918 
BULLETIN OF THE BUREAU OF FISHERIES 
Maine) an admixture of the colder and less saline water from the Nova Scotian cur- 
rent; but the details of this process and the extent to which it influences the tem- 
perature, salinity, and circulation of the northeastern part of the gulf can not be 
worked out until more data are gathered for the critical months of May and June. 
It is much to be regretted that no records on the eastern part of Georges Bank 
have been obtained for June, which might throw light on the expansion of Nova 
Scotian water in that direction; but the fact that we have found the surface salinity 
considerably higher in the Eastern Channel and in the basin of the gulf near by 
than from Browns Bank in to Cape Sable, both in June and in July (figs. 128 and 
136), shows that any movement that may take place along this zone toward the 
southwest in spring had ceased by the beginning of the summer both in 1914 and 
in 1915. 
CIRCULATION OF THE SUPERFICIAL STRATUM AS INDICATED BY 
TEMPERATURE 
The distribution of temperature is by no means as clear an index to the non- 
tidal circulation of the surface waters of the gulf as is its salinity, because any given 
mass of surface water may be warmed rapidly by the sun or cooled by radiation 
when the overlying air is the colder without suffering any alteration in its identity 
by mixture with other water masses. In the deep strata, however, which are more 
or less insulated from these thermal influences from above, regional differences in 
temperature are more easily interpreted in terms of circulation. 
The relationship of temperature to circulation is referred to repeatedly in other 
connections; 84 only the most salient aspects, then, need be referred to here. 
The belt of coldest water, which fringes the shores of the gulf in winter, owes 
its low temperature to the chilling effects of the icy winds that blow out over it from 
the land. The fact that this cold band (as illustrated by the surface charts for mid- 
winter (fig. 80) and for February to March — fig. 1) is comparatively uniform in 
breadth all along the northern and western shore line, is best reconciled with a set 
paralleling the shore. Any considerable movement of surface water either from the 
land out to sea or vice versa would give much more undulatory courses to the criti- 
cal isotherms of 5° in December to January and 2° in February to March. 
Surface water equally cold over the Northern Channel and Browns Bank on 
the February to March chart (fig. 1), giving place, by a rather abrupt transition, to 
readings 1.5° higher over the Eastern Channel, reflects the westernmost bound of 
the Nova Scotian current at the time; and an expansion of water colder than 4° out 
over the channel from the the gulf and across the eastern end of Georges Bank but 
not across the western end of the bank is evidence of a movement in that direction, 
which corresponds to the drifts of bottles set out off Cape Cod in summer (p. 886). 
The undulatory course of the March isotherm for 3° gives a rather clear indica- 
tion of an anticlockwise eddying movement in the central part of the gulf, with 
warmer water moving northward in the eastern arm of the basin and colder water 
drifting out from the land off Penobscot Bay, illustrating one of the varying forms 
81 See the chapter on temperature. 
