Seasonal study of the interrenal function of the European green frog, in vivo and in vitro

Adrenal activity in maned wolves is higher on farmlands and park boundaries than within protected areas

In this study we measured excreted fecal corticoid metabolites (FCM) in maned wolves (Chrysocyon brachyurus) living within a protected reserve, on farmlands or in a boundary zone between the two habitats, and determined the impacts of season and reproductive status on adrenal activity. Feces were collected within a national park (n=191 samples), a park boundary zone (n=39) and on nearby farmlands (n=27), processed and analyzed by enzyme immunoassay. FCM amounts from samples collected on farmlands were higher (P<0.05) than in those collected inside the reserve and from the boundary zone. In relation to seasonality, FCM were elevated (P<0.05) in spring (September-November) when wolf pairs were raising young. We then divided the samples collected during breeding season (March-August) into cycling females and male/non-cycling females based on fecal progesterone: fecal testosterone ratio. FCM concentrations of the former collected inside the park were higher than (P<0.05) than the latter group. However, there were no differences in FCM levels between the two groups for samples collected in the boundary zone and on farmlands. Furthermore, FCM concentrations of male/non-cycling females samples collected on farmlands were 2- to 5-fold higher (P<0.05) than in counterparts collected inside the park. The consistently high FCM concentrations in samples collected on farmlands indicate that, in addition to seasonality, gender and reproductive status, anthropogenic pressures also contribute to elevating adrenal steroid for individuals living in altered habitat.

Are baseline and short-term corticosterone stress responses in free-living amphibians repeatable?

Amphibians respond to environmental stressors by secreting corticosterone, a stress hormone which promotes physiological and behavioral responses. Capture handling can be used to stimulate physiological stress response in amphibians. The use of single blood sampling and presentation of mean data often limits the quantification of within and between individual variation in baseline and short-term corticosterone stress responses in amphibians. It is important for studies of amphibian physiological ecology to determine whether baseline and short-term corticosterone stress responses are consistent or not. We quantified repeatability (r), a statistical measure of consistency, in baseline and short-term corticosterone stress responses to a standard capture and handling stress protocol in free-living adult male cane toads (Rhinella marina). Corticosterone metabolite concentrations were measured entirely non-invasively in male toad urine samples via an enzyme-immunoassay. During the first sampling occasion, urine samples were collected manually from individual male toads (n=20) immediately upon field capture. Toads were handled for 5min then transferred to plastic bags (constituting a mild stressor), and urine samples were collected hourly over 8h in the field. The toads were resampled for baseline (0h) urine corticosterone with hourly urine sampling over 8h (for quantification of the stress induced corticosterone) at 14 day intervals on three consecutive occasions. Within and between sample variations in urinary corticosterone metabolite concentrations were also quantified. All toads expressed a corticosterone stress response over 8h to our standard capture and handling stress protocol. Variations both within and between toads was higher for corrected integrated corticosterone concentrations than corticosterone concentrations at baseline, 3 or 6h. Baseline urinary corticosterone metabolite concentration of the male toads was highly repeatable (r=0.877) together with high statistical repeatabilities for 3h (r=0.695), 6h (r=0.428) and 8h (r=0.775) corticosterone metabolite concentrations, and for the total and corrected integrated corticosterone responses (r=0.807; r=0.743 respectively). This study highlights that baseline and short-term corticosterone stress responses are repeatable in free-living amphibians. Future studies should utilize this non-invasive tool to explore repeatability among seasons and across years, and determine its functional significance in relation to behavioral ecology and reproduction in amphibians generally.

Elevated plasma corticosterone increases metabolic rate in a terrestrial salamander

Plasma glucocorticoid hormones (GCs) increase intermediary metabolism, which may be reflected in whole-animal metabolic rate. Studies in fish, birds, and reptiles have shown that GCs may alter whole-animal energy expenditure, but results are conflicting and often involve GC levels that are not physiologically relevant. A previous study in red-legged salamanders found that male courtship pheromone increased plasma corticosterone (CORT; the primary GC in amphibians) concentrations in males, which could elevate metabolic processes to sustain courtship behaviors. To understand the possible metabolic effect of elevated plasma CORT, we measured the effects of male courtship pheromone and exogenous application of CORT on oxygen consumption in male red-legged salamanders (Plethodon shermani). Exogenous application of CORT elevated plasma CORT to physiologically relevant levels. Compared to treatment with male courtship pheromone and vehicle, treatment with CORT increased oxygen consumption rates for several hours after treatment, resulting in 12% more oxygen consumed (equivalent to 0.33 J) during our first 2h sampling period. Contrary to our previous work, treatment with pheromone did not increase plasma CORT, perhaps because subjects used in this study were not in breeding condition. Pheromone application did not affect respiration rates. Our study is one of the few to evaluate the influence of physiologically relevant elevations in CORT on whole-animal metabolism in vertebrates, and the first to show that elevated plasma CORT increases metabolism in an amphibian.

