Testosterone and Aggression in Birds

Full text

Sigma Xi, The Scientific Research Society
Testosterone and Aggression in Birds
Author(s): John C. Wingfield, Gregory F. Ball, Alfred M. Dufty Jr., Robert E. Hegner,
Marilyn Ramenofsky
Source:
American Scientist,
Vol. 75, No. 6 (November-December 1987), pp. 602-608
Published by: Sigma Xi, The Scientific Research Society
Stable URL: http://www.jstor.org/stable/27854889 .
Accessed: 28/08/2011 16:00
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Sigma Xi, The Scientific Research Society is collaborating with JSTOR to digitize, preserve and extend access
to American Scientist.
http://www.jstor.org

Testosterone and Aggression
in Birds
John
C. Wingfield,
Gregory
F. Ball,
Alfred
M. Dufty, Jr.,
Robert
E. Hegner, Marilyn Ramenofsky
The familiar
spring
sound of birdsongs
heralds the
onset of territory
formation and a complex sequence of
interrelated events that make up the breeding period.
Such songs are an integral part of the repertoire of
aggressive behaviors that
males use to advertise and
defend territorial boundaries and to attract mates (Fig.
1). It is well established that hormones, particularly
testosterone, have stimulatory effects on aggression in
reproductive contexts. The prevailing "challenge" hy
pothesis asserts that testosterone and aggression corre
late only during periods of heightened interactions be
tween males. Under more stable social conditions,
according to the
hypothesis, relation
ships among males are maintained
by other factors such as social inertia,
individual recognition of status, and
territorial
boundaries, and testoster
one levels remain low. Recent re
search has suggested ways in
which
the
hypothesis should be modified or
extended. In this article
we will con- _
sider the complexities of aggressive
behaviors and their regulation, focusing specifically on
species differences in territorial
behavior of male birds as
models for the
multiple interactions of hormones, envi
ronment, and behavior.
The secretion of testosterone by interstitial cells in
the testis is controlled primarily by a glycoprotein,
luteinizing hormone, secreted from the anterior pituitary
gland (Fig. 2). Testosterone stimulates the development
of secondary sex characteristics such as wattles, combs,
spurs, the cloacal protuberance (a copulatory organ), and
John
C. Wingfield is an associate professor at the
University of Washington.
He has combined laboratory techniques in comparative endocrinology with
field investigations to study the responses of birds to their
social and physical
environment. He obtained his Ph.D. from the
University College of
North
Wales in
1973
and
was on the
faculty
of
The
Rockefeller
University
before
moving to the
University of
Washington. Gregory F. Ball obtained his Ph.D.
at the
Institute of
Animal Behavior, Rutgers University, and is now
assistant professor at The Rockefeller University; Alfred
M. Dufty performed
his doctoral work at SUNY Binghamton and is a post-doctoral fellow at The
Rockefeller University; Robert E. Hegner graduated with a Ph.D. from
Cornell and completed postdoctoral work at Oxford, The Rockefeller
University, and the
University of Washington; and
Marilyn Ramenofsky
was awarded a Ph.D. from the
University of
Washington, was visiting
assistant professor at Vassar College, and is
now a research associate at the
University
of
Washington.
Address
for Professor
Wingfield:
Department of
Zoology, NJ-15, University of Washington, Seattle, WA 98195.
Testosterone
may not
trigger aggressive
behavior
but
may
facilitate
responses to it
in some species bright-colored skin and nuptial plum
age. These characteristics are used extensively in sexual
and aggressive displays (Witschi 1961).
Testosterone is also transported in the blood to the
brain, where it influences the expression of reproductive
behaviors. Classical experiments conducted on a variety
of vertebrates, including birds, showed that if the testes
are removed, there is a decline in the frequency and
intensity of aggressive and sexual behaviors such as
singing (or equivalent vocalizations), threat postures,
and actual fights. If exogenous testosterone is given to
these castrates, the frequency of aggressive behaviors
increases again (for reviews on birds
- see Harding 1981; Balthazart 1983).
The extent to which aggressive
behaviors decline after castration or
increase after administration of exog
enous testosterone varies greatly
from species to species, in part be
cause of the different
ways in
which
_ testosterone can influence behavior.
Two mechanisms have been pro
posed involving organizational and activational effects.
