RICHARDSON: LARVAL FISHES OFF OREGON 



Table 1. — Number of deep hauls taken, number of positive hauls, number of taxa taken, volume of water filtered, num- 

 bers of larvae originally in samples, number of larvae/1,000 m^ of total volume of water filtered according to type of 

 gear and cruise. 



results in a loss of rare taxa in the split fraction. 

 The IKMT was towed the longest and filtered 

 the greatest volume of water which increased 

 the probability of capturing rare taxa. This fac- 

 tor outweighed the loss of larvae by escape- 

 ment through the 5 -mm mesh liner which would 

 be expected to be great. 



Bongo samples gave the largest estimate of 

 larval abundance (individuals/1,000 m^) for 

 combined cruises. (Although not done in this 

 paper, subsequent data on larval fishes will be 

 standardized to numbers under a unit area of 

 sea surface as encouraged by E. H. Ahlstrom.) 

 Bongos captured a large number of anchovies 

 on the July-August cruise which mainly 

 accounted for the large estimate for combined 

 cruises. MN samples gave the largest estimates 

 of abundance for five of the six periods sampled. 

 The bongos were towed at a speed of 5 knots 

 which may have resulted in extrusion of small 

 larvae. This would explain the low estimates of 

 abundance on three of the four cruises compared 

 with the MN which was hauled more slowly 



and primarily took small larvae. The IKMT 

 gave the smallest estimates of abundance of lar- 

 val fishes for each cruise and for the combined 

 cruises probably because of a high degree of 

 escapement through the net by small forms. 



Numbers of taxa occurring in 5% or more of 

 the samples for each gear were: 7 for the bon- 

 gos, 5 for the MN, and 19 for the IKMT (Table 

 2). For each gear the four taxa taken most fre- 

 quently were northern anchovy, E)igraidis 

 mordax; northern lampfish, Stetiobrachius 

 leucopsarus; blue lanternfish, Tarletonbeania 

 crenularis, and Sebastes spp. {Sebastes spp. was 

 fifth in bongo samples preceded slightly by Pro- 

 myctophum thompsoni.) However, percent fre- 

 quency of occurrence of each of the four, except 

 S. leucopsarus, varied considerably among 

 types of gear (Table 2). Percent occurrence of 

 E. mordax was similar in bongo and IKMT 

 samples but much lower in MN samples. The 

 IKMT took T. crenularis and Sebastes spp. 

 much more often than either the bongos or the 

 MN. The greater frequency of occurrence of 



699 



