Chimpanzee genital swelling and its role in the pattern of sociosexual behavior.код для вставкиСкачать
American Journal of Primatology 28101-113 (1992) RESEARCH ARTICLES Chimpanzee Genital Swelling and Its Role in the Pattern of Sociosexual Behavior JANETTE WALLIS Department of Obstetrics and Gynecology, University of Oklahoma College of Medicine, Oklahoma City Behavioral observations were made on thirteen female and seven male adult group-living chimpanzees (Pun troglodytes). The behavioral data were analyzed as a function of the day of the females’ menstrual cycles to explore the possible interrelationship between genital swelling and sociosexual behavior of female chimpanzees. Copulatory behavior was confined almost entirely to the period of genital swelling and the occurrence of male-to-female genital inspection (both female- and male-initiated) was negatively correlated with the days from swelling onset, i.e., as the presumed day of ovulation approached, genital inspection decreased, while copulatory behavior increased. In addition, more females groomed their male cagemates during the luteal phase than in the follicular phase of their cycles, whereas male-to-female grooming was positively correlated with the progress of the cycle, with peaks during the time of swelling onset and menstruation. The profile of sociosexual behavior observed throughout the menstrual cycle suggested that, although chimpanzees exhibit an extended period of sexual receptivity and genital swelling, the presumed fertile period is not concealed. The role of genital swelling in chimpanzees is discussed in relation to the possible hormonal effect on female sexuality and the evolution of chimpanzee mating strategies. Q 1992 Wiley-Liss, Inc. Key words: sexual behavior, sociosexual behavior, chimpanzee, Pun troglodytes, genital inspection, genital swelling INTRODUCTION The 36 day menstrual cycle of adult female chimpanzees (Pun troglodytes) is characterized by a cyclical change in the size, shape, and turgidity of the entire anogenital region. There is an initial period of increasing genital swelling that lasts an average of 7.6 days. The period of maximal swelling may last for as much as 17 days, averaging about 10-12 days [McGinnis, 19791. The volume of interstitial fluid contained in the swelling measures almost 1.5 liters [Elder & Yerkes, Received for publication September 10, 1991; accepted April 4, 1992. Address reprint requests to Dr. Janette Wallis, Dept. of Obstetrics and Gynecology,University of Oklahoma College of Medicine, P.O. Box 26901, Oklahoma City, OK,73190. 0 1992 Wiley-Liss, Inc. 102 / Wallis 19361. A rapid decrease in the size and turgidity of the swelling occurs during the next 4 days and the area then remains completely flat for approximately 13.6 days [McGinnis, 19791. Menstruation occurs about midway through the phase of quiescence. Several profiles of chimpanzee urinary [Graham et al., 1972, 1977; Lasley et al., 1980; McArthur et al., 19811 and serum [Reyes et al., 1975; Graham, 1976; Nadler et al., 19851 hormones have been reported [see review by Graham, 19811. The early follicular phase shows high levels of FSH, which decrease gradually for several days thereafter. Estrogen rises in conjunction with the occurrence of genital swelling, peaking just prior to the peak in FSH and decreasing immediately after. A testosterone peak follows that of follicular phase estradiol [Nadler et al., 19851. The estrogen decline, accompanied by detumescence of the genital area, is followed by a rapid increase of progesterone that peaks in synchrony with a midluteal phase estrogen peak. An LH peak is associated with the last day of maximal tumescence and the resulting rapid rise in progesterone verifies that ovulation has occurred-typically on the last one or two days of full genital swelling [Graham, 1981, 19821. Observations of wild chimpanzees have noted the relationship between sexual behavior and the presence of genital swelling [van Lawick-Goodall, 1968, 1969; Nishida, 1968; Sugiyama, 1969; Tutin, 1975,19791.Studies of captive group-living chimpanzees show the same basic pattern: copulatory frequency is highest during genital swelling, although sexual behavior may occur in other reproductive phases as well [Kollar et al., 1968; Lemmon & Allen, 1978; Coe et al., 1979; Tutin & McGinnis, 1981; Wallis, 1982; Goodall, 1986; Wallis & Lemmon, 19861. Many of the early studies of sexual activity in captive chimpanzees involved pairing individuals for brief test periods throughout the female’s menstrual cycle. The data indicated a copulatory increase during maximal genital swelling [Tinklepaugh, 1933; Yerkes & Elder, 1936a,b; Fish et al., 1941; Young & Orbison, 19441. A consistent finding in these studies was that the mid-luteal peak in estrogen [Fish et al., 