contains a brass fitting set in canvas, by which a 

 cod end (the same cod end as for the 1.5-m. net; 

 len^h not included in the 2.6-m. length of net) 

 can be attached. The material and mesh of the net 

 are the same as for the 1.5-m. net and uniform 

 throughout. The ratio, total filtering area to 

 mouth-aperture area (area of nylon thread in- 

 cluded in the filtering area) , is about 7.5 :1. Three 

 plastic cord bridles extend from the circular frame 

 to the swivel at the end of the towing wire (the 

 same a.s for the 1.5-m. net) ; one 20-kg. bronze 

 streamlined depressor is also attached to this swivel 

 on a 2.4-m. length of 6-mm. welded-link chain 

 so that it is below the net when both are in the 

 water. 



This net has been used only for horizontal hauls 

 at cruising speeds of 9 to 12 knots, with 50 m. of 

 towing wire out. Under these conditions it is about 

 10 m. below the surface (as determined with an 

 attached bathythermograph) . Most hauls with the 

 high-speed net were made at night, because day- 

 time hauls caught little. Generally, only one man 

 and a winchman were required to operate the net; 

 when other operations were compatible, the normal 

 routine was to tow the net about 2 or 3 hours, haul 

 it up to change the cod ends, and put it out again 

 for a similar tow, all without stopping or slowing 

 the ship. The main difficulties (in order of magni- 

 tude) were that (a) other operations often did not 

 permit a reasonably uniform time for each tow, (b) 

 the organisms caught were generally in poor con- 

 dition (witli much flesh lost) when removed from 

 the cod end, and (c) wear generally forced re- 

 placement of the net after about 1,000 to 1,500 

 nautical miles (1,850 to 2,780 km.) of towing 

 (sooner, if catches were heavy) . 



The filtration coefficient (at 9 knots), estimated 

 twice in .3 years in the same way as described for 

 the 1.5-m. net, was 0.938 for the first trial and 0.811 

 for the second. This coefficient .shows that the high- 

 speed net filters more efficiently at 9 knots than 

 the 1.5-m. net does at 5 knots; the two estimates 

 do not agree as well as the two for the 1.5-m. net. 

 The value of 0.938 for the coefficient had been 

 used to standardize the volumes of catches from 

 two cruises before the 0.811 value was determined ; 

 these standardized volumes have been retained for 

 the present, since there is no reason to prefer the 

 second value of the coefficient to the first. Volumes 

 were standardized by dividing actual volumes by 



the number of minutes for the haul and multi- 

 plying by 10.0, since I calculated (from the 0.938 

 coefficient) that the net actually strained 1.000 m.^ 

 of water every 9.96 minutes. Tlie standardized 

 volumes are not tabulated in this paper, but the 

 actual volumes for cruises TO-59-2 and TO-60-2 

 have been listed elsewhere (Blackburn, Griffiths, 

 Holmes, and Thomas, 1962 ; Scripps Institution of 

 Oceanography, 1961), and some of the stand- 

 ardized volumes for tho.se cruises are summarized 

 here in figures 18 and 19. 



TREATMENT OF COLLECTED MATERIAL 



The net catclies were preser\-ed in 10 percent 

 buffered formalin (4 peirent fonnaldehyde) . All 

 organisms except sea snakes, which are dangerous 

 to handle, were included. Later I sorted the 

 material (a few days later for the major cruises 

 TO-58-1 and TO-60-2; up to 3 years later for 

 other cruises) into four components — fish, crusta- 

 ceans, cephalopods, and others — and measured the 

 displacement volume of each of the first three com- 

 ponents. I also noted the taxonomic composition 

 of each component, by volume or number of 

 individuals, as far as I was able. The fourth 

 component consisted of timicates, medusae, siphon- 

 ophores, chaetognaths, heteropods, and pteropods, 

 which are not considered to be micronekton and 

 are not significant in the diets of tropical tunas. 



Subsequently, the fish, crustaceans, and cepha- 

 lopods were sorted, generally to family or, for 

 cephalopod.s, to genus and species. Displacement 

 volumes of these groups for each haul were then 

 measured (for large samples of fish and crus- 

 taceans) or estimated from the number and size 

 of the organisms (for cephalopods, and small 

 samples of fish and crustaceans) , and these meas- 

 urements were reconciled with the original meas- 

 urements of the three main groups. Occasionally 

 the reconciliation was impossible as a result of 

 imauthorized removal of specimens between the 

 first and the second sorting; the taxonomic com- 

 position of the catch by volume was then estimated 

 from the notes made at the first sorting and, rarely, 

 from the catch at another station adjacent in space 

 and time. I realize that the measurements of minor 

 groups are not entirely free from error, but the 

 errors are small and scattered and mostly affect 

 the scarcer groups. The sorted material from these 

 hauls has been catalogued and stored at the Scripps 

 Institution of Oceanography. 



80 



U.S. FISH AND WILDLIFE SERVICE 



