SALMON RUNS OF THE COLUMBIA RIVER IN 1938 119 



We will now consider, specifically, certain runs and portions of runs in respect 

 of their importance to the general problems of the preservation of the salmon of the 

 Columbia River, and in particular of those that have derived from the river above 

 Grand Coulee Dam. Although the data have been studied and presented on the 

 basis of time units of 1 week, it is convenient and even more illuminating to consider 

 them also for longer intervals of time which have been selected for various reasons 

 as being of special importance. 



CHINOOK SALMON 



HISTORY OF THE RUN OF 1938 



On account of the dominating importance of this species in the fishing industry, 

 particular attention has been paid to it. The data are presented in tables 12 to 14 

 and are shown graphically in fig. 3. 



The earliest part of the run to the Columbia River above Bonneville does not 

 enter into the commercial fishery — it is past the commercial fishing area before the 

 opening of the season on May 1. The first of the run to contribute to the commercial 

 catch is that which enters the mouth of the river during the week ending April 23. 

 These fish, in general, may be expected to pass Bonneville and to be in Zone 6 during 

 the first week in May — the first week of the spring open season. We have therefore 

 considered as a separate period the weeks up to and including the week ending on 

 April 16. The next period includes the part of the run that provides the peaks in 

 catch and Bonneville count that occur in May. We consider that this period ter- 

 minates with the week ending May 28. The next period includes the succeeding 9 

 weeks ending on July 30, during which the catch and the Bonneville count were both 

 relatively low, while at corresponding weeks the Rock Island count attained the 

 maximum for the year. 



In the original report the last period treated covered only the 4 weeks ending 

 August 27 — the last 4 weeks of the spring fishing season. It was impossible to carry 

 the study beyond this because at the time the report was prepared data were not 

 available for the fall season. But, with the data now on hand, it is obvious that the 

 portion of the run beginning with the week ending August 6 and extending to the 

 end of the year should be considered as forming a single unit rather than two or 

 more units. In table 12 it is apparent that the run from the week of August 6 to 

 the end of the year contains the main mode which, for purposes of study, should 

 certainly not be broken up without good reason. Furthermore, table 12 and fig. 3 

 show that there is a mode in the Rock Island count for this period. In the present 

 report, therefore, we shall take for the final period to be studied the entire remainder 

 of the year after the week ending July 30. 



The data for these selected periods are given in table 13, which, for comparison, 

 also includes the figures for the last period considered in the original report — July 31- 

 August 27. Table 14 gives some of the more significant comparative figures that 

 may be derived from table 13. 



