that monkey’s mother in relation to other adult females. Kawai theorized that offspring-rank is determined by mother’s-rank via the protectiveness of monkey mothers, who typically rush to the aid of squabbling offspring. Thus an infant supported in its encounters with other monkeys by a high-ranking mother would have a high “dependent rank.” A second type of rank called “basic rank” was also hypothesized by Kawai. “Basic r a n k is the order established when animals interact aggressively on a one-toone basis, with the outcome of the encounter being unaffected by other members of the society. Basic rank was viewed as based on physical strength and size, as well as upon kinship in some unknown fashion. Sade (’67) demonstrated that among freeranging rhesus monkeys, the rank of offspring among their peers closely parallels the rankings of their respective mothers. Marsden (‘68) reported that in captive rhesus monkeys, induced changes in the adult female dominance hierarchy produced rises or drops in offspring-rank paralleling the changes in the mother’s rank. Sade (‘67) presented some data indicating that the rank of offspring persists even after the loss of the mother. He suggested that dominance hierarchies are established early in life and persist for several years “either because the first few fights set precedents which cannot be easily broken, or because the offspring continue to associate with their mothers or older monkeys who rank near their mothers, or because of both reasons” (’67:113). The present study group provided a unique test of the relationship of motherrank to offspring-rank. During her 19681969 study, Missakian determined the rank ordering of the adult female genealogy heads for Cay0 Santiago Group A (Missakian, ’72, and personal communication). Using her data, we were able to predict the rank orders with the two ageclasses represented in our study group. Missakian’s genealogical rankings are shown in figure 1. Dominance ranks for the study group monkeys were based on observations of the following types of interaotions. 1. Fights. A fight is an attack of any intensity followed by a flight of any intensity. This definition follows Sade (’67), as 86 JAMES LOY AND KENT LOY does our catalog of attack and flight behavior patterns, with two exceptions. In the present study i t was not considered a definite indication of subordinate rank when an attacked monkey either “glanced away” from or “presented to its attacker. In interactions of these sorts, the attacking monkey was usually awarded a “probable” win. 2. Unprovoked submissive gestures. For example, it was considered a good indication of subordinate rank if monkey “A” grimaced at monkey “B”, who was walking by, without overtly threatening the first animal. 3. Displacements. One animal moved away from the approach of another prior to any noticeable directed threat from the latter monkey (spatial equilibration; Chance, ’56). Interaction type 2 was considered as strong an indicator of rank as type 1, and the totals for both are combined in figure 2 under the category “fights.” Type 3 was considered to be a slightly less reliable measure of dominance, since it depended upon the observers’ interpretation of the motivation prompting movement. Occasionally, some confusion would arise as to the winner of an agonistic interaction. Often when this occurred, the interaction was scored as a “probable” win for one monkey or the other, and was also considered a less than completely reliable indicator of dominance. Only dyadic encounters were utilized in determining hierarchial positions. Testing the predictability of offspring-rank During the study, 1303 dominance interactions were recorded (fig. 2). Of that number, 727 were “within age-class” interactions (either between yearlings or between two-year-olds). Of the “within ageclass” dominance interactions, 686 (94.4%) featured as the dominant animal that monkey predicted from mothers’ ranks to be higher ranking (table 1). Looked at differently, of a total of 276 possible “within age-class’’ dyadic dominance relationships (e.g., A is dominant to B), data were gathered on 229 dyads. Of those 229 dyads, 219 relationships (95.6%) were in the order predicted from knowledge of the mothers’ ranks (table 2). The dominance hierarchies within the two age-classes were thus highly predictable from knowledge of mothers’ ranks, and were also quite stable during the period immediately following removal from the natal group (in other words, immediately following the uniform removal of direct maternal