Changes in urinary testosterone and corticosterone metabolites during short-term confinement with repeated handling in wild male cane toads (Rhinella marina)

Stressors generally decrease testosterone secretion and inhibit reproduction in animals. Urinary testosterone and corticosterone metabolite concentrations were measured in adult male cane toads (Rhinella marina) at the time of capture from the wild and during 24 h of confinement with repeated handling. Mean urinary testosterone concentrations increased 2 h after capture, were significantly elevated above initial concentrations at 5 h, and then declined. Mean testosterone concentrations remained elevated 24 h after capture. Mean urinary corticosterone concentrations increased after capture, were significantly elevated above initial concentrations at 2 h, and remained elevated thereafter. This is the first report in amphibians of an increase in testosterone excretion after capture from the wild, with previous studies showing either no change or decline in testosterone. This finding may be associated with the mating strategy and maintenance of reproductive effort in the cane toad, a species that shows explosive breeding and agonistic male–male interactions during breeding. The finding that testosterone excretion increases rather than decreases after capture in male cane toads shows that it should not be generally assumed that reproductive hormone secretion will decrease after capture in amphibians.

Hormones and acoustic communication in anuran amphibians

Circulating hormone levels can mediate changes in the quality of courtship signals by males and/or mate choice by females and may thus play an important role in the evolution of courtship signals. Costs associated with shifts in hormone levels of males, for example, could effectively stabilize directional selection by females on male signals. Alternatively, if hormone levels affect the selection of mates by females, then variation in hormone levels among females could contribute to the maintenance of variability in the quality of males' signals. Here, I review what is known regarding the effects of hormone levels on the quality of acoustic signals produced by males and on the choice of mates by females in anuran amphibians. Surprisingly, despite the long history of anuran amphibians as model organisms for studying acoustic communication and physiology, we know very little about how variation in circulating hormone levels contributes to variation in the vocal quality of males. Proposed relationships between androgen levels and vocal quality depicted in recent models, for example, are subject to the same criticisms raised for similar models proposed in relation to birds, namely that the evidence for graded effects of androgens on vocal performance is often weak or not rigorously tested and responses seen in one species are often not observed in other species. Although several studies offer intriguing support for graded effects of hormones on calling behavior, additional comparative studies will be required to understand these relationships. Recent studies indicate that hormones may also mediate changes in anuran females' choice of mates, suggesting that the hormone levels of females can influence the evolution of males' mating signals. No studies to date have concurrently addressed the potential complexity of hormone-behavior relationships from the perspective of sender as well as receiver, nor have any studies addressed the costs that are potentially associated with changes in circulating hormone levels in anurans (i.e., life-history tradeoffs associated with elevations in circulating androgens in males). The mechanisms involved in hormonally induced changes in signal production and selectivity also require further investigation. Anuran amphibians are, in many ways, conducive to investigating such questions.

Seasonal alterations in adrenocortical cell function associated with stress-responsiveness and sex in the eastern fence lizard (Sceloporus undulatus)

