¿ìè¶ÌÊÓƵ

Neuroanatomy used to be a nice, quiet discipline. When there were squabbles,
they usually involved no more than two scientists armed with microscopes,
arguing over the precise pattern of connections that they had traced in
a corner of the brain. But last year the academic calm was shattered when
Simon LeVay, a neuroscientist working at the Salk Institute in San Diego,
published research from which he concluded that the brains of homosexual
and heterosexual men were anatomically different.

The research made headlines around the world, prompting a question of
enormous public interest: are sexual orientation, and other characteristics
that are supposedly ‘masculine’ or ‘feminine’, fixed in the developing brain
before a child is born, or are they determined by experiences in childhood
and adolescence?

So far there is no sign of a definitive answer. LeVay argues vigorously
that his research shows that homosexuals are born not made – that sexual
orientation is established in the womb as a result of the action of hormones
on the brain of the developing fetus. And LeVay, who is homosexual, thinks
that conclusion is good news for gays: ‘It’s one more nail in the coffin
of critics who argue that homosexuality is a choice and thus immoral,’ he
told the press earlier this year when commenting on the latest report of
anatomical difference between the brains of homosexual and heterosexual
men. Roger Gorski and Laura Allen, of the University of California Medical
School in Los Angeles, had discovered that a ‘cable’ of nerve fibres connecting
the two halves of the brain was larger in gay men that in heterosexuals.
‘Further progess will require the identification of the genes that influence
sexual orientation,’ said LeVay. Already some biologists claim to have traced
a gene linked to homosexual behaviour in the fruit fly.

But other researchers don’t necessarily agree with LeVay. Psychologists
have lined up on both sides of the debate and philosophers have attacked
the logic of the argument that sexual orientation is fixed before birth.
Not all LeVay’s fellow neuroanatomists agree with him, either. Dick Swaab,
of the Netherlands Institute for Brain Research in Amsterdam, found anatomical
differences between the brains of homosexual and heterosexual men two years
before LeVay. Unlike LeVay, he argues that the sexual differentiation of
the human brain is not fixed at birth, but continues for several years afterwards.
That means, he says, that social factors, not just chemical and hormonal
ones, can influence an individual’s sexual orientation.

The debate is of more than academic significance. If LeVay is right,
it will no longer be tenable to regard homosexuality as freely chosen and
therefore ‘sinful’. However, it might instead be seen as a biological defect,
which can be diagnosed prenatally and ‘cured’. One researcher in the field,
Gunther Dorner, director of the Institute of Experimental Endocrinology
at Humbolt University in what was East Berlin, already suggests that women
bearing male fetuses should have hormone injections to guard against the
risk of having a homosexual son.

There are also important implications for heterosexuals. Psychologists
such as Sandra Witelson of McMaster University in Hamilton, Ontario, argue
that many behaviours which are regarded as ‘masculine’ or ‘feminine’ are
not the result of social experience, as most people believe, but are determined
by hormonal exposure before birth. Taking this philosophy to its extreme,
it implies that women who put a career before motherhood may have brains
that were ‘masculinised’ by exposure to male hormones in the womb.

To understand how the debate over the links between the prenatal environment,
hormones and sexual orientation has taken shape, it is necessary to go back
to the 1940s, to the roots of the idea that brains acquire their ‘sex’ before
birth.

At first, researchers concentrated on physiology, and in particular
on the hormonal factors that controlled whether embryos developed male or
female genitals. Experiments on pregnant laboratory rodents soon showed
that to develop male genitals, a genetically male fetus needs a supply of
androgen hormone. Without this hormonal stimulation, individuals of either
genetic sex develop female internal and external genitalia. Fortunately
for the would-be male fetus, the genes it inherits normally produce the
necessary androgen, in the form of testosterone. This ensures that it develops
testes, which in turn secrete the hormone needed to complete development
of the reproductive equipment of a male.

