FISHERY BULLETIN: VOL. 70, NO. 4 



adaptive significance of the "refractoriness" to 

 20°C in July may be to prevent "premature" 

 initiation of spermatogenesis should tempera- 

 tures decrease somewhat during the regression 

 phase. 



Although the data presented here are by no 

 means conclusive, they suggest that photoperiod 

 has a variable effect at 20°C. Long photoperiods 

 may be more effective in maintaining spermato- 

 genesis whereas short photoperiods seem to pro- 

 mote a faster rate of recrudescence at 20°C. 

 Moreover, active spermatogenesis is maintained 

 at 22°C only with a long photoperiod. Differ- 

 ences are evident between the sexes since vitel- 

 logenesis does not occur at this temperature re- 

 gardless of photoperiod. Since a majority of 

 spawning in GilUchthys occurs when daylength 

 is increasing and recrudescence occurs when 

 daylength is decreasing, the variation in the ef- 

 fects of photoperiod seen here seem reasonable. 



That 20°C is not as effective as lower temper- 

 atures in maintaining gametogenesis or promot- 

 ing gonadal recrudescence is consistent with the 

 phenological and climatic data presented by de 

 Vlaming (1972b). Average daily temperatures 

 are below 20°C from early October to mid-June; 

 it is during this period that most of the gonadal 

 activity occurs in this species. 



Complete gonadal recrudescence in both male 

 and female GilUchthys occurred at 12° and 13°C, 

 suggesting that the decreasing temperatures in 

 autumn are primarily responsible for regulating 

 this process. Constant low temperature treat- 

 ment promoted a faster rate of recrudescence 

 than occurred in the natural population. Diurnal 

 increases in temperature in the natural habitat 

 during autumn may account for the slower rate 

 of gonadal recrudescence. Short photoperiods 

 augmented the rate of recrudescence at low tem- 

 peratures in both males and females. Thus, the 

 decreasing photoperiod in autumn probably fa- 

 cilitates, but is not essential for, the effects of 

 decreasing environmental temperatures in pro- 

 moting gonadal recrudescence. 



With a 12L/12D photoperiod, gonadal recru- 

 descence was initiated within 21 days at 10° and 

 18°C. This suggests that gonadal response in 

 this species is relatively fast; rapid mobilization 

 of energy for gonadal recrudescence may be 



characteristic of species that spawn more than 

 once in a season. Whether recrudescence was 

 more rapid at 10° or 18°C was not investigated, 

 nor are there sufficient data to indicate whether 

 an optimal temperature exists for the comple- 

 tion of gonadal recrudescence. However, the 

 data presented here show that gonadal recrudes- 

 cence will occur over a wide temperature range 

 (10°-20°C). This gonadal responsiveness to a 

 wide range of temperatures may reflect an adap- 

 tation to the labile thermal environment of this 

 species. 



Active gametogenesis was maintained at low 

 temperatures regardless of photoperiod. After 

 treatment at a low temperature in May (Figure 

 4) , more male fish were in the prespawning con- 

 dition at the longer photoperiod. In the same 

 experiment, however, a short photoperiod was 

 more effective in maintaining vitellogenesis 

 (Figure 4) . Perhaps these results reflect a true 

 photoperiod influence, but because of the vari- 

 able gonadal conditions of the beginning controls 

 no conclusive statements can be advanced. In 

 fact, experiments begun in January with fish 

 revealing less variable gonadal conditions indi- 

 cated that long and short photoperiod treatments 

 maintained active gametogenesis equally well at 

 low temperature. Photoperiod influences, how- 

 ever, could vary between January and May. 



The question arises as to why all fish treated 

 at low temperatures did not progress to the pre- 

 spawning condition. One possible explanation 

 is that many of the experiments discussed here 

 were relatively short term. In many of the 

 long-term experiments some of the fish could 

 be stripped of ripe eggs and sperm. It is likely, 

 however, that physical factors other than tem- 

 perature and photoperiod, and also social factors, 

 influence final gonadal maturation and spawn- 

 ing in this species. Indeed, Weisel (1947) 

 has indicated that GilUchthys is territorial and 

 has an elaborate courtship behavior. In addi- 

 tion, de Vlaming (1971b) has shown that sa- 

 linity changes and changes in the availability 

 of food can alter the rate of gametogenesis in 

 GilUchthys. Thus, the failure of many of the 

 experimental fish to completely attain the pre- 

 spawning condition was probably due to the ab- 

 sence of certain conditions in the laboratory sit- 



1150 