influence). This suggests that for rhesus monkeys the mother’s rank is the major factor determining the offspring’s “basic r a n k within its ageclass. Sade (’67) has indicated that these hierarchies remain remarkably stable up to sexual maturity for rhesus monkey males and throughout adulthood for females. Reversals of predicted relationships Figure 2 contains the dominance interactions observed during the study. Among the two-year-olds, two unpredicted dominance relationships were discovered. Female 260 out-ranked female 277, and female 235 out-ranked male 241. Eight unpredicted relationships were found among the yearlings: female 400 over male 355; female 331 over male 383; female 388 over male 383; male 330 over male 383; male 330 over male 301; male 330 over female 388; male 335 over male 383; and male 301 probably dominant to female 367. There were other unpredicted dominance encounters not included in the above relationships. (1) Female 386 possibly displaced female 366 once. However, 366 defeated 386 definitely four times, and “probably” three other times, and 366 was considered dominant. (2) Female 386 defeated male 399 once. However, 399 defeated 386 fourteen times (plus 1 “probable” and 1 displacement) and 399 was considered the dominant. (3) The relationship between females 333 and 386 was unclear. Each won one fight over the other. (4) Likewise, the ranking between females 333 and 354 was unclear, since each displaced the other once. ( 5 ) Although male 330 possibly displaced female 366 once, 366 “probably defeated 330 once. 366 was considered the probable dominant monkey. (6) Female 379 “probably” defeated male 383 once, but 383 definitely defeated 379 once and 383 was considered the dominant animal. When the unpredicted relationships are totalled by sex, it is found that in two cases a male unpredictably out-ranked a female, ” BEHAVIOR O F JUVENILE RHESUS MONKEYS a w (I) 0 J n A a W I 0 t 87 JAMES LOY AND KENT LOY 88 TABLE 1 Dominance predictability: Observed interactions Fights 2-year-olds Yearlings Combined 1 Displacements 153 366 519 Probables 13 47 60 62 86 148 Total Unpredicted 228 499 727 Predicted % % 4.8 6.0 5.6 95.2 94.0 94.4 Within ageclass interactions. TABLE 2 Dominance predictability: Observed dominance relationships Possible dyads 2-year-olds Yearlings Combined 1 Observed relationships 66 210 2 76 61 168 229 Unpredic ted 1 Predicted % % 3.3 4.8 4.4 96.7 95.2 95.6 Within ageclass relationships. in three cases a male out-ranked another male, in one case a female out-ranked another female, and in four cases a female unpredictably out-ranked a male. ity of rising or falling in rank remains unclear. Sibling dominance Dominance relationships were observed Reversals and genealogy for nine of the ten sibling pairs in the Two instances were recorded of all mem- group. In seven of these nine, the twobers of a genealogy attaining higher domi- year-old monkey was the dominant (1 case nance than predicted. (1) Female 260 was of male over female, 1 case of male over dominant to female 277,and 260s yearling male, 3 cases of female over male, and 2 sister 400 was likewise unpredictably dom- cases of female over female). inant to 277’s yearling brother 355. It is Yearling male 389 had risen over his possible that since Missakian’s study there two-year-old brother 230 by the end of the has been a rise in rank of LH (mother of study, although 230 was initially the dom260 and 400)over GS (mother of 277,355). inant. Very few instances of younger Such a rise would explain the dominance brothers rising in rank over older brothers alignment observed in the study group. have been reported for rhesus monkeys. (2) Several members of the 090 genealogy Sade (‘72) describes two such cases, and rose in rank. Female 235 was unpredict- Missakian (‘72)reports on the rise of a ably dominant to male 241 (037 geneal- five-year-old male over both his ten-yearogy). 