FISHERY BULLETIN: VOL. 73, NO. 3 



1-5-73 



1-5-73 



Male - 

 Female- 



100 120 



140 160 180 



TOTAL LENGTH, mm 



T 1 



200 220 240 



Figure lO.-Length frequencies of group selected from selec- 

 tively fished test population. Lengths are from snout to tip of tail. 



the analog computer method of Silliman (1967). 

 Growth in length was essentially identical in the 

 two groups for the females (Table 5, Figure 12), 

 and a single curve was fitted. Constants are given 

 in Table 6. For males, however, growth was sig- 

 nificantly greater in the unselectively fished con- 

 trol group than in the selectively fished test group. 

 The sexual misclassification of one fish in the 

 test group and two in the control group (Table 5) 

 must be considered in relation to possible effects 

 on the results. These fish were misclassified at the 

 beginning of the growth period, when the fish 

 were relatively small (chosen so to provide room 

 for growth) and sex determination was difficult. 

 As the fish grew and determination became easier, 

 the errors were discovered and corrected. To test 

 the effect of the errors it was assumed that they 

 occurred in the manner most contrary to the 

 conclusion adopted-that growth was greater 

 among males in the control than in the test group. 



TOTAL LENGTH (mm) 



Figure ll.-Length frequencies of group selected from unselec- 

 tively fished control population. Lengths are from snout to tip of 

 tail. 



260- 



240 



220- 



200 



180 



P 140 



Figure 12.-Gompertz curves fitted to mean lengths in group 

 selected from selectively fished test population and unselectively 

 fished control population. 



504 



