Also confirming the conclusion that there is no difference in the catch associatAd 

 with type of badt are the results of experiments aboard the Sag ami Maru, a research vessel from 

 Kanagawa Prefecture, conducted during the seventh mothership expedition (Anonymous 1952). 

 Sardines and sauries were alternated on each hook (5 hooks per basket). Slightly more yellowfin 

 (73 versus 65) and black marlin (75 versus 63) were tstken on sardines than on sauries, bigeye 

 and other tunas were taken equally, and more sharks (77 versus 62) were taken on sauries. Thus 

 it appecirs that the catches made during the nine mothership expeditions can be used as indexes 

 of abundance without regard to the type of bait used on the longline. 



RELATIVE CATCHING EFFICIENCY OF THE VESSELS 



The mothership expeditions comprised numerous catcher vessels, many of which 

 participated in several expeditions. Analysis of their catches for evidence of geographical 

 and temporal variations in relative abundance would be considerably simplified if it could be shown 

 either that there are no differences in relative efficiency or that such differences as exist are un- 

 innportant in magnitude. In longline fishing the total catch made by any vessel is related to the 

 number of hooks fished emd the abundance of the fish. This affords ample opportunity for the 

 total catch to be relcited to such factors as vessel size and the skill of the fishermen, because 

 these affect the amount of gear fished each day. On the other hand, the catch per unit of effort-- 

 catch per 100 hooks--should be relatively stable, because once a given number of hooks is 

 placed in the water the number of fish taken on these hooks is largely a function of the availability 

 of fish. There were, however, minor vcU-iaitions in the design of the gear used by the various 

 vessels (Shimada 1951a) and these could conceivably affect the catch per 100 hooks. 



In order to test for difference in vessel efficiency, as it affects the catch per unit of 

 effort, a two-way tabulation by vessels and by nnonths was made of the catch rates for each 

 expedition except those of the Najisei Fisheries Company (3, 4, 6, and 9), which were nnuch 

 smaller in size and scope. The catches of yellowfin, comprising about half the total catch, were 

 then studied by analysis of variance (Snedecor 1948). This classification actually affords some 

 statistical control of the catch rate of vessels by area as well as time, because all of the ex- 

 peditions tended to fish in a group that moved from west to east (and in some of the later ex- 

 peditions from south to north near the end of their operating period). 



The results of the analyses of varijince (table 3) show that while there were differences 

 in the catch rates associated with time (and area implied), there were no significant differences 

 associated with vessels except on the fifth expedition. This single exception is puzzling but 

 perhaps can be attributed to failure to completely control area and time in the cinalysis, a factor 

 that may also account for the several near-significant F values associated with vessels of the 

 first, second, and seventh expeditions. With this in nriind, it appears that in general the catch 

 rates of the various vessels do not have to be adjusted for relative efficiency in order to use them 

 interchangeably as indexes of abundance. Further comfort may be taken from the fact that necirly 

 all of the geographical areas and time periods discussed in the balance of this report are repre- 

 sented by the catches of several vessels, tending to average out any differences in efficiency 

 that might exist. 



