Data supporting the original "organization
hypothesis" were based on our research involving
guinea pigs. In those experiments, genetic
female guinea pigs treated prenatally with
testosterone propionate (TP) displayed
mating behavior as adults similar to that of
castrated males when both were treated with TP.
In addition, responsiveness to female hormones
was suppressed. The "organization hypothesis"
led us to predict that genetic males deprived of
testosterone before the period of sexual
differentiation was complete would have a
diminished capacity to display mating behavior
and an enhanced capacity to display femelle
receptive behavior when appropriately treated in
adulthood. Support for the hypothesis came from
a series of experiments in which male rats
castrated before they were 5 days old displayed
less male mating behavior when treated with
testosterone and more female mating behavior
when treated with estradiol in adulthood than
males castrated after Day 5 of life and
similarly treated.
An important aspect of the organization
concept was the hypothesis that "the masculinity
or femininity of an animal's behavior beyond
that which is purely sexual ... developed in
response to certain hormonal substances within
the embryo and fetus". It was largely this
component of our hypothesis that led us to turn
to the rhesus macaque as a model; we thought it
was better suited than the guinea pig or
rat.
In a series of studies on young genetic
rhesus females between 3 and 48 mo of age that
had been treated prenatally with TP, we found
that they displayed more social threats,
initiated play more frequently, and engaged in
rough-and-tumble play more often than did
control females. Mounting behavior also was
displayed at an early age, with a frequency like
that of same-age males rather than that of
females. In all of those studies, the female
pseudohermaphrodites (hermaphrodites) were
tested without hormone treatment and with
ovaries intact. Most of the tests were carried
out before the animals reached the age of
menarche, which was delayed.
After ovariectomy, when the hermaphrodites
were between 5 and 7 yr of age, they were paired
with ovariectomized, estrogentreated females in
a series of 30 weekly tests. In pretreatment
tests, the hermaphrodites (n = 7) displayed
significantly higher levels of aggression than
did control females (n = 7). The two groups were
then treated with testosterone, and again
aggressive behavior was displayed at a
significantly higher level by the hermaphrodites
than by similarly treated control females. The
mounting rate did not differ between
hermaphrodites and control females before or
after treatment with TP. One hermaphrodite
(Monkey 1640) achieved intromission and
ejaculateu in two separate tests of sexual
behavior. Two other hermaphrodites (Monkeys 1616
and 1239) masturbated to ejaculation in their
home cages but did not achieve intromission or
ejaculation during tests of sexual behavior.
Thus, although evidence for the masculinizing
action of testosterone on the tissues that
mediate nonsexual behavior in adolescent and
young adult rhesus macaques is quite clear,
evidence for the same action on tissues that
mediate adult mating behavior rests largely on
the performance of a single animal (Monkey
1640).
These hermaphrodites were used in two
additional behavioral studies when they were
approximately 9-11 yr old and again when they
were between 13 and 14 yr of age. Other studies
that did not concern behavior but that indicated
masculinization of their anatomy and physiology
also were carried out. In one study, it was
demonstrated that the canine dental complexes of
the hermaphrodites had been masculinized. In
another study, hermaphrodites and males were
less sensitive to the long-term estradiol
inhibition of luteinizing hormone secretion than
were similarly treated females.
We undertook the experiments reported here to
determine how, at the more mature age of 15-17
yr, the behavior of these hermaphrodites would
differ from that of ovariectomized females; and
castrated males in the absence of exogenous
hormones and when treated with TP. In addition,
we wished to compare the effects, if any, of
estradiol on the sexual performance of
hermaphrodites, control females, and castrated
males in pair tests with receptive females. To
our knowledge, the effects of exogenous
estradiol on the sexual behavior of
gonadectomized rhesus males and hermaphrodites
have not been investigated.
[...]
Discussion : It is well established
that in rodents, patterns of mating behavior are
readily masculinized by testosterone treatment
during psychosexual differentiation. In rodents,
dogs, rhesus macaques, and perhaps even human
beings, behaviors other than those that are
purely sexual are likewise readily masculinized
by appropriate testosterone treatment during the
period of differentiation. What is not well
established and needs further delineation is the
extent to which prenatal testosterone
masculinizes the tissues mediating the mating
behavior of adult primates.
In this study, the sexual behavior of
gonadectomized adult females treated prenatally
with testosterone did not differ substantially
frora that of untreated fernales. The
hermaphrodites differed only in grimace rate
from females in the untreated condition.
Although there was a significant overall group
difference in the aggression rate,
hermaphrodites did not differ from females as
determined by the Newman-Keuls test. However,
the proportion of hermaphrodites displaying
aggression was significantly greater than the
proportions of untreated males and females. When
differences in behavior occurred between females
and males, the hermaphrodites resembled females
rather than males. Because these comparisons
involved untreated animals, similarities and
differences reflected something other than
changes in sensitivity to hormonal stimulation
in adulthood. Rather, they indicated the extent
to which prenatal testosterone treatment of
females modified hormonally independent behavior
in adulthood.
