PHYSICAL OCEANOGRAPHY OF THE GULF OF MAINE 
719 
The salinity of the water in the Eastern Channel and its relationship to the 
water over Georges and Browns Banks, which bound it to the west and east, is always 
of interest, because this is the only possible route by which a deep bottom current 
can enter the gulf. During the second week of March, 1920, the saltest water in the 
channel took the form of a definite ridge, with the isohaline for 33 per mille, as rep- 
resented in cross section (fig. 99), paralleling the isotherm for 3° on the correspond- 
ing profile of temperature (fig. 19). The rather abrupt transition from 34 per mille 
to 33 per mille, made evident at the 50 to 80 meter level by closely crowded isoha- 
lines, contrasting with the vertical homogeneity of the shoaler water, marks this as 
the upper boundary of the saline bottom drift. 
The relationship between the vertical distribution of salinity in the trough 
(station 20071) and on the neighboring shallows of Georges Bank (station 20070; the 
o 
o 
<Nl 
Fig. 99. — Salinity profile running from the eastern part of Georges Bank across the Eastern Channel, Browns Bank, and the 
Northern Channel, to the offing of Cape Sable, March 11 to 23, 1920 
former much more saline than the latter at depths greater than 40 meters) is evi- 
dence of a banking up of the saltest water against the eastern side of the channel 
and of an overflow across Browns Bank consistent with the effect of the rotation of 
the earth on any movement of water inward through the channel toward the gulf. 
On the Georges Bank side, however, this indraft was separated from the slope by a 
wedge of water lower in salinity as well as in temperature (p. 541) ; therefore suggest- 
ing a counter drift in the opposite direction — i. e., out of the gulf (p. 938) — by its 
physical character. Unfortunately its lower boundary can not be definitely estab- 
lished from the station data, but the courses of the isohalines in the upper strata on 
the profile (fig. 99), combined with the contour of the bottom, suggest that it bathed 
the western slope of the channel down to a depth of at least 170 meters. 
This profile (fig. 99) also corroborates the evidence of the charts (p. 703) that 
water from the eastward had not yet freshened the upper 50 meters of water as far 
