local, i.e., confined to the area of fishing. 

 This is suggested for skipjack, at least, by 

 comparing our trolling results with those of the 

 commercial pole-and-line fishery. Instead of a 

 downward trend over the period 1951 to 1953, in 

 the commercial fishery both 1951 and 1953 were 

 years of high availability whereas 1952 was a 

 year of low availability (Yamashita, In press). 

 The comparison is vitiated in part by differences 

 in size composition of the catches; the troll- 

 caught fish were mostly small, whereas the 

 commercially caught fish were a mixture of 

 small and large sizes. 



The average monthly availability of troll- 

 catchable fish for the combined data of the years 

 1951 to 1955 is shown in figure 3. Although fish 

 were caught in all months of the year except 

 December (when effort was low), the highest 

 availability occurred from May to October. 

 Skipjack and dolphin were taken throughout 

 most of the year but with a peaking of availabil- 

 ity during the summer months. The commercial 

 skipjack catch also peaks during the summer 

 months (Yamashita, In press). No little tunny 

 were taken during January to April, but this 

 cannot be regarded as significant in view of the 

 small number of hours trolled during this period. 

 Welsh (1949b) reported good catches of this 

 species during March, April, and May, 1948 and 

 during February and April 1949. The peaking 

 of the availability of yellowfin during September 

 and/or October seems to be a common occur- 

 rence, for it was encountered during each of the 

 three years (1951, 1954, and 1955) in which 

 fishing was done during these months. 



LURE PREFERENCE 



During 1951 and 1953 records were kept of 

 the strikes and catches on the several lures 

 which were used. The lures, as shown in figure 

 4, were of three general types: (1) "metal 

 squid" with a heavy blue, silver, or red metal 

 body and white rubber arms, (2) "feather" with 

 a heavy white or red metal head or a hollow sil- 

 ver head to which feathers of variegated color 

 are attached, and (3) "rubber squid," with brown 

 pliable rubber body and arms. Other types used 

 to a small extent included a "plastic squid" and 

 a "rubber fish. " 



Data on catch rate for all species of fish 

 are shown in table 2.—' By comparing the 



— The fish catch for 1953 does not agree 

 with that of table 1 because lure data were not 

 available for a few days of that year. 



4 5 6 7 



MONTHS 



Figure 3. --Average monthly catch per hour's 

 trolling for all species combined and for the 

 four most abundant species in the catch over 

 the period 1951 to 1955. 



strikes per lure per hour with the catch per 

 lure per hour, it is at once apparent that about 

 the same proportion of fish was lost from each 

 lure during hauling of the line. Of the 3 lures 

 usedmost extensively in 1951, the brownrubber 

 squid produced the highest catch rate (0. 61 fish 

 per lure per hour), the metal squid with red 

 body the second highest (0.42), followed by the 

 feather lure with white head (0. 29). The same 

 order persisted when the data were selected to 

 include only days in which the three lures were 

 used simultaneously, with the following catch 

 rates: 0.66, 0.44, and 0. 30. However, in 1953 