235’s yearling brother 330 was un- old and six-year-old brothers. To our knowlpredictably dominant to his cousins, male edge, 389 is the youngest male known to 301 and female 388,and to male 383.Male have risen in rank over an older brother. Yearling female 400 was observed to 301 was dominant to female 367, and female 388 was dominant to male 383. It cause her two-year-old sister 260 to cower is probable, but not demonstrable, that the a few days after the study group was rank changes of any of these monkeys may formed. This was the only dominance interhave affected those of their relatives. action observed between 400 and 260. The Four yearlings were unpredictably dom- rise of 400 over 260 was not unexpected. inant to yearling male 383, who had no Sade (‘72) reports that among female rhematrilineal kin in the study group. Six re- sus siblings, rank order is inversely related versals were recorded in which the unpre- to age by the time all of the monkeys are dictably subordinate monkey had three adults. relatives in the group. The relationship beInter-age-class dominance relationships tween the numbers of relatives present, or Dominance relationships between yearthe absence of relatives, and the probabil- BEHAVIOR OF JUVENILE RHESUS MONKEYS lings and two-year-olds were not predictable, since the rule that “mother’s rank determines rank of offspring” is applicable only within age-classes. However, certain trends regarding inter-age-class dominance were discovered. Observed dominance interactions between yearlings and two-year-olds totalled 576. The older monkeys defeated, probably defeated, or displaced the yearlings in question 334 times. In general, two-yearold monkeys tended to be able to (1) defeat yearlings in all genealogies ranked below their own, (2) occasionally defeat a yearling of their own genealogy whose mother out-ranked their own, and (3) occasionally defeat a yearling belonging to the genealogy ranked immediately above their own (fig. 2). These were only trends, however, and individual variability was great. Yearlings tended to be able to defeat two-year-olds belonging to genealogies more than one rank below their own genealogy. Again, individual variability in yearling dominance over two-year-olds was great. Four yearlings were observed to defeat two-year-olds in the genealogy ranked immediately above their own. Female 400 repeatedly defeated female 277, who was also unpredictably defeated by 400’s older sister 260. Yearling female 388 and yearling males 301 and 330 (all cousins) all out-ranked two-year-old male 241 in the genealogy above their own. Female 235 (sister of 330 and cousin of 301 and 388) also unpredictably defeated male 241. It is conceivable that the yearlings’ ability to defeat the unrelated two-year-olds in the four intances above was related to the unpredicted rise in dominance of their twoyear-old relatives. Grooming Grooming and kinship During the study, 1753 grooming episodes were seen (fig. 3). Of that total, 323 episodes (18.4% ) were self-grooming and 1430 were inter-monkey grooming. Grooming between matrilineally related monkeys (who comprised 6.6% of the total pair combinations of the group) accounted for 491 episodes, or 34.3% of total inter-anima1 grooming. Thus related monkeys were involved in grooming at a frequency which was over five times greater than that expected from random selection of grooming partner. Sade (‘65) reported that within a 89 free-ranging rhesus group on Cay0 Santiago, related pairs (15% of the total possible grooming combinations) accounted for 62% to 64% of observed grooming episodes. The fact that the ratios between percentage of related pairs and percentage of related grooming are remarkably similar for the present study (0.19) and for Sade’s (’65) study (0.24) strongly indicates the normalcy of the grooming behavior in the corral group, and suggests as well the possibility of a mathematical constant between the number of related monkeys in a group and the amount of inter-relative grooming which will occur. Grooming and dominance Mid-ranking and high-ranking monkeys tended to receive more grooming than they gave to other animals. This trend was reversed among the low-ranking members of the group (table 3). In his review of grooming behavior, Sparks (‘67) mentions evidence of the inverse relationship between dominance and grooming in several primate species. N u m b e r of grooming partners It has been suggested that the juvenile period is characterized by an increase in the number of social relationships (Jay, ’65). We decided to utilize grooming data to determine whether or not the two ageclasses represented in our study group differed in the number of social contacts per monkey. Sade (‘65:5) has argued that grooming is the single activity “best suited for demonstrating relations between animals of different sex and ages.” It was found that the two-year-old monkeys were grooming and being groomed by a larger number of individual monkeys than were the yearlings (table 4). There seemed to be no difference between males and females (age-classes lumped) in the mean number of monkeys from whom they received grooming. However, two-year-old females tended to have