have noted the frequent abundance of animals, particularly zooplankton, in 

 oxygen-poor waters and the frequent association of both with sound scattering 

 layers. 



An oxygen deficient layer was found between about 600 and 1000 meters 

 in Area Bravo (Figure 9): the depths encompassed by the middle and deep samples 

 of tow 1-D. Unfortunately, the depth of minimum oxygen concentration and the 

 opening depths of the three sample nets, are not known precisely. Thus, we 

 cannot exactly correlate fish abundance with the O2 minimum, but only note 

 that the deep sample with the heavy fish concentration came from the lower 

 part of the oxygen deficient layer. 



It is clear that in all three regions, the most numerous diurnally vertically 

 migrating fish are the lanternfish or myctophids; although, the bristlemouths or 

 gonostomatids are prominent and, at times, may exceed the myctophids in 

 numbers. Along with Pearcy and Lours (op.cit.) and others we also found a 

 predominanace of only a few species within a collection. To some extent this, 

 as well as the diurnal distribution pattern, is a reflection of gear selectivity, 

 but It also reflects the actual distribution. Pearcy and Lours (ibid) discussed 

 the problem of gear selectivity and net avoidance by fish and attempted 

 quantitative estimates of avoidance. Many others have discussed the problem 

 of gear selectivity in capturing other animals, even including, rather surprisingly, 

 copepods (Grice and Hulsemann, 1968). 



Although our estimates of fish concentrations indicate that the Be Multiple 

 Plankton Sampler catches as many or more fish as the six-foot Isaacs-Kidd Mid- 

 water Trawl, when the volume of water filtered is taken into account, the fish 

 caught by the six-foot I KMT are much larger than those taken by the MPS. 

 The MPS catches relatively more small fish. This may be because the liner of 

 the IKMT is much coarser mesh than the MPS, while the mouth opening of the 

 MPS is too small to catch most large fish. Harrisson (1967), in discussing the 

 methods for sampling mesopelagic fish, pointed out the shortcomings of even the 

 lO-foot IKMT for taking a representative sample of the mesopelagic fish fauna. 

 He also discussed the increased catches that the 10-foot IKMT made during 

 oblique hauls as compared to horizontal hauls. He attributed this largely to the 

 increased speed of the net coupled with the behavior and swimming abilities of 

 the hatchetfish on which he based his discussion. Our data from the Gulf of 

 Mexico indicate that many of the oblique hauls also had substantially larger 

 catches than many of the horizontal hauls. The oblique hauls of short duration 

 moved as much as 12 to 20 percent faster than the horizontal net hauls, although 

 the ship speed remained constant. This increased speed must have some effect 

 on increasing the catch in the oblique hauls. However, this cannot explain the 

 low fish concentrations in some of the oblique hauls, nor the high fish concentration 



29 