We characterized steroidogenic properties of dispersed adrenocortical cells from field-active male and female eastern fence lizards (Sceloporus undulatus) to investigate whether alterations in cell function could, in part, explain seasonal variation in baseline and stress-induced plasma corticosterone (B). Lizards were collected during the breeding and postbreeding seasons and shortly prior to hibernation. Dispersed cells in vitro produced B, aldosterone (ALDO), and progesterone in response to 8-Br-cAMP, 25-(OH)cholesterol, adrenocorticotropin (ACTH; as little as 100 fM), and angiotensin II. Maximal progesterone, B, and ALDO responses to ACTH were roughly 1000%, 500%, and 100% greater than corresponding basal values. Angiotensin II was an effective steroidogenic stimulant but much less so than ACTH. Corticosteroid production exhibited considerable steroid-specific variation among seasons. Maximal ACTH-induced B production was lower in the postbreeding season than at either of the other two measurement points, essentially opposite to the pattern for ALDO. Males and females generally produced B at similar rates, but ALDO and progesterone showed numerous sex differences that usually covaried between the two steroids. Cellular sensitivity to 25-(OH)cholesterol and angiotensin II showed few sex differences or seasonal changes. In contrast, sensitivity to ACTH decreased markedly from the breeding to the postbreeding season in males, corresponding to the decrease in stress-responsiveness, and in both sexes was considerably lower prior to hibernation than during the breeding season. Under some conditions, plasma B may be limited by the production capacity of adrenocortical cells. In summary, seasonal variations in body condition, reproductive activity, and baseline and stress-induced plasma B may be attributed at least in part to alterations in adrenocortical cell steroidogenic function.

Seasonal changes in plasma glucocorticoid concentrations in free-living vertebrates

The vertebrate stress response helps animals respond to environmental dangers such as predators or storms. An important component of the stress response is glucocorticoid (GC) release, resulting from activation of the hypothalamic-pituitary-adrenal axis. After release, GCs induce a variety of behavioral and physiological changes that presumably help the animal respond appropriately to the situation. Consequently, GC secretion is often considered an obligatory response to stressful situations. Evidence now indicates, however, that free-living species from many taxa can seasonally modulate GC release. In other words, the magnitudes of both unstressed and stressed GC concentrations change depending upon the time of year. This review examines the growing evidence that GC concentrations in free-living reptiles, amphibians, and birds, but not mammals, are commonly elevated during the breeding season. This evidence is then used to test three hypotheses with different focuses on GC's energetic or behavioral effects, as well as on GC's role in preparing the animal for subsequent stressors. These hypotheses attempt to place annual GC rhythms into a physiological or behavioral context. Integrating seasonal differences in GC concentrations with either different physiological states or different life history stages provides clues to a new understanding of how GCs actually help in survival during stress. Consequently, understanding seasonal modulation of GC release has far-reaching importance for both the physiology of the stress response and the short-term survival of individual animals.

Hormonal effect on the osmotic, electrolyte and nitrogen balance in terrestrial Amphibia

Two main hormones regulate water balance in amphibian. First, mesotocin (MT) acting as a diuretic agent, and second arginine vasotocin (AVT) being an anti-diuretic hormone. In addition, prolactin (PRL), aldosterone, corticosterone, angiotensin II and atriunatriuretic hormones, play a role too in regulating water and ion balance. The hormones affect the epidermis and bladder permeability to water and ions as well as the kidney through the control of the glomerular filtration rate (GFR). The main questions concern the presence and action of these hormones during the amphibian's life history. Are they present in both larval and adult stages? Are these hormones being synthesized in both aquatic and terrestrial adult phases? Under what circumstances are they being stored or released? Would the target organs (epidermis, bladder, kidney) respond in a similar way during all periods? The problem is the fact that under most circumstances an amphibian while in an aquatic environment responds physiologically differently than when on land. Only partial information concerning hormone presence, release and control of water balance is available at the moment, and even that is fragmentary and based on only a very small number of amphibian species.

Adrenal activity in the female lizard Lacerta vivipara Jacquin during artificial hibernation

The variations of interrenal activity were investigated in captive female Lacerta vivipara submitted to artificial hibernation (4 months at 6 degrees) and compared to data obtained in nonhibernating females. Plasma corticosterone levels reached 25 ng/ml during the prehibernal period. During the first day following the transfer to cold conditions, an initial significant peak of plasma corticosterone was observed (up to 63 ng/ml). A second, more gradual, but also significant increase was observed thereafter and levels remained maximum during the two first months of artificial hibernation (75 ng/ml). The circulating levels of corticosterone then decreased gradually. At the time of transfer to warm conditions, a third significant peak of corticosterone was observed (up to 82 ng/ml). The minimal values (15 ng/ml) previously described during vitellogenesis were reached within 1 week. High corticosterone levels appeared to be actually related to the "hibernation state" since they were also observed in hibernating males and not in nonhibernating females. In order to explain the pattern of plasma corticosterone, variations of adrenal sensitivity to synthetic ACTH 1-39 were examined in vitro, using a perifusion system technique. Surprisingly, ACTH-induced stimulation of corticosterone and aldosterone release was significantly reduced during hibernation, whatever the temperature of the perifusion bath (30 or 6 degrees). Nevertheless, a fourfold increase in the half-life of injected tritiated corticosterone was observed during hibernation which likely contributes to maintain high levels of corticosterone despite a low production rate of the hormone.

