170 
Crews 
is more labile and depends in part upon the 
performance of specific male behavior pat- 
terns and/or the social milieu of the breeding 
population. 
These generalizations appear to be borne 
out by recent research on the control of 
seasonal ovarian growth in the lizard, A. 
caroUnensis. The use of female ovarian ac- 
tivity as a bioassay for pituitary gonado- 
tropin secretion experiments indicates that: 
(1) While an unseasonal environmental re- 
gime will stimulate out-of-season ovarian 
recrudescence in winter-dormant females, 
the presence of an intact conspecific male will 
cause a significantly more rapid rate of 
growth and is necessary for normal gonado- 
tropin secretion; (2) The presence of a 
stable dominance hierarchy is necessary for 
this facilitation to occur (an unstable domin- 
ance hierarchy inhibits ovarian recrudes- 
cence) ; (3) The male behavior patterns 
responsible for the facilitation and inhibition 
of environmentally induced ovarian growth 
are courtship and male-male aggression, re- 
spectively ; (4) Gonadotropin secretion in the 
female is graded in accordance with the 
amount of male courtship behavior to which 
she is exposed; (5) Both male-courtship be- 
havior and courtship facilitation of environ- 
mentally induced ovarian development are 
dependent upon testicular secretions ; and 
(6) The critical component in male-court- 
ship facilitation of environmentally induced 
ovarian recrudescence is the ability of the 
male to extend the dewlap, thus changing his 
body shape in the species-typical manner. 
These findings suggest that, while male 
courtship behavior insures normal gonado- 
tropin secretion in the female, the absence 
of this courtship behavior results in sub- 
normal gonadotropin secretion, and that ag- 
gression between males actively inhibits or 
greatly suppresses gonadotropin secretion. 
Finally, it is possible that the inhibition -of 
environmentally induced ovarian recrudes- 
cence by male-male aggression may have been 
one of the selection pressures favoring the 
differential male-female vernal emergence 
that is observed in many temperate verte- 
brates. Females arriving simultaneously at 
the breeding grounds with the males would 
be exposed to the high levels of male-male 
aggression necessary for the establishment 
of territories which, as these experiments 
indicate, would prevent the female from 
rapidly reaching breeding condition. Females 
arriving after males have established breed- 
ing territories and male-male aggression has 
waned, however, would be exposed to both 
optimal environmental stimuli and male 
courtship and thus be maximally stimulated. 
Lizards appear to be well suited for the 
study of behavioral adaptations associated 
with breeding. During the breeding season, 
female A. caroUnensis undergo regular cycles 
of sexual receptivity which are correlated 
with the cyclic pattern of follicular matura- 
tion peculiar to Anolis. It was demonstrated 
experimentally that mating inhibits further 
sexual receptivity within each follicular 
cycles, but not' later receptivity during sub- 
sequent cycles; the stimuli responsible for 
this inhibition arise from the intromission, 
and possibly the ejaculation, of the male. 
These findings, when joined with observa- 
tions of the social system of this species, in- 
dicate that females mate with the territorial 
male once during each estrous cycle and so 
minimize predation pressure. 
This investigation of the factors control- 
ling different phases of the lizard reproduc- 
tive cycle has also suggested, for the first 
time, the possible physiological basis for the 
sexual refractory period that typically fol- 
lows the breeding season in vertebrates. Ex- 
periments indicate that the degenerating 
follicle secretes a substance (s) which coun- 
teracts the effects of both environmental and 
hormonal stimuli which, at other times of 
the cycle, reliably elicit ovarian growth. 
I believe that the experiments discussed 
here demonstrate that principles originally 
discovered with laboratory species can be 
applied to feral species and shown to have 
clear adaptive significance in the animal’s 
natural history. 
ACKNOWLEDGMENTS 
Parts of this research were supported by P.H.S. 
Grant No. GM-1136 (to Daniel S. Lehrman) and by 
N.S.F. Grant No. GB 36241X (to Paul Licht). 
