Diurnal and annual variation of adrenocortical activity in the squirrel monkey.код для вставкиСкачать
American Journal of Primatology 35:28%292 (1995) Diurnal and Annual Variation of Adrenocortical Activity in the Squirrel Monkey CHRISTOPHER L. COE' AND SEYMOUR LEVINE' 'Department of Psychology University of Wisconsin Madison, Wisconsin; 'Department of Psychiatry Stanford University School of Medicine Stanford, California Diurnal changes in basal cortisol levels and stress reactivity were assessed in male and female squirrel monkeys. Blood samples were collected at four-hour intervals throughout the day and night in the mating and nonmating seasons. Basal cortisol levels in females were similar in both seasons, but males tended to have higher cortisol levels during the mating season, especially at night. For both sexes, cortisol secretion was highest between 0400 and 0800 preceding the onset of the diurnal activity period. Assessment of cortisol responses following brief handling and anesthesia indicated that stress responses were relatively stable across the year, but cortisol increments were slightly higher in the nonmating season. Cortisol levels post-stress were generally related to prior baseline values. Thus, a knowledge of biorhythmic changes in basal hormone levels was important for predicting hormone levels after acute stressors. Males also underwent marked seasonal variation in their basal testosterone levels, which markedly altered the nature of their testosterone responses 30-minutes poststress. 0 1995 Wiley-Liss, Inc. Key words: squirrel monkey, adrenal, cortisol, testosterone, diurnal cycle, annual cycle INTRODUCTION It is well known that hormone levels vary across the day and, especially in the seasonally breeding primates, show marked annual variation. Despite this knowledge, however, only a few primate studies have directly evaluated how the daily cycle of a particular hormone varies within the context of the annual cycle [e.g., Wilson et al., 1978; Wiebe et al., 1984; Michael & Zumpe, 19933.Because many hormones, including the adrenal and gonadal steroids, are higher at night, one might anticipate more dramatic annual variation during the nocturnal period. This question was considered in the following study of the squirrel monkey, which assessed 24-hour rhythms in cortisol levels at two points in the annual reproductive cycle. In addition, the possible effect of this variation in basal hormone secretion on adrenal reactivity to stress was investigated. The pituitary-adrenal axis of the squirrel monkey is of special interest because Received for publication September 17, 1993; revision accepted August 11, 1994 Address reprint requests to Christopher L. Coe, Department of Psychology, University of Wisconsin, 1202 West Johnson, Madison, WI 53706. 0 1995 Wiley-Liss, Inc. 284 I Coe and Levine this species is one of the small New World monkeys that secretes extremely high levels of cortisol [Brown et al., 19701. Even under basal conditions, cortisol levels are 5-10 times higher than found in Old World monkeys and humans, and have been reported to reach between 200-1,000 pg/dl following stressful events [Coe et al., 19921. Two hypotheses have been advanced to account for the evolution of this unusual endocrine pattern: (1) reduced numbers of receptors or ones with low affinity necessitated the increased hormone secretion [Chrousos et al., 19821, or (2) alternatively, a n inefficient metabolism and clearance of steroid hormones has resulted in the maintenance of high hormone levels in circulation [Klosterman et al, 19861. Regardless of which interpretation is correct, one consequence for extant squirrel monkeys would be marked diurnal changes and possibly annual variation in cortisol levels. An early study reported that diurnal cycles in adrenal and gonadal hormone levels were difficult to discern [Wilson et al., 19781, but seasonal variation was subsequently described for many hormones, including gonadal, thyroid, and some pituitary hormones [Diamond et al., 1984; Mendoza et al., 1978; Wiebe et al., 1988; Yeoman et al., 19881. A subsequent study also demonstrated marked