FISHERY BULLETIN: VOL. 70, NO. 4 



Nearly 50 9f more fish larvae were obtained 

 per haul, on the average, in ETP II (347.0 lar- 

 vae) as compared with ETP I (231.9 larvae). 

 The larvae of most families of fishes were taken 

 in larger numbers on ETP II than on ETP I, 

 and larvae of several families were taken in 

 markedly larger numbers. The striking excep- 

 tion to this trend was afforded by scombrid 

 larvae; only 13.5% as many scombrid larvae 

 were obtained in ETP II collections as in sim- 

 ilar coverage of ETP I. 



Families showing the largest increase in total 

 numbers of larvae on ETP II compared with 

 ETP I included Engraulidae (1,360 to 205 lar- 

 vae), Scomberesocidae (153 to 1 specimen), 

 Scopelosauridae (390 to 13 larvae), Everman- 

 nellidae (67 to 13 larvae) , Astronesthidae (75 to 

 13 larvae), Trachichthyidae (70 to larvae), 

 Sciaenidae (127 to 12 larvae), Trichiuridae (186 

 to 48 larvae), Gempyhdae (370 to 110 larvae), 

 and Clupeidae (270 to 81 larvae). For the ma- 

 jority of these, the increase in relative abun- 

 dance of larvae was most marked in the inner 

 pattern occupied by Rockaway. The species 

 compositions involved in these increases, when 

 known, are discussed later under the respective 

 families. 



For the majority of families, however, the in- 

 crease in abundance of larvae on ETP II was 

 moderate, seldom as much as double; for a third 

 of the families, counts of larvae were not much 

 different during the two contrasting periods of 

 the year; thus the similarity in abundance of 

 larvae during the two periods is the striking 

 feature of this comparison. 



COMPARISON WITH EASTROPAC 

 MONITORING CRUISES 



The portion of the eastern tropical Pacific pat- 

 tern that could be covered by a single research 

 vessel on surveys averaging 45 days was occu- 

 pied at bimonthly intervals on four monitoring 

 cruises by David Starr Jordan. The cruises 

 were numbered as follows: 20.000 series for 

 the April-May 1967 monitoring cruise, 30.000 

 series for the June-July monitoring cruise, 

 50.000 for the October-November 1967 moni- 



toring cruise, and 60.000 for the December 1967- 

 January 1968 monitoring cruise. The monitor- 

 ing pattern is shown superimposed on ETP II 

 stations in Figure 2; it consisted of four station 

 lines all ending at lat 3°S. The outer line along 

 long 119°W extended from lat 20°N, the inner 

 line along long 98°W, off the Mexican coast (ca. 

 lat 17°N) . The two middle lines along long 105° 

 and 112°W were doglegs veering coastward 

 fi^om about lat 12° or 13°N — one line ending up 

 off Manzanillo, Mexico, and the other off Aca- 

 pulco, Mexico. 



Coverage equivalent to the monitoring pattern 

 was determined for ETP I and II. For ETP I 

 the following stations were occupied: 11.022 

 to 11.118 (35), 12.002 to 12.109 (50), 12.209 to 

 12.264 (24), and 13.187 to 13.265 (28); total 

 137 stations. For ETP II comparable coverage 

 was obtained from Stations 45.016 to 45.114 

 (41), 45.191 to 45.365(37), 46.002 to 46.069 

 (36), and 46.079 to 46.132(27) ; total 141 sta- 

 tions. 



COMPARISON OF LARVAL 



COMPOSITION IN MONITORING 



PATTERN VERSUS LARGER 



EASTROPAC PATTERN 



In Table 11 a comparison is made for both 

 ETP I and ETP II of the average number of 

 larvae per haul and percentage contribution of 

 larvae of the more important fish families in the 

 monitoring pattern as compared with the total 

 pattern occupied on ETP II. 



The correspondence between relative abun- 

 dance and composition of larvae in the monitor- 

 ing pattern as compared with the larger ETP 

 pattern is closer for ETP I collections than ETP 

 II. The average abundance of larvae in the moni- 

 toring pattern on ETP I was 92% as large as for 

 the larger ETP I pattern (equivalent to the cov- 

 erage obtained on ETP II) . The more abundant 

 kinds of larvae — myctophids, gonostomatids, 

 and sternoptychids — had similar average abun- 

 dances per haul and similar percentage contribu- 

 tions in the monitoring pattern as compared 

 with the more extensive ETP I coverage. Of 

 the remaining seven families used in this com- 



1166 



