H0GUE and CAREY: FEEDING ECOLOGY OF 0-AGE FLATFISHES 



Citharichthys stigmaeus; and sand sole, Psettichthys melanostictus. Numbers in parentheses under the heading "No. fish examined" 

 numerical percent composition and average frequency of occurrence (in parentheses). " — " indicates prey item <0.01. 



Large benthic prey 



Pelagic prey 



Species and 

 date 



No tows 

 exam- 

 ined 



No. fish 

 examined 



Length 

 range 

 (mm) 



Amphi- 

 pods 



Cumacea Decapods 



Poly- 

 chaetes 



Mysids 



Calanoid 

 copepod 



Veliger 

 larvae 



ii--A-6-6--A--rj— g — 6— g--g 



— • P. VETULUS 

 a--a I. ISOLEPIS 



I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 

 NO. FISH 



Figure 2. — Cumulative number of prey categories sampled as 

 a function of sample size for Parophrys vetulus and Isopsetta 

 isolepis. Each data point represents the mean of two separate 

 trawl collections. Vertical bars are ±1 standard deviation. 



contents varied little among individuals of a spe- 

 cies within a trawl. For each species studied, food 

 items from fish caught in the same trawl tended 

 to have the same rank order of numerical abun- 

 dance. This consistency of results was statistic- 

 ally significant; the null hypothesis of inde- 

 pendence among prey rankings was rejected 

 (Friedman's nonparametric randomized block 

 ANOVA, P<0.005; Gibbons 1971). On several 

 sampling dates the number of fishes available 

 for study was small, e.g., Parophrys vetulus on 5 

 September 1978 (three fish), /. isolepis on 30 May 

 1979 (four fish). The small within-sample vari- 

 ability of these species, however, indicates that 

 such small sample sizes will not unduly bias de- 



559 