diurnal variation in adrenal and gonadal hormones during the mating season, based on assessments a t 0300, 0900, and 2300 [Wiebe et al., 19841. In the current study, cortisol levels were assessed in male and female squirrel monkeys during both the mating and nonmating seasons, anticipating that higher hormone levels or greater reactivity might occur during the mating season. The occurrence of extensive interactions between gonadal and adrenal physiology has been known for several decades [Parkes & Deanesley, 19661, and our a priori prediction was that there would be higher levels of cortisol in the mating season. Formation of male-female groups has been shown previously to increase gonadal and adrenal activity in male squirrel monkeys for several weeks [Mendoza et al., 19791. In addition, we had demonstrated that estrogen is a potent stimulator of the pituitary-adrenal axis in the squirrel monkey, and probably accounts for the large increase in cortisol levels during early pregnancy [Coe et al., 19861. However, in this study, females were not allowed to conceive and, thus, the issue was whether significant annual variation in adrenocortical activity would still be apparent. A related issue was whether adrenal responsivity would shift across the day and year. Studies in rodents have indicated that there is circadian variation in pituitary and adrenal hormone responses to stressors [Sapolsky, 1992; Dallman et al., 19921. In general, larger hormone increments have been observed when basal values are low, although there are many exceptions to this general conclusion. The handling and brief anesthesia used €or the first sample functioned as a n acute stress manipulation, and allowed us to evaluate variation in stress reactivity by collecting a second sample 30 minutes later. METHODS Subjects The subjects for this research were 11male and 12 female squirrel monkeys of the Guyanese variety (Saimiri sciareus). The monkeys had been imported as adults from Guyana several years prior to this study and were well acclimated to laboratory conditions. They were housed in four social groups, each consisting of 5-6 monkeys of the same sex, established for several months prior to the onset of the study. Housing Each group was housed indoors in a wire-mesh pen (1.9 x 1.9 x 1.9m high) at the Stanford Primate Facility. Within each pen there were three horizontal Adrenocortical Rhythm in Saimiri / 285 perches, two food hoppers, and automatic watering devices. High protein monkey chow (Purina) was available ad libitum, and supplemented each morning. Each pen also had a small Plexiglas window in the side wall permitting the monkeys to view members of the opposite sex in the adjacent cage. This facilitated heterosexual interactions normally associated with the annual breeding cycle, although physical contact was not permitted to prevent conceptions. The occurrence of a normal annual cycle was also facilitated by exposing the monkeys to a natural light rhythm through large windows. The monkeys experienced the typical annual fluctuations in day and night length that occur a t this latitude-sunrise: sunset times averaged 0650:1659 and 0535:1845 for the November and April sampling periods, respectively-which results in a mating season from December to April in Guyanese squirrel monkeys [Coe et al., 1985; Mendoza et al., 19781. This timing is 1-3 months earlier than that exhibited by monkeys of Peruvian or Bolivian derivation, housed under identical conditions in this colony. Procedures The experiment was designed to determine base and stress hormone values for one male and one female group a t the end of the mating period (April), and from a second pair of groups at the end of the birth season (November). This design obviated the experimental order effect that would have occurred from a longitudinal study of the same subjects living in one group. For mating season values, samples were collected beginning in April, a time of high gonadal activity in this colony. For the nonmating season values, samples were collected at the end of births in November, the nadir of gonadal activity. Small heparinized blood samples (1ml) were