Organizational effects of testosterone occur early in
development, often immediately after hatching, and
once adulthood is reached the neurons involved can
operate independently. Activational effects require the
immediate presence of testosterone for the sensitive
neurons to function normally. Whether organizational or
activational effects
predominate depends on context and
stage in the breeding period. However, in birds it
appears that testosterone may have important activa
tional effects regulating short-term changes in territorial
aggression within the breeding season.
Over the
past 15
years, radioimmunoassay has been
used to determine circulating levels of testosterone. If
testosterone does activate aggressive behavior, plasma
levels should correlate with the
behavior in reproductive
contexts. Recent work on rodents (Schuurman 1980;
Brain 1983; Sachser and Prove 1984) and primates (Eaton
and Resko 1974; Dixson 1980; Phoenix 1980; Bernstein et
al. 1983; Sapolsky 1984) suggests that there are such
correlations, but that they depend to a great extent on
taxonomic class, age, experience, social context, and
other environmental influences. The mechanisms under
lying such variation are still largely unknown.
In birds, the evidence for correlations of testoster
one and aggression is
more convincing, although not
completely so. Once again, social context
must be taken
602 American Scientist, Volume 75

Figure 1. As part of the annual ritual of establishing territories and attracting mates, male birds engage in a variety of aggressive behaviors.
In the photograph on the left, a
male song sparrow (Melospiza melodia) assumes the posture that heralds an attack on an intruder, in this case
a decoy in a cage. Recent research has shown how the steroid hormone testosterone stimulates aggression in response to such perceived
threats. Mist nets stretched between aluminum poles are used to catch birds in the held. After removing a bird from the net {right), the
scientist collects a blood sample from a wing vein. The bird is then released unharmed. (Photographs by ). C. Wingfield.)
into
account, as
well as environmental influences such as
length of day, presence of a mate, and nest sites
(Wingfield and Ramenofsky 1985). At least some of this
confusion can be eliminated by bringing a comparative
approach to
bear on a variety of avian species. Birds are
ideal for this kind of research because there is
much
diversity in social systems across species. They are also
relatively easy to study under free-living conditions,
enabling us to conduct parallel field and laboratory
investigations.
Seasonal changes
If testosterone is as intimately involved with territorial
aggression in
birds as is
usually presumed, testosterone
levels in the blood should parallel the expression of
seasonal territoriality.
This relationship has been investi
gated in several species of birds under free-living condi
tions, thus reducing possible artifacts of captivity (see
Wingfield and Farner 1976).
It is crucial when analyzing these kinds of data to
deterrnine the
precise stage in the reproductive period at
which each individual is sampled. This point is illustrat
ed in
Figure 3,
which depicts plasma levels of luteinizing
hormone and testosterone in free-living house sparrows
(Passer domesticus). If plasma levels of a number of
individuals are organized by calendar date, several
stages of reproductive activity (prelaying, laying, incu
bating, renesting) are averaged out on any given date,
and the result is a pair of curves, with luteinizing
hormone and testosterone rising in spring, remaining
relatively high during the breeding season, and then
declining to basal as reproduction ends in
August and
September. If the data are reorganized according to the
phase of the
breeding cycle, the true
pattern of
hormone
variation is revealed, making allowance for the average
time it
takes a pair to
progress through each stage (about
4 to
6 days to lay the first
egg, 5 days to
produce a clutch,
and 11 to 14 days to incubate).
Figure 4 compares levels of testosterone in several
monogamous species sampled in free-living conditions.
Typically, testosterone is highest when territories are
first established and aggressive interactions among
males are
most frequent. For the song sparrow (Melospiza
melodia), there are two peaks of testosterone, the first
associated with the establishment of territory
and the
second with the egg-laying period for the first clutch,
when the male guards his sexually receptive mate.
Plasma levels of testosterone decline markedly just prior
to
or during the
parental phase (incubation) and gradual
ly dirninish to basal concentrations by the end of the
breeding season.
There is
no increase in
plasma levels of testosterone
during the egg-laying period of the second brood for
many of the species with open-cup nests, such as the
song sparrow and the European blackbird (Turdus mer
?ld), because there are virtually unlimited sites for these
nests, and competition focuses on maintaining territorial
boundaries and guarding mates. However, species such
as the
house sparrow and the
European starling (Sturnus
vulgaris) that
nest in
holes, a limited resource for
which
there often is intense competition (in addition to
guard
ing
mates), do show an increase in testosterone level
with each egg-laying period (see Figs. 3 and 4).