slightly more individual groomers than like-aged males (table 4). Females groomed greater numbers of other monkeys than males (ageclasses lumped). This trend occurred in both age groups (table 4). Measures of spatial relations Two measures of inter-monkey spacing were used to determine further the strength 90 JAMES LOY AND KENT LOY GROOMEE 266 370 334 246 381 387 37I 277 316 260 400 366 272 o= 377 230 W 369 P 399 0 366 383 g f9 231 331 273 333 242 367 241 314 301 388 231 330 379 335 Fig. 3 Grooming Interactions. Matrix cells contain the totals of observed grooming episodes. Interactions between matrilineal relatives are shown within the heavy black lines. Episodes of self-grooming appear along the diagonal. of the matrilineal kinship bonds. Animals were recorded “sitting close” (i.e., close enough to touch one another, but not in actual physical contact) and “sitting touching.” Both are non-agonistic activities and presumably reflect either a mutual attraction between animals or else a oneway attraction coupled with tolerance for the presence of the attracted monkeys. Sitting touching was exclusive of other activities, such as grooming, which by definition involve physical contact between animals. A total of 658 “sitting close” dyads were recorded. (In those occasional instances in which a large number of monkeys sat close, the clusters were broken down into dyads for analysis.) Of that total, 224 dyads (34.0% ) involved related monkeys. This figure is over five times greater than expected, based on random selection of “sitting close” partners. Observed “sitting touching” dyads totalled 359. Of that number, 160 dyads (44.6% ) involved related animals. This figure is 6.8 times larger than expected by chance. BEHAVIOR OF JUVENILE RHESUS MONKEYS TABLE 3 Grooming and dominance ~~~~ ~ Dominance ranks 1 Episodes groomee High Mid Low 1.14 1.30 0.67 0.88 0.77 1.50 Episodes groomer Episodes groomer Episodes groomee 1 Study group divided into thirds (the 11 highest-ranking, the 11 mid-ranking, and the 11 lowest-ranking monkeys). TABLE 4 Grooming relationships Mean no. of indi- Mean no. of individuals who viduals groomed groomed subject by subject Males Females Sexes combined 2-year-olds 12.7 14.2 13.4 10.8 14.3 12.6 Males Females Sexes combined Yearlings 9.2 8.6 8.9 7.5 11.5 9.4 (ages combined) All males 10.4 8.7 (ages combined) All females 10.7 12.6 Play Play and kinship Social play was frequently observed among the juveniles. Play consisted primarily of mutual chasing, wrestling, and mouthing (see Southwick et al., ’65). Play typically appeared to be a pleasurable experience for all participants, but in several instances, vigorous wrestling erupted into aggression with the dominant monkey biting and/or pursuing the subordinate. Loizos (‘67) reviewed play among primates, including theories regarding the nature and function of play. We will direct our attention to the frequency of play in different juvenile age-sex classes and to the selection of play partners. A total of 479 play dyads were recorded in the study. Play occurred most frequently between pairs of monkeys, although larger play groups were not unusual. Play inter- 91 actions involving more than two monkeys were divided into multiple dyads for analysis. Eighteen play groups of three animals and six groups of four animals were recorded. These observations accounted respectively for 11.3% and 7.5% of the total play dyads. Related monkeys were involved in 127 dyads (26.5% of the total). There is little doubt that related monkeys were seeking each other out as play partners, since the observed frequency is slightly over four times greater than the figure expected from random selection of partners. The remaining 352 dyads (73.5% of the total) involved unrelated pairs. Age-sex differences in play Table 5 lists the raw frequencies of play interactions by the age-sex classes involved. Even assuming some differential observation, certain trends are clear. The age-sex class most active in play was the yearling males, followed by two-year-old males and yearling females, with the two-year-old females being least playful. Lindburg (’71) notes that among wild rhesus monkeys in India, older juvenile females play much less than like-aged males. The combined males showed a much higher mean frequency of play (4 1.7 episodeslmonkey) than combined females (1 5.6 episodes/ monkey). Combined yearlings showed more play (mean of 37.9 episodeslmonkey) than combined two-year-olds (1 3.6 episodes/ monkey). Yearling males and females played with all available age-sex