obtained at 1-2 week intervals from each monkey at six timepoints across the day and night (0400, 0800, 1200, 1600, 2000, 2400). The order of the sample collection was completely counterbalanced across subjects such that values were obtained for every timepoint on each of the six sampling days. On each sampling day, blood was obtained from a different male and female monkey a t each timepoint. To ensure the attainment of basal hormone values, the samples were collected rapidly after momentary immobilization with ethyl ether (time to sample collection less than two minutes). Thirty minutes later, a second blood sample was collected to determine the acute effect of the ether (a standard stressor) and handling on hormone levels. Approximately five minutes after the second sample, the one male and one female monkey that had been sampled were returned to their respective social groups. Hormone assays Plasma cortisol levels were determined by the radioimmunoassay methods of Klemm and Gupta [19751 using antiserum F3-314 from Endocrine Sciences (Tarzana, CA). Two aliquots were taken from each plasma sample, and each aliquot assayed in duplicate. The aliquots were extracted with ethyl acetate, and evaporated under N, a t 37 "C. The dried extracts were stored a t -15 "C until assayed. On the day of assay, the extracts were brought to room temperature and borate buffer was added to each tube. An appropriate dilution for each sample was chosen to fall within the linear portion of the standard curve (r = .998). The cortisol assay had a n average inter-assay variation of 15.3% and a n intra-assay C.V. of 9.8%. Plasma testosterone levels were determined in duplicate from the males' samples a t the laboratory of Drs. J. Davidson and E. Smith using the radioimmunoassay procedures of Frankel et al.  and antiserum CEO No. 1. A preliminary screening assay was performed in order that samples could be correctly diluted. All determinations in the second assay were within the linear portion of 286 I Coe and Levine Mating Season Nonmating Season T r 150 Hours i * Base Hours Fig. 1. Mean ( + S.E.) basal and stress levels of testosterone during the mating and nonmating seasons. The second sample was obtained 30 minutes after the baseline sample. the standard curve. The inter-assay C.V.determined between 2 0 4 0 % of the curve was 2.7%, and the intra-assay C.V. for this study averaged 5.4%. Statistical analyses The data were evaluated with analyses of variance assessing Season (Mating, Nonmating) and Gender (Male, Female) as independent factors, and the six timepoints (Time of Day) and sampling conditions (Base, Stress) as repeated measures. Each monkey was sampled at all six times of the day and night in the Basal and Stress conditions. RESULTS Assessment of male testosterone levels verified the selection of April and November as representing two distinct periods in the annual reproductive cycle. In keeping with our expectation, male monkeys had significantly higher levels of testosterone in April a t the end of the mating period than during November, the normal infancy period when sexual activity does not occur (F[1,9] = 13.5,p < .005) (Fig. 1).During the nonmating season, testosterone levels were lowest during the daylight hours (% = 23 ngiml) and then increased at night (2 = 69 ngiml). Although these values were relatively high compared to other species, even the nocturnal values a t this time of year remained far below mating season levels (X = 122 and 132 ng/ml, for day and night, respectively). The highest testosterone values in both seasons occurred between 2400 and 0400 (F[5,451 = 3.68, p < .008). When evaluating Fig. 1, it should be noted that the anomalously low mean value evident at 0800 in the mating season was due to the fact that three lower-ranking males showed a marked decrease at this timepoint, but the average was still above the nonmating range. There was also a dramatic seasonal shift in the testosterone response 30 minutes after the disturbance and anesthesia associated with the initial base samples. During the mating season, there was typically a decrease in testosterone when Adrenocortical Rhythm in Saimiri I 287 l50 1 100 - 50 - MALE '--g + , 1 " Nonmating Season I 2000 2400 150 Mating Season * ......