19411not only lacked a corresponding increase in genital swelling, but also failed to correspond with an increase in sexual behavior. It became apparent, then, that copulatory behavior is more closely correlated with the appearance of genital swelling than with high levels of estrogen, per se. Thus the inhibitory influence of progesterone during the luteal phase is similar with respect to copulation and genital swelling. Although the pair-test design allows the reliable recording of all sexual activity between a pair, there are certain drawbacks to studying chimpanzees in this way, as was well recognized by the early investigators. Yerkes and Elder [1936al noted that a multiparous female will typically present to a male with whom she is on friendly terms during periods of maximal genital swelling. During other phases, she will tend to ignore him or avoid his advances. If, however, the female is recently matured or is a stranger to the male (or if the male is hostile or aggressive), Yerkes and Elder reported that mating behavior may occur a t any time of the cycle. It is difficult to measure accurately the sexual initiative of female chimpanzees. In pair-test studies, the male may be conditioned to expect the female t o be receptive. A male chimpanzee has the estimated capacity to ejaculate four times a day [Short 19811; given only brief opportunities to do so, his sexual drive, particularly in the presence of an unresponsive female, may lead him to behave somewhat more aggressively than is normally observed in chimpanzees. Yerkes [19391, therefore, predicted that when a reliable measurement of sexual initiation is assessed, the female will tend to restrict mating to the periovulatory period, whereas Chimpanzee Genital Swelling and Behavior / 103 males-while also initiating copulations most frequently during maximal swelling-will tend to initiate outside this phase more often than females. Yerkes [19391 also predicted that as the period of genital swelling progressed (or as the time of ovulation approached) copulatory frequency would increase. Coe et al.  found peaks in copulatory activity, but the peak varied for individual females. Not until field data were analyzed did it become apparent that there may, indeed, be a change in frequency of copulation during the period of maximum genital swelling. Tutin and McGinnis  published a record of observed possessive mating for nine female chimpanzees in Gombe, compiled over several months and calibrated into one “cycle.” They found that the highest frequency of male possessive behavior occurred during the last three days of maximal genital swelling. Similar findings were reported by Goodall [19861 and Takasaki  both of whom also noted that, as genital swelling increased, females mated with adult males more than with immatures. A detailed study of sexual activity throughout the menstrual cycle has never been conducted on captive group-living chimpanzees. The purpose of this study was to test Yerkes’s  hypotheses that 1) the pattern of female initiation of sexual interaction may differ from that of males when the day of the female’s reproductive cycle is considered and 2) as the day of ovulation approaches, sexual activity will be more likely t o occur, i.e., the presumed day of ovulation will not be hidden. Data on sexual interactions, grooming, and genital inspection were inter-related with the day of the cycle to examine the relationship between sociosexual behavior and changing genital swelling levels in captive adult female chimpanzees. METHODS Subjects At the onset of the study, the subjects were 13 sexually mature, female chimpanzees, ranging in age from nine to 22 years (mean = 14.3, SD -I 4.6).They were members of three social groups, each containing at least one adult male. A total of seven males, ages nine to 20 years, was present. All animals were wild-caught as infants, captive reared, and had lived a t the Institute for Primate Studies, Norman, OK, for a t least 3 years. The groups were housed in indoorloutdoor cages (totalling 148.8 m2) and fed twice daily. Visual inspection of the genitalia allowed determination of genital swelling size, recorded daily for each female. The number 1 represented no swelling; 2, partial swelling; and 3, maximal swelling. Maximal swelling was characterized by a tight, shiny appearance; when a maximally swollen female sits and/or allows palpation of the genitalia, the skin immediately returns to full size, leaving no indention. Although some researchers insist on using many more stages when measuring level of female chimpanzee genital swelling, a simple system of only three stages was chosen for this study. It is easily replicated and provides the necessary information requested here, i.e., is swelling present? (If so, is it maximal?) Behavioral Observations Behavioral data were collected from May to August, 1980, and June to October, 1981. Only the behavioral observations of cycling females were included in these analyses; all observations of pregnant and acyclic animals were omitted. Females exhibiting the first postpartum estrous (PPE) cycle were included. The 20th day before the last day of first PPE maximal swelling was arbitrarily chosen as the point at which observational data for those subjects were included in this study. 