CONTROL OF ANNUAL GONADAL CYCLES 739 



body weight appeared to rule out any appreciable nutritional differ- 

 ences. Appropriate modifications were made for the small effect of en- 

 vironmental temperature. The date of attainment of testicular weight 

 of 100 mg was selected for prediction since, at this stage of develop- 

 ment, the increase in testicular weight still conforms closely to equa- 

 tion (1). The date predicted for 100-mg testes on the basis of 

 experimentally established rates using equation (3) was about 10 days 

 earlier than the date at which this is attained in nature. The "pre- 

 dicted" date is within the statistical error of the basic measurements. 

 Therefore, it can be concluded that for this species the photoperiodic 

 mechanism is of predominant importance in timing the testicular cycle. 

 Since the responses obtained with many other Temperate Zone species 

 (Table I) are of a similar order of magnitude, a generalizing exten- 

 sion of this conclusion appears to be in order. 



It appears appropriate to conclude this evaluation of the photo- 

 periodic control of gonadal cycles with a brief consideration of a 

 phenomenon which has caused some miseivings concerning the gen- 

 eral validity of the theories of photoperiodic control. I have in mind 

 here the phenomenon of autumnal sexual activity which has been 

 extensively noted, particularly in resident species. (See, for example, 

 Benoit, 1950b; BuUough, 1942, 1943; Marshall, 1952.) Although 

 the necessary experimental evidence is lacking, it appears to me that 

 existing data in no way argue against a fundamentally important 

 photoperiodic timer if one makes the following very plausible assump- 

 tions: (1) that the curve representing /: as a function of p (Fig. 2) 

 is displaced somewhat to the left or that it has a definite lower tail 

 with k being a more substantial positive function of p for values of 

 p down to about 9 hr; (2) that the refractoriness of the photoperiodic 

 response mechanism is ended relatively early (August or early 

 September) while day lengths are still sufficiently long to cause 

 appreciable stimulation of the photoperiodic mechanism, (3) and, as 

 is quite obviously the case from field studies, that these species must 

 be somewhat more temperature sensitive and possibly more sensitive 

 to "psychic" factors. These characteristics would then permit an 

 autumnal gonadal development which varies from year to year de- 

 pending on weather conditions. Thus also the differences in vernal 

 gonadal development, which can be correlated well with weather 