_. 0400 0800 1200 FEMALE 1 1 T .........0 " 1600 I 2000 2400 0400 0800 Mating Season Nonmating Season I I 1200 1600 Hours Fig. 2. Mean ( + /- S.E.) basal levels of cortisol for males and females during the mating and nonmating season. Every monkey in the 4 social groups was sampled a t each of the 6 timepoints. basal levels were high or no consistent change when the baseline values were lower. In contrast, during the nonmating season, testosterone levels were typically elevated 30 minutes after the base sample. This seasonal shift in the gonadal response to stress resulted in a significant interaction between Season and the Base/Stress factor in the analysis of variance (F[1,9] = 7.68, p < .03). Basal levels of cortisol also showed significant diurnal and annual variation (Fig. 2). Adrenocortical activity increased in both sexes during the latter portion of the night and, for most monkeys, the highest cortisol levels were found at 0400 (F[5,95] = 16.93, p < .001). The nadir of the adrenal rhythm typically occurred between the late afternoon and midnight. There also was a modest seasonal shift in basal cortisol levels in males, with higher levels found during the mating season. An ANOVA evaluating just the male cortisol values indicated that there was a significant interaction between Season and Time of Day (F[1,9] = 5.29, p < .047). The highest cortisol levels in male monkeys occurred a t 0400 during the mating season. Contrary to the a priori prediction, an equivalent seasonal shift was not evident in female cortisol values, possibly because the sampling days did not always coincide with the follicular phase of the ovarian cycle. As expected, plasma cortisol levels increased in all monkeys 30 minutes after the initial sample (F[1,19] = 149.14, p < .0001) (Fig. 3). In general, the increments tended to be fairly consistent across the two seasons and time of day and, thus, 288 I Coe and Levine NONMATING SEASON MATING SEASON 2oo 1 r Male 1 1 2oo - 150 Male 2oo 150 4 Fc 2oo 150 - 100 -50 0 01 0 2000 2400 0400 0800 1200 1600 2000 2400 0400 0800 12001600 200 - -0 Female - 200 2oo 1 Female r - 150 2 150- r 0 0 I 100- - 100 50- - 50 -5m 01 0 2000 2400 0400 0800 1200 1600 Hours 2000 2400 0400 0800 1200 1600 Hours Fig. 3. Mean basal and stress levels of cortisol for males and females during the mating and nonmating season. The stress samples (x---x) were obtained 30 minutes after the baseline sample. stress values followed a circadian pattern similar to that of the basal values. However, there were several small differences that are noteworthy. Overall, the stress increment tended to be higher in females than in males (delta value = 45.3 versus 31.3 Fg/dl, F[1,191 = 4.92, p < .04).The cortisol increase also tended to be higher in the nonmating season than in the mating season (51.5 versus 25.1 Fgldl, F[1,19] = 17.71, p < .001). Larger increments were typically observed when the prior basal values were lower. The smallest increments occurred at 0800 and, across all monkeys sampled at this hour, the magnitude of the cortisol response was inversely related to the prior basal value (r = -0.55, p < .007). DISCUSSION This study verified the occurrence of a clear daily rhythm in cortisol levels in the squirrel monkey as had been reported previously [Wiebe et al., 19841. The temporal orientation was similar to that found for humans and other diurnal animals, with peak adrenal output preceding the onset of the active period of the day. However, the impact of this biorhythm was greatly magnified by the high steroid hormone levels of the squirrel monkey such that changes on the order of 50-75 pgldl occurred across the day. The annual and diurnal changes in testos- Adrenocortical Rhythm in Saimiri / 289 terone were also very dramatic because of the high level of gonadal steroids secreted by this species [Diamond et al., 1984; Mendoza et al., 19781. Even in the nonmating season, male squirrel monkeys maintained levels of testosterone that would be considered extremely high for Old World