FISHERY BULLETIN: VOL. 71, NO. 2 



majority of species, the juveniles tended to 

 occur shallower than adults. The trend was 

 more obvious in the night samples due to closer 

 sample spacing, but probably holds for the 

 daytime depth distribution in most cases. 

 Hygophum proximum and H. reinhardti were 

 anomalous in that at night during new moon 

 the adults tended to occur shallower than the 

 juveniles. 



The changes in size composition with depth 

 may be an artifact caused by larger fish avoid- 

 ing the trawl better at lesser better-lighted 

 depths. However, many small species such as 

 Benthosema suborbitale, Diaphus .'^schmidti, 

 Triphoturus nigrescens, and Notnlychnus val- 

 diviae are unlikely to avoid the trawl even as 

 adults, and the trends with depth were as 

 evident with these species as with the larger 

 species. Also the same trends were evident 

 from CT data for species which the CT appeared 

 to sample larger individuals relatively better. 

 Furthermore, avoidance by larger fish would 

 only partly explain the changes noted for many 

 species, i.e., the smaller fish were absolutely 

 less abundant with depth. 



The ecological significance of the trends in 

 size with depth is not clear. The juveniles are 

 removed to some extent from predation by 

 adults, but this is of doubtful significance since 

 other predators are probably more important. 

 Possibly the smaller fish are less visible to 

 predators than are the adults in the better- 

 lighted shallow layers. In those species where 

 adults occur farther below the juveniles in the 

 day than at night, it may be related simply to 

 the juveniles being unable to make longer 

 migrations. This pattern could be related to 

 distribution of food. It is quite likely that 

 juveniles require greater concentrations and 

 smaller average particle size of food than adults, 

 and optimal conditions for juveniles may occur 

 shallower than for adults. 



Sex Ratios 



Sex ratios of mature and juvenile fish of 

 several species are compared in Table 3. The 

 difference is significant for only one species, 

 Diaphus schmidti, and in that case, the ratios 

 did not differ significantly when data for mature 



Table 3. — Proportions of females among juveniles and 

 mature fish for nine species of myctophids. Number of 

 specimens examined is given in parentheses. 



fish from a single sample with a very large 

 proportion of females was not included. For 

 most species there was a tendency for higher 

 proportions of females among immatures, but 

 the figures are suspect in the sense that the 

 data were not properly weighted to account for 

 possible trends with depth or season. Legand 

 and Rivaton (1970) have noted opposite trends 

 in sex ratio with size for three species of 

 myctophids. (However, their criteria for separat- 

 ing "larger" from "smaller" fish is not entirely 

 unbiased and the proportions given for one 

 species, Benthosema simile, do not differ 

 significantly.) 



Given that there may be trends in sex ratio 

 with size, recognition of trends with depth or 

 season was made difficult because there were 

 only a few series where sufficient numbers of 

 similar-sized fish were captured at several 

 depths or several seasons. Thus, significant 

 trends in sex ratio with depth were observed for 

 mature fish of only a few species. Quite possibly, 

 more data would show that these trends are the 

 rule rather than the exception. When data 

 were pooled for all mature fish examined from 

 each series (Table 4), some significant dif- 

 ferences in sex ratio were observed. In most 

 cases one series showed significantly higher or 

 lower percentage of females than the rest with 

 no suggestion of seasonal trends. Again, how- 

 ever, differences in depth composition of the 

 samples or size composition of the fishes pooled 

 for a series may have either obscured trends or 

 have been responsible for the few significant 

 differences observed. 



Seasonal Trends 



In considering seasonal trends, it was assumed 



426 



