Sex differences in the play behavior of immature spotted hyenas, Crocuta crocuta.

Social, locomotor, and object play were studied in a colony of five male and five female peer-reared spotted hyenas during 12 1-hr tests while the animals were 13-19 months of age. Animals were tested in both same-sex and mixed-sex groups and were stimulated to play by the introduction of fresh straw and sawdust bedding. Each test was videotaped and the frequency of each type of play was determined by a time sampling procedure. Females played more frequently than males, however, the category of play which was elevated depended upon the social context during testing. In same-sex tests the frequency of vigorous social play displayed by females markedly exceeded that by males, but no comparable sex difference appeared in mixed-sex tests. Females engaged in locomotor play more frequently than males in mixed-sex tests, but no comparable sex difference appeared in same-sex tests. No sex difference in object play was observed. Two male and two female hyenas were gonadectomized prior to the initiation of the present sequence of tests. The results suggest that gonadectomy during the prepubertal period does not affect the frequency of play behavior. However, the small sample sizes preclude any conclusive determination of the effects of these gonadectomies on play.

Effects of testosterone on ontogeny of urinary behavior in male and female dogs.

Development of urinary behavior from birth to adulthood was observed in six groups of beagles including normal males (NM), normal females (NF), males castrated soon after birth (CM), males castrated soon after birth and treated with testosterone (T) for the next 90 days (CMT), females exposed to T in utero (FTU), and females exposed to T in utero and during infancy, i.e., the first 30-40 days postpartum (FTUI). Prenatal treatment with T had masculinizing effects on juvenile urinary behavior in FTU and FTUI. On the other hand normal development of fully adult masculine urinary patterns in males and females necessitated both prenatal and postnatal androgenic stimulation. It was not necessary that T be present at the time the overt behavior developed. For example, adult male behavior appeared in FTUI at the same time as in NM, i.e., 6-10 months, although the supply of exogenous androgen in FTUI had been exhausted within 30-40 days after birth. CMT showed precocious development of all components of the adult male pattern. Development of adult responses was markedly retarded in most CM, and their performance did not equal that of NM at 23 months. They were then injected with TP which promptly evoked completely normal male urinary behavior. It is tentatively concluded that T acting before birth and during the juvenile period "prepares" critical CNS mechanisms so that when general maturation reaches the appropriate point adult male behavior develops. Although the preparatory role of T is essential, the behavior is not dependent on T after it has developed.

Hormonal modulation of genital reflexes in male and masculinized female dogs.

When testosterone propionate (TP) is administered to adult, neonatally castrated male dogs and to adult females with masculinized genitalia produced by prenatal and neonatal exposure to androgen, both types of animals are unsuccessful in their attempts to copulate with receptive females. They mount and thrust vigorously but do not achieve intromission and establish a copulatory "lock." The deficit could be due to incomplete "organization" of neuromuscular mechanisms mediating erection or to the fact that the penis of such dogs are abnormally short. In this experiment, neonatally castrated males and genitally masculinized females were tested before and after TP treatment for responses to manual stimulation of the genitalia. An additional experimental group consisted of males castrated as adults, and there was a control group of normal males. After a series of TP injections, neonatally castrated males, adult castrates, and genitally masculinized females exhibited complete and strong erectile and ejaculatory reflexes. Erect penis lengths of neonatally castrated males and masculinized females were significantly shorter than those of normal males or of males castrated as adults. It is tentatively concluded that the failure of males castrated at birth, and of genitally masculinized females, to insert and lock when mounting receptive females is due to incomplete penile development and not to incomplete "organization" of spinal reflex mechanisms. However, because artificial stimulation was employed, results of this study do not prove that central nervous system mechanisms necessary for insertion and locking in copula were normally developed.

Development of attraction to estrous females in male dogs.

In simultaneous choice tests male beagles were allowed to visit a caged female in estrus, or caged, spayed female not in estrus. Males were tested periodically from 1-3 to 22-24 months of age. The 3 subject groups were normal males (Group N), males castrated 4-7 days postpartum and injected with testosterone propionate (TP) until they were 3 months old (Group TPTC), and males castrated 4-7 days but given no hormone treatment, i.e., untreated castrates (Group UC). A statistically reliable preference for visiting the estrous rather than the nonestrous female first appeared in N males at 4-6 months, in TPTC males at 1-3 months, and in UC males at 10-12 months. In N males attraction to the estrous female (measured by time spent visiting her) increased progressively from 4-6 to 16-18 months. In the same period concentration of plasma testosterone rapidly increased, reached a peak at 10-12 months, and then declined. UC males exhibited no significant increase in attraction to the estrous female from 10-12 months (when a preference first appeared) to 22-24 months. They then received 10 injections of TP after which their visiting time to the estrous female was equal to that of N males. TPTC males exhibited a precocious preference for the estrous female at 1-3 months, while they were receiving TP, and there was no decrease in strength of attraction in the following 9 months during which time no hormone was administered. Between 10-12 and 13-15 months, still without exogenous androgen, visiting to the estrous female began to increase and continued to do so until 19-21 months.

Sexual cycles in female dogs treated with androgen during development.