primates. Indeed, the primary decrease during the nonmating season appeared to be in the daytime testosterone levels, because fairly high nocturnal surges were still apparent in this season. While the annual change in basal testosterone levels was anticipated, the factors underlying the marked shift in the direction of the 30 minute stress response are less obvious. Nevertheless, the results do concur with earlier stress research on gonadal hormone responses in the male squirrel monkey [Coe et al., 19831, and with findings on the male baboon [Sapolsky, 19871, and men [Davidson et al., 19781. The initial response of the pituitary-gonadal axis to a stressor is usually a transient increase in output, often lasting 15-45 minutes, before the commonly described stress-related decrease in secretion. However, the increase tends to persist longer when the preceding basal levels are low, resulting in elevated responses a t 30 minutes (i.e., in the nonmating season or subordinate males), whereas the rise and subsequent fall is more rapid when basal testosterone levels are high (i.e, in the mating season or dominant males). Similar phenomena have also been described for other physiological systems and reflect both the rapidity of negative feedback and “ceiling” set points. This effect of baseline on the subsequent course of reaction has been described historically as the “law of initial values” [Lacey, 1956; Wilder, 19561. Diurnal and annual variation in the cortisol stress response was not as marked, although the smallest increments tended to occur: (1)in males, (2) at night, (3) during the mating season, or (4) when basal levels were higher. The similarity of the cortisol increments may have been due to the magnitude of the acute response to ether, a potent stimulus, especially at 30 minutes. Cortisol levels following ether exposure usually increase quickly and then subside by two hours [Coe et al., 19781, so a 60-minute timepoint might have been more sensitive to biorhythmic variation. Studies with rodents suggest that larger adrenal responses to a stressor should have occurred when the baseline corticosteroid levels were lowest [Dallman et al., 19921. In keeping with this general conclusion, higher increments were in fact observed in the nonmating season and lower increments were observed at a time of high basal activity (0800). We also had anticipated more dramatic seasonal changes in female cortisol levels because of the strong influence of estrogen on adrenal activity [Coe et al., 19861, but the selection of a timepoint late in the mating season may have resulted in less annual variation in ovarian activity than has been reported previously [Mendoza et al., 1978; Diamond et al., 19841. In addition, it may require sampling specifically during the follicular phase to reveal the estrogenic stimulation of adrenal activity. Several brief comments on the experimental design and sampling procedures are also warranted. A biorhythm study requires multiple samples obtained from the same subject either on the same day or over time. We opted not to employ the catherization approach typically utilized in studies of Old World monkeys [Perachio et al., 19771 because of the squirrel monkey’s prolonged endocrine responses to restraint [see Brown et al., 19701. The utilization of ether to facilitate the collection of the first sample did serve the primary purpose of permitting us to obtain low baseline values. Typically, samples were collected in less than one minute following exposure t o the ether. In contrast, much higher basal cortisol values have been reported for this species when blood samples were obtained from conscious monkeys using manual restraint. Nevertheless, it should still be acknowledged that the clarity of the circadian rhythm may have been partially obscured by the approach 290 I Coe and Levine of sampling multiple animals from the group, each a t a different timepoint on the same day. Cortisol levels are known to be strongly affected by recent changes in housing conditions and social stimuli [Brown et al., 1970; Wilson et al., 1978; Mendoza et al., 1979; Gonzalez et al., 19801,but this type