832 
BULLETIN OF THE BUREAU OF FISHERIES 
from Halifax, and one from Gloucester Harbor, where it was picked up on August 
14. 31 Although two of the bottles from this line drifted to Newfoundland, showing 
a division, this does not detract from the evidence of the Gloucester recovery. 
Clearer evidence that the cold tongue that skirts Nova Scotia and flows past 
Cape Sable into the Gulf in Maine in spring is actually an overflow from the icy 
pool that develops from Cabot Strait out over Banquereau Bank, when the ice is 
coming out of the Gulf of St. Lawrence, could hardly be asked than results from the 
temperatures, salinities, and bottle drifts just discussed. 
I believe it now sufficiently demonstrated that while this cold pool (fig. 167) 
owes its low temperature, to some extent, to the direct outflow of icy water from 
the Gulf of St Lawrence via the Cape Breton side of Cabot Strait, it more directly 
mirrors the chilling effect of the field ice from the Gulf of St. Lawrence as this 
melts in the region between Banquereau Bank and Sable Island. Consequently, 
cold water that reaches the Gulf of Maine from the east is, in fact, ice-chilled, 
though this takes place 300 miles or more to the eastward of the eastern portal to 
the gulf. 
It is to this cold band skirting Nova Scotia that the name “Nova Scotian cur- 
rent” is applied in the preceding pages. During the spring a large volume of water 
enters the eastern side of the Gulf of Maine from this source, producing the effects on 
salinity and temperature described in the chapters on those physical features; and 
this is certainly the chief source that contributes cold water of northern origin to the 
Gulf of Maine — almost certainly the only source making a contribution of this sort 
sufficient in amount and cold enough to exert any appreciable effect on the tempera- 
ture of the gulf (p. 682). 
This current flows into the gulf in volume during only a few weeks in spring — 
earlier in some years, later in others. As its fluctuations are referred to repeatedly 
in the preceding pages a summary will suffice here. 
In 1920 (a late season) icy water (<1°) from this source had spread west- 
ward as far as the offing of Shelburne, Nova Scotia, by the last week in March; but 
neither the temperature nor the salinity of the eastern side of the Gulf of Maine give 
any evidence that it had commenced to flood past Cape Sable up to that date, nor 
do the isohalines for that April suggest any drift of water of low salinity into the gulf 
from the east. The coastal zone, also, warmed about as rapidly in the one side of 
the gulf as in the other during that month (p. 553). Conditions seem, then, to have 
remained comparatively static off Cape Sable through the first two months of the 
spring of 1920, and if the Nova Scotian current discharged at all into the gulf in that 
year this did not happen until May or later. In 1919, however, an early season, its 
western expansion culminated before the last of March, and had slackened, if not 
ceased, by the end of April (p. 558). In this respect 1915 seems to have been inter- 
mediate (so may be taken as a representative spring), with the Nova Scotian current 
exerting its chief chilling effect on the eastern side of the gulf before the first week 
in May (p. 560), and slackening from May to June, as indicated by the contraction 
(to the eastward) of the area inclosed by the surface isohaline for 32 per mille (cf. 
fig. 120 with fig. 128). 
si Information kindly supplied by Dr. A. G. Huntsman. 