104 / Wallis Day of Menstrual Cycle 0 2 4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4 2 6 2 8 1 3 2 3 4 3 6 2 4 6 8 101214161820 Days from Detumescence .... . .-1&1&14-12-10 DaysfromSwellingOnset -8 -6 -4 -2 0 2 4 6 8 10 1214 1618.. ....... . - 8 - 6 - 4 -2 0 2 4 6 8 1 0 1 2 1 4 1 6 1 8 . . - 1 4 - 1 2 - 1 0 - 8 - 6 - 4 -2 0 2 4 6 8 10 Fig. 1. Numbering systems of the chimpanzee menstrual cycle. See text for detailed explanation. Two or three females were observed simultaneously, if only one male was present in a group, using a multi-focal sampling technique. Subjects were chosen randomly and observed for an average of 30 minutes during afternoon sessions. Only the behavioral interactions between the focal female and male cagemates were recorded. Major behavioral categories included sexual behavior, grooming, genital inspection, and other affiliative contact, as detailed further in Wallis and Lemmon [19861. More than 700 hours of observation were made during data collection. However, this study includes only the 332.9 hours of data collected from cycling females. Method of Behavioral Analysis The behavioral data from all female subjects were later pooled by day of the menstrual cycle. Because of the high variability in cycle length and duration of swelling phases, the data were analyzed twice: 1) in relation to the day of genital swelling onset and 2) in relation to the day of detumescence (probable time of ovulation). When the data were analyzed according to swelling (S) onset, the first day of observable swelling was identified as day S(O), with the preceding days numbered negatively and succeeding days numbered positively. According to average calculations, therefore, maximal swelling usually occurred around day S(4), whereas menstruation appeared roughly ten days before swelling onset [S( - lo)]. Similarly, when the data were analyzed around the day of detumescence (D), the last day of maximal swelling was labelled D(O), etc. Using this system, day DUO) was the approximate day of menstrual onset. The large variability that necessitates two numbering systems must be considered when interpreting the results; although both systems predict when menstruation is likely to occur, it is not necessarily accurate to equate behavioral data from, for example, D(10) to that of S(- 10). Figure 1 presents a stylized version of these two numbering systems along with the traditional system of considering onset of menses as the start of the cycle. For analysis and illustration, behavioral profiles were determined by calculating the percent of the sampled subjects that exhibited each behavioral category during two day blocks of the cycle. For example, during days D(- 10) and D(-9), 11 of the 13 females were observed. Of those, five exhibited at least one femaleto-male grooming bout. Thus, the profile shows that 45% 6/11) of the sampled females were observed to groom males on day D( - 10). Due to violations of distributional assumptions, nonparametric tests were used to analyze the data. By using either day S(0) or D(O) as a midpoint and by matching the correspondingly numbered days before and after, a Wilcoxon test for correlated samples was performed on the data to test the null hypothesis that sociosexual behaviors would occur before and after swelling onset (or detumescence) with identical probability. In addition, Pearson product-moment correlation coefficients Chimpanzee Genital Swelling and Behavior / 105 were computed and tested to examine the null hypothesis that the duration of swelling (or the approach of ovulation) had no effect on the percent of females exhibiting each behavior. Finally, chi-square goodness-of-fit tests were used to test the null hypothesis of uniformity in the behavioral profiles. RESULTS Behavioral Interactions Correlated With Genital Swelling Behavioral data included analysis of 53 cycles. The average cycle length was 36.7 days. Pooling the data according to days from detumescence, genital swelling began on about day D(-15) (mean = 15.4 days before detumescence, SD +- 4.4 days) and reached maximal size on day D(-12) (mean = 11.9, SD 5 4.4). Calculated from swelling onset, maximal genital swelling occurred on about S(4) (mean = 3.6 days after swelling onset, SD ? 