classes, Two-year-old males played with all age-sex classes except two-year-old females, and the latter played only with yearlings of both sexes (table 5). Observations following the study period have shown that play does occur, although apparently rarely, between twoyear-old males and females, and among two-year-old females. Sexual behavior Single mounts Although single mounts are often dominance (rather than sexually) motivated, they are included under the heading “sexual behavior” since the motor patterns involved are essentially sexual in derivation. Single mounts occurred between almost all age-sex classes. Two-year-old females 92 JAMES LOY AND KENT LOY TABLE 5 Play Two year male Two year female Yearling male Yearling female Total 1 Mean play episodes/ monkey 120 20.0 15 43 7.2 65 589 53.5 61 * 206 20.6 I 2 year female - 0 Yearling male - - Yearling female - - 28 1 212" - 1 I 1 Dyads involving animals of like agesex class contributed t w o episodes to the class total. Those subtotals which were doubled in calculating the grand total for the category are marked with an asterisk. Class totals are summed left to right for two-year-old males, down and to the right for two-year-old females and yearling males, and down for yearling females. were not seen to mount yearling females, and yearling females were not observed mounting males of any age. Of the 230 total single mounts observed, 2 13 were between monkeys whose dominance relationships were known. Of the latter total, 183 mounts (85.9%) featured the dominant monkey as the mounter. Mounts between related monkeys accounted for 12.6% of the total observed single mounts. Estrous behavior Carpenter ('42) was the first to describe the behavioral characteristics of estrus in free-ranging rhesus' monkeys. Our behavioral definititions closely follow Carpenter, plus the few refinements presented in another publication (Loy, '71). A seriesmounting is equivalent to a mating, and is defined as at least two mounts, closely spaced temporally, and connected by grooming, sitting close together or sitting touching. Copulation refers to a seriesmounting which ends in ejaculation. Records were kept of sexual swellings and sex skin coloration. The latter was judged by eye on a five-point scale from zero (no color) to four (bright red). Sexual swelling, color, and menarche During the study, four females, all twoyear-olds, showed sexual swelling and reddening of the sex skin (fig. 4). Undoubtedly these were the first swellings and coloration for all four monkeys. Van Wagenen ('66) has reported that these two phenom- ena are reliable immediate precursors of the first vaginal bleeding, and that in the laboratory, rhesus females reach menarche between one and one-half to two and onehalf years. Of the four study females mentioned above, three showed probable vaginal bleeding 36-43 days after their swellings were initially noticed (fig. 4). The fourth female was initially recorded as swollen only 19 days before the study ended, and no vaginal bleeding was observed. Series-mountings Twenty-two male-to-femaleseries-mountings were observed (fig. 4). Two were between two-year-old males and females, one between a two-year-old male and a yearling female, five between yearling males and females, and 14 between the same yearling male and two-year-old female pair. Average number of mounts per series was 5.6, with a range from 2 to 18. In no instance was a series-mounting observed to end in the motor patterns characteristic of ejaculation (Loy, '71), or in the female clutching reaction (Zumpe and Michael, '68). No female was seen with a vaginal plug of coagulated semen. These results were not surprising, since Conaway and Sade ('65) reported that freeranging rhesus males on Cay0 Santiago do not reach sexual maturity until about three and one-half years of age. The juvenile series-mountings were awkward and amateurish in comparison to adult behavior. Female posturing, male BEHAVIOR OF JUVENILE RHESUS MONKEYS FEMALES II IA sw 277 260 272 273 242 238 400 388 367 Fig. 4 Sexual Behavior. Data are presented for all two-year-old females and for the three yearling females who were series-mounted by males. Vertical dashed lines indicate missed days of observation. Symbol key: SW ( A ) indicates the presence of swelling of the perineal area; SM ).( indicates a day with a t least one male-to-female series-mounting; and M ( 0 ) indicates possible menstrual bleeding. Sex skin color ( S S C ) is the mean of the face, posterior thighs, and pen-anal skin. Values for SSC range from zero (no color) to two (light red). 