660 PHOTOPERIODISM IN VERTEBRATES 



spermatogonia! mitoses, production of primary and secondary sperma- 

 tocytes and even spermatids, and tlie formation of islets of proliferating 

 cells within the previously formed sperm mass. These contrasting 

 testicular states have their close if not identical counterparts in Ennea- 

 canthus but under almost opposite conditions. In Enneacanthus, 

 spermatogenesis is completed early in autumn, and a quiescent testis 

 persists thereafter under short day lengths whether at very low or at 

 high temperature, whereas long days combined with high temperature 

 cause a recrudescence of spermatogenesis coincident with the onset of 

 precocious spawning. The interpretation of these results with Gastero- 

 steus is complicated by the fact that the end of the experimental period 

 virtually merged with the onset of the natural breeding period, whereas 

 the results with Enneacanthus were obtained in mid-December far 

 from the breeding period. Egg maturation in Gasterosteiis was arrested 

 by the constant short day length, but progressed with increasing day 

 length, more rapidly when the same total amount of illumination was 

 administered as multiple rather than as single daily exposures (see be- 

 low). Only under continuous illumination did males mature completely 

 and build nests, and females become very gravid. 



Kazanskii (1951, 1952) in contrast to the findings of vanden Eeck- 

 houdt, reports a lack of responsiveness in the same species (during 

 December- January) at temperatures as low as 10°C. In experiments 

 begun December 7 with fish fresh from the wild, when vitellogenesis 

 is just commencing in the most advanced oocytes and yolk has not yet 

 aggregated, continuous lighting at 20-22 °C induced nest building in 

 16 days, complete egg maturation in 18, and spawning followed by 

 parental care in 20. Maturation was slower at temperatures moderately 

 below this. Even in natural daylight during winter, at 19-20°C ma- 

 turation followed by spawning occurred within 37 to 45 days. The 

 ovaries of females under continuous light at temperatures below 12- 

 13°C remained unchanged from their condition at the onset of the 

 experiment. In virtually day-long darkness at 17-1 9 °C during two 

 winter months there was ovarian status quo. 



A fresh approach has recently been made in the study of the annual 

 sex cycle of Gasterosteiis (Baggerman, 1957). Starting at various 

 times throughout the calendar year, different groups of fish were 

 exposed simultaneously to contrasting combinations of day length and 