3.9 days). Figures 2-4 illustrate the profiles of sociosexual behavior occurring throughout the subjects’ menstrual cycles, as determined by calculating the percent of sampled females that exhibited each behavior. Median smoothing by three, as described by Tukey [19771, was performed on some profile curves. According to Tukey [p. 2181, “The purpose of the smooth is to show off general behavior,” thus better illustrating the trend of the results. Sexual behavior. Males initiated sexual interaction more often than did females. The seven males made a total of 74 copulatory attempts, with 37 (50%) resulting in successful copulations. The 13 females solicited 40 times, and succeeded 16 (40%) times. Thereewas no significant difference in the copulatory success rates as a function of gender-initiation (x2 = .68, df = 1, ns.). Sexual interactions were confined almost entirely to the follicular phase of the cycle (Fig. 2a,b). A high correlation between the days of genital swelling and the percent of subjects exhibiting both female initiated (r = 5 8 , n = 12, P<.025) and male initiated (r= .86, n = 12, P<.005)behaviors is revealed in the follicular phase data (Fig. 2b). Grooming. The 13 female subjects initiated grooming of male cagemates in a total of 77 bouts and they received grooming from 106 male-initiated bouts. Females solicited male-to-female grooming on 27 occasions, but male’ subjects never solicited female grooming. The percent of females that initiated grooming bouts with males deviated significantly from uniformity throughout the cycle (x2= 180.4, df = 22, P<.OOl, Fig. 3a). There was a larger percentage of female subjects that groomed their male cagemates during the luteal phase than in the follicular phase (t=7.5, n = l l , P<.05, Fig. 3b). Male-to-female grooming increased consistently as the cycle progressed from day S(-22) to S( 22)(r = 567, n = 23, P<.005, Fig. 3a). However, when the data were analyzed according to days from detumescence, male-to-female grooming peaked during the days of swelling onset and again around time of menstruation (Fig. 3b). Genital inspection. The profile of male-to-female genital inspections, according to both the days from swelling onset and detumescence, are illustrated in Figure 4. A chi-square test rejected the hypothesis of uniformity, indicating a nonrectangular distribution (x2= 334.6, df= 22, P<.OOl). The percentage of females receiving inspection during genital swelling was higher than before swelling onset (t = 7, n = 11, P<.05). There was a negative correlation between the incidence of genital inspections and the days from swelling onset ( r = -.58, df=9, P<.05). Days S(20) and S(22) + 106 I Wallis a SEXUAL BEHAVIOR 50 Female Initiated: (Prowptivity) 40 - 30 E fm 6- 2o 10 P f H a 9 ln - 50 40 830 20 10 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 Days from Swelling Onset b 50 I - t ! SEXUAL BEHAVIOR I Female Initiated: (Pmceptivity) Male Initiated: (Attnctivily) 20 - -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 Days from Detumescence Fig. 2. a: Profile of total copulatory attempts and successful copulations occurring throughout the menstrual cycle, according to swelling onset. b Profile of total copulatory attempts and successful copulations occurring throughout the menstrual cycle, according to days from detumescence. F = follicular phase; L = luteal phase. were excluded from this analysis because, due to their high number, they were more likely to represent post-detumescence than swelling. DISCUSSION The results of this study confirm earlier reports that chimpanzee sociosexual behavior varies across the menstrual cycle and appears to be affected by the pres- Chimpanzee Genital Swelling and Behavior I 107 GROOMING BOUTS I Female Initiated - smoothed curve a a a a a a a a a Male Initiated a a a a a 0 , I a I I t , , -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 : . 1 -2 0 , 2 4 , I 6 8 , 1 I , . . I 10 12 14 16 18 20 22 Days from Swelling Onset GROOMING BOUTS Female Initiated a a a a Male Initiated -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 Days from Detumescence Fig. 3. a: Profile of grooming bouts occurring throughout the menstrual cycle, according to days from swelling onset. Solid line represents median smoothing by three. See text for details. b Profile of grooming bouts occurring throughout the menstrual cycle, according to days from detumescence. ence of genital swelling. Social grooming was initiated by a larger percent of females in the luteal phase than in the follicular phase, whereas males appeared to initiate grooming more as the cycle progressed from luteal to follicular phases. Male-to-female genital inspections occurred more when swelling was present than 108 I Wallis I il 10 4 Male-to-Female GENITAL INSPECTIONS a IP Days from SwellingOnset v) I !