140 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



August 20. These 2 modes are clearly indicated in each of the component elements 

 into which it has been possible to divide the run as a whole — except possibly the catch 

 above Bonneville Dam (fig. 7). The exact significance of these 2 modes is not ap- 

 parent, but it is evidently a real phenomenon so far as the run of 1938 is concerned. 



In addition to these 2 major modes there are at least 3 minor modes; one centering 

 about the week of March 26, another about the week of May 7, and a third about 

 the week of December 10. It is quite probable that each of these modes, both major 

 and minor, represent races (stocks) or groups of races that dominate the run at those 

 times. Only future observations will show how constant these modes are from year 

 to year, and to which part of the Columbia Basin the fish go for spawning. 



The run that centers about the week of March 26 evidently enters and passes 

 through the lower river before the commercial fishing season opens. Up to the 

 week ending Aprd 16 the escapement of steelhead, as shown by the estimate of fish 

 passing Bonneville, amounted to 13,934 fish, of which 724 were later counted over 

 the dam at Rock Island — slightly over 5 percent. This part of the run is practically 

 untouched by the commercial fishery. It should be kept in mind that, for this and 

 the following period, the records of the number of fish of each species passing Bonne- 

 ville is only an estimate based on "spot counts." It was not until May 7, 1938, 

 that actual counting through gates placed in the fish ladders was begun. 



For purposes of study we have separated the portion of the run from which the 

 commercial catch is made into four parts — dividing them, first, at about the center 

 of the period of scarcity that includes the latter half of May and the first half of June; 

 secondly, between the 2 major modes, and finally separating the late fall run from the 

 preceding portion that contained the second major mode. For the first period, 

 April 17 to May 28, the total catch was 30,800 pounds, or 3,080 fish, on the basis of 

 an average weight of 10 pounds. The estimated escapement past Zone 6 was 6,496, 

 and the Rock Island count 660. The ratio of catch to escapement was only 0.48:1, 

 and the percentage of the total run that may be referred to the upper Columbia 

 races is 10.1. The estimated number of Rock Island fish in the total commercial 

 catch is 311, with an aggregate weight of 3,110 pounds. 



The second period extends from May 29 to July 30, roughly the months of 

 June and July, and includes the first major mode. The total catch from the fish 

 that entered the river at this time was nearly 750,000 pounds — some 72,500 fish. 

 The estimated net escapement was less than 30,000 fish — the ratio of catch to escape- 

 ment 2.57:1. Of this escapement only about 1 percent can be referred to the races 

 breeding in the tributaries above Rock Island Dam. By inference only some 7,000 

 pounds of the total catch for the period can be considered as deriving from these 

 races. 



In the period from July 31 to September 24, the ratio of total catch to escape- 

 ment was 2.16:1. This was not greatly different from that of the preceding period, 

 but the catch was very differently distributed. Whereas, in the period from May 29 

 to July 30, only 4.5 percent of the total catch was made above Bonneville, in the 

 next period, covering roughly the months of August and September, over 36 percent 

 of the total catch came from the river above the dam. The estimated net escape- 

 ment was over 37,000 fish. This was an increase of some 9,000 over that of June 

 and July — a little over 30 percent. 



It might have been expected that the closed season from August 25 to September 

 10 would have had a more favorable effect upon the escapement of those stocks of 