of stimulation proved not to be a confounding factor in the current study. The failure of the acute disturbance to affect the next subject’s basal value may have been due to a “social buffering” phenomena described previously; i.e., when squirrel monkeys live in social groups with three or more members, they do not show the typical adrenocortical response to aversive stimuli and psychological disturbance [Vogt et al., 19811. In addition, the 1-2 week intervals between sampling days and the randomization of timepoints may have obviated any anticipatory responses by the monkeys. Most significantly, this study does emphasize the important methodological point that underlying biorhythms should always be taken into account when conducting a physiological or psychoendocrine study. Primate studies have usually focused on diurnal values, even though higher levels of several hormones often occur during the latter part of the nocturnal period. Similarly, discussions on the health significance of stress-induced changes in diurnal values often don’t take into account the fact that equivalently high levels may occur normally during the daily cycle. In this study undisturbed basal levels of cortisol at night frequently exceeded the levels found after the acute stressor during the daytime. Further research is needed to resolve the different consequences of high hormone levels or stress responses at various timepoints across the day. CONCLUSIONS 1. Plasma cortisol levels showed a clear daily rhythm in male and female squirrel monkeys, with the highest levels occurring between 0400 and 0800. Annual variation was less apparent, although males had higher levels during the night in the mating season. Evaluation of annual changes in female cortisol levels may require a more systematic scheduling of samples according to the ovarian cycle. 2. The 30-minute stress response to the handling and anesthesia was fairly consistent, but slightly higher responses were found in the nonmating season, and in females as compared to males. 3. Males had much higher levels of testosterone in the mating season, which significantly affected the temporal trajectory of the gonadal response to a stressor. 4.Steroid hormone output is particularly high before monkeys wake in the morning and thus, hormone levels at this timepoint may be extremely sensitive to environmental and psychological influences. A greater appreciation of endocrine biorhythms may be important for predicting and evaluating the consequences of stress-related changes in hormone secretion. ACKNOWLEDGMENTS This research was supported primarily by NICHD grant (HD02881). SL also has a Research Scientist Award (MH19936).CLC currently receives partial salary support from an NIMH grant (MH41659) and from the Wisconsin Regional Primate Center (RR00167-28). REFERENCES Brown, G.M.; Grota, L.J.; Penney, D.P.; Reichlin, S. Pituitary-adrenal function in the squirrel monkey. ENDOCRINOLOGY 8 6 5 1 9 4 2 9 , 1970. Chrousos, G.P.; Renquist, D.; Brandon, D.; Barnard, D.; Fowler, D.; Loriaux, D.L.; Lipsett, D. The squirrel monkey: Receptor mediated end-organ resistance to proges- Adrenocortical Rhythm in Saimiri 1 291 terone. JOURNAL OF CLINICAL ENDOCRINOLOGY AND METABOLISM 55(2): 364-368, 1982. Coe, C.L.; Smith, E.R.; Levine, S. The endocrine system of the squirrel monkey. Pp. 191-218 in HANDBOOK OF SQUIRREL MONKEY RESEARCH, L.A. Rosenblum; C.L. Coe, eds. New York, Plenum Press, 1985. Coe, C.L.; Mendoza, S.P., Davidson, J.; Smith, E.R., Dallman, M.; Levine, S. Hormonal response to stress in the squirrel monkey. NEUROENDOCRINOLOGY 26: 367-377, 1978. Coe, C.L.; Murai, J.T.; Wiener, S.G.; Levine, S., Siiteri, P. Rapid cortisol and CBG responses during pregnancy and after estrogen administration in the squirrel monkey. ENDOCRINOLOGY 118:435-440,1986. Coe, C.L.; Savage, A.; Bromley, L.J. Phylogenetic influences on hormone levels across the primate order. AMERICAN JOURNAL OF PRIMATOLOGY 28(2):81100, 1992. Coe, C.L.; Smith, E.R.; Mendoza, S.P.; Levine, S. Varying influence of social status on hormone levels in male squirrel monkeys. Pp. 7-32 in HORMONES, DRUGS AND SOCIAL BEHAVIOR IN