Table 7. — Adjustment of 1959 catches of pink salmon 

 for effect of ail! net selectivity 



samples of 50 or more fish, as was done for pink 

 salmon. Sockeye salmon catch ratios for each 

 pair of mesh sizes showed approximate linearity 

 over the greater part of the range of fish lengths, 

 but not at the extremes of the range. Approxi- 

 mate linearity extended from 29 cm. to 38 cm. 

 for the 3}4/2}2-inch mesh sizes, from 38 cm. to 

 53 cm. for the 4}2/3K-inch mesh sizes, and from 

 47 cm. to 62 cm. for the 5^/4}o-inch mesh sizes. 

 Least squares lines were fitted in these ranges. 

 The procedure of discarding extremes in line- 

 fitting previously was used by Garrod (1961). 



Figure 5 shows a similar picture for 1957 

 (data are in table 9). Catch ratios were ap- 

 proximately linear except at the extremes. In 

 1957 and 1959 similar length ranges were used 

 in line-fitting, except for extending the range to 

 41 cm. for the 3^/2K-inch mesh sizes in 1957. 



32 



38 44 50 56 



FORK LENGTH (CM.) 



62 



68 



Figure 4. — Catch ratio of adjacent mesh sizes by fork 

 length, sockeye salmon, 1959. 



+ 6 

 + 5 



+ 4 



o +3 



H 



< +2 



5 +l 



i- 



q - 2 



-3 

 -4 



26 



32 



38 44 50 56 



FORK LENGTH (CM) 



62 



68 



Figure 5. — Catch ratio of adjacent mesh sizes by fork 

 length, sockeye salmon, 1957. 



Table 8. — Catch by mesh size and catch ratio of adjacent 

 mesh sizes, sockeye salmon, 1959 



1 Original catches of the 4££-inch mesh were 6 times as large as shown; 

 they were divided by 6 to equalize fishing effort between mesh sizes. 



Note. — Catch ratios in parentheses were not used. See text. 



Catch ratios showing the linear relation represent 

 mainly fish which were enmeshed (gilled) around 

 the head and gill cover by the net twine. Catch 

 ratios departing from the straight line at either 

 end were discarded because they represent large 

 fish snagged or small fish tangled in the gill nets. 



In the 1959 sockeye catch ratios of the 4>2/3^- 

 inch mesh sizes (fig. 4), the discarded catch ratios 

 at 32 cm. and 35 cm. and at 56 cm. and 59 cm. 

 curve away from the fitted line, giving the effect 

 of a tipped S-shaped curve. At 32 cm. and 35 cm. 

 the fish in the 4}£-inch mesh were probably 



GILL NET MESH NET SELECTION CURVES FOR SALMON 

 774-711 0—66 7 



385 



