922 BULLETIN OF THE BUREAU OF EISHEEIES 



component of the eddy had shifted southward to sku't the northern slopes of 

 Georges Bank more closely. 



Information as to the movement of water along the bottom of the Northern 

 Channel is much to be desired at the season when the Nova Scotian current is flood- 

 ing in greatest volume into the gulf. Some drift may be assumed to take place into 

 the gulf by this route as deep as 100 meters in 1915, to account for the concen- 

 tration of the most saline water in the western side of the basin at the 100-meter 

 level in May (fig. 127), instead of in the eastern side, as at other times of year. 

 It is probable, therefore, that when the drift past Cape Sable is at its maximum it 

 causes a westerly shift in the vortex of the general eddy in the mid depths, though 

 not essentially altering the anticlockwise type of circulation, however. Any west- 

 erly drift that may have taken place along the bottom of the Northern Channel in 

 1915 had ceased by June; on this basis, alone, is the abrupt east-west transi- 

 tion that appears there on the 100-meter chart of temperature for that month expli- 

 cable (fig. 43). 



In midsummer the transition from lower salinities and temperatures in the 

 western side of the gulf to higher in the eastern, at the 100-meter level, and the 

 sweep of the successive isohalines and isotherms from east to west along the northern 

 slope of the gulf, again give evidence of a general set northerly past Nova Scotia 

 and westerly along the coast of Maine in the mid depths, paralleling the dominant 

 circulation at the surface. The nontidal movement of water of the southern side of the 

 basin at this level is not so clear, the picture being confused by an area of relatively 

 high salinity and temperature off the northern slope of Georges Bank near the 

 entrance to the Eastern Channel, which is not easy to account for. 



In spite of this and of other apparent anomalies the distribution of temperature 

 and salinity in the mid depths, as exemplified by the 100-meter level, are, as a whole, 

 compatible with the domination of the basin by the general Gulf of Maine eddy, 

 anticlockwise in character. 



The horizontal circulation of the gulf at greater and greater depths is more and 

 more directed by the contour of the bottom, which gives the basin the outlines of a Y, 

 with two arms uniting and open to the Eastern Channel (p. 784) at 175 meters, but 

 entirely inclosed at 200 meters and deeper. 



With temperatures and salinities recorded at one deep station or another for so 

 many months and years, it can be stated confidently that the movement of bottom 

 water inward into the gulf takes place in pulses, the secular fluctuations of which 

 have only been glimpsed as yet (p. 850). On the other hand, dynamics (fig. 204) and 

 the distribution of temperature and salinity point to some outgoing drift via the 

 western (Georges Bank) side of the Eastern Channel between these pulses in 

 summer (pp. 789, 852). 



The presence of water of high salinity (34 per mille) in both arms of the trough 

 but never (so far as yet recorded) over the submarine ridge that separates them is 

 good evidence that the latter divides the slope water as it drifts inward in the deep- 

 est stratum of the gulf. 



Two separate anticlockwdse eddying drifts are indicated in the bottoms of the 

 two arms of the trough, at depths of 175 meters and deeper, by salinities and tem- 

 peratures averaging somewhat higher on the side that would be to the right, for an 



