As the cold war ideology that praised heterosexuality and ranted about the homosexual menace came to dominate the national scene during the 1950s, more restrictive readings of animal sexuality gained visibility and strength. By 1959, a new rodent emerged that was distinctly heterosexual and far more bound by gender roles than were Beach’s rats. A new theory implied that indi­vidual variation resulted from early hormone exposure;76 it also made little attempt to develop the kind of integrated account of behavior so evident in Beach’s work. Instead, biologists divorced life experience from biological ex­planations of behavior, leaving it as a kind of embarrassing little sister—al – ways mentioned, but never really included in the big kids’ game. And, finally, as researchers applied Jost’s account of genital development to animal behav­ior, femininity became an absence, masculinity a struggle.

One key figure through whom we can trace this progression is William C.

Young, who obtained his Ph. D. at the University of Chicago for work on sperm transport (from the testis to the outside world). During the 1930s and 1940s, with CRPS funding, Young focused on mating behaviors in the guinea pig.77 His motto was, ‘‘Observe, measure, andrecord!’’ and he did just that.78 He noted the cyclic nature of the female’s mating responses, detailed exactly when during the estrus cycle particular behaviors came and went, and figured out the relationship between cyclical changes in estrogen and progesterone and the waxing and waning of the female mating response. Like the rat, the female guinea pig exhibited lordosis when in heat, ‘‘frequently. . . accompa­nied by a guttural vocalization and by the pursuit and mounting of other fe­males and even males.’’79

Even though the females went ‘‘through the motions of copulation except that they do not afterwards roll dorsally and clean their genitalia,’’ Young and his colleagues displayed a certain ambivalence about this untoward female be – havior.80 On the one hand, they described such mounting as a normal part of the female sex drive.81 On the other, they labeled it ‘‘homosexual behavior by normal females.’’82 In one set of experiments, Young and colleagues found that a combination of estrogen and progesterone induced female mounting. Testosterone—much to their surprise—had little effect.83

Young’s 1941 review of research on mating behavior in female mammals covered much of the same territory as Beach’s 1942 synthesis. Young hesi­tated, however, to posit overarching theories about such complex behavior. ‘‘Endocrine, neural, genetic, ontogenetic, nutritional, environmental, psy­chological, pathological and age factors. . . and doubtless others’’ all com­bined to produce mating behaviors. Figuring out the part played by any partic­ular factor seemed nearly impossible. ‘‘Nevertheless. . . some starting point must be chosen. . . the ovarian hormones have been selected, not because they are necessarily the exclusive limiting factor, but because they are the means to the induction of heat by experimental procedures and the means by which the role of the other factors can be elucidated.’’ Hormones, in other words, were ‘‘the hook,’’ the entry point to the understanding of sexual be – haviors.84

During the first part of his career, Young worked mostly on female guinea pigs, but in 1930 he and his colleagues turned their attention to males. First they carefully described and measured five aspects of male mating behavior: nuzzling, sniffing, nibbling, mounting, intromission, and ejaculation.85 Time and again they observed individual variation in male behavior. Some males had high sex drives, while others seemed hardly interested in mating. Did the low – drive males have less circulating testosterone? No. When these researchers castrated high – and low-drive individuals and then injected them each with the same concentration of testosterone, the differences remained. The guinea pig who had been an enthusiastic stud before castration returned full throttle when given back his missing hormones. But the originally lackluster fellow retained his disinterest even when given high levels of testosterone. Because the amount of circulating hormone didn’t explain the differences, Young pos­tulated that in different animals the tissues mediating sexual behavior must differ in their ability to respond to the hormone.86

How, though, did these mediating tissues come to differ from one male to the next? For several years, Young and his students studied both genetic and experiential factors. Genetic differences between inbred and outbred animals told some of the story. And early social experiences mattered a lot. In some experiments they separated newborns from their siblings—housing them only with their mothers for the first ten to twenty-five days of life and thereaf­ter rearing them in total isolation. In an inbred strain in which the males always have a low sex drive, isolation after twenty-five days of nursing caused drastic drops in sexual performance. In stud strains, weaning at ten days fol­lowed by isolation severely depressed the mating response. The conclusion: ‘‘Contact with other animals has an organizing action on the development of the copulatory pattern of the male guinea pig-’’87

