KARLUK EIVER RED-SALMON INVESTIGATION 



37 



done easily if all the progeny of a given spawning matured at the same age and thus 

 constituted the run of a given year; but, as we have seen, maturity in the Karluk 

 race is attained at any age from 3 to 8 years. The returns from the 1921 spawning, 

 for instance, are distributed throughout the runs from 1924 to 1929, and those of 

 1922 from 1925 to 1930. If our aim is to be accomplished, each year's run must be 

 analyzed into its various age components, and each of these must be determined 

 quantitatively. By this method we can be prepared to add together the 3-year 

 fish of 1924, the 4-year contingent of 1925, the 5-year quota of 1926, the 6-year 

 group of 1927, the 7-year group of 1928, and the rare 8-year individuals of 1929, 

 and obtain a total that will represent the complete returns of the 1921 spawning; 

 and so on, by the same method, for each succeeding year. 



This analysis of the runs is conducted by the customary method of random 

 sampling. A certain number of individuals are taken at random, without selection, 

 to form what is considered an adequate sample of the commercial catch on a given 

 day. The sample must be large enough to furnish reliable data concerning the 

 constitution of the run on the day on which it is taken, and successive samples must 

 be obtained at close enough intervals to present a record of such changes in the con- 

 stitution of the run as may develop from time to time throughout the season. 

 Believing it wiser to err on the safe side of unnecessarily extensive sampling, rather 

 than the reverse, we are attempting in each sample to secure data from 100 or more 

 individuals taken by the random method. With regard to each individual, we obtain 

 length, weight, sex, and a specimen of scales for the determination of age. The 

 results have been such as to assure us of the adequacy of our method and the suf- 

 ficiency of our sampling to present a reliable cross section of the entire run. 



Although no returns from the 1921 spawning — the first for which we had 

 numerical data — could be looked for until 1924, we undertook an analysis of the 

 run of 1922, to survey the field and secure a background for the determinations of 

 future years. The characteristics of the Karluk race were not wholly unknown to 

 us at that time, for as a part of our general investigation of red-salmon races we had 

 previously examined limited samples of the runs of 1916, 1917, 1919, and 1921. 

 For each of these years, except, perhaps, 1917, the samples were too small to permit 

 generalizations as to the characteristics of the entire run, being confined to one or two 

 days of the season only. We have material gathered on seven different days in 

 July and August in 1917. We include returns from each of these years for such 

 value as they may have for purposes of comparison. 



Having established our base line in 1922, our series of complete consecutive 

 analyses begins in 1924 and should continue without interruption as long as the 

 investigation is in progress. The results obtained are not to be confined to the 

 segregation and enumeration of age groups, although this is our prime object. Many 

 other important facts in the life history of the Karluk salmon will emerge, it is hoped, 

 from the data obtained. These should include, among others, unexplored phenomena 

 in the growth of the salmon and in the quantitative relations of the various age 

 groups that develop in the broods of successive years. The latter is a most important 

 subject, from the practical standpoint, and is one concerning which no information 

 has been obtainable hitherto. In the line of our present experiment, we shall ascer- 

 tain the numbers of each age group that develop from the eggs of a given brood and, 