MOUNTING EVIDENCE

Links between hormones, brain and behaviour came next. By analogy, researchers
speculated that a person’s sexual behaviour could be determined, just as
the sex organs are, before birth, by hormones acting on the fetus’s brain
as it develops in its mother’s womb.

The classic experiment linking sexual behaviour and prenatal hormones
was published in 1959 by Charles Phoenix and his colleagues at the University
of Kansas. They injected testosterone into pregnant guinea pigs, and produced
genetically female offspring which not only developed a penis, but behaved
sexually rather like males. They were more likely to mount other guinea
pigs than untreated females, and were less likely to adopt the female sexual
posture, an arching of the back known as lordosis.

Phoenix concluded that the infant females’ brains, not just their bodies,
had been irreversibly ‘masculinised’ by his hormonal manipulations before
they were born. Prenatal exposure to androgens has an ‘organising action’
on the brain, he claimed, permanently altering adult sexual behaviour.

At the time, scientists readily accepted this ‘organisational’ theory.
And, by analogy with the earlier experiments on the differentiation of the
genitalia in the womb, the hormonal programming of the brain was deemed
to be irreversible. Later in life, hormonal changes – such as a surge of
testosterone at the onset of sexual maturity – could only ‘activate’ the
prenatally imprinted programme of sexual behaviour.

In rats, researchers found that the critical period, when sexual behaviour
is imprinted in the brain in this way, lasts from a few days before birth
to a few days after. During that time, the story goes, the testosterone
surging out of the fetal testes makes its way to the brain, where it is
converted to oestradiol. This compound, the most potent of the oestrogen
hormones, then ‘masculinises’ the rat’s brain. (It is one of the ironies
of this research that the masculine brain is apparently preserved only through
the intervention of a supposedly ‘female’ sex hormone; but in fact both
sexes make all the ‘sex’ steroids, albeit in different ratios.)

Following this line of thought, researchers went on to look for specific
parts of the brain where there were differences between males and females.
Their first successes came in the early 1970s when Geoffrey Raisman and
Pauline Field then at the University of Oxford found that the sexes differ
in the number and pattern of synaptic connections in a part of the hypothalamus
known as the preoptic area – an area known to be vital to the sex life of
male rats, and the oestrus cycle of females. The researchers could reverse
these differences by treating newborn females with testosterone or by castrating
newborn males.

Solid links seemed to be emerging between rats’ prenatal exposure to
hormones and their sexual behaviour. But the picture looked less clear when
researchers turned to humans. In 1988, Swaab and his colleagues discovered
that humans, too, have a region of the brain known to differ between the
sexes, a cluster of nerve cells known as the sexually dimorphic nucleus
(SDN). But the researchers established that the differences arise after
birth, not in fetal life after all. At birth this nucleus contains only
18 per cent of the cells that will have developed by the time a child is
two years old.

WILTING CELLS

In the first year after birth the number of cells in the SDN rises rapidly,
and remains similar in boys and girls. Only after two to three years of
age does a difference between males and females appear in the SDN. This
difference is the result of a decrease in cell number in the female’s SDN,
but not in the male’s. In men, the number of cells in the SDN remains unaltered
up to the age of 45, when it declines sharply – hence Swaab’s conclusion
that sexual differentiation of the human brain continues after birth, and
so can be influenced by social factors.

Interpreting these findings remains difficult. In rat and human alike
‘the function of this nucleus is not known’, says Dennis Kelly of the New
York State Psychiatric Institute. Nor, he says, does anyone understand the
significance of the anatomical sex differences reported in other regions
of the brain – the superior cervical ganglion, the amygdala, the dorsal
hippocampus and the orbital frontal cortex – in a variety of rodents and
sometimes humans.

So far, the only clear example of a neural difference between the sexes
linked with sexual behaviour lies in the rat’s spinal cord. Adult male rats
possess neurons known as the motor nucleus of the bulbocavernosus, whose
fibres run to the muscles of the penis. Not surprisingly, adult female rats
do not possess the same nucleus.

