FISHERY BULLETIN: VOL 76, NO. 4 



of vertical distribution of species. Thus in Taylor's 

 study, carried out not far from Station Q, P. 

 thonipsojti and D. thcUi were more abundant than 

 S. leiHopsarus, but this was probably because 

 Taylor's net either did not efficiently catch 

 juvenile ( <35 mm) S. lei/copsarus which were the 

 numerically dominant size class of that species in 

 our samples (Figure 4), or they were much less 

 abundant during the time he sampled. Further, 

 Taylor obtained some of the largest catches of D. 

 theta and S. Icucopsariis below 90 m at night. This 

 probably also reflects the sampling bias of his net 

 for larger sizes offish, which at night tend to be 

 more broadly spread over the water column than 

 smaller fish (Figure 5C: Table 6, night series). 

 Unfortunately, Taylor did not report the sizes of 

 fish captured. Except for probable sampling bias 

 toward larger sizes offish, Taylor's results on ver- 

 tical distribution of the nonmigratory P. 

 thompsoni and other species of fish agree with 

 ours. 



Pearcy et al. ( 1977) described patterns of verti- 

 cal distribution of mesopelagic fishes and crusta- 

 ceans off the coast of Oregon. The mesopelagic 

 assemblage there is essentially subarctic in 

 faunistic affinity and the vertical distributions of 

 species are similar to those observed at Stations P 

 and Q. The only notable departure from our re- 

 sults was the finding by Pearcy et al. that sig- 

 nificant numbers of all sizes of S. leucupsarus did 

 not participate, at least on a regular basis, in the 

 diel vertical migrations. Our observations at both 

 Stations P and Q indicate that virtually all S. 

 leucopsarus smaller than about 80 mm performed 

 extensive diel vertical migrations (Figure 5). 

 However, in our studies, S. leucopsarus was also 

 very rare below 400 m (Figure 5, Table 4), whereas 

 Pearcy et al. found large concentrations below 

 that depth. Thus there may be major differences in 

 the vertical distribution and migration behavior 

 of S. leucopsorus in different parts of its geograph- 

 ical range (Paxton 1967). Significantly, Pearcy et 

 al. ( 1977) detected no seasonal variations in verti- 

 cal distributions and migrations for any species, 

 which may also be true for subarctic waters to the 

 north (Taylor 1968). 



Perhaps the most remarkable feature of the 

 mesopelagic fauna of the area sampled was its 

 simplicity. Only four species of myctophid fishes 

 were abundant in the upper 700 m. Two of these 

 species, S. leucopsarus and D. theta, undertook 

 diel migrations of substantial vertical extent; the 

 other two, P. thompsoni andS. nannochir, did not. 



766 



Other taxonomic groups also showed low diver- 

 sity. Among the micronektonic crustaceans there 

 were single species of abundant euphausiid, E. 

 pacifica, and decapod shrimp, Sergestes similis, 

 and both were vertical migrators. The contrast 

 between this relatively simple mesopelagic mi- 

 cronekton fauna and that, for example, in the sub- 

 tropical North Pacific (Brinton 1962a; Clarke 

 1973; Walters 1977) or subtropical North Atlantic 

 (Badcock 1970; Foxton 1970a, b) is striking, but 

 not atypical. Low taxonomic diversity of the 

 mesopelagic micronekton is found in other subpo- 

 lar oceans, such as the Boreal Atlantic (e.g.. Back- 

 us et al. 1971; Zahuranec and Pugh 1971). 



Associated with the taxonomic simplicity of the 

 mesopelagic fauna herein reported, was a rela- 

 tively simple structure of the sound-scattering 

 layers. Generally, both the number and depth of 

 sound-scattering layers change with latitude in 

 the deep ocean; fewer and shallower layers are 

 found in subpolar oceans than in tropical- 

 subtropical oceans (Haigh 1971; Cole et al. 1971; 

 Donaldson and Pearcy 1972; Tont 1976). Our un- 

 published observations on deep sound-scattering 

 layers ( 12-kHz echosounder), taken in September 

 1972 along long. 155°W between Alaska and 

 Hawaii, showed this trend. Subarctic waters had 

 the relatively simple sound-scattering structure 

 illustrated in Figure 2, with single migratory and 

 nonmigratory layers occurring shallower than 

 400 m. In the subtropical waters near Hawaii, at 

 least three sound-scattering layers were observed 

 in the daytime at depths ranging from 260 to 625 

 m, and three to four migratory layers were re- 

 corded. 



It is unlikely that the correlation between 

 taxonomic diversity of the mesopelagic micronek- 

 ton and complexity of the sound-scattering struc- 

 ture in the water column was fortuitous. Attempts 

 to causally relate deep sound-scattering layers to 

 aggregations of mesopelagic organisms were 

 stimulated by hypotheses advanced more than 

 three decades ago (for a review see Hersey and 

 Backus 1962). However, field studies based on net 

 samples taken simultaneously with echosounder 

 records tend to be inconclusive for a variety of 

 reasons. A major difficulty is that different 

 taxonomic groups tend to occur together at the 

 same depths and may even show similar migra- 

 tory behavior. For example, all four of the migra- 

 tory mesopelagic species in our study (Stenobrach- 

 ius leucopsarus , D. theta , E. pacifica , and Sergestes 

 similis) ascended towards the surface layer after 