By the late 1950s, Young and his co-workers had completed exhaustive studies of male and female mating behaviors. In the many experiments Young, Beach, and others had performed since the 1930s, hormones seemed to be­have much as Beach had postulated. In one way or another, they could stimu­late the expression of potentials ‘‘previously organized or determined by ge – netical and experiential factors.’’88 But other experiments suggested that hormone exposure early in development might have long-term effects on be­havior, effects not evident until animals reached maturity. The discrepancy between such data and Beach’s theory was unresolved. Young and his col­leagues decided to reopen the question of long-term hormonal effects and in doing so started a new chapter in the history of the manly rat.

In 1959, when cold war rhetoric about homosexuality, communism, and the family was at its peak, Young and his three younger colleagues published their now classic paper ‘‘Organizing Action of Prenatally Administered Tes­tosterone Proprionate on the Tissues Mediating Mating Behavior in the Guinea Pig’’ (hereafter referred to as Young’s 1959 paper). The stakes were high, and they knew it. A finding that prenatal exposure to androgens or estro­gens had ‘‘an organizing action that would be reflected by the character of adult sexual behavior’’ might mean that a whole range of adult behaviors could be traced largely to prebirth hormonal chemistry. It would also suggest a par­allel between the importance of hormones for behavior and their importance for anatomical development. Finally, positive results would ‘‘direct attention to a possible origin of behavioral differences between the sexes which is ipso facto important for psychologic and psychiatric theory.’’89

This last comment, which referred to John and Joan Hampson’s work on the development of sexual difference in humans, carried a subtle but impor­tant signal. Remember (from chapter 3) that during the 1930s the Hampsons and John Money had studied the development of human intersexuals raised as either males or females. Unlike Beach, who clearly accepted human homosex­uality as part of a natural range of sexual behavior, the Hampsons considered homosexuality and transvestism to be abnormal.90 By citing their work, Phoe­nix et al. implied their disagreement with Beach’s view of underlying bisexu­ality, while at the same time suggesting that the guinea pig studies might lead to the finding of a biological basis for homosexuality.91

The publication of Young’s 1939 paper shaped the study of hormones and sexual behavior for decades to come. The writers proposed a theory—the organizational/activational (O/A) model of hormone activity—that rele­gated Beach’s 1942 synthesis to the back drawers of history. What did Young and his colleagues find? What was their initial statement of the O/A model? How did the heterosexual rodent—the manly guinea pig or the feminine rat—replace the bisexual animal as the center of attention?

Young and his colleagues suggested that pre – or perinatal hormones orga­nized central nervous tissue so that at puberty hormones could activate specific behaviors. They injected pregnant guinea pigs with testosterone.92 The in­jected mothers produced female intersexes (which researchers called her­maphrodites in the article). All of the testosterone-exposed offspring had inter­nal anatomical signs of masculinization. Some also developed masculinized external genitalia. When these females grew up, they took longer to come into heat after stimulation by injections of estrogen and progesterone. Their lordosis responses were much weaker than unexposed controls, and ‘‘the low gutteral growl which is so characteristically a part of the pattern of lordosis in normal females was commonly, and in some individuals always, lacking.’’ They also vigorously mounted other guinea pigs when injected with testoster­one. Except for the growl emitted by the female during lordosis, quantity, not quality, distinguished feminine from masculine. In one experiment, for example, 89 percent of control females came into estrus after spaying and hormone injection, compared with 63 percent of the prenatally treated fe­males with normal external genitalia, 22 percent of externally masculinized females, and 38 percent of castrated males (a second type of control group).93 A lack of estrus, longer latency to achieve estrus, shorter length of heat, shorter lordotic response, and mounting in the absence of estrogen/proges­terone injection all signaled a decrease in femininity and an increase in mascu­linity. Masculinity and femininity became mutually exclusive. An increase in one implied a decrease in the other.

