ALLEN and DeMARTINI: NEARSHORE DISTRIBUTION AND ABUNDANCE OF PELAGIC FISHES 



ferent (Fig. 7). Members of Group I (especially E. 

 mordax, Seriphus politus, and G. lineatus) dispersed 

 offshore, resulting in a much more even distribution 

 of individuals among net-hauls. Certain species in 

 Groups III and IV also moved offshore at night. 

 Group II persisted over the deeper portions of the 

 study area, with Sarda chiliensis absent at night. 



Interrelations of Depth Block, 

 Water Clarity, and CPUE 



The relations between CPUE and water clarity 

 were significantly negative for five of seven major 

 taxa in one or more depth blocks during the day (Ta- 

 ble 5). Water clarity differed significantly among 

 depths (Shallow block: x = 4.4 m; middepth block: 

 6.8 m; deep block: 9.6 m; Kruskall-Wallis H = 33.8, 

 df = 2, P < 0.001). ANCOVA results, however, in- 

 dicated that, although water clarity in addition to 

 depth block were in general negatively related to 

 daytime catches, there were no depth by water clarity 

 interactions (all P > 0.10). On average, only an 

 estimated 12-16% of the total variation in the CPUE 

 of major taxa was attributable to variations in water 

 clarity with any depth block (Table 5). 



Diel Effects on Catch Efficiency 



The percentage recapture of fin-clipped Seriphus 

 politus on average was significantly (P < 0.05) greater 

 {x± SE = 13 ± 3%,JV= 40 net-hauls) duringthe day 

 versus night (25 ± 3%, N = 24 hauls). Parametric 

 estimators for the percentage recapture versus water 

 clarity relation were not possible for daytime tests 

 due to the nonnormality of these data; however, 

 about 30% of the variation in the ranks of daytime 

 recapture data was attributable to water clarity (r = 

 —0.54, P <0.001). Queenfish recaptures at night 

 were insignificantly (Pearson's r = — 0.31,P= 0.14) 

 related to water clarity. 



Day-Night Feeding Comparison 



No significant differences in stomach fullness (CI) 

 were found between day- and night-caught speci- 

 mens of E. mordax, G. lineatus, and P. simillimus (Ta- 

 ble 6). Specimens of Seriphus politus, however, had 

 significantly greater foregut fullness at night, while 

 those of Atherinopsis calif orniensis had significantly 

 greater fullness during the day (Table 6). 



TABLE 5. — Summary of Spearman's rank correlation (r s ) values and significance levels 

 for comparisons of CPUE and water clarity within depth blocks for two categories and 

 four major species plus one taxon (* = P < 0.05; ** = P < 0.001). Dashes indicate 

 insignificant (P> 0.05) results. Also indicated is the estimated percentage of total varia- 

 tion (ft 2 ) of daytime CPUE explained by variation in water clarity in each depth block./? 2 

 approximations reflect results of parametric regressions performed in an ANCOVA with 

 water clarity as covariate and depth block as treatment. 



Table 6. — Summary of results of day versus night comparisons of the contents 

 index ( CI) of foregut samples of the five most abundant fishes present in lampara 

 collections (* denotes statistical significance). 



TEST — Wilcoxon signed-rank test of median CI for paired day-night samples 



Species 



N 

 (day-night) 



Paired 



N 



Ts 



581 