An interesting contrast is provided by the
western
gull (Larus occidentalis
wymani). Individuals of this species
are long-lived, may breed for
20 years or
more, usually
mate for life,
and return to the same breeding territory
year after year. Furthermore, there is an excess of
females and no shortage of nest sites at one of the
breeding colonies, Santa Barbara Island (Hunt et al.
1980). As a result, competition between males is mini
1987 November-December 603

mal, and it is not surprising, given the low level of
aggression, that the cycle of plasma testosterone in
male
western gulls is of very low amplitude (Wingfield et al.
1982).
As Figure 5 shows, the same relationship of testos
terone levels and aggression can be found in polyga
mous and promiscuous species, but males of these
species have high levels of testosterone for longer peri
ods than do monogamous species.
For example, male red-winged black
birds (Agelaius phoeniceus) generally
do not feed young but rather display
at one another throughout the breed
ing season in an attempt to maintain
territorial boundaries and retain fe
males.
Both monogamous and polygy
nous males are found within popula
tions
of the
pied flycatcher
(Ficedula
hypoleuca). Monogamous males
have testosterone levels similar to
those of
monogamous males in
other
species, but polygynous males main
tain high levels of testosterone until
the second female has begun incu
bating. Only then do levels decline
rapidly, followed by a return of the
male to his first
mate, whose young
he helps to feed (see Silverin and
Wingfield
1982).
Male brown-headed cowbirds
(Molothrus ater) are unusual in that
they do not defend a territory but form dominance
hierarchies for access to females. They are brood para
sites, showing no parental care. Males spend the entire
breeding season guarding females from the attentions of
other males. Accordingly, we see prolonged high levels
of testosterone that decline only gradually during the
season (Dufty and Wingfield 1986a).
Laboratory tests
of the challenge
hypothesis
As we have seen, field investigations of free-living
birds
suggest that testosterone is elevated during periods of
elevated competition between males, and that parental
behavior in
males is preceded by a decline in testoster
one. Only in species in which males do not feed young
or are exposed to intense competition do plasma levels of
testosterone remain elevated. These results have led to
the challenge hypothesis.
What is the experimental evidence in support of the
hypothesis? A positive correlation of aggressive displays
with plasma testosterone was found when Japanese
quail (Coturnix coturnix) were paired in a tournament
lasting several days, but the correlation was apparent
only immediately
prior to the first
fighting day and
during the following three days (Ramenofsky 1984).
From the fifth
day onward, levels of testosterone in
quail
that
won fights
were indistinguishable from levels in
those that lost. By that time, dominance relationships
had been established. This may explain why Balthazart
and his colleagues (1979)
and Tsutsui and Ishii (1981)
could find no correlation of plasma testosterone level
and dominance status in
groups of
male quail with well
established social relationships.
Other experiments confirmed these findings. Cap
tive flocks of
house sparrows formed social hierarchies in
which dominant individuals had higher plasma levels of
testosterone than subordinates only during the first
week after the
birds were grouped. Before grouping, and
more than one week after, there
were no correlations of
testosterone level and social status (Hegner and Wing
field 1986). This is consistent with the challenge hypoth
esis, since testosterone levels were elevated only for a
short period as relationships were established. Similarly
in the brown-headed cowbird, three
males grouped with
a single female formed social relationships, and the
dominant male gained access to the female. Plasma
levels of testosterone in dominant males were elevated
one day after
grouping, but not before or one week after
(Dufty
and
Wingfield
1986b).
What happens if
exogenous testosterone is
given to
individuals? Do they rise in status, gain a territory,
or
enlarge an existing one? If a testosterone implant was
given to
an identified subordinate of a regularly
matched
pair of Japanese quail, he became more aggressive and
fought more persistently with other males. Neverthe
less, these subordinates did not win a sufficient
number
of fights to
be considered dominant (Ramenofsky 1982).
This suggests that testosterone is
not sufficient in itself
to
heighten aggressive displays to the point of overthrow
ing previously established relationships. Similar results
have been obtained in
dominance hierarchies of
Califor
nia quail (Lophortyx californica), free-Uving
Harris's spar
rows (Zonotrichia querula), and sharp-tailed grouse (Tym
panuchus phasianellus) (Emlen and Lorenz 1942; Trobec
and Oring 1972; Rohwer and Rohwer 1978).
