60 



BULLETIN OF THE BUREAU OF FISHERIES 



In 1924 the weirwas not operated after August 21. Another heavy run of pink 

 mon entered the Karluk River that year, and the weir was washed out as it had 

 been in 1922. The records for 1924, furthermore, are made less reliable than those of 

 the other years as the result of a system of estimating which, through a misunder- 

 standing, was put into practice for a part of the time that the weir was in operation. 

 The following notes were made by the senior author at the time: 



Reds were actually counted through the weir up to and including July 26. From that time 

 until the 21st of August the number was estimated as follows: There were six gates, ail of which were 

 opened when the run was so heavy that they were all necessary, or if the run slacked a little, two of 

 the six were closed and fish permitted to pass through the other four. If the numbers were still 

 further reduced, for a time only two were open. The counting was done at two gates only, and the 

 number of red salmon that passed through was multiplied by 3, if six gates were open, or by 2 if 

 four gates were open. The gates at which the counting was done were changed so that any inequality 

 was thus taken care of. The water was clear and there was no difficulty in distinguishing the reds 

 and counting them. Any inaccuracy in this method was due to unequal numbers passing the 

 different gates and not to difficulty in distinguishing reds and counting them. When the r\m was 

 heavy enough to make it necessary to open other gates than those at which counting was in progress, 

 they ran about equally at all the gates, in the opinion of Mr. Wood. 



On account of the lack of complete escapement records for 1924 it has been 

 impossible, of course, to get directly the total run; but for purposes of comparison 

 it is important that we know approximately what this was. The following pro- 

 cedure, therefore, was adopted: In the last column of Table 26 are given the aver- 

 age percentages of the total runs that had accumulated up to the given dates. 

 These averages have been based on the records for the years 1921, 1922, 1923, 1925, 

 and 1926 only. If we assume that the accumulated total for the run up to July 16 

 (the last date on which the weir was operated) was, in 1924, approximately the same 

 percentage of the total run as in the other years, and from these data calculate the 

 total run, this proves to be very close to 2,000,000 fish. The accumulated total to 

 August 16, 1924, was 1,203,000, and the average percentage of the total runs that 

 had accumulated to that date in the other years was 62.8. Dividing this accumu- 

 lated total by 0.628 gives us 1,920,000 as the total run. Similar calculations, based 

 on the accumulated totals up to August 9 and 2, give the total run as 1,965,000 and 

 2,020,000, respectively. We believe, therefore, that the total run in 1924 was, for 

 all practical purposes, close to 2,000,000, and have used this figure in calculating 

 the number of fish in the run subsequent to August 16 and in calculating the accu- 

 mulated percentages during the part of the season previous to August 16. On 

 account of our method of calculation, the percentages after that date are the same 

 as those in the last column of the table — the average percentages based on the five 

 years for which we have reliable data. 



The "run," as we have used the word here, is the catch plus the escapement. 

 In the natural course of events the fish first approach the beach at Karluk Spit, 

 where the commercial fishery is carried on. Those that escape the seines enter the 

 river and may remain in the estuary or passing back and forth between the estuary 

 and the ocean for some time before passing on up to the lake. The weir is situated 

 just above the estuary, and the fish do not pass the weir until they are definitely 

 starting their migration to the lake; and once they have started they do not turn 