This experiment was conducted to study hormonal and behavioral cycles in female dogs exposed to androgen during development. The four groups, each consisting of five subjects, were "Normal" females, "Prenatal" females exposed to testosterone propionate (TP) before birth, "Postnatal" females injected with TP for 3 months postpartum, and "Prepost" females exposed to TP in utero and again to testosterone for 4--6 weeks after birth. All Normal females had two estrous cycles during which they showed vaginal bleeding, ovulation, progesterone (P) secretion, sexual receptivity, and were attractive to males. All androgen-treated females showed at least 1 cycle with ovulation and P secretion. Three Postnatal females developed the delayed anovulatory syndrome. Genital bleeding through the vagina (Postnatal group), or through the "penis" (Prenatal and Prepost groups), occurred in 4 Postnatal, 3 Prenatal, and 1 Prepost female. Attractiveness for males was observed in all Prenatal and Postnatal females, but in no Prepost subject. Receptivity was present in four of five Prenatal females but not in any Postnatal or Prepost female. Occurrence of cycles in treated females indicated functional integrity and sensitivity to estrogen in brain mechanisms producing gonadotropin-releasing hormones. Absence of receptivity is referred to diminished responsiveness to ovarian hormones in brain mechanisms mediating receptive behavior. Absence of genital bleeding and lack of attractivity are due to lowered response to estrogen in uterine and vaginal epithelium which results in reduced extravasation of blood, and in failure to secrete the putative "pheromonal" agent normally responsible for attractiveness.

Competitive behavior in male, female, and pseudohermaphroditic female dogs.

Male, spayed female, and pseudohermaphroditic female dogs were tested in pairs for competition over a large bone. Pseudohermaphroditic females had been exposed to testosterone proprionate before and immediately after birth. Two types of tests were used. In equal opportunity tests (EO tests), both members of a pair had equal chance to seize the bone when it was tossed into the arena. Each EO test was followed by an established possession test (EP test). During an EP test, the loser of the preceding EO test was given possession of the bone before the former winner was returned to the test arena. Every dog was tested against all other members of its own and both of the other groups. Males won 78 of 100 EO tests against normal females but only 61 of 100 against pseudohermaphroditic females. Pseudohermaphroditic females won 70 of 100 EO tests against normal females. When either normal or pseudohermaphroditic females were given possession of the bone in EP tests they were equally ineffective in defense against male challengers, losing 75% and 71% of their tests, respectively. Both types of females were much more successful in defending possession during EP tests against each other than they had been against males. When normal and pseudohermaphroditic females were treated with testosterone propionate, their performance in competition tests was not altered. Androgenic stimulation before and immediately after birth permanently modified the social behavior of pseudohermaphroditic females as well as the behavior of males and normal females toward the experimental dogs. However, the changes could not appropriately be described as "masculinization." Developmentally androgenized females differed from members of their own sex, but they differed just as greatly from normal males.

Animal models for human sexuality.

The value of animal models in biomedical research is firmly established, and many basic principles of human psychology have been explicated as the result of comparative studies. There is pressing need for non-human models in the behavioural sciences as represented by psychiatry, psychology and ethology; and such models should be constructed, provided their validity can be assured. Valid models cannot be based exclusively on similarity in the formal properties of behaviour. Commonality of descriptive terms as applied to different species does not guarantee identity of the concepts to which the terms apply. Model builders must evaluate interspecific similarities and differences in the causes, mediating mechanisms and functional outcomes of behaviour. The validity of interspecific generalization can never exceed the reliability of intraspecific analysis; and the latter is an indispensable antecedent of the former. Existing and potential models for homosexuality and other psychosexual characteristics of human beings are evaluated within the perspective provided by the foregoing generalizations.

Hormonal control of sexual attraction in pseudohermaphroditic female dogs.

Groups of female dogs exposed to different degrees of androgenic stimulation during development and a control group of ovariectomized females were tested for their attraction to tethered male and female stimulus animals. Attraction to the male was measured before and after administration of estradiol, and attraction to an estrous female was tested before and after administration of testosterone propionate (TP). Time spent visiting the tethered male was approximately equal for all groups prior to hormone treatment, but after receiving estradiol, control females exhibited a pronounced increase in visiting time; the second longest visits were paid by females that had received moderate amounts of androgen before birth; more heavily androgenized females exhibited no increase in attraction to the male despite estradiol injections. Visits to the estrous female before administration of TP were longer for some groups than for others, but there was no relation between the degree of perinatal androgenization and mean visiting time. After injections to TP the most pronounced increase in visiting was shown by females that had received the largest amounts of androgen during development, and the second largest increase occurred in the prenatally androgenized group. Control females showed the smallest increase in visiting time.

Male rats with inherited insensitivity to androgen show reduced sexual behavior.

Pseudohermaphroditic Stanley-Gumbreck male rats showed infrequent and incomplete copulatory responses to receptive females. Administration of testosterone propionate produced no increase in this behavior. Injections of estradiol and progesterone induced the pseudohermaphrodites to exhibit lordosis when mounted by stimulus males, but feminine responses were no more frequent than those of normal males given the same hormonal treatment. The hypothesis is suggested that early in development sufficient endogenous testis hormone is produced to cause normal desensitization of feminine behavioral mechanisms to estradiol, but that mechanisms for male behavior are not normally sensitized to testosterone.

Variables affecting "spontaneous" seminal emission in rats.

The frequency of spontaneous seminal emission (SE) by isolated rats (X = 2.05 +/- 0.64/day) was not increased when estrous females were housed in adjacent cages. Coital ejaculation during mating tests temporarily inhibited the production of SE, and inhibition was more marked after 5 than after 1 ejaculation with a female. Males experiencing only 1 coital ejaculation tended to resume production of SE somewhat sooner if they were exposed to stimuli from estrous females than if such stimuli were lacking. The possible functional significance of spontaneous and self-induced noncoital seminal emission which is common in a variety of mammals is discussed.