Another laboratory experiment sheds more light
on
the challenge hypothesis. Castrated male white-crowned
sparrows (Zonotrichia leucophrys
gambelii) were given im
plants of testosterone that
maintained circulating levels
very similar to those observed during the spring (see
Wingfield and Farner 1978a, 1978b). Castrated controls
Figure 2. The system through which testosterone influences aggressive behavior begins with
the secretion, in response to environmental stimuli, of the glycoprotein luteinizing hormone
from the anterior pituitary gland at the base of the brain. Luteinizing hormone in turn
stimulates secretion of testosterone by interstitial cells in the testis, where testosterone is
produced. In addition to arousing aggressive behavior, testosterone contributes to the
development of secondary sex characteristics, such as combs, spurs, and bright plumage.
604 American Scientist, Volume 75

were given empty implants. Songs and threat displays
often seen during the establishment of territories in the
field increased in
both groups after treatment, but there
was no significant difference in the frequency of these
actions between the two groups despite the
wide differ
ence in testosterone level (Wingfield 1985a).
This apparent contradiction of the challenge hy
pothesis can perhaps be attributed to the fact that the
birds had been housed together for
over six
months. It
was thus likely that social relationships among individ
uals had been established for some time. When a new
male was introduced in an adjacent cage, there
was an
immediate increase in aggression in both groups, and
the males with higher levels of testosterone showed
significantly
more aggressive displays than did the con
trols. By the next day, the frequency of aggression had
dropped dramatically, illustrating how quickly social
relationships can be established and emphasizing the
ephemeral nature of the correlation between testoster
one and aggressive behavior.
It is of little surprise that some investigations have
identified hormone-behavior relationships and some
have not, particularly since social contexts vary across
the studies. Experiential factors such as the
development
of dominance relationships among individuals can exert
a strong influence on the
degree to
which the circulating
levels of testosterone affect frequency and intensity of
aggressive behavior. Nevertheless, there is little doubt
that testosterone is requisite for increased frequency of
aggressive behavior when an individual is challenged in
a territorial
or other reproductive context.
Environmental cues and testosterone
What controls the tinting and amplitude of changes in
plasma levels of testosterone so that they occur at
appropriate stages in the reproductive period? Clearly,
environmental cues play a major role, and one obvious
candidate is the annual change in the length of day. It is
well known that the vernal increase in length of day
promotes secretion of luteinizing hormone and steroid
hormones such as testosterone (e.g., Farner and Follett
1979; Wingfield and Farner 1980). Experiments with
male white-crowned and song sparrows demonstrated
that spermatogenesis is completed, secondary sex char
acteristics are developed, and the full repertoire of
reproductive behaviors (both territorial and sexual) are
expressed when birds are transferred from short to long
days (see Wingfield and Moore 1986). However, the
seasonal changes in testosterone in free-living
males are
dramatically different from those generated solely by
exposing captive males to long days in the laboratory,
and the absolute levels can reach an order of
magnitude
higher than those of
males maintained in captivity. Since
it
has also been shown that high circulating levels of
testosterone are not required for the expression of sexual
behavior (Moore and Kranz 1983), it is possible that
elevated levels in
free-living males are involved solely in
the regulation of aggression.
What other environmental cues influence the secre
tion of testosterone and aggressive behavior? Two possi
bilities spring to
mind: stimuli from the territory
itself
or
signals emanating from a challenging male. To evaluate
these possibilities, male song sparrows were captured
and removed from their territories, thus creating a
vacant spot
within the local population. Usually another
male claimed the spot within 12 to 72 hours. The result
was an increase in conflict between the replacement
male and the neighbors, who reestablished territorial
boundaries with the newcomer. During this period of
social instability, blood samples were collected from
replacement males and neighbors. Samples were also
Stage of
breeding
cycle
Figure 3. It is important when analyzing seasonal changes in
luteinizing hormone and testosterone to distinguish between
organization of data by calendar date and by stage in the breeding
cycle. If plasma levels of a number of individuals are organized by
calendar date (top), the various stages in the breeding cycle average
out, and the result for both luteinizing hormone and testosterone is
a curve with two peaks. If on the other hand the data are organized
by stages in the breeding cycle (bottom), a
much more complicated
pattern of hormone variation appears. The data displayed here are
from free-living male house sparrows (Passer domesticus). (After
Hegner and
Wingfield 1986.)
collected from control
males in a separate area in
which
boundaries had been stable for
some time.
