Stream complex. The approximate limits of this loop water, as delineated by the 

 22°C isotherm for 50 meters, in the vicinity of collecting stations 1 through 7 

 ore indicated in Figure 1 . The loop water is characterized by higher salinity 

 and temperature with a pronounced peak centered at around 200 meters on T-S 

 diagrams. Stations 1 and 5 are outside the loop and stations 2, 3, 4, and 7, 

 as well as stations Lima and Lima 2 are in the loop. 



As expected, all of the shallow tows taken in the daytime contain 

 practically no fish, even when the tow was taken exactly at a depth that corres- 

 ponded to a weak scattering layer on the depth recorder. The fish that were 

 taken were predominantly small, mostly juveniles or larvae, but included a few 

 small myctophids, most often Notolychnus valdiviae and a few species of Diaphus , 

 and a few small, silvery gonostomatids, mostly Bonapartia pedaliota . 



The shallow tows taken in the surface scattering layer at night, 

 after the DSL had made its evening ascent, all contained much heavier catches 

 than the daytime tows. Four of the tows are directly comparable. Tows 1-T2 

 and 5-Tl were on either side of the loop water while tows 3-T1 and 7-Tl were 

 in the loop water. 



An examination of the catches of fish (excluding larvae and post- 

 larvae) in the deeper portions of the four tows mentioned above (Figure 3), shows 

 that four of the samples have relatively high concentrations ranging from about 

 2 to 3.3 fish per 1000 m^ of water, as opposed to about 1 fish per 1000 m^ of 

 water for most of the other samples. A diagrammatic depth distribution and 

 configuration of the tow paths with the four higher concentration samples shaded 

 (Figure 4) shows that these four samples were all collected between the depths 

 of 80 to 170 meters. Because three low concentration collections were between 

 depths of 140 and 170 meters, the heavier concentration of fish may be charac- 

 terized as occurring between the depths of 80 and 140 meters. The fact that 

 samples 3-Tl-C and 1-T2-C both were taken in the same depth range and the 

 latter had a higher fish concentration may be related to the fact that it was taken 

 in cooler water. Indeed, other possible evidence for a rise to shallower depths 

 in cooler water is that both deep horizontal samples outside the loop water, 1-T2-A 

 (60.5° to 61 .5°F) and 5-Tl-A (59° to 62°F), as well as sample 5-Tl-B (62° to 70°F), 

 contained specimens of a hatchetfish Argyropelecus aculeatus . Samples 1-T2-A and 

 5-Tl-B also contained a melamphaid, Melamphaes simus. Both of these species 

 tend to be deeper living midwater fish. In the loop water, the deep sample 3-Tl-A 

 (73.5° to 76°F) contained neither of these species. Thus, in the cooler water out- 

 side the loop, these species were caught at relatively shallower depths. 



The great majority of the fish in the four samples that contained the 

 heavy concentrations were myctophids. From 62 percent to 100 percent of the 