Without a clear verdict from studies of the brain’s anatomy, other researchers
have looked for links in humans between prenatal hormones and sexual orientation
by turning to ‘experiments of nature’ – people born with endocrine disorders
such as androgen insensitivity syndrome and congenital adrenal hyperplasia
(CAH). Research into these conditions has, however, inspired many alternative
interpretations of the data.

Children with adrenal insensitivity syndrome are genetically male, but
are born with female or ambiguous external genitalia. When reared as girls,
they are usually not discovered until they fail to menstruate. They have
abdominal testes that produce testosterone, but their tissues are completely
insensitive to it. They can convert testosterone to oestradiol, however,
and in amounts sufficient to develop breasts at puberty. Given that it is
oestradiol, not testosterone, that masculinises fetal male rat’s brains,
shouldn’t these children’s brains have been ‘masculinised’ in the womb?

Two American researchers, John Money at Johns Hopkins Hospital in Baltimore
and Anke Ehrhardt of Columbia University in New York, have studied these
androgen-insensitive children since the 1950s. They discovered that the
affected individuals develop a female gender identity and a sexual orientation
towards men – no one could fault their ‘femininity’. This suggests that
upbringing is more important than prenatal hormone exposure in determining
sexual identity.

CAH is an enzyme defect, which results in most of the steroid hormone
produced by the adrenal cortex being transformed from corticosteroid into
androgen. Either genetic sex can have this syndrome: infant girls with CAH
are born with large clitorises sometimes mistaken for penises and usually
surgically altered in later life; boys with CAH are anatomically normal.
All people with the condition require lifelong treatment with cortisone
to compensate for their nonfunctioning adrenal glands.

Ehrhardt and Money found that females with CAH develop female gender
identity if brought up as girls. But the researchers said most of them exhibited
‘tomboyism’ – defined as having an preference for outdoor active play rather
than indoor, less active play. These girls also showed relatively more interest
in a public career than in being a housewife, and less interest in tending
small infants and playing with dolls, than ‘normal’ young females. In a
later study, Money and Ehrhardt reported that 37 per cent of these girls
with CAH rated themselves as homosexual or bisexual. Those genetically female
children brought up as boys had a male gender identity and apparently conventional
sexual orientation.

Critics of this research have highlighted various shortcomings. One
difficulty is that observations about the girls’ behaviour came from parents
and teachers who knew of their abnormal physiological condition, and from
the girls themselves. Hence these supposedly objective descriptions could
have been biased by their expectations. The children certainly did not have
a ‘normal’ childhood. Some did not have plastic surgery on their genitals
until they were seven, and most had further vaginal surgery at adolescence.

In a striking study in 1984, a Dutch researcher, Froukje Slijper, gave
psychological tests designed to measure ideas of gender among children with
CAH as well as a group of young diabetic patients. There was no difference
between the two groups. Slijper suggests that this may be related to their
experience of chronic illness and hospitalisation, and that both groups
had rebelled against the intrusion of medical authority into their lives.
She also argues that these children were more aware of their situation and
more insecure about themselves than the average child.

There are further reasons to be circumspect about concluding from these
data that any girl who likes to play outdoors must have been masculinised
by prenatal exposure to androgens. As the late Ruth Bleier, a neurobiologist
at the University of Wisconsin at Madison, pointed out, all this work
accepts at face value ‘the idea of tomboyism as an index of a characteristic
called ‘masculinity’, presumed to be as objective and innate a human feature
as height and eye colour’. An alternative view is that masculinity is a
cultural construct which has changed historically as ideas about what is
appropriate to a particular sex have altered. According to Bleier, the
uncritical acceptance of such studies ‘places the stamp of science on a
set of unexamined social values and judgments concerning gender’.