94 JAMES LOY AND KENT LOY body orientation and grasping of the fe- in rhesus monkeys was found to be characmale’s waist and ankles were almost always terized by much play, expansion of social done correctly. In most cases, it was not interaction spheres and the development of possible to determine whether or not the skill in motor patterns (such as copulamale had an erection during a mount, al- tion) characteristic of adult monkeys. though several mounts were seen that The juveniles in the study group were obwere definitely accompanied by an erec- viously immature, physically and behaviortion. Males seemed rarely to achieve intro- ally. In many ways, however, their behavmission, and although pelvic thrusts were ior and organization paralleled those of often given, the male seemed usually to be intact rhesus groups. Data on agonistic patterns, dominance, grooming and spatial “searching” for the vagina. As noted above, yearling male 399 was relationships all bear strong resemblance seen series-mounting two-year-old female to reports from intact groups. The key to 242 on 14 occasions. These matings oc- these parallels is a behavioral base comcurred between December 28, 1971 and mon to both the study group and intact January 15, 1972, and were accompanied free-ranging social units, the matrilineal genealogy. by a peak in 242s sex skin color (fig. 4). Sade (’65, ’67, ’68, ’72) has outlined in Two examples of peri-menstrual estrus were seen (fig. 4). Male 399 series-mount- detail the interactions within a rhesus ed female 242 on January 15th, two days monkey social group, and has emphasized before 242 was seen with probable men- the importance of the mother-offspring strual bleeding. Two-year-old male 230 bond and the matrilineal genealogy to series-mounted two-year-old female 235 on rhesus behavior and organization. With the second day of probable menstruation the development of mother-infant specificon January 21st. Loy (‘70a) discussed peri- ity, the infant monkey is linked to its mamenstrual estrus among rhesus monkeys, triline and, to a large extent, its future behavior is determined. The infant develincluding its possible hormonal basis. ops close relationships with mother, sibHomosexual behavior lings and other matrilineal kin, and its Between December 23, 1971 and Jan- kin ties then serve as the basis for its inuary 11, 1972, two-year-old female 272 tegration into the remainder of the social was observed to mount two-year-old female group. The strength of the matriline is demon242 fifty-one times. Most of these mounts were in the form of series, with the indi- strated by its adaptability. Sade has shown vidual mounts separated by grooming, or its importance to a medium-sized rhesus the females sitting close together or sitting group, and Missakian (‘72) has presented in body contact. Many of the mounts were similar data for a group with over 230 posturally identical to male-to-female members. Loy (‘70b) has shown that the mounts; i.e., mounter grasps mountee’s matriline (and behavior centered on the waist with her hands and mountee’s ankles matriline) remains stable during periods or calves with her feet. However, on many of extreme food shortage. The most significant result of the pres“mounts,” 272 would sit on 242’s lower back and scoot forwards and backwards ent study was the discovery of a stable, while she (272) yawned repeatedly. The viable social organization among the imlatter type of “mount” possibly functioned mature members separated from a naturas an auto-erotic stimulus for the mounter. ally occurring rhesus monkey social group. It is of interest that the homosexual be- Upon separation from their natal group, havior described above coincided with 242s the juveniles did not fall into behavioral sex skin color peak and her most intense chaos, but rather maintained that organiheterosexual mounting activities. Beach zation and those relationships which they (‘68) has presented evidence indicating had known as a segment of the larger that in several mammalian species, an unit. This study demonstrated that the efestrous female is an effective sexual stim- fects of the matriline were strongly seen as early as the juvenile stage of developulus for female conspecifics. ment. Months after being separated from DISCUSSION their natal group, monkeys only 20-30 As reported by other workers, juvenility months of age were able to interact in an BEHAVIOR OF JUVENILE RHESUS MONKEYS orderly and predictable fashion because of genealogically determined relationships which