42 BULLETIN OF THE BUREAU OF FISHERIES 



Table 8. — Random samples of the Karluk red-salmon run of 1921, distributed by age groups, sex, and 



length 

 [M=male; F = female] 



Length, in inches 



Age groups and sex 



Total 



5, 



5| 



6i 



64 





M 



F 



M 



F 



M 



F 



M 



F 



F 



21. 









1 















2iy 2 - 







1 















22 







5 

 6 

 25 

 19 

 20 

 14 

 9 

 3 

 2 















22M- 





















23 







2 

 5 

 8 

 9 

 12 

 15 

 17 

 2 

 8 

 2 

 1 









1 

 2 

 1 

 1 







23>£ 









1 

 2 



1 



2 







24 













24H— 







1 

 2 

 1 

 3 







25. 



1 



1 



1 



1 



1 





roy 2 









26- 

















26^ 









1 

 1 









27 - 



















27^ _ 



















28 





















Total 





















1 



1 



82 



104 



7 



4 



6 



5 



1 



211 





KARLUK RED-SALMON RUN OF 1922 



During the season of 1922 for the first time a fairly satisfactory series of sam- 

 plings was attempted, although these were smaller and less frequent than in sub- 

 sequent years. Two samplings per week were planned, of 75 individuals each, the 

 first taken June 5 and the last September 18. There were 33 samples in all, evenly 

 distributed, 3 or 4 days apart, between the above-mentioned dates. The tota 

 number of individuals examined and included in this report is 2,469. 



The year 1922 offers one of the conspicuous failures in correlation between the 

 packs of a brood year and the year in which the progeny of the brood year largely 

 return. As we have already seen, the Karluk red salmon mature principally at 

 the age of 5 years, and the race has established, therefore, a well-marked 5-year 

 cycle. The only other age that participates to any considerable degree in the run is 

 six, the five and six year fish together constituting usually over 95 per cent of the run. 

 This being the case, the brood years for the 1922 run were 1916 and 1917, in each 

 of which over 165,000 cases of red salmon were packed. Using size of pack as a basis 

 for prediction, it would have seemed highly probable that the year 1922 would produce 

 one of the largest packs of recent years, for it could be expected to include the 5-year 

 fish from 1917 and the 6-year fish from 1916 — two outstanding years, each of which 

 had produced a pack larger than that of any of the intervening years between 1916 

 and 1907. The results in 1922, however, were not at all in accord with such a 

 prophecy, for the red-salmon pack of that year amounted only to 46,000 cases. 



The analysis of our samplings of the run, presented in Table 9, shows beyond 

 question that the failure in this instance was caused by the very meager returns from 

 the brood year 1917. Not only was the total pack of 1922 very small, but the pro- 

 portion of this poor pack produced by the progeny of 1917 was extrordinarily low, 

 consisting of only 63 per cent, whereas in normal years the 5-year fish comprise 80 

 per cent or more of the total run. If it had not been for the unusually large pro- 

 portional contribution to the run made by the other brood year (1916), the results 

 would have been even more serious. For, whereas the 6-year fish in normal years 

 make only 10 to 15 per cent of the run, in 1922 these constituted 36 per cent. From 



1 



