FISHERY BULLETIN: VOL 77. NO. 4 



opening/closing bongo net samples during a strat- 

 ified sampling program in the North Pacific cen- 

 tral gyre (Scripps Institution of Oceanography 

 1974). I describe the vertical abundance and size- 

 depth distributions of the more abundant families 

 (Gonostomatidae, Sternoptychidae, and Mycto- 

 phidae) and species. I also present estimates of the 

 predescent stages of development of species for 

 which there are sufficient catch data. 



METHODS 



All depth stratified samples were obtained dur- 

 ing Climax I expedition (19-28 September 1968; 

 Scripps Institution of Oceanography 1974) near 

 lat. 28° N, long. 155° W. The physical, chemical, 

 and biological properties at this locale have been 

 shown to be stable over large time and space scales 

 (McGowan and Walker in press) and are expected 

 to be representative of the central gyre (Scripps 

 Institution of Oceanography 1974). During this 

 cruise, round-the-clock sampling was done with 

 opening/closing bongo nets (Scripps Institution of 

 Oceanography 1966) between a set of parachute 

 drogues placed at 10 m depth about 10 km apart. 

 The drogues traveled 346 km during the 10 days of 

 sampling. The main depth intervals sampled 

 were: 0-25 m, 25-50 m, 50-75 m, 75-100 m, 100-350 

 m, and 350-600 m; other intervals were also sam- 

 pled. Paired 505 /urn nets on the 70 cm diameter 

 frames each had a mouth area of 0.396 m^. They 

 were opened at the bottom of the desired depth 

 range and fished obliquely upward, closing near 

 the top of the range. Maximum tow depths were 

 determined by Benthos^ depth-telemetering 

 pinger and/or wire angle. The nets were closed 

 automatically by a calibrated flowmeter after 400 

 m^ of water had been filtered per net. Ship speed 

 was nominally 2.5 kn. 



All fishes were sorted from 60 samples repre- 

 senting 38 separate tows (one sample equals the 

 catch from one of the paired nets). Samples in- 

 cluded: ten each from 0-25 m, 25-50 m, 50-75 m, 

 and 75-100 m; six from 100-225 m; six from 100- 

 350 m; and eight from 350-600 m (Table 1). 



To reduce biases due to net avoidance by larger 

 individuals and more agile species (Bridger 1956; 

 Ahlstrom 1959) most analyses were of "night" 

 samples (taken between 2000 and 0600 local 

 time). To provide enough replicate samples to 



allow statistical analyses of strata deeper than 50 

 m, I found it necessary to include 1 1 "day" sam- 

 ples; these were selected from tows taken as close 

 to dawn or dusk as possible. As a result, this study 

 does not include aspects of diurnal changes in 

 depth distributions. Ahlstrom ( 1959) and Badcock 

 and Merrett (1976) showed that the larvae of some 

 midwater fishes (size or stage of development 

 unreported) do undergo limited diurnal migra- 

 tions. 



I identified all fishes caught to the lowest taxon 

 possible. These were categorized to stage of de- 

 velopment (i.e., larval, metamorphic, post- 

 metamorphic or juvenile, adult), enumerated and 

 measured to the nearest 0.1 mm standard length 

 (SL) (notochord length was measured for preflex- 

 ion larvae). The data presented in this paper (un- 

 less otherwise noted) are based on larval to early 



Table l. — Opening/closing bongo net samples used for larval 

 fish depth distribution analyses. Samples taken near lat. 28° N, 

 long. 155° W in the North Pacific central gyre during September 

 1968. L and R designate left or right sample from paired net 

 assembly. Mean and standard deviations of temperature at 

 upper iTfj \ and lower iTi^ ) limits of each interval based on 10 

 day and 7 night 0-500 m STD lowerings. 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



778 