$ 53 70 - Days from Detumescence Fig. 4. Profile of gential inspections occurring throughout the menstrual cycle. when it was not and were received by more females during the first few days of swelling. Whether increased attention to the genitalia a t swelling onset was due to the presence of an olfactory cue or to the novelty of the visual cue is uncertain. Sexual interactions were confined almost entirely to the follicular phase of the cycle. Both female- and male-initiated copulatory attempts were positively correlated with the day of genital swelling. This finding supports Yerkes’s  hypothesis that as ovulation approaches, sexual activity gradually increases. However, the peak in percent of subjects exhibiting these behaviors occurred slightly before the probable day of ovulation. Yerkes [19391 also hypothesized that female chimpanzees may tend to restrict mating attempts to the periovulatory period, whereas males may exhibit a less restricted pattern of attempts. Although the results of this study indicated that males initiated copulatory bouts during phases when females did not, i.e., during the luteal or early follicular phases (Fig. 2b), there was no overwhelming support for this second hypothesis. Of interest is the very restricted period of male response to female initiation; although male subjects occasionally attempted copulations with non-swollen females, they responded to female solicitations only when the female exhibited maximal genital swelling, on days D(-12) to D(0). The Function of Female Chimpanzee Genital Swelling Several authors have offered hypotheses to explain the function of genital swelling in female primates [see discussions by Harcourt, 1981; Dixson, 1983; Hrdy and Whitten, 19871: 1. Hrdy 119813 and Harcourt [19811 argued that genital swellings function to Chimpanzee Genital Swelling and Behavior / 109 increase a female’s opportunity to mate with several different males, thereby increasing the amount of total male care her infant receives. 2. Clutton-Brock and Harvey 119761argued that the swelling enhances female attractivity and generates intermale competition for access to females. This hypothesis suggests that the “best” male will win the competition and ultimately become the father of the female’s offspring. 3. Hamilton  hypothesized that genital swellings simply signal the timing of ovulation, thereby increasing paternity confidence and possible paternal investment by one or two males. These hypotheses are not mutually exclusive: the second hypothesis emphasizes the benefit to females, whereas the third emphasizes the benefit to males. However, it appears that, for chimpanzees, all three hypotheses may be valid explanations not only for the occurrence of genital swelling, but for the lengthy duration of the swelling and the corresponding increase in sexual attractivity and proceptivity. The following sections discuss evidence that genital swelling in chimpanzees is indeed associated with intermale competition and female choice as suggested in the many male, best male and ovulation advertisement hypotheses. Many-Male Hypothesis For a female chimpanzee, there is a large number of mates from which to choose because, unlike males, a female is not confined to a single territory. She may copulate with any of the males in her community or she may cross into a neighboring community and mate with any of those males [Kawanaka & Nishida, 1974; Pusey, 19791. Apparently, the only requirement for passage by females between communities is the presence of swollen genitalia; there have been no reports of attacks on immigrant females exhibiting genital swellings [Goodall, 19861. Females thus may use their genital swelling as a “passport” to cross territorial borders [Nishida, 1979; see Wallis, 19821. Whether a female chooses to transfer to another community or stay within her own, the extended phase of swelling may allow more travel time and may allow greater opportunity for female choice by providing the ability to mate with several males, essentially obscuring paternal assurance, and gaining paternal care from many males. One result of a non-investing male may be infanticide. In most cases of maleperpetrated infanticide the infanticidal male was not likely to be the father of the infant [Suzuki, 1971; Goodall, 19771. Takahata [19851 reported infanticide of infants within the group, however, the mothers had been ranging near the territorial border and may therefore have been associating with males of another community. In both intergroup and intragroup encounters, therefore, male chimpanzees may use a recent history of interaction with the female as a cue either to attack or tolerate the infant. (In fact, this recent sexual history can occur through copulations during pregnancy [see Wallis, 1982; Wallis & Lemmon, 19861.) Thus a lengthy duration of genital swelling may facilitate female