594 
BULLETIN OF THE BUREAU OF FISHERIES 
not more than half a dozen miles wide by the end of the summer, persisting only as a 
reflection of purely local activity of tidal stirring. 
Our Gulf of Maine cruises have not crossed the southeastern part of the area in 
August; hence the isotherms for this region (fig. 46) are only tentative for that 
month, combined from the July cruise of the Grampus in 1914, the Canadian Fish- 
eries Expedition stations off southern Nova Scotia for July, 1915, temperatures taken 
by the Albatross in August, 1883, and July, 1885 (Townsend, 1901), and from scat- 
tering records from other sources. These combine to show a rather abrupt transi- 
tion in surface temperature in the region of the Northern Channel between the cool 
area along western Nova Scotia (12°) and somewhat higher readings (14° to 16°) on 
Browns Bank, but make it unlikely that the surface normally warms above 16° over 
the latter at any season. It is probable, too, that much local variation in tempera- 
ture exists on Browns Bank, with cool and warm streaks caused by tidal mixings, 
especially along its southwestern edge fronting the Eastern Channel, where the 
Albatross had surface readings of 12.78 to 13.3° at four stations on August 31, 1883. 
The surface temperature in the center of the Eastern Channel was 15.1° on July 
24, 1914 (station 10227), but readings of 12.8°, 16.1°, 14.2°, and 13.3° at four succes- 
sive stations on a line crossing the deep water from Georges Bank to Browns Bank 
on August 31, 1883, suggest that while the central core of the channel is usually 
fractionally warmer than 16° by the end of the summer, vertical stirrings or upwell- 
ings are sufficiently active along the edges of the two banks to maintain narrow lanes 
there colder than 16° on the surface. 
It is probable that the surface is from 1 to 3 degrees cooler over the eastern, 
northern, and central parts of Georges Bank, as a whole, than in the basin of the gulf 
to the north throughout the summer, and certainly it is considerably cooler than the 
oceanic waters outside the edge of the continent to the south, just as it is in June 
(fig. 39). Thus, Dr. W. C. Kendall had surface readings of 12.8° to 15.3° (averag- 
ing about 14.5°) at 55 stations along the northwestern edge of the bank on August 21 
to 25, 1897, and the isotherm for 16° for this region is located on the chart (fig. 46) 
from these observations. 
This part of the bank offers an excellent illustration of the chilling of the surface 
that follows when cooler water from below is brought up over and around shoals by 
the tides, with the surface averaging 1° to 3° cooler over the shoal ground than 
elsewhere on the bank and (generally) coldest (13° to 14°) over the shoalest part, 
where the water is less than 50 meters deep. Even small isolated shoal spots may cause 
cool pools at the surface in this region, and the effect of projecting submarine prom- 
ontories or ridges may be made evident for some miles by lowered surface temper- 
ature. Where the water is not only shoal, but the topography of the bottom is 
broken and tidal currents run strong, considerable variations in surface temperature 
also are to be expected from ebb to flood, as Dawson found to be the case near Cape 
Sable (p. 593). Doctor Kendall records several such alterations on Georges Bank, 
notably a drop of about 1.5° at a station on its northern edge during a period of a 
few hours on August 21. A few yards’ sailing may also be enough to bring the 
vessel from a cool streak into a warm one, or vice versa, the explanation for which 
is apparent enough on calm days when the lines of contact between different runs 
of tide are often made visible by miniature rips, oily slicks, or by the accumulation 
