48 BULLETIN OF THE BUEEAU OF FISHERIES 
similar to those obtained in this study of the fish taken off the mouth of the Colum- 
bia River, as may be seen from the following table: 
Table 11. — Percentages of fish with ocean nuclei in the various age groups 
Age 
Locality 
Second 
Third 
Fourth 
Fifth 
Sixth 
year 
year 
year 
year 
year 
Off mouth of Columbia River 
20.5 
40.7 
30.5 
7.50 
0.8 
Straits of Georgia (from Fraser) 
28. 1 
42.9 
27.8 
1.2 
The correspondence is remarkably close, considering the fact that the data 
were taken at different places, in different years, and were handled by different 
observers. 
It is interesting to compare these data with those obtained from a study of 
the collections made of fish taken inside the river. Among the fish with ocean 
nuclei taken in the river, 4 or 5 year old fish were found to be the most numerous. 
Among the fish taken in the ocean, 3-year fish were most abundant. Also, in the 
case of the fish with stream nuclei, it was found that 4 and 5 year fish were most 
numerous in the river, while 3 and 4 year fish were more common in the catches 
made outside. It has already been pointed out that the data bearing on the pro- 
portions of fish of the different age groups present in the ocean can not be compared 
in detail with the similar data for the fish taken in the river, but in a general way 
it may be stated that the fish taken outside tend, on the whole, to be about 1 year 
younger than those taken inside. 
GROWTH 
A study has been made of the increase in length of fish of different age groups 
taken during the season of 1919, both outside and inside of the river. The study 
was confined to the collections made during a single season in order that any yearly 
fluctuations in growth rate might be excluded. It must be emphasized that an 
observed increase in the average size of the individuals contained in a series of 
collections is not necessarily due to growth, particularly in the case of fish taken 
inside the river. The constant migration upstream after fish have entered fresh 
water results in a constant change in the content of the run passing any given 
point along the river. Gilbert (1914, 1915, 1916, 1918, 1919, 1920, 1922, and 
1923), in his intensive study of the sockeye salmon of British Columbia, has 
conclusively demonstrated the existence of distinct races characterized, among 
other things, by very different sizes at maturity. Considerable unpublished evi- 
dence is in the hands of the writer, which indicates that similar races are to be 
found among the chinook salmon of the Columbia River. The result of such differ- 
ences in size of the races running in the river at various times will necessarily be to 
mask, more or less completely, the results of growth. The extent to which the 
effect of growth will be thus masked will depend upon a nmnber of factors, such 
as the relative sizes and numbers of the various races found together within the 
river. If large races were running early in the season and smaller races later, the 
