PHYSICAL OCEANOGRAPHY OF THE GULF OF MAINE 769 
as water is brought up by vertical currents of some sort, not in any direct indraft 
from offshore. 
This salt pool had no counterpart in June (fig. 128) or in May (fig. 120) of 1915, 
but much smaller phenomena of the same sort were recorded off Lurcher Shoal in 
April, 1920 (station 20101, 32.9 per mille), in the southeastern part of the gulf and 
in the eastern part in that March (fig. 91). Thus, following the freshening charac- 
teristic of May (p. 745), the eastern side of the surface of the gulf is once more as 
salt by the end of August as at any time during the early spring. 
Much lower values prevail along the west Nova Scotian shore all summer, 
Vachon (1918) having recorded 31.34 to 32.09 per mille on a line from Brier Island 
to Yarmouth on September 7, 1916, with readings of 31.17 per mille at high tide, 
31.12 per mille at low tide, in Yarmouth Harbor on the 8th. It is on the strength 
of his data that the isohaline for 32 per mille is represented on the August chart 
(fig. 136). 
To the eastward of Cape Sable the water next the coast is still less saline (31.7 
to 31.6 per mille) in summer, with rather an abrupt west-east transition from higher 
to lower values off the cape. Essentially this is the same regional distribution as in 
June, except that the successive isohalines shift to the eastward during the early 
summer as the Nova Scotian current loses head. The constancy of this Nova 
Scotian water from month to month and from year to year also deserves mention, 
the lowest values recorded in the offing of Shelburne (including Bjerkan’s (1919) 
data) ranging only from 30.9 to 32.1 per mille for the months of March, June, July, 
and September of the years 1914, 1915, and 1920. Sometimes these lowest values 
have been close in to the land off Shelburne, as was the case in July, 1915 (Bjerkan, 
1919), and in September of that year (fig. 137) ; sometimes farther out, with higher 
values next the coast, as in July, 1914, and in March, 1920 (p. 703); but no definite 
seasonal succession is yet established in this respect. 
The narrow band of water less saline than 32 per mille, which probably skirts 
the western coast of Nova Scotia every summer, is separated from the equally low 
salinities (31.2 to 32 per mille) of the northern side of the Bay of Fundy by consid- 
erably more saline surface water (32.3 to 32.4 per mille) along the southern (Nova 
Scotian) shore of the latter; such, at least, was the case in the summers of 1916 
(Vachon, 1918) and 1919 (Mavor, 1923). 
In each midsummer of record (1912, 1913, 1914, 1915) we have found the least 
saline surface water as a narrow but continuous band skirting the coast of Maine, 
and so southward to the region of Massachusetts Bay, usually 31 to 32 per mille in 
actual value. Inside the outer islands, and in the estuaries, still lower surface salin- 
ities are to be expected locally (e. g., 30.61 per mille in the western entrance to 
Penobscot Bay, August 3, 1912, station 10021a), grading, of course, to brackish water 
in the mouths of rivers. The definite boundary of this coastal water of low salinity 
(32 per mille) can not be laid down along the coasts of Maine and Nova Scotia on 
the chart for August, 1914 (fig. 136), because most of our stations for that year were 
located outside the 100-meter contour. In this respect the chart for 1913 (fig. 135) 
is more instructive. 
