570 



Fishery Bulletin 89(4). 1991 



♦12° C 

 H 

 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 



DAYS AFTER HATCHING 



Figure 1 



Survival (%) over a 17-day period of California halibut Para- 

 Iwhthys californicus larvae subjected to various temperatures. 

 Groups of larvae were either (A) starved or (B) fed rotifers. 

 Points are replicates (re 2) combined. Each replicate originally 

 consisted of 75 individuals. The fed experiment did not include 

 an 8°C trial because larvae at 8°C did not develop functional 

 mouths or eyes and died with undepleted yolksacs. 



tional to temperature: 74, 50, and 34 hours at 12, 16, 

 and 20°C, respectively. Although not quantified, most 

 eggs remaining (the initial number was reduced by 

 sampling) at each temperature hatched. No abnormal 

 yolksac larvae were observed. 



Larval growth and survival 



Replicates (n 2) of survival curves at each temperature 

 for starved and fed larvae were combined for analysis 

 since they were not significantly different (P<0.05). 

 Survival curves of 12, 16, 20, and 24°C starved trials 

 were significantly different (P<0.01), with starvation 

 time (6-17 days) inversely proportional to temperature 

 (Fig. 1A). Total mortality occurred first at temperature 

 extremes, 8 and 24°C, which had similar mortality 

 curves. Larval development at 8°C never progressed 



Table 2 



Mean notochord lengths of larval California halibut Paralich- 

 thys californicus 17 days after hatching at four temperature 

 treatments. Two replicates were maintained at each tempera- 

 ture, with initially 75 late-stage eggs per replicate. N = com- 

 bined number of surviving larvae in the two replicates. 



Temperature (°C) 

 Mean SE 



Notochord length (mm) 

 Mean 



beyond the yolksac stage with unpigmented eyes and 

 nonfunctional mouths (thus the fed experiment did not 

 include an 8°C trial). At 24 °C, however, dead larvae 

 had pigmented eyes, functional mouths, and completely 

 depleted yolks. 



Survival curves of 20 and 24 °C fed larvae were 

 statistically similar, while the 12 and 16°C fed trials 

 differed (P<0.05). At 24°C, mortality initially was 

 high, perhaps due to temperature acclimation prob- 

 lems, but then mortality was gradual, paralleling the 

 20°C curve. At 16°C, survival initially was highest, but 

 dropped after 12 days. Survival was high for fed larvae 

 at 12 °C until the end of the second week when the mor- 

 tality rate increased, resulting in almost total mortal- 

 ity by day 17 (Fig. IB). Although this survival pattern 

 approximates the starved 12 °C counterpart, the two 

 curves are significantly different (P<0.01; more larvae 

 in the fed trial survived longer). Fed larvae at 12°C 

 initiated feeding (since food was observed in their guts) 

 but survival enhancement was only short-term. 



Final survival of fed larvae at 16°C was intermediate 

 to final survival values at 20°C and 24°C; final survival 

 was highest at 20°C and nearly zero at 12°C (Fig. IB). 

 Final survival did not significantly differ between tem- 

 perature treatments (P 0.14), however. At experimen- 

 tal end, mean notochord lengths of surviving fed larvae 

 were directly proportional to temperature (Table 2) 

 and significantly different between all temperatures 

 (P<0.05). 



Juvenile growth and survival 



Replicates (n 3) did not differ and were combined for 

 analysis. Survival of 97-day-old fish at 16°C was sig- 

 nificantly (P<0.05) lower than survival at 20, 24, or 

 28°C (Fig. 2); there were no significant survival dif- 

 ferences between the three higher temperatures. Final 

 mean standard length at 16°C did not differ signif- 



