growth as developed from the study of Little 

 White Salmon Hatchery mark recovery speci- 

 mens. For study material, we used scales col- 

 lected from unmarked fall chinook at Spring 

 Creek Hatchery and those collected from un- 

 marked spring chinook at Carson Hatchery. 



Scales of 50 fall chinook and 109 spring 

 chinook were measured. The frequency dis- 

 tributions of the ratio of second year's to first 

 year's marine growth A/B of these two groups 

 of fish are shown in figure 15. These distribu- 

 tions show clearly that fall and .spring chinooks 

 are distinctly separate groups as far as the 

 character of marine growth is concerned. The 

 difference between the two groups is similar to 

 that between May to July release group and 

 October to February release group of Little 

 White Salmon Hatchery fall chinook. In other 

 words, the fall chinook are similar to May to 

 July release group in having a large A/B ratio, 

 and the spring chinook are similar to October 

 to February release group of fall chinook in 

 having a small A/B ratio. 



The outstanding feature of spring chinook 

 scales is that the marine growth of the fir.st 

 year is nearly as good as that of the second year, 

 so that its A B ratio approximates 1.0, as 

 compared with 1.2 for the fall chinook released 



Figure 16. — A scale of a spring chinook salmon. 

 178 



in October to February. In fact, this character 

 alone is often sufl!icient to distinguish a spring 

 run from a fall run of chinook salmon. An 

 example of a spring chinook scale is shown in 

 figure 16, in which the circuli inside of the 

 first marine annulus (II) are as widely spaced 

 as circuli outside of it. 



SUMMARY AND CONCLUSIONS 



Scales of Columbia River chinook salmon 

 were studied to find answers to the following 

 questions : 



1. Is it possible, from structures of adult 

 chinook scales, to identify whether a fish has 

 originated from fry or fingerling migrant ? 



The answer is negative. Scales from marked 

 fish recoveries showed that there was no sig- 

 nificant difi'erence in the mean values of cir- 

 culus spacing in nuclear growth part between 

 chinook salmon released as fry and those 

 released as fingerlings, although the spacing is 

 more vai'iable in the fry than in the fingerling 

 group. In the first marine growth, circuli in 

 the group released as fry are more widely 

 spaced than in the group released as finger- 

 lings. While the difference is statistically sig- 

 nificant, there was too much overlap so that 

 identification of individual scales was not pos- 

 sible. 



2. Can the effect of marking, if any, on 

 growth of chinook salmon be detected by scale 

 studies? 



The an.swer is positive. At the Spring Creek 

 Hatchery, fall chinook fingerlings were marked 

 by removal of adipose and right pectoral fins. 

 When scales from marked fish recoveries were 

 compared with those from unmarked fish re- 

 turns, circulus spacing in marine growth of the 

 marked group was found narrower than the 

 unmarked group. 



3. How do the scales of early season (May- 

 July) releases of fall chinook differ from those 

 of later season (October-February) releases? 



In answering the above question, we found 

 some interesting relations between scale growth 

 patterns and times of release. Early season 

 releases of fall chinook resulted in an ocean 

 nucleus type (sub-one age) that is typical of 

 fall chinook scales. Late season releases, how- 

 ever, resulted in a stream nucleus type (sub- 



U.S. FISH AND WILDLIFE SERVICE 



