638 



BULLETIN OF THE BUBEAU OF FISHERIES 



Unfortunately, it is not known whether autumnal cooling proceeds at as rapid a 

 rate during October out over the basin of the gulf in general as it does along the 

 western shore, nor are data available for Georges or Browns banks during that 

 month; but Ratiibun's (1887) tabulations show the surface almost as cool at Pollock 

 Rip, off the southern angle of Cape Cod, on October 1 (11° to 13.5°) as it is in 

 Massachusetts Bay at that same date. This applies also to the whole region of 

 Nantucket Shoals, where the Halcyon had surface temperatures of 11.6° to 12.2° on 

 October 1, 1925, showing that a decided regional equalization had taken place 

 since midsummer, when surface readings in the same region have ranged from 

 11.6° to 16.4° (p. 1012). 



The autumnal cycle of temperature to the southward of Marthas Vineyard lags 

 several weeks behind that of the waters to the north and east of Cape Cod. Thus, 

 the surface was 13.3° to 14.4° across the whole breadth of the continental shelf off 

 Marthas Vineyard on October 22, 1915 (stations 10331 to 10333), with 15.5° a few 

 miles outside the continental edge, while the Halcyon had 13.3° near No Mans Land 

 on the 28th of the month in 1925. This corresponds closely with Rathbun's aver- 

 ages of 15° for October 1 and 11.7° for November 1, 1881 to 1885, 22 miles off Nan- 

 tucket (the old situation of Nantucket South Shoals lightship, which has since been 

 relocated) . 



Average and extreme surface temperature, °C., 1881 to 1885, from Rathbun's {1887) graphs, to the 



nearest half degree only 



• For years 1884 and 1886 only, the readings (or 1881 and 1882 being omitted because so irregular that their reliability is 

 doubtful. 



> Omitting one reading of 0.66°, which was obviously an error. 



SUBSURFACE 



At first the autumnal cooling of the surface, which accompanies the cooling of 

 the air, is due not only to an actual loss of heat by radiation (p. 692) but reflects mix- 

 ture with the cooler underlying water, a process that correspondingly warms the 

 latter. The result is that the annual maximum is attained later and later in the year 

 as the depth of observation increases down to about 100 to 150 meters, or to the lower 

 boundary of the stratum, the temperature of which is controlled by solar warming 

 alternating with winter chilling. Consequently the wide vertical range of temperature 

 that characterizes most parts of the gulf in summer gradually gives place to a state 

 of vertical homogeneity as the autumn progresses. In 1915 (a tjj-pical year) autum- 

 nal cooling had affected only the uppermost stratum of Massachusetts Bay up to the 

 end of September, the 20 to 25 meter temperature having continued virtually sta- 

 tionary at the midsummer value (11° to 12°) up to that date, with a rise of 2° to 3° at 