transfer in chimpanzees and may reduce inbreeding [Kawanaka & Nishida, 1974; Pusey, 1979,19801 and reduce probability of infanticide [Moore, 1984; Takahata, 19851, while increasing the number of possible mates. The Best-Male Hypothesis A chimpanzee male may increase his ability to monopolize a female if he leads her to consort and copulate away from the rest of the group. Because the maintenance of a consortship requires the cooperation of the female, the male is at an 110 / Wallis advantage if he has maintained an affiliative relationship with the female. In fact, Tutin [19801 found that females prefer consort partners that have spent large amounts of time with them, grooming and sharing food during non-sexual contacts. Although there is a considerable amount of intermale competition in establishing the dominance hierarchy, there is little intracommunity aggression in chimpanzee societies. The affiliative traits of grooming and food sharing, etc., seem to be necessary for establishing and maintaining close male-male bonds, and the ability to rise subsequently in the hierarchy appears to require continued use of these behavioral characteristics [Goodall, 19861. Nishida [19831 noted that the social skills (or social intelligence) leading to successful alliances in agonistic relationships might have been favored by individual selection, as well as by kin selection. Because consort copulations tend to result in conception more often that do the other mating patterns [Tutin, 19791, it appears that having affiliative traits may serve to increase male reproductive success. Therefore, in chimpanzees, intrasexual selection and intersexual selection appear to have produced some of the same behavioral characteristics in males. If the characteristics that make a male attractive are inherited, then a female mating with attractive males will have attractive sons who will be more likely (than unattractive sons) to provide her with many grandchildren and, thus, increase her inclusive fitness [see Robinson, 19821. The lengthy duration of genital swelling in chimpanzees provides adequate time for females to experience affiliative interactions with many males and thus may have the effect of allowing females to exercise partner preference and choose the “best” male [Tutin, 19801. Ovulation Advertisement Hypothesis It has been suggested that, due to the increased period of both receptivity and the huge genital swelling, in addition to relatively high levels of testosterone in females, the genus Pan may have evolved a type of “concealed ovulation” [regarding P. paniscus, see Turke, 19841. It appears, however, that although chimpanzees exhibit an extended period of genital swelling, ovulation is not concealed. In the present study, the results were analyzed to determine if behavior during the 10-12 day maximal swelling period occurred at a constant level or if they appeared to vary in concert with follicular development. As noted earlier, both male- and female-initiated copulatory attempts increased as the days of swelling progressed. Similarly, Tutin and McGinnis [19811 found that the highest frequency of possessive behavior in wild chimpanzees occurred on the presumed day of ovulation and the two preceding days [i.e., D(-2)-D(0)1. Furthermore, whereas the early days of maximal swelling were marked by increased male-to-female genital inspection, incidence of inspection declined as the swelling phase progressed (present study). These data suggest that, although one day of maximal genital swelling may be visually indistinguishable from the next, hormonal and olfactory changes may affect the behavior of females and, to an even greater extent, the behavior of their male conspecifics. Thus, rather than concealing the occurrence of ovulation, the lengthy period of genital swelling in chimpanzees may simply increase the time of advertising the approach of ovulation. Chimpanzees live in very large territories and, when detumescent, females tend to spend time travelling with their mothers, siblings, and/or own offspring; i.e., they are not typically travelling with adult males. A lengthy period of genital swelling, therefore, is a visual indicator of sexual receptivity Chimpanzee Genital Swelling and Behavior / 111 several days before ovulation that allows females extra time for finding, approaching, and maintaining proximity to males [Nadler, 1977; Harcourt, 19811. CONCLUSIONS 1. Sexual interactions were confined almost entirely to the follicular phase of the cycle. Furthermore, as the presumed time of ovulation approached and the female genitalia became more tumescent, both proceptivity and attractivity increased. 2. Male-to-female genital inspections occurred most often around the time of genital swelling onset. As the follicular phase progressed, and genitalia became more tumescent, frequency of inspection decreased. 