e.g. Nakamura (1949). These conclusions were necessarily based on a consideration of relative 

 hook depth, since no nneans were available to measure the actual depth at which the line fished. 

 Subsequent effort on this problem has been devoted to a continuation of the study of the relative 

 depth of capture, auid attempts have also been made to measure the absolute depth at which the 

 line fishes. 



The most successful method of ascertaining the line depth has been the measurennent 

 of the depth of the center of each section of the main line with a Bendix Depth Recorder, a method 

 suggested by W. F. Royce. The depth of the line was usually measured about 3 or 4 hours after 

 setting by conning the ship parallel to the line at various distances on either side of the line of 

 buoys iintil a trace of the main line was obtained on the recording paper. After the first trace 

 appeared, it was usually possible to locate other sections of the line by conning the ship so that 

 it passed over the same position relative to other buoys. A series of good records is shown in 

 figure 12, consisting of traces from a series of baskets near the end of a 40-ba8ket set. The pro- 

 gressively increasing depth of the traces is a reflection of the increasing sag of the main line, 

 caused by the closing in of the end buoys. 



In connection with the problem of measuring the depth of the line, attempts were made 

 to measure the distance between buoys in order to ascertain the relationship between the buoy 

 distance and the depth fished. A' On Smith cruise 18 the average buoy interval was estimated 

 from the setting speed and time, dividing the number of baskets into the distance covered over 

 the sea surface. On this cruise the average buoy interval was also measured by radar. This was 

 accomplished by placing a target on one of the buoys 2 or 3 baskets fronn the end of the entire set 

 3Lnd nneasuring the total distance between the center of the set and the radar target. This meas- 

 urement appeared to be fairly reliable as the line was usually quite straight. The measurements 

 by radar taken some 4 hours after the gear was set were in close agreennent with the distance be- 

 tween buoys as estimated fronn the setting speed and tinne (table 11). Thus, the latter method 

 furnishes a good estimate of the average buoy interval during the course of the fishing day and 

 provides a useable figure from which to estimate the theoretical maximunn depth. 



It is quite evident from comparison of the theoretical depths and the measured deptha 

 (tables 11 and 12) that on these two cruises there was considerable streaming of the line, which 

 probably was caused by a current differential between the surface water and the thernnocline. 

 This was also evident in the field, as the center of the sag of the main line was nearly always lo- 

 cated at a considerable distance from a straight line between any two buoys. 



The correlation between the theoretical depth as determined from the average buoy 

 interval and the actual depth measured with the depth recorder is not particularly close (r = 0. 315 

 for the combined data of tables 1 1 and 12). Part of this can be laid to a lack of precision in nneas- 

 uring the two variables (line depth and buoy interval), and part can be laid to changing conditions 

 with respect to the speed of the surface current from station to station, i. e. , failure to control 

 the factors other than buoy distance that determine the actual depth the line fished. Another diffi- 

 culty, which nnay in part be responsible for the low correlation between buoy distance and depth, 

 is indicated by the suggestion of a negative relation between the number of successful depth re- 

 corder traces and the average depth at each station (table 11). This suggests that as the line goes 

 deeper the chances of obtaining a legible record on the depth recorder are lessened. This was 

 also indicated in the field by the progressive faintness of the traces as the apparent depth of the 

 line increased. That there is a real relation between buoy distance and line depth is indicated by 

 the progressive deepening of the traces in figure 12. These traces were obtained near the end of 

 a set, where the buoys supporting successive baskets were progressively closer together. 



It has not yet been possible to establish any relationship between the catch and the data 

 on line depth, nor between the relative depth of the greatest catch and either the depth of the 



2/ The theoretical fishing depth is, of course, fixed by the buoy interval, since the line should 

 hang in a catenary. Curves of theoretical fishing depth versus buoy interval are given in 

 Murphy and Shomura (1953a). 



22 



