WATSON and DAVIS: LARVAL FISH DIETS IN SHALLOW COASTAL WATERS 



Table 1. — Number of fish larvae dissected on eacfi survey date. Pr = preflexlon stage larvae; FP 



flexion and postflexion stage larvae. 



Table 2. — Feeding cfironology of larval Paralabrax spp., Paralichthys californicus, and Seriphus 

 politus. Survey A samples were collected at thie 8 m isobath on 25 July 1978: Survey B samples 

 were collected at the 13 m isobath on 21 September 1978. The time of day when samples were 

 collected is given as Pacific Standard Time. N, = total number of specimens dissected; N, = 

 number of specimens containing food; % Fl = percent feeding incidence (100 > N,); 95% confi- 

 dence limits (C.L.) are given for % Fl. 



during the day and 12% at night. Feeding inci- 

 dence was highest (86%) during midafternoon 

 (1345-1420 PST), dropped to 50% within an hour 

 after sunset, and was 0% by about midnight 

 (Table 2). The 95% confidence limits about %FI 

 were rather broad owing to the small sample 

 sizes, and it is unclear whether there were any 

 real differences in feeding incidence from morn- 

 ing through early evening. However, the large 

 differences in %FI and nonoverlapping confi- 

 dence limits between the morning through even- 

 ing and midnight sampling episodes do indicate a 

 real reduction in feeding incidence at night. 



Paralichthys califor7iicus larvae displayed a 

 low feeding incidence: 38% during the day and 

 22% at night. More larvae contained food items 

 during midmorning that at any other time; none 

 contained food in the midafternoon samples (but 

 only seven larvae were available for disssection 

 in the afternoon samples). At night, feeding in- 

 cidence decreased from 27% about an hour after 

 sunset to 15% by about midnight (Table 2). How- 

 ever, owing to the small sample sizes and conse- 

 quent broad confidence limits about %FI, it is 

 unclear whether there were any real differences 

 in feeding incidence over this period. 



571 



