AGE AND GROWTH OF THE CISCO 
275 
question. Since, as will be brought out presently, the calendar years 1924 to 1927, 
inclusive, were all below average as to goodness of growth, while the years 1929, 
1930, and 1931 were all above average, the limitation of the estimate for a poor year, 
as for example 1926, to comparisons with two other poor years, 1925 and 1927, or of 
a good year as 1930 to comparisons with other good years, 1929 and 1931, would 
lead inevitably to erroneous conclusions concerning the goodness of growth in these 
years. To avoid this difficulty the evaluation of the different years was based on 
the comparison of total growth in each calendar year with that of the preceding year. 
An illustration will be given of the method used in the evaluation of goodness of 
growth of the Clear Lake cisco in different calendar years. (See table 53.) In 1924 
the first year’s growth of the VUI-group fish was 108 milhmeters. Tins growth was 
10 millimeters or 8.5 percent less than the first year’s growth of the preceding year 
(118 millimeters in 1923). Thus the calendar year 1924 shows a —8.5-percent devia- 
tion from 1923 as to goodness of growth. Similarly the 1925 growths of 87 (VIII 
group, second year of life) and 102 (VII group, first year of life) show a total of 189 
or a deviation of —3.1 percent from the total corresponding growth of 195 
millimeters in 1924. The position of 1925 with respect to 1923 is the sum of the 
deviations, —8.5 and —3.1 or —11.6 percent. A continuation of this procedure 
shows the position of each calendar year with respect to the year, 1923. In order to 
make the relative positions of the different years describe their deviations from aver- 
age growth rather than from 1923 growth, the mean value of all the deviations as 
computed from 1923 as the starting point was subtracted from the individual devia- 
