280 
BULLETIN OF THE BUREAU OF FIHSERIES 
The existence and the progression through the catch of dominant year classes is 
more clearly shown by the age distributions than by those of length. The 1921 year 
class (cross hatched) is dominant over that of 1922 (solid black) for the three years in 
which it enters the catch. The 1923 year class (vertically barred) is also slightly dom- 
inant over that of 1922. The 1924 year class (stippled) appears to be much more nu- 
merous than those of 1923 and 1922, but is dwarfed in 1928 by the overwhelming 
length-frequency curves (fig. 8) from the 7-year average 
abundance of the 1926 year class (horizontally barred), which maintains its dominant 
position through 1930. 
The 3-year-olds of 1925 and the 3-year-olds of 1926 make a sharp rise in percent- 
age taken between their third and fourth years and the same would have been true 
of the 3-year-olds of 1927, but for the great abundance of the 1926 year class. These 
fish do not enter the catch in true proportion to their actual abundance as 3-year-olds 
partly on account of differential schooling and partly on account of selection of sizes 
by the fishing gear. The 2-year-olds of 1927 made a sharp rise in the percentage taken 
between their second and third years. In fact, this year class is the only one which 
entered the catch as 2-year-olds. This difference in their availability to the fishermen 
as compared to the other year classes may have been caused by the unusual abundance 
of the 1926 year class inducing these young fish to school with the older and more 
