PHYSICAL OCEANOGKAPHY OF THE GULF OF MAINE 771 
high surface salinity (presumably about 32.5 per mille) is also to be expected over 
the shoal part of the bank and near its northern edge. 
Very considerable fluctuations are to be expected in the salinity of the surface 
along the edge of the continent abreast of the Gulf of Maine, as well as in its tem- 
perature (p. 596), as the oceanic water of high salinity approaches the banks or 
recedes from them. 
In the southwestern part of the area, in the offing of Marthas Vineyard, the 
data for July, 1916, August, 1914, and for autumn (p. 801) make it reasonably certain 
that surface water as saline as 33 per mille normally drifts in over the outer part of 
the shelf during July and the first three weeks of August, but seldom (perhaps never) 
approaches much nearer the shore than is represented on the chart for 1914 
(fig. 136). 
Farther to the east the isohaline for 33 per mille may be expected to skirt the 
southern edge of Georges Bank in July, lying a few miles farther in in some summers, 
farther out in others, and crossing the oceanic triangle between Georges and Browns 
Bank, but not, in our experience, encroaching at all over the latter. Still farther 
eastward surface water as saline as 33 per mille overflows the edge of the continent 
in July or August of some years, as in 1915, when Bjerkan (1919) had still higher 
readings (34.27 per mille) at the 400-meter contour in the offing of Cape Sable on 
July 22. In 1914, however, the surface water near by was only 31.22 per mille a 
week later in the season (station 10233), though the difference in date would suggest 
a difference in salinity of just the reverse order, evidence of considerable fluctuation 
in this respect from summer to summer. 
It is doubtful whether surface water as salt as 34 per mille ever encroaches on 
the edge of the continent abreast of the Gulf of Maine; certainly we have no record 
of such an event at any season, but the surface charts for the winter, spring, and 
summer (figs. 93, 127, and 136) show that it is to be expected only a few miles out from 
the 200-meter contour south of Marthas Vineyard and off the western end of Geor- 
ges Bank by the first half of July in early seasons, but perhaps not until August in 
late seasons. In some summers, as in 1914, water of this high salinity lies farther 
out from the edge of the continent to the eastward. In other summers, however, 
it evidently spreads shoreward over the slope off Shelburne as early in the season 
as it does farther west — witness the records obtained by the Canadian Fisheries 
Expedition in 1915, mentioned above (Bjerkan, 1919; Acadia station 41). 
None of our lines have run far enough out, abreast the gulf, to reach surface 
water of full oceanic salinity (35 per mille and upwards) ; nor is it known how far 
out from the edge of the continent water of 34 per mille withdraws in winter and 
spring. 
ANNUAL VARIATIONS IN SURFACE SALINITY IN SUMMER 
Passing reference has been made in the preceding pages to the variations that 
have been observed in the salinity of the surface from summer to summer. The 
most interesting fluctuation of this sort that has come to our attention is that surface 
values averaged much lower in the southwestern part of the region in July, 1916, 
than in that same month in 1912, 1914, or 1915; the surface of Massachusetts Bay, 
for instance, was about 1 per mille less saline on July 19 to 20, 1916, than at about 