An ‘experiment of medicine’ has added to the controversy. Between 1940
and 1971, American doctors gave the oestrogenic drug diethylstilboestrol
(DES), to several million pregnant women in the mistaken belief that this
would reduce their risk of miscarriage. This treatment resulted in a much
higher incidence of certain cancers of the reproductive tract in the exposed
children. Studies on rats had suggested that prenatal DES should act like
androgens and ‘masculinise’ females. So American psychiatrists were keen
to know whether the drugs had also altered their children’s sexual behaviour.

In 1984, Ehrhardt and her colleague Heino Meyer-Bahlburg compared 30
women who had been exposed prenatally to DES with 30 unexposed women who
were referred to the same clinic because they had abnormal cervical smears.
They could find no difference in their ‘psychosexual milestones’ – events
such as age at first boyfriend or first sexual intercourse. After further
study of the same women, they concluded in 1989 that DES-exposed women show
‘less orientation toward parenting’ than the controls, although the differences
were not statistically significant. When the women were asked whether they
would prefer being a mother to having a career, ‘most subjects in both groups
wanted to have both’. Moreover, according to their mothers, the DES-exposed
women engaged in less, not more, rough-and-tumble play during childhood
than the controls did.

The researchers report, however, that more of these DES-exposed women
said they were lesbians or bisexual than did the controls. The fact that
more of the control group were married Catholic women, while the DES-women
tended to be Jewish, unmarried or students did not undermine the researchers’
conviction that sexual orientation was influenced by prenatal hormones.

Research in this field has been consistently dogged by methodological
shortcomings of this kind. Consider, for instance, the difficulties of deciding
what constitutes a genuine ‘control’ group for a hormonal or psychological
study of women who live with other women and have sexual relationships with
them. The usual solution is to compare these women to the wives of the researchers,
who inevitably must lead very different lives.

‘It would be helpful,’ writes June Reinisch of the Kinsey Institute
for Research in Sex, Gender and Reproduction in Bloomington, Indiana, ‘if
researchers employed double-blind designs and had control groups matched
on a wide range of variables.’

Humans are, in any case, so complicated that it may be useless to equate
sexuality with reproductive behaviour. This was certainly the conclusion
reached by Louis Gooren of the Free University, Amsterdam, after carrying
out studies of transsexuals. Physically, transsexuals are demonstrably ‘normal’
men and women, who may have established conventional heterosexual relationships.
Yet they think of themselves as ‘trapped’ in a body of the wrong sex. Something
as complex as this experience, says Gooren, depends on human cognition and
language.

But Witelson is convinced that there are differences in the way men
and women think, and that these differences reflect permanent brain alterations
caused by prenatal exposure to sex hormones. Her key finding, she says,
is that the brains of men and women differ in their ‘cerebral asymmetry’
– the way the two sides of their brains function.

The phenomenon – still a matter of much controversy – is also known
as cerebral dominance, brain lateralisation and hemispheric specialisation.
It is linked by several researchers to the tendency for boys and men to
score higher in tests of mathematical reasoning and spatial relations, and
girls and women to perform better in tests of verbal fluency or the interpretation
of facial expression.

But such work is open to criticism. It ‘floats on a sea of assumptions’,
says Helen Longino, a philosopher at Mills College in California – not
least the unsubstantiated assertion that test scores reflect inherent ability
rather than acquired knowledge and skill.

Doreen Kimura, a psychologist at the University of Western Ontario in
Canada, and a leading researcher in this field, cautions against quick generalisations
about sex differences in the organisation of the human brain. An example
from Kimura’s studies of people with brain damage bears this out. Because
damage to the left hemisphere appears to result in a higher incidence of
speech disorders in men than it does in women, many researchers have concluded
that speech control in men is more heavily concentrated in this hemisphere.
In other words, speech asymmetry is more marked in men than in women.

Kimura, however, has uncovered a different explanation. Women, she has
found, are more likely to suffer speech disorders when the front part of
the brain is damaged (see diagram above). The twist here is that damage
restricted to just one hemisphere is far more common at the back of the
brain. So maybe, argues Kimura, the relatively low incidence of speech disorders
in women with brain damage reflects the fact that the critical area is less
often affected, not that speech control is less asymmetric.

