FISHERY BULLETIN: VOL. 85, NO. 2 



(12%), and amphipods (10%, predominantly hyperi- 

 ids and caprellids) contributed the greatest volumes 

 of prey. Copepods (70%), mollusk larvae (47%), am- 

 phipods (30%), and decapod larvae (25%) occurred 

 most frequently. 



Some differences were found in stomach contents 

 of fishes of different size (Table 6). Small fish preyed 

 almost exclusively on copepods. Medium-sized fish 

 preyed predominantly on copepods, but less fre- 

 quently and to a lesser extent than small fishes. 

 Large fish fed on a variety of prey and consumed 

 large prey items (such as chaetognaths). 



Mollusk larvae and copepods dominated the diets 

 in all seasons except spring (Table 7). In spring, 

 round scad fed on copepods, ostracods, chaeto- 

 gnaths, and barnacle cyprids. The preponderance 

 of copepods in diets of round scad in summer is due 

 in part to the large number of juveniles included in 

 the analysis. The mean size of fish in the summer 

 sample v^ras smaller than the mean size of fish in all 

 other seasons (Student-Nev^man-Keuls tests; ^453 4 

 = 14.1 for winter vs. summer, ^453 3 = 13.2 for 

 summer vs. spring, and ^453 9 = 11.1 for summer 

 vs. fall). In all other seasons, dietary analyses (Table 

 5) were based on fish samples with similar size 

 distributions (Student-Newman-Keuls tests; 94533 

 = 0.7 for winter vs. fall, ^453 2 = 0.7 for winter vs. 



spring, and ^453 2 = 0.1 for fall vs. spring). The 

 mean number of prey items showed considerable 

 seasonal variation from 4.6 in fall to 104 in spring. 

 The percentage of empty stomachs varied as a 

 function of time of day (Fig. 6). Few empty stomachs 

 (2-7%) were collected from midmorning to early 

 evening, whereas 13-29% of stomachs were empty 

 from early evening to midmorning. Size effects are 

 unlikely to have caused the observed differences in 

 the percentages of empty stomachs: samples with 

 lower (2, 5, and 7) and higher (13, 20, and 29) per- 

 centages were comprised of fish of similar size 

 (ANOVA, F,i,4) = 0.01, P > 0.75). 



Age and Growth 



Validation of the daily growth marks on otoliths 

 of round scad was provided in two ways. Examina- 

 tion of marginal increments of lapilli from small 

 specimens (13-55 mm FL) collected at different 

 times of day suggested daily periodicity of increment 

 formation. The margin consisted of the transparent 

 incremental zone from midafternoon until early 

 morning and the dark discontinuous zone in mid- 

 morning (Table 8). The allometric relationship be- 

 tween otolith and fish length also validates the use 

 of otoliths for age determination. Otolith length (OL) 



Table 6.— Index of relative importance (IRI), frequency of occurrence (%F0), 

 volumetric displacement (%VOL), and relative abundance of prey (%A/) by sizes. 

 N = sample size (# empty stomachs), x = mean fork length in mm, s = vari- 

 ance of fork lengths, and n = total number of prey. 



258 