Neuroendocrine, behavioral, and morphological changes associated with the termination of the reproductive period in a natural population of male rough-skinned newts (Taricha granulosa)

Male rough-skinned newts (Taricha granulosa) were collected from the same natural population every second week from early April to mid-June. They were either field-tested for their sexual responsiveness or used to measure the plasma concentrations of androgens and corticosterone, the brain concentrations of immunoreactive (ir) gonadotropin-releasing hormone (GnRH) and arginine vasotocin (AVT), and morphological parameters. During the experimental period, the percentage of sexually responsive males gradually declined from 100 to 4%, concurrent with a decrease of plasma concentrations of androgens, but not corticosterone. Concentrations of irGnRH in two brain regions (medial septum; ventral telencephalon containing the nervus terminalis) did not change significantly during this time. In the infundibulum, irGnRH concentrations increased from the end of May to mid-June, which coincided with an increase in plasma androgen concentrations, a marked increase in testis weights, and a decrease of the proportion of males with spermatozoa in their vas deferens. During this period, no changes in irAVT concentrations in four brain regions (infundibulum; pars distalis of the pituitary; interpeduncular nucleus; cerebrospinal fluid) were detected, but significant changes were observed for irAVT in the dorsal preoptic area that were not correlated with the seasonal changes in behavior. Also, during this period, there were decreases in mean body weight and tail height, and in the proportion of males with smooth skin and dark nuptial pads. These results are discussed in view of our current knowledge of the endocrine mechanisms that regulate sexual behaviors and secondary sex characteristics in male amphibians.

Adrenal activity in the female lizard Lacerta vivipara jacquin associated with breeding activities

Variations of adrenal activity were studied in captive viviparous females Lacerta vivipara, in relation to breeding activities. The study was restricted to the period of active life which includes both the phase of annual reproduction and a phase of sexual inactivity. Significant seasonal changes in plasma corticosterone levels were measured with a peak during the second half of gestation followed by an abrupt fall at parturition. No significant variations in plasma aldosterone levels were observed. A limited extraovarian production of progesterone was detected which might be of adrenal origin. The half-life of injected tritiated corticosterone was not longer in pregnant than in nonreproductive females, suggesting that the peak of circulating corticosterone in pregnant females corresponds to an increase in the production rate of the hormone. The functional importance of the pituitary-adrenal axis was demonstrated in vivo: plasma corticosteroid levels dropped to the detection limit after adenohypophysectomy. Seasonal variations of adrenal sensitivity to synthetic ACTH 1-39 were examined in vitro, using a perifusion system. No significant variations were observed throughout the period of active life. These results suggest that the peak of plasma corticosterone during gestation can be ascribed to activation of the pituitary-adrenal axis. Experimental modifications of circulating corticosterone level during late gestation altered the timing of parturition, thus indicating that the fall of corticosterone just before term may be involved in the process of parturition in the female L. vivipara.

Nychthemeral variations of plasma corticosteroids in captive female Lacerta vivipara Jacquin: influence of stress and reproductive state

This work was designed to study nychthemeral variations of plasma corticosterone and aldosterone in captive female lizards Lacerta vivipara. In preliminary experiments, the possible alterations of plasma corticosteroids by various stress factors were researched. A prolonged blood sampling (up to 8 min) did not alter plasma corticosterone levels but a significant increase of plasma aldosterone levels was observed. Confinement (1 or 18 hr) in small individual cages before blood collection resulted in a significant increase of both corticosterone and aldosterone. Whatever the period investigated (vitellogenesis, gestation, 2 months after parturition), plasma corticosterone levels showed a unimodal daily rhythm correlated with the activity of the females in the laboratory. No shift of the peak was observed according to season but the mean minimal and maximal levels were lower during vitellogenesis than during the other periods tested. Nychthemeral variations of plasma aldosterone levels were similar to those of corticosterone but of lower amplitude. Adrenal response to a short confinement (less than 1 hr) before blood sampling varied during a 24-hr period (period tested: vitellogenesis). Only minimal levels of corticosteroids were significantly increased. The possible effects of a long duration of captivity under optimal thermal conditions are discussed.

