PHYSICAL OCEANOGBAPHY OF THE GULF OF MAINE 
631 
was slightly colder on July 25, 1916, than it had been on July 14, 1915, and nearly two degrees (C.) 
colder on August 16, 1916, than it had been on August 27, 1914. Also, in the Bay of Fundy, east 
of Grand Manan, the temperature of the body of the water was nearly one degree (C.) lower on 
July 24, 1916, than on July 15, 1915, and more than two degrees (C.) lower oil August 16, 1916, 
than on August 27, 1914. This shows that in the Bay of Fundy the water was colder in the 
summer of 1915 than in that of 1914, and still colder in that of 1916. 
Enough data have thus been gathered to class 1916 definitely as an abnormally 
cold year in the gulf. 
It is interesting to consider whether climatic conditions during the preceding 
months will account for this abnormal] ty. Unfortunately, no observations were taken 
in the gulf during the preceding winter, but the deep temperature of the western side 
changes so little from February to June that its July state gives an indication of the 
temperatures that have prevailed there in spring. Judged from this viewpoint, the 
July temperatures of Massachusetts Bay and of the neighboring parts of the gulf 
for 1916 do not suggest that the sea temperatures of the preceding winter were 
abnormally low. 
This conclusion is corroborated by meteorological conditions, for the early part 
of the winter of 1915-16 was warmer than usual (mean temperature for January 
about 6.7° F. higher than normal at Boston, 2.7° F. higher than normal at Province- 
town) ; but the temperature was about 2.5° F. below normal at Boston in February, 
4.4° F. below normal in March, with unusually heavy snowfall in both these months 
(30.3 and 33.3 inches, respectively). Consequently, there is every reason to suppose 
that the temperature of the water of Massachusetts Bay did not commence to rise 
until a month or even two months later than usual that spring, and that vernal 
warming proceeded more slowly at first than in more normal years, because the 
weather continued abnormally cool and cloudy throughout May and June. Fur- 
thermore, it is in just such a spring as this, when the surface stratum warms very 
slowly at first, but then rapidly, that the deeper water is most effectively blanketed 
from the penetration of heat from above by the sudden development of a state of 
high stability. Indeed, a better illustration of how slowly the deeper water warms 
under such circumstances could hardly be found than by the very small rise in tem- 
perature that took place off Cape Cod from July 22 (station 10344) to August 29 of 
that year (10398) at 40 to 50 meters. 
Thus the difference in temperature between the cold summer of 1916 and the 
warm summers of 1913, 1914, and 1915, in the western side of the gulf, was no wider 
than can be accounted for on the basis of the local weather. 
I may point out that a cold winter and spring in 1916 were similarly followed 
by low summer temperatures in the coastal water all along the continental shelf, 
westward and southward to Chesapeake Bay during that same year (Bigelow, 1922), 
not alone in the Gulf of Maine. 
It is possible that the low gulf temperatures of 1916 also reflected some unusual 
expansion of the Nova Scotian current, because even a temporary offshoot of that 
icy-cold stream crossing the gulf at any time during the spring would chill the sur- 
face of its western side 2° to 3° or more below normal (p. 680). Had the Grampus 
made a general survey of the gulf in 1916, as she did in 1914 and 1915, this question 
would have been cleared up; but the few stations for that cold year were all located 