PRIMATES. H.D. Steklis; A S . Kling, eds. New York, Spectrum Publ., 1983. Dallman, M.F.; Akana, S.F.;Scribner, K.A.; Bradbury, M.J.; Dominique Walker, C.; Strack, A.M.; Cascio, C.S. Stress, feedback and facilitation in the hypothalamic-pituitary-adrenal axis. JOURNAL OF NEUROENDOCRINOLOGY 4(5):517526, 1992. Davidson, J.M.; Smith, E.R.; Levine, S. Testosterone. Pp. 57-62 in PSYCHOBIOLOGY OF STRESS: A STUDY OF COPING MEN. H. Ursin; E. Baade; S. Levine, eds. New York, Academic Press, 1978. Diamond, E.J; Aksel, S.; Hazelton, J.M.; Jennings, R.A.; Abee, C.R. Seasonal changes of serum concentrations of estradiol and progesterone in Bolivian squirrel monkeys. AMERICAN JOURNAL OF PRIMATOLOGY 6:103-133, 1984 Frankel, A.I.; Mock, E.J.; Wright, W.W.; Kamel, F. Characterization and physiological validation of radioimmunassay for plasma testosterone in the male rat. STEROIDS 25:73-98, 1975. Gonzalez, C.A.; Coe, C.L.; Levine, S. Cortisol responses under different housing conditions in female squirrel monkeys. PSYCHONEUROENDOCRINOLOGY 7:209216, 1982. Klemm, W.; Gupta D. An outine method for the radioimmunoassay of plasma cortisol without chromatography. Pp. 143 in RADIOIMMUNOASSAY OF STEROID HOR- MONES. D. Gupta, ed. Weinheim, Germany, Verlag Chemie. Klosterman, L.L.; Murai, J.T.; Siiteri, P.K. Cortisol levels, binding, and properties of corticosteroid-binding globulin in the serum of primates. ENDOCRINOLOGY 118: 424-434,1986. Lacey, J.I. The evaluation of autonomic responses: Toward a general solution. ANNALS OF THE NEW YORK ACADEMY OF SCIENCE 67:125-163,1956. Mendoza, S.P.; Coe, C.L.; Lowe, E.L.; Levine, S. The physiological response to group formation in adult male squirrel monkeys. PSYCHOENDOCRINOLOGY 3:221-229, 1979. Mendoza, S.P.; Lowe, E.L.; Resko, J.A.; Levine, S. Seasonal variations in gonadal hormones and social behavior in squirrel monkeys. PHYSIOLOGY AND BEHAVIOR 20:515-522, 1978. Michael, R.P.; Zumpe, D. A review of hormonal factors influencing the sexual and aggressive behavior of macaques. AMERICAN JOURNAL OF PRIMATOLOGY 30: 213-241, 1993. Parkes, A.S.; Deanesley, R. Relations between the gonads and the adrenal glands. Pp.1,064-1,111 in MARSHALL'S PHYSIOLOGY OF REPRODUCTION, vol. 3. A S . Parkes, ed. Boston, Little Brown & Co., 1966. Perachio, A.A.; Alexander, M.; Marr, L.D.; Collins, D.C. Diurnal variations of serum testosterone levels in intact and gonadectomized male and female rhesus monkeys. STEROIDS 29~21-33,1977. Sapolsky, R.M. Stress, social status, and reproductive physiology in free-living baboons. Pp. 291-322 in PSYCHOBIOLOGY OF REPRODUCTIVE BEHAVIOR. D. Crews, ed. Englewood Cliffs, NJ, Prentice Hall, Inc., 1987. Sapolsky, R.M. STRESS, THE AGING BRAIN AND THE MECHANISMS OF NEURON DEATH. Cambridge, Mass., MIT Press, 1992. Vogt, J.L.; Coe, C.L.; Levine, S. Behavioral and adrenocorticoid responsiveness of squirrel monkeys to a snake: Is flight necessarily stressful? BEHAVIORAL AND NEURAL BIOLOGY 32:391-405,1981. Wiebe, R.H.; Diamond, E.J.; Akesel, S.; Liu, P.; Williams, L.E.; Abee, C.R. Diurnal variations of androgens in sexually mature male Bolivian squirrel monkeys (Saimiri sciureus) during the breeding season. AMERICAN JOURNAL OF PRIMATOLOGY 7:291-297,1984. Wiebe, R.H.; Williams, L.E.; Abee, C.R.; Yeoman, R.R.; Diamond, E.J. Seasonal changes in serum dihydroepiandrosterone, androstenedione, and testosterone levels in 292 I Coe and Levine the squirrel monkey (Saimiri boliviensis boliviensis). AMERICAN JOURNAL OF PRIMATOLOGY 14:285-291, 1988. Wilder, J. The law of initial value in neurology and psychiatry. JOURNAL OF NERVOUS AND MENTAL DISEASES 125: 73-86,1956. Wilson, M.I.; Brown, G.M.; Wilson, D. Annual and diurnal changes in plasma androgen and cortisol in adult male squirrel monkeys (Saimiri sciureus) studied longi- tudinally. ACTA ENDOCRINOLOGICA 87:424-433, 1978. Yeoman, R.R.; Aksel, S.; Hazelton, J.M.; Williams, L.E.; Abee, C.R. In vitro bioactive luteinizing hormone assay shows cyclical seasonal hormonal changes and response to luteinizing-hormone releasing hormone in the squirrel monkey (Saimiri boliviensis boliviensis). AMERICAN JOURNAL OF PRIMATOLOGY 14:167175, 1988.