SALMON RUNS OF THE COLUMBIA KIVER IN 1938 107 



trend. It suggests, however, that the slightly reduced averages for the past two 

 5-year periods may signify some real reduction in abundance. 



The general trend for both silver and chum salmon is distinctly upward (fig. 1) 

 despite rather wide fluctuations. This doubtless reflects an increased usage of these 

 2 less desirable species that has come with the reduced abundance of the other species, 

 especially the chinook. 



DATA FOR THE RUNS OF 1938 



In this study of the 1938 salmon runs to the Columbia River, data have been 

 available for the first time in the history of the fishery that have made it possible to 

 evaluate the intensity of the fishery as a whole, the relative intensity at different 

 times and in different parts of the river, and the proportion of the total that is formed 

 by the run to the upper Columbia River (Clarks Fork). These data include the 

 following series: (1) Daily commercial catch in pounds and by species in each of 

 6 districts corresponding to the 6 counties of the State of Washington that form the 

 northern shore of the Columbia; (2) daily counts, by species, of the salmon passing 

 Bonneville Dam beginning with May 7, and estimates for the period from February 

 15 to May 6; and (3) daily counts, by species, of the salmon passing Rock Island 

 Dam across the upper Columbia near Wenatchee, Wash., about 100 miles below the 

 site of the Grand Coulee Dam. The latter have been available since the season of 1933. 



The importance of the data on the Bonneville count and the total daily catches 

 to the proper development of a sound program for the conservation of the salmon 

 of the Columbia River should be emphasized. Without them an intelligent con- 

 sideration of the problems raised by the Grand Coulee Dam would have been im- 

 possible, and they will be of equal importance in the study of any other problems 

 dealing with the maintenance of this valuable resource. For the previous three seasons 

 the Washington Department of Fisheries had collected records of the daily deliveries 

 of each species of salmon in each of the counties of the State bordering on the Columbia 

 River. The Fish Commission of Oregon also had collected data on the daily de- 

 liveries of salmon, but not until 1938 were these presented in such form as to make 

 it possible to combine them with the data from Washington so as to give a record 

 of the total daily deliveries by species and by locality. For no other year are such 

 data available, although figures for 1939 will be in suitable form for study when 

 they are available. Now that a uniform system for presenting the catch data has 

 been started by the two States, it probably will be continued so that in the future 

 data will be available showing the total daily deliveries in each of the six districts. 



Of equal importance has been the record of counts of fish passing the dams at 

 Bonneville and Rock Island. Since 1933 there have been counts, more or less com- 

 plete, at Rock Island, but the Bonneville Dam was not finally closed to the passage 

 of fish previous to 1938, so that this year marks the beginning of the count at this 

 point. The tremendous value in the conservation program of the count of salmon 

 passing over the Bonneville Dam cannot well be overstated. This count should, 

 by all means, be made a permanent feature and should be in the hands of competent 

 men familiar with the fish and with the techniques of fishery research, and having a 

 primary interest in the fishery problems upon which these data will bear. 



In presenting these data it has been found expedient to sum them for the smallest 

 practical time interval. The unit of 1 week was selected as the shortest period that 

 would avoid insignificant fluctuations, particularly the disturbing effect of the Sunday 

 closed period. For special purposes the data have also been arranged relative to 