Given the uncertainty surrounding the basic evidence for cerebral asymmetry,
what are we to make of the proposed link with prenatal exposure to hormones?
‘Sex differences in the cerebral asymmetry of the mature brain have not
yet been related to perinatal hormonal events,’ says Kelly. Witelson argues
the opposite, drawing on the notion popularised by the late Norman Geschwind
of the Massachusetts Institute of Technology that handedness is a marker
of ‘lateralisation’ of the brain (‘The left and right of brains at work’,
¿ìè¶ÌÊÓƵ, 11 February 1989). Left-handed people are meant to be more
likely to have ‘unusual brain organisation’ – dealing with language, for
instance, with the right, rather than the left side of the brain, or with
both sides at once. Convinced that ‘there is a neurobiological factor related
to sexual differentiation in the aetiology of homosexuality’ she has searched
for left-handed homosexuals.

Based on interviews with 32 lesbians, for instance, Witelson concludes
that ‘the majority of female homosexuals have some left-hand preference’.
She hastens to add that she does not mean to imply that most left-handed
women are homosexual, as 35 per cent of all women apparently are not consistently
right-handed. But she concludes that her data suggest an ‘atypical pattern
of hemispheric specialisation’ among the lesbians. This in turn suggests,
she says, that there is ‘a neurobiological difference between homosexual
and heterosexual women, likely present from birth’. She puts it down to
a hypothetical excess of testosterone in the womb.

In Witelson’s scheme, homosexual men should be more often left-handed
than heterosexual men, as a result, this time, of being exposed to unusually
low levels of prenatal masculinising hormones. Sadly for her theory, Witelson
could find no statistically significant difference between handedness in
38 homosexual men and her sample of ‘the general population’.

FLEXIBLE NERVE CELLS

While psychologists have been busy with a flurry of speculative human
research, neurobiologists have continued to study the long-suffering laboratory
rat. Recent experiments with ‘animal models’ of human sexuality and gender
have lent strength to the idea that neural structures are not fixed at birth.
‘Natural fluctuations in steroid levels in adulthood can produce significant
neural changes comparable to those seen during the early organisational
period,’ says Janis Weeks of the University of Oregon in a recent review.

Akira Matsumoto of the Juntendo University School of Medicine in Tokyo
finds that even in adult rats, sex steroids can cause neurons in areas such
as the hypothalamus to alter their structure and connections with other
neurons. Nerve cells can increase in number and size, grow new projections,
and remodel their synaptic junctions with other neurons.

Elizabeth Gould, Catherine Woolley and Bruce McEwen at Rockefeller University
in New York find that a range of hormones – ovarian steroids, thyroid hormone
and glucocorticoids – can alter neuronal structures in an adult rat’s hippocampus.
This banana-shaped structure plays a role in learning and memory (see ‘These
cells were made for learning’, supplement to ¿ìè¶ÌÊÓƵ, 21 November).

The researchers maintain that these neuronal cells can modify their
structure in adulthood just as much as they did when the nervous system
was developing. No one yet understands what these observations might mean
in terms of behaviour, but they suggest that the interplay of hormones and
the nervous system is far more flexible than many have supposed.

A rat’s environment can also profoundly alter regions of the brain purported
to differ inexorably between the sexes. Janice Juraska of the University
of Illinois at Urbana-Champaign says that simply housing rats in groups
with a selection of novel objects to examine, rather than singly in uniform
cages can influence both ‘the degree and even the direction’ of differences
between the sexes in many areas of the brain involved in cognition, such
as the hippocampus and the cerebral cortex.