The results
were quite clear: plasma levels of testos
terone
were higher in the replacement males and in their
otherwise untreated neighbors than in the controls. Both
the neighbors of the replacement and the controls had
territories, yet the latter had much lower levels of
testosterone. These two groups differed only in that the
neighbors were reestablishing territorial boundaries
whereas the controls were not. This suggests that the
stimulus for increased secretion of testosterone may be
not the territory per se (although the data do not
disprove a possible effect) but the challenging behavior
of the replacement male as he attempts to establish new
territorial
boundaries (Wingfield 1985b).
To test this further, intrusions were simulated with
a decoy male song sparrow in
a cage placed in the center
1987 November-December 605

Figure 4. The plasma levels of testosterone in males of four other
monogamous species are quite different from those of the house
sparrow shown in Figure 3, although in all cases a relationship
between testosterone and aggressive behavior is discernible.
(Periods when confrontations between males are most frequent are
indicated by bars.) There are two peaks for the song sparrow
(Melospiza melodia), the first associated with the establishment of
territory and the second with the egg-laying period of the first
brood. The European blackbird {Turdus menda) has a single peak
during the first brood. Neither the song sparrow nor the European
blackbird has a peak during subsequent broods, because these
species nest in open cups, for
which there are unlimited sites and
thus little competition. In these species, competition between males
is most intense early in the season. The European starling {Sturmis
vulgaris), on the other hand, nests in holes, for
which competition
is keen, and so starlings with double broods have a second peak
during their second brood; those with single broods have the
expected single peak. The western gull (Larus occidentalis wymani)
has a distinctively different pattern because of the relative lack of
conflict in its breeding colonies. (After Schwabl et al. 1980;
Wingfield et al. 1982; Dawson 1983; Wingfield 1984a; Ball and
Wingfield
1986.)
of a territory (see Fig. 1). Tape-recorded songs also were
broadcast through a speaker placed alongside the
decoy.
The territorial
male almost invariably attacked and at
tempted to drive the simulated intruder away. He was
captured after skirmishing with the intruder for
5 to 60
minutes, and a blood sample was drawn. Controls Were
captured at the same time of day as the simulated
intrusions. Males exposed to a challenge from a simulat
ed intruder showed an increase in testosterone com
pared with controls. Essentially the same result was
obtained in early April and in May through June,
indicating that this effect
could occur at any time during
the breeding period.
It is important to note that the response required
about ten minutes before the increase in testosterone
was significant.
We know that
males tend to trespass on
other territories regularly and are quickly chased out
when seen by the owner (Wingfield 1984b). The confron
tations usually last only a few seconds, and thus an
increase in testosterone level is
unnecessary. However, if
an intruder persists and attempts to take over the
territory,
prolonged fights lasting several hours or even
days may result. In such cases an increase in plasma
testosterone is appropriate.
There is a third line of evidence suggesting that
encounters between males can result in an increase in
plasma levels of testosterone. Implants of testosterone in
free-living
male song sparrows resulted in heightened
aggression for longer periods than in control males. In
turn, plasma levels of testosterone were elevated in
neighbors of testosterone-implanted males compared
with neighbors of controls. This effect
was most appar
ent early and late in the season. At other times no effect
was noted, because factors such as the presence of
young possibly overrode the effect of the aggressive
male neighbor.
It
was also found that
males who had a territory
at
least one removed from a testosterone-implanted male
did not have elevated levels, even though they could
hear and see encounters between their immediate neigh
bors and the testosterone-implanted males (Wingfield
1984b). It appears that an individual male must be
involved directly in
an agonistic encounter for
a hormon
al response to
be initiated. This blocking of a ripple effect
may be adaptive; otherwise, a wave of responses would
pass mdisoiminately through the local population, af
fecting males that were not involved in the original
skirmish. Moreover, functionally irrelevant surges of
testosterone could interfere severely with other repro
ductive activities such as the feeding of young.
The environmental stimuli generated in the course
of an agonistic encounter could enter the central nervous
system by several routes: visual, auditory (songs and
other vocalizations), tactile (fights), or chemical (phero
mones). We can rule out tactile stimuli, because several
of the experiments outlined above show that testoster
one levels increase in response to a caged male with
whom contact is precluded. Also we can probably rule
out pheromonal cues, since these are largely regarded as
being absent in birds (although it is important to note
that this point has not been rigorously investigated).
Thus we are left
with visual and auditory information
influencing secretion of testosterone.
Are both components required for the response?
