The proportion of schools of large skipjack (over 10 pounds) which we fished appears to 

 be similar to that reported by the fishery, except during June _', when only one of the eight 

 schools we sampled was large. During April one of nine schools was large, and during Septennber 

 all of the six schools sampled were composed of large fish. 



SUMMARY 



Three scouting cruises of the CHARLES H. GILBERT were made in February, April, and 

 June 1953. These were followed by three cruises of the HUGH M. SMITH in September, October, 

 and November 1953. Cruises primarily for other purposes but on which some scouting was done 

 were made in November and December 1952, and in late February and May 1953. 



Six scouting flights were made in U. S. Navy PBY amphibian planes between November 

 1952 and June 1953. 



Skipjack schools are located alnnost entirely by accompanying bird flocks, which frequently 

 by their behavior indicate the size of schools and species of the fish. The flocks are chiefly com- 

 posed of dark-colored terns and shearwaters. 



Bird flocks were located up to 4 miles from the vessel, but most were seen within 2 

 nniles. The average proportion spotted at 4 miles was only 0. 2 percent of the schools estimated 

 to be present. At 2 miles only 5. 9 percent had been seen and at 1 mile 42.4 percent. Flocks 

 were seen at nearly random bearings in the quadrants forward of the vessel. 



Few were seen astern of the vessel in the spring scouting but after instructions to search 

 astern as much as ahead, in the fall scouting the fishermen located 22 percent more flocks. 



Larger bird flocks were seen at greater distances from the vessel. 



Poor scouting conditions were mostly due to spray over the bridge, which interfered with 

 the use of binoculars. Thirty-eight percent less schools per hour were seen under poor condi- 

 tions than under good and none were seen beyond 2 miles under poor conditions. 



The hours of the day during which most flocks were seen varied somewhat with the season. 

 In April nnore were spotted between 7 and 8 a. m. and between 2 and 4 p. m. ; in June most 

 between 9 a.m. and noon; during the autumn the distribution was nearly random. 



A fairly constant scouting efficiency was maintained despite frequent changes in personnel. 



Airplane scouting revealed far fewer fish schools per unit of distance than vessel scouting. 

 The principal reasons probably are the prevailing choppiness of Hawaiiain waters and the difficulty 

 of locating dark-colored birds against the background of the ocean. 



Skipjack schools were more abundant northeast than southwest of the islands in April and 

 June. In February, September, October, and November they were more abundant to the south- 

 west, and in November none were seen in 4 days of scouting to the northeast. This suggests a 

 seasonal migration. 



The movements of schools, while being observed from the vessel, suggested in some cases 

 a concerted movennent which might lead to local abundance or scarcity, but there was no evidence 

 of a general movement of the schools sufficient to cause a "Doppler effect". 



The Hawaiian skipjack fleet obtained three-fourths of its catch from 1948 to 1952 within 

 20 miles of land, but both our spring and fall scouting indicated that skipjack schools were about 

 equally abundant out several hundred miles to the limit of scouting, except for the difference 

 between the northeast and southwest sectors, and except for a single station where on two of the 

 three fall cruises substantially greater numbers of schools were located. 



— At this time the catch of the fishery was about 93 percent large fish. 



22 