And Gooren has shown that a supposedly permanent feature of the human
central nervous system is remarkably labile. He found that transsexual men
– who think of themselves as women – do not show an surge in luteinising
hormone (LH) from the pituitary gland after an injection of oestrogen before
treatment. Yet after a sex-change operation, which involves castration
and treatment with oestrogen, the same individuals do develop a luteinising
hormone surge. This establishes, he says, that in humans the presence
or absence of this hormone surge ‘is not definitively and irrevocably designated’
before or just after birth.

It’s too early to say whether there is a ‘paradigm shift’ in the air,
but certainly neuroscientists are learning the hard way that even a rat’s
brain is more complicated, and more flexible, than anyone supposed. Meanwhile,
the old linear model of brain organisation, with its simple vision of prenatal
hormones determining behaviour in adulthood in both rats and humans, grows
increasingly threadbare. Neuroanatomists may yet find themselves handing
the search for the roots of homosexuality back to social psychologists and
sociologists.

* * *

1: Gay brains, gay minds?

Are some people born with a gay brain? This notion gained the imprimatur
of science earlier this year, thanks to a paper published in Science. The
excitement centred on a tiny cluster of cells, known as the third interstitial
nucleus of the anterior hypothalamus, or INAH 3, which earlier research
had found to be bigger in men than in women. Simon LeVay of the Salk Institute
in San Diego reported that this bit of the brain is bigger in heterosexual
men than in homosexual ones. He claimed that his data suggest that ‘sexual
orientation has a biological substrate’.

LeVay’s conclusions rest on a series of assumptions. First, because
the hypothalamus is crucial to the sexual behaviour of rats, LeVay argues
that it plays a vital role in typical ‘masculine’ sexuality in humans. Secondly,
because INAH 3 had previously been reported by Roger Gorski at the University
of California at Los Angeles to be larger in men than in women, LeVay
says it ‘could be involved in the generation of male-typical sexual behaviour’.

He concludes that INAH 3 engenders sexual interest in women, specifically.
Hence lesbians and heterosexual men should have large clusters, LeVay predicted,
and heterosexual women and gay men should have small clusters.

LeVay could not get hold of the brains of any lesbians, but deaths from
AIDS provided a supply of gay men’s brains for analysis. His data rest on
the autopsy of the brains of 19 men presumed to be homosexual, 16 presumed
heterosexual men and 6 presumed heterosexual women.

As the editor of Nature, John Maddox, pointed out in an article entitled
‘Is homosexuality hard-wired?’, the measurements LeVay made are not technically
easy. ‘LeVay makes the technique sound much simpler than it is,’ he wrote.
But there are problems even if his data are taken at face value. The range
of measurements in gay men is huge – spanning a twenty-fold difference.
Some gay men had an INAH 3 nucleus as large as the heterosexual men. ‘The
scatter of the measured sizes suggest that nuclear size, if in any sense
a ’cause’, is neither a unique nor an unambiguous determinant of homosexual
behaviour,’ says Maddox.

Anne Fausto-Sterling, a developmental biologist at Brown University
in Rhode Island, sees another problem with LeVay’s research – the fact that
he has ‘no specific information about the sexual behaviour of the men in
his study’. It is conceivable, for instance, that ‘the brain differences
he found, if confirmed, correlate with frequency of sexual activity rather
than with orientation’, she argues. And again, the direction of causality
remains unknown. Neuroanatomical differences could be the result, rather
than the cause, of different lifestyle. We can’t understand the origins
of human sexual expression, Fausto-Sterling says, without knowing more about
how we actually behave.

In 1990, Dick Swaab and his colleagues at the Netherlands Institute
for Brain Research reported the first anatomical difference between the
brains of homosexual and heterosexual men. They found that a part of the
brain that influences circadian rhythms, the suprachiasmatic nucleus, is
twice as large in gay men as in heterosexual men. And this year Gorski and
his colleage Laura Allen claimed that a bundle of nerve cells connecting
the left and right sides of the brain, the anterior commissure, is larger
on average in gay men than in heterosexuals. It is not clear how these parts
of the brain could control sexual behaviour, so few see them as serious
candidates for a biological cause of homosexuality.