?iifop?ari
starting
(single-brooded>
606 American Scientist, Volume 75

Recent field experiments showed that if male song
sparrows are exposed to a playback of tape-recorded
songs (auditory but no visual stimulus), a devocalized
male (visual but no auditory stimulus), or a playback
plus a devocalized male (visual and auditory stimuli),
only those males exposed to both visual and auditory
cues have elevated levels of testosterone. Auditory or
visual cues alone do not result in
a significant increase in
testosterone. It
was also found that the response is
specific: captive male song sparrows showed an increase
in plasma levels of testosterone following a challenge
from
another song sparrow but not following a challenge
from a house sparrow (a heterospecific).
Now that the external receptors for
environmental
cues have been identified and the endocrine response
and the specificity of that response deterrnined, we can
investigate the neural pathways by
which environmental
information controls reproductive function.
What is testosterone doing?
This may appear to be an odd question, since it is well
established that testosterone has direct effects
on aggres
sive territorial behavior. There is
no doubt that testoster
one has organizational effects insofar as it influences the
formation of song control nuclei in the brain during
development and seasonal breeding (e.g., Nottebohm
1981). It is also clear that high levels of testosterone
during establishment of a territory are playing some
activational role, at least early in the breeding season.
However, many of our observations do not fit
neatly into
these categories (see also Arnold and Breedlove 1985).
Responses to challenges outside the normal season
al pattern suggest another role for testosterone in the
arousal of aggressive behavior. The initial response to a
challenging male is to attack vigorously even though the
circulating level of testosterone may be much lower than
in early spring. Only after the attack does testosterone
increase, and this appears to take at least ten
minutes.
Clearly testosterone cannot be playing an activational
role in the literal sense of the
word, since it increases
after the fact. Is it
possible that testosterone is
playing a
facilitative role for the
neurons involved during extended
periods of intense aggression?
This role would require a very rapid action of a
steroid hormone on a target cell. The classical mode of
action is through the genome, a process that can take
many hours (typically 16 to 30). But recently compiled
evidence from mammals suggests that steroid hormones
can also have very rapid effects. For example, steroids
have been shown to influence rates of gene transcription
in rats
within 15
minutes, and estradiol can have mor
phological effects on neuronal cell nuclei within two
hours Qones et al. 1985;
McEwen and Pfaff 1985). Even
more striking is the demonstration in
vitro that estradiol
injected directly onto the membrane of an excitable cell
induces action potentials within one rninute (Duty et al.
1979). Furthermore, Towle and Sze (1983) found that
several steroid hormones, including testosterone, bind
to synaptic membranes in the rat
brain with high speci
ficity
and affinity.
Thus the
potential exists for
very rapid
actions of testosterone on the central nervous system
through
membrane receptors, although more research is
required to confirm this in avian systems.
brown-headed cowbird :
prebreeding sexual
^?t?ig? of
breeding
cycle
Figure 5.
Males of polygynous and promiscuous species have their
own characteristic patterns of circulating testosterone: levels remain
high for longer periods than they do in
monogamous species (see
Fig. 4). These correlate with the greater amount of time that such
males must spend defending territories and females. (Periods of
frequent conflicts between males are again indicated by bars.) In
red-winged blackbirds (Agelaius phoeniceus), the breeding period
includes both sexual and parental stages, since each male may have
several females on his territory, some of
which may be in the sexual
or parental stage at any one time. The pied flycatcher (Ficedula
hypoleuca) includes both monogamous (gray line and bar) and
polygynous males (colored line and bar). The brown-headed
cowbird (Moluthrus ater) is a brood parasite that has no territory
and performs no parental duties but rather spends the breeding
season guarding females from competing males. (After Silverin and
Wingfleld 1982;
Dufty
and
Wingf?eld
1986a.
Additional data
supplied by W. A. Searcy.)
Such a concept is speculative, but the possibility
arises that in addition to the two classical modes of
genomic action of steroid hormones, involving organiza
tional and activational effects, a third
mode of action?
supporting or facilitative?could arise during periods of
heightened agonistic encounters. Mediated either
through rapid-acting membrane receptors or genomical
ly, the third mode would influence the function of brain
nuclei involved in the control of aggression. Whether
1987 November-December 607

this may ultimately prove to be simply a form of
activational effects of testosterone, or indeed a separate
mode of action, remains to
be seen.
erences
Arnold, A. P., and S. M. Breedlove. 1985. Organizational and activa
tional effects of sex steroids on brain and behavior: A reanalysis.
