CLARKE: ECOLOGY OF LANTERNFISHES 



in September and December 1970 and were 

 significantly higher in March and June 1971. 



The decline in abundance observed for T. 

 nigrescens in June and September 1971 was the 

 most dramatic change observed among all the 

 species during the study period. The difference 

 between the series in July and September 1970 

 and those of June and September 1971 suggest 

 that either spawning success or larval survival 

 was markedly lower in 1971. The only obvious 

 factor correlated with this change was a sig- 

 nificant increase in the percentage of females 

 among mature fish in June 1971. Whether 

 there was a causal relationship between the 

 two changes cannot be determined from the data. 



Boliiiichthys lo)igipes 



Apparently almost all B. lo)igipes of the sizes 

 sampled regularly migrate. The only indication 

 of nonmigration was a catch of seven small 

 (12-14 mm) individuals at 650 m during the 

 night in June 1971; individuals of these sizes 

 were caught frequently during the day tows 

 but rarely at night in the upper layers suggest- 

 ing that some small individuals do not migrate. 

 Within each series there were relatively few 

 significant differences between size-frequency 

 cur\^es of the separate tows. These indicated 

 that the smaller fish tended to be found shal- 

 lower at night. Differences between day tows 

 clearly indicated (Figure 10) that few fish less 

 than 20 mm occurred deeper than 625 m and 

 few larger than 30 mm occurred above this 

 depth. 



In the March 1971 night samples, the per- 



100 n 



3 



so- 



lo 20 30 40 



STANDARD LENGTH (mm) 



Figure 10. — Cumulative size-frequency curves for samples 

 of Bolinichthvs longipes collected during the day during 

 June 1971 at depths of 560 m, 11 individuals (A); 600 m, 

 18(B): 620 m, 21 (C); and 640 m, 11 (D). 



centages of females among mature fish de- 

 creased with depth. The values were 73% at 80 

 m, 59% at 95 m, 32% at 120 m, and 46% at 135 m. 

 The first two values were significantly different 

 from the latter two. Insufficient mature fish 

 were caught to compare values at different 

 depths in other series. 



The calculated totals agreed well for the day 

 and night series, but calculated size-frequency 

 curves differed for almost all pairs. Depth 

 coverage was best for the March and June 1971 

 series, and the calculated curves were closest 

 for the March 1971 series. The curves for the 

 June 1971 series differed mostly because of a 

 greater frequency and number of small indi- 

 viduals in the day. The numbers caught were 

 quite low for all tows of this series, and 

 inclusion of seven small individuals from a deep 

 night tow (see above) in the calculations brought 

 both the calculated totals and curv^es into very 

 close agreement for the June 1971 series. 



The CT tows indicated that peak depth was 

 at 75-100 m and caught fair numbers of B. 

 longipes at 25-50 m where the IK caught few 

 or none. The IK peak depth was at 100 m and 

 substantial numbers were caught at 125 and 

 135 m where CT catches were relatively low. 

 The ratios of CT to IK numbers were 9.4 and 

 5.6 at 75 m and 100 m, respectively, but only 

 1.1 at 125 m. The calculated CT total was 5x 

 that for the IK. The IK and CT size-frequency 

 curves were similar at 75 m but differed sig- 

 nificantly at 100 and 125 m due to larger 

 percentages of fish over 35 mm in the CT 

 catches. The calculated CT curve was displaced 

 to the right also due to higher percentages of 

 fish over 35 mm. Thus it appears that the CT 

 sampled B. lougipei^ as well or better than the 

 IK above 100 m, but in the deeper zones the 

 IK gave higher estimates of abundance in spite 

 of missing some larger fish. 



At new moon, B. longipes was abundant 

 between 60 and 125 m and peaked at 80-100 m. 

 At full moon, practically none were caught 

 above 130 m and the peak was at 170-190 m. 

 Full moon tows in other series indicated that 

 few occurred deeper. The new moon catch at 

 60 m and the full moon catches at 130 and 

 170 m were similar in size composition; almost 

 all fish were <20 mm. The 190-m curve at full 



421 



