800 



Fishery Bulletin 91(4). 1993 



of the leukocyte population and were excluded from 

 the total leukocyte count. Although Alexander et al. 

 (1980) described the distinguishing features of alba- 

 core blood cells, we were only able to differentiate eosi- 

 nophils clearly from the remainder of the leukocytes. 

 Leukocytes were, therefore, not further classified. 



By using data from unmarked fish only, regressions 

 of body weight on fork length, maximum girth on fork 

 length, and otolith weight on body weight were fitted 

 by a least squares procedure (Statgraf Statistical Soft- 

 ware) to the exponential equation: 



o-x" 



Regression parameters a and b were estimated for each 

 case. 



Relative condition factor, relative girth, and relative 

 otolith weight for individual fish were then calculated 

 with the regression parameters: 



K=W/(a-X"). 



When K = relative condition factor, W = body weight and 

 X = fork length; when K = relative girth, W = maximum 

 girth and X = fork length; and when A' = relative otolith 

 weight, W = otolith weight and X = body weight. 



Relative condition factor, relative girth, and relative 

 otolith weight were calculated to permit direct com- 

 parison of groups containing individuals of different 

 body sizes (Pollard, 1972; Brill et al„ 1987). 



The standard condition factor (C) was calculated 

 separately for unmarked and net-marked fish: 



C = W/FL 3 



where W = body weight and FL = fork length. 



Fish were grouped only as net-marked (damage codes 

 1-4, Table 1) and unmarked (damage code 0). Differ- 

 ences in mean values for unmarked and net-marked 

 fish were evaluated by unpaired Student's i-tests with 

 P < 0.05 taken as the maximum level for statistical 

 significance. 



Results and discussion 



The only significant differences between net-marked 

 and unmarked fish were in measures of absolute size 

 (mean fork length, body weight, and maximum girth), 

 which were greater for unmarked albacore (Table 2). 

 These data suggest that smaller fish escape drift nets, 

 survive, and are recaptured by the troll fleet more 

 frequently than large fish. This appears to be con- 

 firmed by size-frequency distributions for the 217 net- 

 marked and 114 unmarked fish included in this study 

 (Fig. 3A). The two peaks in Figure 3A (at approxi- 

 mately 65 and 80 cm) represent 3- and 4-year-old fish, 

 respectively' and imply that 4-year-old individuals pre- 

 dominate the unmarked group. In the net-marked 

 group, 3- and 4-year-old fish are approximately equally 

 represented. This result is, however, more likely due 

 to either a sampling bias of the observers when choos- 

 ing fish for inclusion in this study, or to time-area 

 variations in fish size composition and the percentage 

 of net-marked fish. The size-frequency distributions of 