The extent to which behavior was modified by
EB or TP in adulthood also indicated that the
hermaphrodites as adults resembled females
rather than males. Yawning was the only
behavior to show a significant rate increase
after TP treatment for females and for
hermaphrodites. The effect, of TP on the
yawning rate in males and fernales has
been well documented. The percentage of tests in
which hermaphrodites showed penile erections
increased significantly under TP treatment, but
for all treatment conditions the percentage was
significantly lower than in males. Some
investigators have suggested that the malelike
behavior displayed by hermaphrodites as
youngsters was the result of genital development
comparable to that of males. It was argued that
as a consequence of each possessing a penis and
scrotum, the hermaphrodites were treated as
males by their mothers and were perceived to be
males by their siblings; therefore, they acted
as males. Yet, as adults with penile erections,
and having presumably received differential
treatment as infants and juveniles, they failed
to display male behavior. Explaining their early
masculine behavior as the result of alterations
in the central nervous system rather than as
changes in genital structure does not eliminate
the need to account for the absence of male
behavior in adulthood.
In their rates of contacting, a specifically
sexual behavior (interpreted as an invitation to
copulate), females and hermaphrodites did not
differ, but in this behavior both groups
differed from males under all treatment
conditions. Mounting was rarely displayed by
fernales and hermaphrodites. In the 12 tests
without hormone treatment, one hemaphrodite,
Monkey 1616, mounted once; when treated with EB,
the same animal mounted five times in 12 tests.
In the 12 tests when given TP, none of the
hermaphrodites mounted. The one hermaphrodite
that mounted in this study had received 750 mg
of TP prenatally, but two other hermaphrodites
that had received the same prenatal treatment
failed to mount. Both fernale and male groups
had been given more tests than the
hermaphrodites over the past decade, although in
all other respects they were treated alike. The
TP treatment did not result in an increase in
mounting by females, although they had as much
experience in the test situation as males. Both
fernales and hermaphrodites differed
significantly from the males; neither group
displayed a significant increased mounting rate
as a result of TP treatment. This finding is the
same as in a previous study when the amount of
testing experience of the hermaphrodites was
approximately the same as in the comparison
group.
When the six hermaphrodites tested in this
study were 1, 2, 3, and 4 yr of age, their mean
mounting rates per test were .10, .15, .18, and
.15, respectively. Those testing conditions were
not comparable with the ones reported here, and
we mention them only to aid in the
interpretation of results. The hermaphrodites at
that time were tested with ovaries in situ, and
they were tested in mixed-sex groups of four or
five peers. Their mean mounting rates were based
on all mounts, including those without a
foot-clasp. The mounting rate of the
hermaphrodites was intermediate between those of
males and females of the same age given
comparable tests.
When the hermaprodites were between 5 and 7
yr of age, ovariectomized and paired with
receptive females, as in the experiments
reported here, the mean mounting frequency per
test in the absence of hormonal stimulation was
.43, and in the presence of TP treatment it was
.91. The mounting frequency did not differ
significantly from that of control fernales
before or after treatment, and it did not
increase significantly with TP treatment. Two
intervening unpublished studies of the
hermaphrodites as adults gave no indication of
masculinization of sexual behavior. In the study
reported here, none of the hermaphrodites
achieved intromission or ejaculated; one
hermaphrodite (Monkey 1640) had done so in the
1973 study.
In sharp contrast to the few mounts displayed
by the hermaphrodites was the mounting rate of
males that had been castrated 11 yr earlier. One
male failed to mount when given no hormone
treatment but did mount under EB and under TP
influence. Two of the nine castrated males
failed to ejaculate with TP treatment, but they
had ejaculated 10 yr earlier when similarly
treated; this difference may have been due to
their advanced age in the later study (21 yr or
older), but such an explanation is unlikely in
the case of the hermaphrodite, Monkey 1640,
which was 15 yr old at the tîme of
testing. Differences in the ditration of
treatment and the total dose of TP are more
probable explanations for this particular
difference. In the 1973 study, the
hermaphrodites were given a total of 1,470 mg of
TP over 30 wk, whereas in the present study they
received 510 mg over 7 wk; hence the difference
in performance may be primarily one of
threshold.