Reproductive cycle, thyroxine and corticosterone in females of the giant swamp frog Dicroglossus occipitalis at the equator

Females of the giant swamp frog Dicroglossus occipitalis were captured in the tributaries of the Zaïre River near Kisangani, a town situated close to the equator (00 degree 31' N) with virtually constant conditions as to photoperiod, temperature, precipitation, and relative humidity. An annual variation in the reproductive capacity of this species could be demonstrated. A reproductive quiescence with a low gonadal somatic index (GSI), undeveloped eggs, and minimal plasma concentrations of estradiol-17 beta (E2) and progesterone (P4) was found from October to January. E2 and P4 and thyroxine (T4) contents of the thyroids were positively correlated with GSI and egg size. Plasma T4 concentrations, however, were negatively correlated with GSI, whereas variations in corticosterone correlated only with E2. No correlation between plasma concentrations of glucose and GSI or all other hormonal parameters could be found. It is concluded that during egg maturation, increased amounts of E2 and P4 are produced. The observed variations in thyroidal and plasma T4 may be the result of an activation of thyroid function by the peripheral conversion of T4 into triiodothyronine.

Stress-induced inhibition of reproduction: evidence of suppressed secretion of LH-RH in an amphibian

Male rough-skinned newts (Taricha granulosa) were used to investigate the hormonal responses associated with stress-induced inhibition of reproduction. When male newts were confined for 1 hr, using a procedure that previously had elicited physiological stress responses, androgen concentrations decreased in the plasma and immunoreactive (ir) LH-RH concentrations increased in the infundibulum and rostral hypothalamus. Likewise, when male newts were injected with 25 micrograms of corticosterone, androgen concentrations decreased and hypothalamic irLH-RH concentrations increased. These data, which are from experiments in February, support the hypothesis that in this amphibian, exposure to acute stress or to exogenous corticosterone can suppress plasma androgen titers by inhibiting the release of LH-RH from the hypothalamus. The effects of the confinement procedure and the injection of corticosterone on the concentrations of irLH-RH and androgens were different for newts in September than for newts in February.

Effects of permanent deafferentation of the anterior preoptic area on serum aldosterone levels in the crested newt (Triturus cristatus carnifex Laur.)

By cutting off the fiber systems running along the medial forebrain bundle of the urodele amphibian Triturus cristatus, a wide deafferentation of the preoptic area was evoked. This operation elicited a decrease in aldosterone serum level, probably through a reduction of ACTH secretion. At present we are not able to ascertain whether such reduction was prompted by changes in the hypothalamic production of the neurohypophysial hormones or the corticotropin-releasing factor (CRF).

Seasonal changes in luteinizing hormone-releasing hormone concentrations in microdissected brain regions of male rough-skinned newts (Taricha granulosa)

Luteinizing hormone-releasing hormone (LHRH) concentrations were measured in specific brain areas of male rough-skinned newts collected from a single population throughout the reproductive cycle. Plasma androgen and corticosterone (B) concentrations were also measured. Androgen concentrations were highest during the breeding season (winter) and lowest during the summer. In contrast, plasma B was lowest during the breeding season and highest in the summer. Concentrations of LHRH in the infundibulum (I) and rostral hypothalamus (RH) were positively correlated throughout the reproductive cycle; LHRH was always higher in the I than in the RH. Concentrations of LHRH in the ventral preoptic area (POA) fluctuated independently of concentrations of LHRH in the I and RH. However, LHRH concentrations decreased to undetectable levels in all three brain areas in March and April, which was before the end of the breeding season and before plasma androgen concentrations had decreased. An injection of LHRH into postbreeding males resulted in a significant increase in plasma androgen concentrations, indicating that the pituitary-gonad axis was still functional at the end of the breeding season. These results support the hypothesis that an abrupt decline in LHRH secretion is the initial endocrine event that signals the end of the breeding season in this amphibian.