Horm. Beh. 19:469-98.
Ball, G. F., and J.
C. Wingfield. 1986. Changes in plasma levels of sex
steroids in relation to
multiple broodedness and nest site density in
male starlings. Physiol. Zool. 60:191-99.
Balthazart, J. 1983. Hormonal correlates of behavior. In Avian Biology,
ed. D. S. Farner, J.
R. King, and K. C. Parkes, vol. 7, pp. 221-366.
Academic Press.
Balthazart, J.,
R. Massa, and P. Negri-Cesi. 1979. Photoperiodic control
of testosterone metabolism, plasma gonadotropins, cloacal gland
growth, and reproductive behavior in the Japanese quail. Gen. Comp.
Endocrinol. 39:222-35.
Bernstein, I. S., T. P. Gordon, and R. M. Rose. 1983. The interaction of
hormones, behavior, and social context in non-human primates. In
Hormones and Aggressive Behavior, ed. B. S
vare, pp. 535-62. Plenum.
Brain, P. F. 1983. Pituitary-gonadal influences on social aggression. In
Hormones and Aggressive Behavior, ed. .
Svare, pp. 3-26. Plenum.
Dawson, A. 1983. Plasma gonadal steroid levels in wild starlings
(Sturnus vulgaris) during the annual cycle and in relation to the stages
of breeding. Gen. Comp. Endocrinol. 49:286-94.
Dixson, A. F. 1980. Androgens and aggressive behavior in
primates: A
review. Aggressive Beh. 6:37-67.
Dufty, A. M., and J. C. Wingfield. 1986a. Temporal patterns of
circulating LH and steroid hormones in a brood parasite, the brown
headed cowbird, Molothrus ater. I.
Males. /.
Zool. London f>U
208:191
203.
-. 1986b. Endocrine changes in
breeding brown-headed cowbirds
and their implications for the evolution of brood parasitism. In
Behavioural Rhythms, ed. Y. Qu?innec and .
Delvolv?, pp. 93-108.
Toulouse: Universit? Paul Sabatier.
Dufy, B., et al. 1979. Membrane effects of thyrotropin-releasing hor
mone and estrogen shown by intracellular recording from pituitary
cells. Science 204:509-11.
Eaton, G. G., and J.
A. Resko. 1974. Plasma testosterone and male
dominance in Japanese macaque troops with repeated measures of
testosterone in laboratory males. Horm. Beh. 5:251-59.
Emlen, J.
T., and F. W. Lorenz. 1942. Pairing responses of free-living
valley quail to sex-hormone pellets. Auk 59:369-78.
Farner, D. S., and B. K. Follett. 1979. Reproductive periodicity in
birds.
In Hormones and Evolution, ed. E. J.
W. Barrington, pp. 829-72.
Academic Press.
Harding, C. F. 1981. Social modulation of circulating hormone levels in
the
male. Am. Zool. 21:223-32.
Hegner, R. E., and J.
C. Wingfield. 1986. Behavioral and endocrine
correlates of
multiple brooding in the semi-colonial house sparrow
Passer domesticus. I.
Males. Horm. Beh. 20:294-312.
Hunt, G. L., Jr., J.
C. Wingfield, A. Newman, and D. S. Farner. 1980.
Sex ratio of western gulls on Santa Barbara Island, California. Auk
97:473-79.
Jones, K. J.,
D. W. Pfaff, and B. S. McEwen. 1985. Early estrogen
induced nuclear changes in rat
hypothalamic ventromedial neurons:
An ultrastructural and morphometric analysis. /. Comp. Neurol.
239:255-66.
McEwen, B. S., and D. W. Pfaff. 1985. Hormone effects on hypotha
lamic neurons: Analysing gene expression and neuromodulator
action. Trends Neurosci., March, pp. 105-10.
Moore, M. C, and R. Kranz. 1983. Evidence for androgen indepen
dence of male mounting behavior in white-crowned sparrows
(Zonotrichia leucophrys gamhelii). Horm. Beh. 17:414-23.
Nottebohm, F. 1981. A brain for all seasons: Cyclical anatomical
changes in song control nuclei of the canary brain. Science 214:1368
70.
Phoenix, C. H. 1980. Copulation, dominance, and plasma androgen
levels in adult rhesus males born and reared in the laboratory.
Archives Sexual Beh. 9:149-68.
