328 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



tober through April for the years 1939 to 1949. 

 If we take the logarithm of the catch and correlate 

 it with the temperature, we obtain a correlation 

 coefficient of 0.854 with 72 degrees of freedom, a 

 highly significant result. 



In figure 25, the Maine lobster catches for the 

 months of May through September, 1939 to 1948, 

 are plotted against the corresponding mean month- 

 ly surface water temperatures. The relation be- 

 tween catch and temperature for this period was 

 similar to that for the months of October through 

 April, 1939 to 1949, although obviously the 

 monthly catches in summer occurred at a consid- 

 erably lower level than would be predicted from 

 the winter relationship. The correlation between 

 log catch and temperature is 0.548, which is also 

 highly significant. 



50 60 



DEGREES FAHRENHEIT 



Figure 25. — Monthly catches of lobster in Maine plotted 

 against corresponding mean surface water temperatures 

 at Boothbay Harbor for the months of May through 

 September, 1939 to 1948. 



These data are inadequate to show that lobsters 

 are less available during the summer months when 

 temperatures are highest, or that an optimum 

 temperature is exceeded during the summer 

 months. Lobsters begin molting in the western 

 part of Maine in May, the process continuing 



more or less progressively eastward along the coast 

 until late September. Males molt earlier than 

 females, marked changes in the proportion 

 of the sexes are observed in the catch during the 

 molting period, and the molting of females is 

 attended by mating activities. During this period 

 availability decreases. Although availability is 

 usually high following molting, principally because 

 of the recruitment of lobsters that were under 

 legal size before they molted, the increased catch, 

 combined with the soft-sheUed condition, following 

 the molt, usually results in a decline in market 

 price. For these reasons, fishing effort and the 

 resulting catches may not reflect true availability 

 during the summer months. 



A reasonable objection to the foregoing analysis 

 is that the high degree of correlation between 

 lobster catch and water temperature is based on 

 total catches by months and not on catch per unit- 

 of-efTort. The correlation may, therefore, reflect 

 the greater amount of fishing during the warmer 

 months rather than the greater availability of the 

 lobster. 



Although some data on fishing effort by months 

 have been collected in Maine since 1939, the 

 reliability of the data has never been established 

 through adequate analysis. If the data on effort 

 for the period 1939-46 are used, no correlation be- 

 tween catch per unit-of-effort and temperature is 

 apparent; however, for the years 1939-42, a highly 

 significant correlation between monthly catch per 

 unit-of-effort and temperature is obtained. The 

 discrepancy appears to result from a rapid rise in 

 the abundance of lobsters in Maine over the period 

 1943-46, so that the catch per unit-of-effort cor- 

 responding to a given temperature is changing each 

 year. During the period 1939-42, the annual 

 catches did not vary greatly, thus the monthly 

 catches indicated availability rather than abun- 

 dance. Although the annual catches since 1946 

 have attained a fair level of stability, monthly 

 effort data are not available. 



Whatever the reasons for the significance of the 

 relation between monthly catch and monthly 

 temperature, it is, of course, evident from figure 24 

 that the relationship cannot be expected to hold 

 for temperatures much higher than 50° F. A 

 month with an average surface temperatiu-e of 

 65° F. would, for example, yield a catch about 

 equal to the average annual catch in recent years. 

 Disregarding the effect of molting on availability. 