3. The results of this study suggest that the long period of genital swelling in female chimpanzees does not conceal ovulation, rather it serves to advertise its occurrence. ACKNOWLEDGMENTS This paper was based on a study that fulfilled, in part, the requirements for a combined doctoral degree in psychology, anthropology, and zoology from the University of Oklahoma. I wish to thank the members of my doctoral committee, Drs. O.R. Kling, J. Lancaster, D. Mock, J. Rodgers, and P. Schwagmeyer for their comments on earlier drafts of this manuscript. B.J. King, W.C. McGrew, R.D. Nadler and D.A. Collins offered additional advice. Finally, I thank the late Dr. William Burton Lemmon, who first encouraged me to pursue a career in primatology and who offered advice, guidance and, most importantly, unrestricted access to the study subjects. This paper is dedicated to his memory. REFERENCES Clutton-Brock, T.H.; Harvey, P.H. EvoluBE: PATTERNS OF BEHAVIOR. Camtionary rules and primate societies. Pp. bridge, Mass., Belknap Press, 1986. 195-237 in GROWING POINTS IN Graham, C.E. The chimpanzee: a unique ETHOLOGY. P.P.G. Bateson, R.A. Hinde, model for human reproduction. Pp. 29-38 in THE LABORATORY ANIMAL IN THE eds. London, Cambridge University Press, 1976. STUDY OF REPRODUCTION. T. Antikatzides, S. Erichsen, A. Spiegel, eds. Coe, C.L.; Connolly, A.G.; Kraemer, H.C.; Levine, S. Reproductive development and New York, Gustav, Fischer, and Verlag, behavior of captive female chimpanzees. 1976. Graham, C.E. Menstrual cycle of the great PRIMATES 20~571-582, 1979. apes. Pp. 1-43 in REPRODUCTIVE BIOLDixson, A.F. Observations on the evolution OGY OF THE GREAT APES: COMPARAand behavioral significance of “sexual TIVE BIOMEDICAL PERSPECTIVES. skin” in female primates. ADVANCES IN THE STUDY OF BEHAVIOUR 13:63C.E. Graham, ed. New York, Academic Press, 1981. 106,1983. Elder, J.H.; Yerkes, R.M. The sexual cycle of Graham, C.E. Ovulation time: a factor in ape fertility assessment. AMERICAN the chimpanzee. ANATOMICAL RECORD JOURNAL OF PRIMATOLOGY, SUP671119-143, 1936. PLEMENT 151-55,1982. Fish, W.R.; Young, W.C.; Dorfman, R.I. Excretion of estrogenic and androgenic sub- Graham, C.E.; Collins, D.C.; Robinson, H.; stances by female and male chimpanzees Preedy, J.R.K. Urinary levels of estrogens with known mating behavior records. ENand pregnanediol and plasma levels of proDOCRINOLOGY 28:585-592,1941. gesterone during the menstrual cycle of the Goodall, J. Infant killing and cannibalism in chimpanzee: relationship to the sexual free-living chimpanzees. FOLIA PRIMAswelling. ENDOCRINOLOGY 91:13-24, 1972. TOLOGICA 28:259-282, 1977. Goodall, J. THE CHIMPANZEES OF GOM- Graham, C.E.; Warner, H.; Misener, J.; Col- 112 / Wallis lins, D.C.; Preedy, J.R.K. The association between basal body temperature, sexual swelling and urinary gonadal hormone levels in the menstrual cycle of the chimpanzee. JOURNAL OF REPRODUCTION AND FERTILITY 50~23-28,1977. Hamilton, W.J. Significance of paternal investment by primates to the evolution of male-female associations. Pp. 309-335 in PRIMATE PATERNALISM. D.M. Taub, ed. New York, Van Nostrand Reinhold, 1984. Harcourt, A.H. Inter-male competition and the reproductive behavior of the great apes. Pp. 301-318 in REPRODUCTIVE BIOLOGY OF THE GREAT APES: COMPARATIVE BIOMEDICAL PERSPECTIVES. C.E. Graham, ed. New York, Academic Press, 1981. Hrdy, S.B. THE WOMAN THAT NEVER EVOLVED. Cambridge, Harvard University Press, 1981. Hrdy, S.B.; Whitten, P.L. Patterning of sexual activity. Pp. 370-384 in PRIMATE SOCIETIES. B.B. Smuts, D.L. Cheney, R.M. Seyfarth, R.W. Wrangham, T.T. Struhsaker, eds. Chicago, University of Chicago Press, 1987. Kawanaka, K.; Nishida, T. Recent advances in the study of inter-unit-group relationships and social structure among chimpanzees of the Mahali Mountains. Paper presentation at 5th CONGRESS OF INTERNATIONAL PRIMATOLOGICALSOCIETY, Nagoya, Japan, 1974. Kollar, E.J.; Beckwith, W.C.; Edgerton, R.B. Sexual behavior of the ARL colony chimpanzees. JOURNAL OF NERVOUS AND MENTAL DISORDERS 1471444-459, 1968. Lasley, B.L.; Hodges, J.K.; Czekala, N.M. Monitoring female reproductive cycles of the great apes and other primate species by determination of estrogen and LH in small volumes of urine. JOURNAL OF REPRODUCTION AND FERTILITY, SUPPLEMENT 28:121-129, 1980. Lemmon, W.B.; Allen, M.L. Continual sexual receptivity in the female chimpanzee (Pun troglodytes). FOLIA PRIMATOLOGICA 30:80-88, 1978. McArthur, J.W.; Beitins, I.Z.; Gorman, A.; Collins, D.C.; Preedy, J.R.K.; Graham, C.E. The interrelationship between sex skin swelling and the urinary excretion of LH, estrone, and pregnanediol by the cycling female chimpanzee. AMERICAN JOURNAL OF PRIMATOLOGY 1:265270,1981. McGinnis, P.R. Sexual behavior in