Ramenofsky, M. 1982. Endogenous plasma hormones and agonistic
behavior in
male Japanese quail, Coturnix coturnix. Ph.D. diss., Univ.
of
Washington.
-. 1984. Agonistic behavior and endogenous plasma hormones in
male Japanese quail. Animal Beh. 32:698-708.
Rohwer, S., and F. C. Rohwer. 1978. Status signalling in Hams'
sparrows: Experimental deceptions achieved. Animal Beh. 26:1012
22.
Sachser, .,
and E. Prove. 1984. Short-term effects of residence on the
testosterone responses to fighting in alpha male guinea pigs. Aggres
sive Beh. 10:285-92.
Sapolsky, R. M. 1984. The endocrine stress-response and social status
in the wild baboon. Horm. Beh. 16:279-92.
Schuurman, T. 1980. Hormonal correlates of agonistic behavior in adult
male rats. Prog. Brain Res. 53:415-520.
Schwabl, H., J.
C. Wingfield, and D. S. Farner. 1980. Seasonal variation
in
plasma levels of luteinizing hormone and steroid hormones in the
European blackbird, Tur dus merula. Vogelwarte 30:283-94.
Silverin, .,
and J.
C. Wingfield. 1982. Patterns of breeding behaviour
and plasma levels of hormones in a free-living population of pied
flycatchers, Ficedula hypoleuca. }. Zool. London (A) 198:117-29.
Towle, A. C, and P. Y. Sze. 1983. Steroid binding to synaptic plasma
membrane: Differential binding of glucocorticoids and gonadal ste
roids. /. Steroid Biochem. 18:135-43.
Trobec, R. J.,
and L. W. Oring. 1972. Effects of testosterone propionate
implantation on lek behavior of sharp-tailed grouse. Am. Midland
Nat. 87:531-36.
Tsutsui, K., and S. Ishii. 1981. Effects of sex steroids on aggressive
behavior of adult male Japanese quail. Gen. Comp. Endocrinol. 44:480
86.
Wingfield, J.
C. 1984a. Environmental and endocrine control of repro
duction in the song sparrow, Melospiza melodia. I.
Temporal organiza
tion of the breeding cycle. Gen. Comp. Endocrinol. 56:406-16.
-. 1984b. Environmental and endocrine control of reproduction in
the song sparrow, Melospiza melodia. II. Agonistic interactions as
environmental information stimulating secretion of testosterone.
Gen. Comp. Endocrinol. 56:417-24.
-. 1985a. Environmental and endocrine control of territorial
behavior in birds. In Hormones and the
Environment, ed. . .
Follett,
S. Ishii, and A. Chandola, pp. 265-77. Springer-Verlag.
-. 1985b. Short-term changes in plasma levels of hormones
during establishment and defense of a breeding territory in
male
song sparrows, Melospiza melodia. Horm. Beh. 19:174-87.
Wingfield, J.
C, and D. S. Farner. 1976. Avian endocrinology?field
investigations and methods. Condor 78:570-73.
-. 1978a. The endocrinology of a naturally breeding population of
the white-crowned sparrow (Zonotrichia leucophrys pugetensis). Phy
siol. Zool. 51:188-205.
-. 1978b. The annual cycle in
plasma irLH and steroid hormones
in feral populations of the white-crowned sparrow, Zonotrichia
leucophrys
gambelii. Biol. Reprod. 19:1046-56.
-. 1980. Environmental and endocrine control of seasonal repro
duction in temperate-zone birds. Prog. Reprod. Biol. 5:62-101.
Wingfield, J.
C, and M. C. Moore. 1986. Hormonal, social, and
environmental factors in the reproductive biology of free-living male
birds. In Psychobiology of Reproductive Behavior: An Evolutionary Per
spective, ed. D. Crews, pp. 149-75. Prentice-Hall.
Wingfield, J.
C, A. Newman, G. L. Hunt, and D. S. Farner. 1982.
Endocrine aspects of female-female pairing in the
western gull (Larus
occidentalis wyman?). Animal Beh. 30:9-22.
Wingfield, J.
C, and M. Ramenofsky. 1985. Testosterone and aggres
sive behavior during the reproductive cycle of male birds. In
Neurobiology, ed. R. Gilles and J. Balthazart, pp. 92-104. Springer
Verlag.
Witschi, E. 1961. Sex and secondary sexual characters. In Biology and
Comparative Physiology of Birds, ed. A. J.
Marshall, pp. 115-68.
Academic Press.
608 American Scientist, Volume 75