PHYSICAL OCEANOGBAPHY OF THE GULF OF MAINE 
831 
more complex than the simple stream flow from northeast to southwest suggested by 
the earlier evidence. 
The track followed by the ice drifting out of the Gulf of St. Lawrence is especially 
instructive here in this connection, because this discharge takes place in spring (chiefly 
in April and May) just when the Nova Scotian current is flooding past Cape Sable 
into the Gulf of Maine in greatest volume; whereas most of the drift-bottle experi- 
ments have been carried out in summer, when this current is usually inactive or at 
least is carrying so small a volume of water past Cape Sable that it is no longer an 
important cooling agent for the Gulf of Maine. According to Johnston (1915), the 
ice that comes out along the Cape Breton side of Cabot Strait does not tend to 
follow the Nova Scotian coast around to the southwest, as it would if the outflowing 
current hugged the coastline, but divides. Part drifts out to the southeastward; 
but the ice that emerges from the gulf nearest the Cape Breton coast moves south- 
ward across Banquereau Bank, where it fans out, to the offing of Halifax. 
These lines of dispersal correspond very closely with the icy water which Bjerkan’s 
(1919) data for May, 1915, show spreading out from the southern side of Cabot 
Strait to the region of Misaine and Banquereau Banks (fig. 167), but separated from 
the still colder ( — 1°) water on the Newfoundland Banks by a warmer (0°) core in 
the axis of the Lauren tian Channel, and with much higher temperatures off the mouth 
of the latter. Especially suggestive, from the standpoint of the Gulf of Maine, is the 
narrow icy tongue (0° to —0.2°) that then extended westward along Nova Scotia 
past Halifax; a band comparatively uniform, also, in salinity from east to west (31.5 
to 32.5 per mille) and considerably less saline than the still colder water on the 
Newfoundland side of the Laurentian Channel (temperature lower than — 1°; 
salinity 32.7 to 33.2 per mille). This the Ice Patrol cutter had also crossed on her run 
in to that port about a week earlier (United States Coast Guard stations 26 and 27, 
May 20, 1915). 
Lacking data in the offing of Cape Sable, it is not possible to state whether this 
cold tongue actually extended to the Gulf of Maine that May, though it may have 
done so earlier in the season and certainly does so during the spring in some years 
(p. 681). 
A similar concentration of cold water close in to Nova Scotia appears from the 
temperatures taken by the Ice Patrol along a line from Halifax toward Sable Island 
in spring in other years. The records for 1919 are especially instructive, showing 
this band widest at the end of March, when the whole column of water next the land 
was fractionally colder than zero from the surface to bottom; smaller in volume in 
April, when it was overlaid by slightly warmer (0° to 1°) water; and shrunken to a 
narrow tongue on the bottom not more than about 20 miles broad in May. 30 
Drift bottles set out by the United States Coast Guard cutter Tampa (Capt. W. 
J. Wheeler) on April 18, 1924, along a line running 119° (about SE x EJ^ E.) true from a 
point about 18 miles southeast of Sable Island (43° 48' N.; 59° 26' W.) for 50 miles, 
likewise show a drift from this region first northward toward the land and then west- 
ward toward the Gulf of Maine, three out of the seven returns (all from the inner 
end of the line) being from Sable Island, one from the Nova Scotian coast not far 
30 The March profile also cut across the southwestern edge of the icy Cape Breton- Banquereau pool near Sable Island. 