freeliving chimpanzees: consort relationships, Pp. 429-439 in THE GREAT APES. D.A. Hamburg, E.R. McCown, eds. Menlo Park, California, Benjamin Cummings, 1979. Moore, J. Female transfer in primates. INTERNATIONAL JOURNAL OF PRIMATOLOGY 5:537-589,1984. Nadler, R.D. Sexual behavior of the chimpanzee in relation to the gorilla and orangutan. Pp. 191-206 in PROGRESS IN APE RESEARCH. G.H. Bourne, ed. New York, Academic Press, 1977. Nadler, R.D.; Graham, C.E.; Gosselin, R.E.; Collins, D.C. Serum levels of gonadotropins and gonadal steroids, including testosterone, during the menstrual cycle of the chimpanzee. AMERICAN JOURNAL OF PRIMATOLOGY 91273-284,1985. Nishida, T. The social group of wild chimpanzees in the Mahali Mountains. PRIMATES 9:167-224,1968. Nishida, T. The social structure of chimpanzees of the Mahale Mountains. Pp. 73-121 in THE GREAT APES. D.A. Hamburg; E.R. McCown, eds. Menlo Park, California, Benjamin Cummings, 1979. Nishida, T. Alpha status and agonistic alliance in wild chimpanzees (Pun troglodytes schweinfurthii). PRIMATES 24:318-336, 1983. Pusey, A.E. Inter-community transfer of chimpanzees in Gombe National Park, Pp. 465-479 in THE GREAT APES. D.A. Hamburg, E.R. McCown, eds. Menlo Park, California, Benjamin Cummings, 1979. Pusey, A.E. Inbreeding avoidance in chimpanzees. ANIMAL BEHAVIOUR 28543552,1980. Reyes, F.I.; Winter, J.S.D.; Faiman, C.; Hobson, W.C. Serial serum levels of gonadotropins, prolactin, and sex steroids in the non-pregnant and pregnant chimpanzee. ENDOCRINOLOGY 96:1447-1455, 1975. Robinson, J.G. Intrasexual competition and mate choice in primates. AMERICAN JOURNAL OF PRIMATOLOGY, SUPPLEMENT 1:131-144,1982. Short, R.V. Sexual selection in man and the great apes. Pp. 319-341 in REPRODUCTIVE BIOLOGY OF THE GREAT APES: COMPARATIVE AND BIOMEDICAL PERSPECTIVES. C.E. Graham, ed. New York, Academic Press, 1981. Sugiyama, Y. Social behavior of chimpanzees in the Budongo Forest, Uganda. PRIMATES 101197-225,1969. Suzuki A. Carnivority and cannibalism observed among forest-living chimpanzees. JOURNAL OF ANTHROPOL SOCIOLOGICA NIPPON 79:30-48,1971. Takahata, Y. Adult male chimpanzees kill and eat a male newborn infant: newly observed intratroup infanticide and cannibalism in Mahale National Park, Tanzania. Chimpanzee Genital Swelling and Behavior / 113 FOLIA PRIMATOLOGICA 44161-170, Graham, ed. New York, Academic Press, 1981. 1985. Takasaki, H. Female life history and mating van Lawick-Goodall, J. The behaviour of free-living chimpanzees in the Gombe patterns among the M group chimpanzees Stream area. ANIMAL BEHAVIOUR of the Mahale National Park, Tanzania. MONOGRAPHS 1:161-311,1968. PRIMATES 26:121-129,1985. Tinklepaugh, O.L. Sex cycle and other cyclic van Lawick-Goodall, J. Some aspects of preproductive behavior in a group of wild phenomena in a chimpanzee during adolescence, maturity, and pregnancy. JOURchimpanzees, Pan troglodytes schweinNAL OF MORPHOLOGY 54521-547, furthii, at the Gombe Stream Reserve, Tanzania, East Africa. JOURNAL OF RE1933. PRODUCTION AND FERTILITY, SUPTukey, J.W. EXPLORATORY DATA PLEMENT 6~353-355,1969. ANALYSIS. Reading, Mass., Addison-WesWallis, J. Sexual behavior of captive chimley, 1977. panzees (Pan troglodytes):pregnant vs. cyTurke, P.W. Effects of ovulatory concealcling females. AMERICAN JOURNAL OF ment and synchrony on protohominid matPRIMATOLOGY 3:77-88,1982. ing systems and parental roles. ETHOLOGY AND SOCIOBIOLOGY 5~33-44, Wallis, J.; Lemmon, W.B. Social behavior and genital swelling in pregnant chimpan1984. zees (Pan troglodytes). AMERICAN JOURTutin, C.E.G. Exceptions to promiscuity in a NAL OF PRIMATOLOGY 10:171-183, feral chimpanzee community. Pp. 445-449 1986. in CONTEMPORARY PRIMATOLOGY. S. Kondo, M. Kawai, A. Ehara, eds. Basel, Yerkes, R.M. Sexual behavior in the chimpanzee. HUMAN BIOLOGY 11:78-111, Karger, 1975. 1939. Tutin, C.E.G. Mating patterns and reproductive strategies in a community of wild Yerkes, R.M.; Elder, J.H. The sexual and rechimpanzees (Pan troglodytes schweinproductive cycles of chimpanzees. PROCEEDINGS OF THE NATIONAL ACADfurthii). BEHAVIORAL ECOLOGY AND EMY OF SCIENCES 22276-283,1936a. SOCIOBIOLOGY 6~29-38,1979. Tutin, C.E.G. Reproductive behavior of wild Yerkes, R.M.; Elder, J.H. Oestrus, recepchimpanzees in the Gombe National Park, tivity, and mating in the chimpanzee. Tanzania. JOURNAL OF REPRODUCCOMPARATIVE PSYCHOLOGY MONOTION AND FERTILITY, SUPPLEMENT GRAPHS 13:1-39,1936b. 28:43-57, 1980. Young, W.C.; Orbison, W.D. Changes in selected features of behavior in pairs of Tutin, C.E.G.; McGinnis, P.R. Sexuality of the chimpanzee in the wild. Pp. 239-264 in oppositely-sexed chimpanzees during the REPRODUCTIVE BIOLOGY OF THE sexual cycle and aRer ovariectomy. JOURGREAT APES: COMPARATIVE AND NAL OF COMPARATIVE PSYCHOLOGY BIOMEDICAL PERSPECTIVES. C.E. 37:107-143, 1944.