formed the core of their interaction systems. ACKNOWLEDGMENTS The authors are indebted to Dr. Clinton H. Conaway, Scientific Director of the Caribbean Primate Research Center. Dr. Conaway originally proposed the castration-ovariectomy study for which the juvenile group was assembled. Dr. D. S. Sade kindly provided the juveniles from Cay0 Santiago. Thanks go to Dr. Elizabeth A. Missakian for her permission to use unpublished data, to Miss Jeanie DeRousseau for her assistance in securing data stored at Cay0 Santiago, and to the staff of the La Parguera Primate Colony for their assistance throughout the study. Drs. Conaway, Sade and Gershon Berkson critically reviewed the manuscript. Any errors remaining in the paper are attributable solely to the authors. LITERATURE CITED Beach, F. A. 1968 Factors involved in the control of mounting behavior by female mammals. In: Perspectives in Reproduction and Sexual Behavior. M. Diamond, ed. Indiana University Press, Bloomington, pp. 83-131. Carpenter, C. R. 1942 Sexual behavior of free ranging rhesus monkeys (Macaca mulatta). I. Specimens, procedures and behavioral characteristics of estrus. Jour. Comp. Psychol., 33: 113142. Chance, M. R. A. 1956 Social structure of a colony of Macaca mulatta. The British Journal of Animal Behaviour, 4 : 1-13. Conaway, C. H., and D. S. Sade 1965 The seasonal spermatogenic cycle in free ranging rhesus monkeys. Folia Primat., 3: 1-12. Jay, P. 1965 The common langur of north India. In: Primate Behavior: Field Studies of Monkeys and Apes. Chapter 7. I. DeVore, ed. Holt, Rinehart and Winston, Inc., New York, pp. 197-249. Kawai, M. 1965 On the system of social ranks i n a natural troop of Japanese monkeys: (I) Basic rank and dependent rank. In: Japanese Monkeys: A Collection of Translations. K. Imanishi and S. A. Altmann, eds. Published by S. A. Altmann, pp. 6 6 8 7 . Kawamura, S. 1965 Matriarchal social ranks i n the Minoo-B troop: A study of the rank system of Japanese monkeys. In: Japanese Monkeys: A Collection of Translations. K. Imanishi and S. A. 95 Altmann, eds. Published by S. A. Altmann, pp. 105-112. Koford, C. B. 1965 Population dynamics of rhesus monkeys on Cay0 Santiago. In: Primate Behavior: Field Studies of Monkeys and Apes. Chapter 5. I. DeVore, ed. Holt, Rinehart and Winston, Inc., New York, pp. 160-174. Lindburg, D. G. 1971 The rhesus monkey i n north India: An ecological and behavioral study. In: Primate Behavior: Developments in Field and Laboratory Research, Vol. 2. L. A. Rosenblum, ed. Academic Press, New York, pp. 1-106. Loizos, C. 1967 Play behavior i n higher primates: A review. In: Primate Ethology. D. Morns, ed. Aldine, Chicago, pp. 176-218. Loy, J. 1970a Pen-menstrual sexual behavior among rhesus monkeys. Folia Primat., 13: 2 8 6 297. 1970b Behavioral responses of free-ranging rhesus monkeys to food shortage. Am. J. Phys. Anthrop., 33: 263-271. - 1971 Estrous behavior of free-ranging rhesus monkeys (Macaca mulatta). Primates, 12: 1-31. Marsden, H. M. 1968 Agonistic behaviour of young rhesus monkeys after changes induced i n social rank of their mothers. Animal Behaviour, 16: 38-44. Missakian, E. A. 1972 Genealogical and crossgenealogical dominance relations i n a group of free-ranging rhesus monkeys (Macaca mulatta) on Cay0 Santiago. Primates, 13f2): 169-180. Sade, D. S. 1965 Some aspects of parent-offspring and sibling relations in a group of rhesus monkeys, with a discussion of grooming. Am. J. Phys. Anthrop., 23: 1-18. 1967 Determinants of dominance in a group of free-ranging rhesus monkeys. In: Social Communication Among Primates. S. A. Altmann, ed. The University of Chicago Press, Chicago, pp. 99-114. 1968 Inhibition of son-mother mating among free-ranging rhesus monkeys. Science and Psychoanalysis, 12: 18-38. 1972 A longitudinal study of social behavior of rhesus monkeys. In: The Functional and Evolutionary Biology of Primates. R. Tuttle, ed. Aldine-Atherton, Chicago, pp. 378-398. Southwick, C. H., M. A. Beg and M. R. Siddiqi 1965 Rhesus monkeys in north India. In: Primate Behavior: Field Studies of Monkeys and Apes. Chapter 4. I. DeVore, ed. Holt, Rinehart and Winston, Inc., New York, pp. 111-159. Sparks, J. 1967 Allogrooming in primates: a review. In: Primate Ethology. D. Morris, ed. Aldine, Chicago, pp. 148-175. Van Wagenen, G. 1966 Studies in reproduction (M,acaca mulatta). Proc. Conf. on Non-human Primate Toxicology, HEW, Washington, D.C., pp. 103-113. Zumpe, D., and R. P. Michael 1968 The clutching reaction and orgasm i n the female rhesus monkey (Macaca mulatta). Jour. Endocnn., 40: 117-123.