486 



Fishery Bulletin 94(3), 1996 



of the natural logarithms of the recovery weight and 

 of the release weight by the change in time 

 (Weatherley and Rogers, 1978). Because of limited 

 samples of CWT fish, growth rates for ocean- or 

 stream-type fish were pooled between years, regions, 

 waters, and stock groups for each growth period. 

 Ocean- and stream-type fish in each growth period 

 were then tested for significant differences (P<0.05) 

 in growth rates with /-tests. 



Results 



Gear selectivity 



For the 2-d gear comparison conducted in inside 

 waters of the southern Alaska region in September 

 1987, both sizes of hooks and lures caught similar 

 numbers and mean fork lengths of age -.0 chinook 

 salmon. In this comparison, 68 age -.0 chinook salmon 

 were caught on our "standard" small gear and 59 on 

 the smaller gear; numbers caught did not differ sig- 

 nificantly (X 2 ,P> 0.10) by gear size. Mean fork length 

 (27.6 cm) of age -.0 chinook salmon caught on "stan- 

 dard" small gear also did not differ significantly from 

 mean fork length (27.0 cm) offish caught on smaller 

 gear (/-test, P=0.36). However, age -.0 chinook salmon 

 caught on the "standard" small gear were larger 

 (range, 22-37 cm FL) than those caught on the 

 smaller gear (range, 15-34 cm FLl. 



Age, length, and sex composition 



Of the 539 CWT chinook salmon sampled, 486 had 

 scales that could be aged. Agreement between scale- 

 and CWT-age was highest in September (95%) and 

 lowest in February (79%); overall agreement was 90% 

 (Table 2). The most common disagreement of age 



designations was the assignment of ocean-type 

 chinook salmon (as determined by CWT's) as stream- 

 type. Of the 51 disagreements in chinook salmon age 

 determinations, 43 (84%) were ocean-type fish as- 

 signed stream-type designations; some of these dis- 

 agreements may have been a result offish that were 

 released as ocean-type but that overwintered in 

 freshwater, and therefore bore a "stream-type" scale 

 pattern. 



Age data for 5,838 chinook salmon were obtained 

 from scales and CWT's in all sampling periods and 

 regions in 1986-87 (Table 3). CWT's were represented 

 in 36 of the 39 age-group-time-period strata. Age -.0 

 chinook salmon were caught only during September, 

 and most were stream type (age 1.0); only a few 

 ocean-type (age 0.0) chinook salmon were caught, all 

 in the southern region. In February and May, age -.1 

 chinook salmon were primarily stream type (age 1.1); 

 a few ocean-type fish (age 0.1) were caught only in 

 the southern region. In September, however, both 

 stream- and ocean-type age -.1 chinook salmon were 

 caught in all regions. Age -.2 chinook salmon of the 

 ensuing year were also a mixture of ocean- (age 0.2) 

 and stream-type (age 1.2) fish in each season and 

 region. Only four age -.3 CWT chinook salmon were 

 recovered and all were ocean type (age 0.3). 



Chinook salmon were larger and older in the north- 

 ern than in the southern region in each sampling 

 period (Fig. 2; Table 3). Length modes of chinook 

 salmon within each region generally increased be- 

 tween September and May. The proportion of small 

 fish (<30 cm FL) was highest in inside waters of both 

 regions in September. 



Mean length of chinook salmon increased with each 

 successive period; and for each ocean-age group, 

 stream-type fish were consistently longer than ocean- 

 type fish for each period ( Fig. 3 ). Differences in mean 

 length of ocean- and stream-type chinook salmon 