Young and his colleagues began by studying masculinized female guinea pigs, but soon turned to the feminization of males. Following Jost’s presence/ absence logic, they reasoned that if adding testosterone imposed masculinity, removing it should permit underlying femininity to emerge. They castrated young rats or rabbits ‘‘prior to completion of the organizing action of the androgen,’’ and in adulthood injected a mixture of estrogen and progesterone in an effort to elicit ‘‘feminine behavior in response to mounting by intact males.’’ They found that males castrated before the age of ten days showed a higher frequency of feminine behavior, defined in rats as estrus and lordosis, ear wiggling, darting, and crouching. Castration affected male lordosis more dramatically than the wiggling, darting, and crouching, suggesting that not all aspects of rat femininity were similarly organized.94

What made Young’s 1959 paper special was not the particular results; he and other colleagues had published similar findings a full nineteen years ear­lier, and Beach was getting similar data in dogs.95 Rather, it was the scientists’ explanation of their findings that proved important. Did exposing embryos to sex hormones, these authors wondered, affect the neural substrates underly­ing sexual behavior, substrates they assumed were found in ‘‘central nervous tissues’’?96 If so, might fetal hormones permanently fix an individual’s behav­ioral potential as either masculine or feminine? The authors drew heavily on Jost’s work describing how, in the embryo, testosterone promoted the differ­entiation of male genitalia, while Mullerian Inhibiting Substance caused the female parts to disintegrate. In adulthood, the ovaries or testes, the uterus or epididymis, all responded to the hormones of puberty. This second response was functional, rather than developmental. Young and colleagues thought that something similar must happen to ‘‘the neural tissues mediating mating be­havior.’’ In the embryo, such tissue differentiated, or ‘‘organized’’ in ‘‘the di­rection of either masculinization or feminization,’’97 while in the adult, hor­mones ‘‘activated’’ the previously organized tissues.

The ideas developed in the 1959 paper extended Jost’s account of hor­mones and anatomy to behavior. Prenatal testosterone ‘‘heightened’’ the ‘‘re­sponsiveness’’ to adult testosterone, while simultaneously suppressing the ability ‘‘to display the feminine components’’ after estrogen/progesterone treatment. Testosterone, the researchers theorized, played a dual role. First, it heightened masculinity by increasing the frequency of mounting behavior. Second, it suppressed femininity by decreasing the frequency and duration of lordosis. Estrogen and progesterone played roles in the adult as hormonal

activators. The unstated implication: feminine behavior underlay all develop­ment. Testosterone suppressed it and imposed masculine capabilities on an underlying feminine system.98

Young’s group drew further on the anatomical analogy to refute Beach and others’ earlier claims of adult bisexuality. ‘‘These investigators stressed the increased responsiveness of their masculinized guinea pigs and mice’’ to in­jected androgens, seeming ‘‘to regard the change as the expression of an in­herent bisexuality. . . . The existence of bisexuality is assumed. We suggest, however, that this adult bisexuality is unequal in the neural tissues as it is in the. . . genital tissues.’’99 Even though it was possible to elicit cross-gender behavior in adults, it was hard to do so. Again drawing the analogy with genital anatomy, they noted that both males and females contained vestiges of organs present in the embryo, that such organs could respond to adult hormones, but that the responses of the vestige and the fully formed organ were rarely identical. By extending the anatomical model of hormone action to behavior, the authors acknowledged the widespread existence of cross-gendered behav­iors but downgraded their importance, paving the way for a biological account of male and female as heterosexual.100

Boldly, they proposed that their findings might extend well beyond the highly stylized reproductive behaviors on which they had gathered data.101 Rejecting the psychologists’ arguments about ‘‘shaping behavior by manipu­lating the external environment,’’ they proposed that all behavior patterns had underlying biological causes. In this instance, they had demonstrated that testosterone ‘‘acts on the central nervous tissues in which patterns of sexual behavior are organized.’’102