770 
BULLETIN OF THE BUREAU OF FISHERIES 
In the odd-numbered years, which have abundant pink salmon runs, usually 
three or four times as many sockeyes are seined as in even years, because there are 
more seine boats fishing, and they are largely concentrated during the late summer 
in the areas where the pink salmon are migrating on their way to the Fraser River 
and other streams in the Gulf of Georgia. 
The average size of the purse-seine delivery is not a good measure of sockeye 
abundance. In the even-numbered years it tends to be high, as the boats fish only 
during the height of the run. In the odd-numbered years it tends to be low, as the 
boats often made a large number of catches, containing few sockeye per catch, while 
fishing primarily for pink salmon. The purse seine catches are thus not as reliable 
as a measure as either the trap or gill-net catches, but they do show how the purse 
seines have fared under varying conditions of abundance. 
In making this index the number of sockeye taken each year during each 7-day 
period was divided by the weighted number of deliveries. The weights were given 
according to the size of the boats making the catches in accordance with the efficiency 
weighting for all species described in the purse seine section of this report. Data 
were available for every year, except 1920, from 1911-34. Of the 23 years remaining, 
the data for 1918 cover such a short period of time that they were not used in comput- 
ing a normal curve for each week From the other 22 years a normal average daily 
delivery was made for each of the 6 weeks between July 15 and August 25, by merely 
dividing the sum of the averages for all years by the number of years. No week 
had less than 19 years data. 
For each year the sum of all the average daily deliveries for the six 7-day periods 
between July 15 and August 25, or as many of these six periods as there were data 
for, was divided by the sum of the average daily deliveries for the same periods for 
the noimal. The resulting index then is a measure of the annual abundance ex- 
pressed as a peicentage of the normal. 
The purse-seine index of abundance differs from the trap index in a number of 
years, but before deciding on the meaning of these differences several factors must 
be considered. Thus the actual catch of sockeye in 1918, 1922, 1924, 1926, and 1928 
by purse seines in Puget Sound was less than 100,000 fish. In 1918 it was only 
45,000 and in 1928 it was but 62,000. In such years the total quantities caught by 
purse seines were very low in relation to the actual abundances. 
In ceitain other years the purse-seine index is very high in relation to that for 
traps, as the purse seines may make catches out of all proportion to the abundance 
when the fish are heavily concentrated, as they were at Point Roberts in 1930. 
Although it has seemed unwise to lay any stress on the purse-seine index as an accu- 
rate measure of abundance, yet, considered in relation to the trap and gill-net indices, 
it portrays the fluctuations in availability of sockeyes to the purse seines, and is thus 
necessary to an understanding of the fishery. 
