14 U. S. BUREAU OF FISHERIES 
In the more complete records and refined analysis for late years, 
a number of minor difficulties were solved. The validity of variations 
in our records of “lost time,” the most subjective of the factors used in 
our computation of “catch per unit of fishing effort,” was established 
through comparing the percentage lost time with percentage wind 
force of 7 or more, as reported by lightships. A correlation of 0.91 
between these variables indicated that the variations in lost time were 
due to weather and not to personal factors. Shifts within the season 
between Georges Bank and the more distant grounds were found to be 
closely correlated with water temperatures, apparently due to the 
difficulty in keeping fish on long trips during the summer, when water 
temperatures are high. This work completed the primary analysis of 
our data to show catch per day by months. 
The next major step in the abundance analysis was to summarize the 
data by seasons and years. Preliminary attempts, using the usual 
arithmetic averages and direct combination, proved unsatisfactory due 
to monthly differences in level of availability, combined with great 
variations in the amount of fishing carried on in the several areas and 
from month to month. The availability through the season was found 
to follow a consistent pattern throughout the 24 years studied. How- 
ever, it was found that in years of scarcity the winter fishery suffered 
much more than the summer, since at such times the winter catch de- 
clined 40 percent more than the summer. It was determined that the 
best seasonal grouping was by 3-month periods, beginning in February. 
Using this grouping and a weighted-ratio method of combining 
monthly data which takes care of variations from month to month in 
the amount of fishing, the seasonal and yearly summarization was com- 
pleted for the Georges Bank area. 
The final step in the preparation of the index curve is an adjustment 
for the various changes in fishing methods and apparatus that were 
introduced from 1914 to 1937 in order to make the data on catch per 
day comparable over the entire period. The adjustment for these 
changes is of prime importance, since some of the most vital conclu- 
sions concerning the productivity of the stock at different levels of 
abundance depend on a correct determination and comparison of the 
levels of abundance in early and in recent years. 
Most of the work on evaluation of the above changes was completed 
during 1939. The increased size and power of the trawlers was taken 
into account by using specific groups of boats over a protracted pe- 
riod, and when forced through obsolescence of boats to change from 
one group to another the activities of the two groups were overlapped 
to obtain a measure of relative effectiveness. There also was a change 
from steam to Diesel power involving a decrease in both gross tonnage 
and horsepower, but no appreciable change in effectiveness. The effect 
of Vigneron-Dahl gear, adopted by most of the boats between 1927 
and 1931, was evaluated by comparing the productivity of groups of 
boats before and after installation of this gear. Results showed that 
the new gear increased performance by at least 29 percent during the 
winter, and probably more in summer. Evaluations of the influence 
of the radio, fathometer, and gear competition were partly completed. 
A study of the effect of changes in the proportion of other species in the 
catch was partially completed. 
