specimens and from 57 percent to 100 percent of the species were myctophids. 

 In the collections with lower fish concentrations the shallower samples also 

 contained a large proportion of myctophids. However, in the deep collections 

 outside the loop water the percentages of specimens and species of myctophids 

 declined and larger number of gonostomatids as well as other groups were 

 represented. 



The entire increase in numbers of fish in 7-Tl-A, the heavy concen- 

 tration sample of Tow 7 , is due to an increase in the numbers of three species: 

 Diaphus splendidus , Lepidophanes guntheri , and Notolychnus valdiviae (Figure 5). 

 These same species occur in all the other samples in that tow, but in small numbers 

 only. This same pattern is repeated in two out of the three other heavy concen- 

 tration samples: a few species make up the great percentage of the specimens 

 though the species involved are different. In sample 1-T2-C, Lepidophanes 

 guntheri , Notolychnus valdiviae , and Netoscope I us resplendens , account for 

 74 percent of the specimens but only 38 percent of the species, and in sample 

 5-Tl-B, Notolychnus valdiviae , Diaphus mollis , and Vinciguerria poweriae 

 account for 50 percent of the specimens but only 17 percent of the species. All 

 of the above mentioned species are myctophids except the gonostomatid V. 

 poweriae . 



An examination of the depth recorder record from Tow 7-Tl (Figure 6) 

 shows no prominent feature at the depth of sample 7-Tl-A, 106 to 108 meters. 

 Examination of the depth recorder records taken in conjunction with the other 

 tows indicates the same situation. There are also no differences between the 

 depth recorder records made in the loop water and those made outside the loop 

 water. 



There are less data available from the shallower portions of the night 

 tows in the Gulf of Mexico and analysis of the samples indicates a more complex 

 situation. Three of the six samples have high concentrations of fish per 1000 

 cubic meters of water sampled (Figure 7). Two of these three samples are oblique 

 tows to the surface while the other sample is a horizontal sample (Figure 8). 

 There is some indication of a layer of greater fish concentration between 50 

 meters and the surface, but if such a concentrated layer exists, it must be dis- 

 continuous or patchy to account for the low fish concentration in sample 3-Tl-D. 

 The very high concentration of fish in sample 1-T2-D may be the result of a 

 handling error in the laboratory. 



There were four comparable night tows taken by the LYNCH in 

 July 1966 in the Gulf of Mexico at two stations: Lima and Lima 2. All were 

 made using the open MPS and were oblique net hauls from 90 meters to the 

 surface. The fish concentrations estimated from the LYNCH data are in fair 