Recently, two American psychologists reported evidence for the existence
of ‘gay genes’. Michael Bailey of Northwestern University in Illinois and
Richard Pillard of Boston University School of Medicine interviewed 28 pairs
of male twins. They report that if one identical twin is gay, the other
is almost three times more likely to be gay than if the twins are not identical.
This is not a rigorous demonstration of genetic influence, however, because
identical twins are particularly likely to have shared similar experiences
in childhood.

‘In order for such a study to be meaningful, you’d have to look at twins
raised apart,’ says Fausto-Sterling. Moreover, the genetic theory fails
to explain why only one of many pairs of identical twins was gay. Most of
the discordant identical twins scored at opposite ends of the ‘Kinsey scale’
which ranks individuals on a seven-point scale ranging from strictly homosexual
to exclusively heterosexual.

* * *

2: Liberating desire from hormones

Despite popular preconceptions about hirsute Latin lovers and their
testosterone levels, no one has ever been able to find consistent links
between hormone levels and sexual behaviour in people.

Women turn out to be remarkably resistant to any attempt to relate their
hormone levels to libido. For instance, in 1989 Patricia Schreiner-Engel
of Mount Sinai School of Medicine in New York and her colleagues could find
no hormonal differences between healthy, apparently heterosexual women who
had ‘a persistent, pervasive and severe lack of sexual desire’ and those
who were enthusiastic about having sex with men.

Nor do levels of oestrogens, or any other hormone, correlate with desire.
Several studies have found that women tend to show peaks of sexual activity
just before and after their periods, rather than during ovulation at mid-cycle.
So far, no one has come up with a convincing hormonal or evolutionary explanation
for this observation.

Men have partially fulfilled endocrinologists’ expectations. In men,
androgens are ‘necessary (though not sufficient) for the maintenance of
normal sexual desire’, says John Bancroft of the Medical Research Council’s
Mammalian Reproductive Unit in Edinburgh. But researchers have failed to
find consistent links between testosterone levels and sexual behaviour in
men with normally functioning testes.

Researchers have also searched for evidence that homosexuals are a sort
of intersex, with a ratio of sex hormones in between those of heterosexual
men and women. But as Bancroft says: ‘Most of these studies have been notable
for their naivety and for the almost total lack of control of the many variables
that can influence isolated hormone levels.’

He points out, for instance, that earlier claims that homosexual men
have a more feminine physique than heterosexual ones also came to grief
on the rocks of faulty sampling. The supposedly weedy gay men examined
in the 1950s were all psychiatric patients; other ‘neurotic’ men in mental
hospitals of the day turned out to be similarly androgynous. ‘As yet no
endocrine pattern distinguishes male homosexuals from heterosexuals,’ Bancroft
says.

A recent attempt, in 1987, to find a hormonal difference between lesbians
and heterosexual women also failed. Anke Ehrhardt of Columbia University
in New York and her colleagues could find no differences in blood levels
of testosterone, androstenedione and cortisol, nor any significant differences
in ‘sexual behaviour’ apart from the obvious one.

Even monkeys are not slaves to their hormones, says Barry Keverne, head
of the sub-department of animal behaviour at the University of Cambridge.
His studies of reproduction in monkeys reveal how their behaviour has been
freed from the neuroendocrine events that determine ovulation. Indeed, he
says, most of the so-called reproductive behaviour of monkeys happens outside
the fertile period of their cycle and even during pregnancy.

Such behaviour has become possible, Keverne says, thanks to the development
of a large cortex to cope with a complex social life. This expanded cortex
‘exerts control over the neuroendocrine, limbic brain’. In so doing it diminishes
the influence of both the physical environment and the internal, hormonal
environment. Complex social interactions become of prime importance for
reproduction. In monkeys, and presumably humans too, behaviour has become
’emancipated’ from the gonadal environment, Keverne says.