32 
BULLETIN OF THE BUREAU OF FISHERIES 
From the total curve it may be concluded that for Group II, 1926 was a year of 
greater abundance than 1927 and 1928; 1929 may have been a year of as great abun- 
centages that the average weekly catch per boat is of ! 
standard curve (see text) 
, nr, i t i • ii ,i Figure 17— Showing for Group II the geometric means 
elusive. The decline during all months IS SO (express ed as percentages of the 1928 mean) of the per- 
tremendous as to leave no doubts concerning 
the validity of the decline in abundance and 
shows good cause why both of the plants in this group discontinued operations at 
the end of 1928. The only reason that 1928 did not show a further decline over 1927 
is that in the latter year these 
plants extended their operations 
to distant areas, taking over 50 
per cent of their catch in areas 8, 
9 , and 20 . Why this grou p , alone , 
of the three under consideration 
should show such a tremendous 
decline is perhaps best explained 
by comparing Figure 20, giving the 
standard curve for this group, with 
Figures 16 and 18. Group III, as 
is shown, depends chiefly on the 
early portion of the season, taking 
practically nothing after mid- 
August, while Groups I and II ob- 
tain a very considerable portion of 
their season’s catch after the middle of August. Another reason why Groups I and 
II have not declined so rapidly as Group III is fouDd in their exploitation during 
-The standard curve for average weekly catch per boat for 
Group II (see text) 
