PHYSICAL OCEANOGRAPHY OF THE GULF OF MAINE 



865 



Analysis of these sets shows a dominant drift toward the north and west (aver- 

 age direction about NW. by W.) during the spring, summer, and early autumn, 

 averaging about 3.4 miles per day; but about as strong a southeasterly set (3 miles 

 daily) during the late autumn, winter, and early spring, averaging about S. 50° E. in 

 direction. If January and February be credited with about the same dominant 

 drift as is recorded for December and March, the average set of water for the year 

 works out at about 1.3 miles per day toward the N. 74° W. The rate has aver- 

 aged lowest (less than 0.1 knot) from March through June, and drifts as strong as 

 0.2 knot have been recorded only during the months from August to December, a 

 fact of some interest in connection with the discharge of surface water from the gulf 

 (p. 974). This series of observations gives evidence of a considerable balance of 

 movement of water toward the WNW. past the southern slopes of Nantucket Shoals, 

 and whether the set be in that direction or toward the southeast, it is away from 

 the gulf in either case. 



This seasonal reversal in the direction of the dominant current is probably 

 caused by the wind, with the southeasterly drift of winter reflecting the prevalence 

 of strong northwest winds at that season ; but the fact that the summer drift toward 

 the west or northwest is not parallel with the prevailing southerly and southwesterly 

 winds, but at right angles to them, reveals the dominant tendency for the water here 

 to move westward. 



Current measurements taken at eight stations along the southern rim of the 

 the Gulf of Maine by the United States Coast and Geodetic Survey in 1877 show 

 in each case a considerable nontidal resultant; and the indicated drift at any one of 

 these may have been affected by the wind, for all were of short duration. However, 

 they prove so consistent with the theoretic expectation of a clockwise movement 

 around a shoal (p. 972) that they are probably representative of the prevalent smn- 

 mer state. The resultant drifts, as calculated by Mitchell (1881, p. 189, table 8), 

 are as follows: 



• The U. S. Coast and Geodetic Survey writes that "resultant," in Mitchell's (1881, p. 189) original account, refers to the 

 set for a tidal day of 24 hours and 60 minutes. This is reduced here to the set per 24 hours. 



2 The dominant drift is given as southeasterly at station 2, northeasterly at station 3, by Harris (1907, chart 7), and in the 

 1912 edition of the Coast Pilot (1912, chart to face p. 9); but a fresh calculation of the nontidal set at these stations by the 

 Coast and Geodetic Survey shows a very good agreement with Mitchell's results. 



These drifts indicate a general movement of the water northwestward around 

 the western side of Georges Bank and southeastward over the eastern side, which is 

 corroborated by bottle drifts (figs . 1 74, 1 76) . They also suggest a subsidiary clockwise 