In evaluating the effects of prenatal
hormones on the patterns of mating behavior
displayed in adulthood, we did not overlook the
importance of rearing conditions. Goy pointed
out that the display of social behaviors in
early life is entirely unrelated to copulatory
success in young adulthood. Although the
patterns of social behavior (especially play) we
noted in hermaphroditic monkeys were indeed
masculinized in early life (years 1-4), such
behavior was not correlated with the copulatory
ability at that age, and we now know it was not
predictive of the display of male patterns of
mounting behavior in adulthood. Evidence for the
masculinization of behavior in primates by
prenatal treatment with androgens rests largely
on the display of malelike social behavior
during infancy and the juvenile period of
development. Evidence for the masculinization of
sexual behavior in adult primates is based on
the performance of one animal. The behavior of
this single animal should not be dismissed; the
intromissions and ejaculations by Monkey 1640 at
6 yr of age have been well documented, but
additional adult hermaphrodites reared under
optimal conditions need to be studied as
adults.
injected with 19-hydroxytestosterone (the
hydroxylated metabolite of testosterone), a
hormone that is readily converted to estradiol.
Blood levels of estradiol increased
significantly after treatment and LH levels
decreased, but the only behavior that changed
signif icantly was the frequency with which
males sat close to their female partners. There
was no change in mean latency to the first
mount. Those males with blood levels of
estradiol above the group mean failed to achieve
intromission or to ejaculate. On the basis of
that study, we would not have predicted the
increase in mounting with estradiol treatment
that we observed. It bas been reported that the
majority of castrated, sexually experienced rats
treated daily with 100jug of EB for an extended
period show the complete pattern of sexual
behavior (Sôdersten, 1973). Our males had
been castrated 11 yr earlier and received only
20 Mg of EB a day. Although dose comparisons
across species are not necessarily appropriate,
comparable dose in terms of body weight would
have involved injecting each monkey with no'
less than 2,000 jug of EB per day. The entire
question requires additional study, but old
long-terrn-castrated monkeys are not appropriate
subjects for such an investigation. We can
conclude that, at least in the dose used, sexual
behavior was not suppressed by EB and that EE
may have had a facilitating, albeit limited,
effect.
Few significant differences in behavior were
differentially displayed by partner females
toward the three groups of experimental animals;
one difference of note was that the greatest
PAD, threat-away, and prox rates occurred when
they were paired with males. Within each group
the greatest PAD rate occurred when the
experimental monkeys were untreated. It was
under this condition that the experimental
animals, especially the males, displayed the
lowest level of sexual behavior. This fact may
accourit in part for the greater rates of
presenting by the partner females during the
no-treatment tests. The level of aggression was
highest when partner females were paired with
hermaphrodites, and in turn the hermaphrodites
displayed more aggression toward their partners
than did males or experimental fernales. When
partner fernales were paired with
hermaphrodites, their behavior resembled that
displayed in pairings with females rather than
males.
Repeated observations of the hermaphrodites
implanted with testosterone from adolescence to
adulthood and housed in groups with other males
and females would be needed to provide
conditions (hormonal and enviromnental) that
simulate more closely those of normal males in
nature.
Although treatment with EB did not produce
any significant changes in the behavior of
females or hermaphrodites, males displayed a
higher percentage of tests with mounts, and
their mount latency was shorter with EB
treatment than without hormone treatment. The
findings are not readily understandable since no
other behavioral measures changed, and in a
previous study, mounting rate was negatively
correlated with estradiol levels in the blood.
In that study, castrated rhesus males were
injected with 19-hydroxytestosterone (the
hydroxylated metabolite of testosterone), a
hormone that is readily converted to estradiol.
Blood levels of estradiol increased
significantly after treatment and LH levels
decreased, but the only behavior that changed
signif icantly was the frequency with which
males sat close to their female partners. There
was no change in mean latency to the first
mount. Those males with blood levels of
estradiol above the group mean failed to achieve
intromission or to ejaculate. On the basis of
that study, we would not have predicted the
increase in mounting with estradiol treatment
that we observed. It bas been reported that the
majority of castrated, sexually experienced rats
treated daily with 100 µg of EB for an
extended period show the complete pattern of
sexual behavior. Our males had been castrated 11
yr earlier and received only 20 µg of EB a
day. Although dose comparisons across species
are not necessarily appropriate, comparable dose
in terms of body weight would have involved
injecting each monkey with no' less than 2,000
µg of EB per day. The entire question
requires additional study, but old
long-terrn-castrated monkeys are not appropriate
subjects for such an investigation. We can
conclude that, at least in the dose used, sexual
behavior was not suppressed by EB and that EE
may have had a facilitating, albeit limited,
effect.
Few significant differences in behavior were
differentially displayed by partner females
toward the three groups of experimental animals;
one difference of note was that the greatest
PAD, threat-away, and prox rates occurred when
they were paired with males. Within each group
the greatest PAD rate occurred when the
experimental monkeys were untreated. It was
under this condition that the experimental
animals, especially the males, displayed the
lowest level of sexual behavior. This fact may
accourit in part for the greater rates of
presenting by the partner females during the
no-treatment tests. The level of aggression was
highest when partner females were paired with
hermaphrodites, and in turn the hermaphrodites
displayed more aggression toward their partners
than did males or experimental fernales. When
partner females were paired with hermaphrodites,
their be havior resembled that displayed in
pairings with females rather than
males.