Changes in hormone levels in relation to breeding behavior in male bullfrogs (Rana catesbeiana) at the individual and population levels

Fluctuations in plasma androgen (testosterone and 5 alpha-dihydrotestosterone), corticosterone (B), and luteinizing hormone (LH) of male bullfrogs in central California were measured during the spring mating season. Androgen and LH levels generally increased in the population prior to the initiation of chorusing and establishment of territories, whereas plasma B peaked in a 2-week period at the start of heavy chorusing; this coincided with a transitory, 1 week decline in androgen. Individual males showed fluctuations in plasma androgen and LH levels throughout the breeding season, often within 1 day, but there was no clear correlation between changes in the two hormones. No time of day effect was apparent on any of the hormones. B and androgen levels were significantly but weakly correlated, (r = 0.35) but LH and androgen were not. Although the seasonal trend of increasing androgen corresponded with the start of intense chorusing (and presumably sexual activity) by the population, behavior of individuals and their circulating androgen levels did not correlate. After acquiring territories, males showed no overall trend of increased plasma androgen. Moreover, males that showed no vocal or territorial behavior had significantly higher androgen and lower B levels than calling males. Increased B levels suggest that territorial behavior and especially direct agonistic encounters represent stresses that could have an inhibitory effect on androgen secretion.

Circannual changes in plasma aldosterone levels in Bufo japonicus formosus

Plasma aldosterone levels of Bufo japonicus formosus captured in their natural environment were determined each month throughout 1 year by radioimmunoassay. Low values were recorded in winter and spring, and the maximum occurred in July or August. High summer levels of aldosterone and increases in live weight were correlated. This phenomenon could reflect the gluconeogenetic action of aldosterone. In females but not in males, relatively high levels of plasma aldosterone were also observed in autumn; the relatively high levels of estrogen during this season may stimulate interrenal activity. This could be important to provide additional energy for ovogenesis which occurs in the autumn. On the other hand, high levels of androgen in males may reduce interrenal activity. Aldosterone seems to play no direct role in emergence from water after aquatic breeding activity in spring, as plasma levels showed no significant change during this season.

Seasonal and stress related changes in plasma gonadotropins, sex steroids, and corticosterone in the bullfrog, Rana catesbeiana

Studies of seasonal gonadal cycles combined with direct measurements of plasma levels of the two gonadotropins (FSH and LH), several gonadal steroids (estrogen, E; progesterone, P; testosterone, T; and 5 alpha-dihydrotestosterone, DHT), and the interrenal steroid corticosterone (B) were made in male and female bullfrogs in central California over a 5-year period (between 1976 and 1981). During the course of these studies, we discovered that levels of plasma gonadotropins and steroids are highly labile and particularly sensitive to the effects of captivity, especially in males. In animals captured and sampled repeatedly in the field over a 3-day period, hormone levels remained constant, but if held in collecting sacks, gonadotropin and gonadal steroids began to drop within 2--4 hr and usually reached "baseline" levels within 20 hr. This effect was apparent in all seasons, except occasionally in early spring when hormone levels dropped by only about half. Hormone levels were also generally depressed by the time commercially collected frogs reached local supply houses. Plasma B increased within 30 min of capture and remained high for days in captive animals. Blood samples taken from several hundred animals at the time of capture reveal pronounced seasonal cycles in all hormones measured. These changes are discussed in connection with other gross changes in gonadal condition and with regard to possible interactions among gonadotropins and steroids. Both sexes show a general elevation of hormone levels in spring and early summer, but the sexes differ somewhat both in timing and in magnitude of the changes, as well as in the nature of the dominant steroids. Pronounced "surges" in gonadotropins are evident around the time of gamate release in bot sexes, but the temporal pattern of these surges is not the same for ovulation and spermiation; an elevation in plasma P is associated with the periovulatory surge in gonadotropins. Results were not entirely consistent with expectation of pituitary--gonadal relationships. Levels of plasma gonadotropins and steroids did not show the reciprocal relationship expected from a simple negative feed-back between gonadal and pituitary secretion, nor did changes in gonadotropins and gonadal activities show the consistent positive correlation expected from a direct dependence of gonadal function on circulating gonadotropins. In females, plasma T, but not E, correlated with ovarian growth. Plasma T in females reached much higher levels than in males, but DHT was higher in males. Androgens were generally elevated during the period of sexual activity in males, but absolute levels did not correlate well with individual differences in sexual behavior. Thus, seasonal changes in testicular and ovarian activities cannot be accounted for solely by seasonal cycles in circulating gonadotropin levels.