156 MAraON EXPEDITIOX to DAVIS strait and BAFFIlSr BAY 



Many direct observations on the rate of movement of icebergs 

 south of Newfonndhincl have been made by the ice patrol, and, sup- 

 ported by numerous current surveys run simultaneously, furnish us 

 Avith accurate information. Any student interested in examining 

 this matter in detail is referred to the file of annual reports of the 

 patrol. (See p. 150.) The records may be summarized as follows: 

 During March the average rate is to 7 miles per day, and 10 to 17 

 miles per day from April to June around the southern parts of the 

 bank. The bergs on the northern edge of the Gulf Stream move 

 fastest of all, i. e., at rates of 19 to 30 miles per day. In late summer 

 the ice drifts more slowly, but at all seasons the rates are subject to 

 considerable variations from week to week or even from day to day. 

 The velocities of ocean currents, especially at this junction region, are 

 constantly changing, changes reflected by the drifts of the icebergs. 

 For exaui])le, on one occasion a group of bergs on the east side of the 

 Grand Bank just north of the Tail suddenly increased their rate of 

 drift overnight from 7 miles to 28 miles })er day. (Smith, 1921, p. 

 76.) If the curi'ent can accelerate suddenly to such a high rate off 

 the Grand Bank, may not such events occur equally along the Lab- 

 rador shelf ? 



The ice patrol's investigations have revealed the following features 

 in the circulation south of Newfoundland affecting the behavior of 

 icebergs: 



(a) An ellii)tical dei)ressi()n in the sea surface makes a slow 

 cyclonic vortex in the surface layers off the southwest slope of the 

 Grand Bank west of the Tail. This " low,"' covering an area of over 

 2,000 square miles, was first discovered by chance in 1921 by follow- 

 ing an iceberg as it made the circuit. 1926 the "" low," apparently 

 similar in character to an atmosphere cyclonic depression, was accu- 

 rately charted by several successive dynamic surveys (see Smith, 

 1927, pp. 109, 112, and 115) prevailing throughout the season. It 

 was present also in Ai)ril, 1927. l)Ut disappeared early that May. 

 Bergs are often carried northwestward along the sh)pe in the nortli- 

 ern semicircle of this eddv and sometimes a complete circuit of 365° 

 in its periphery. (See fig. 101.) 



(b) Sometimes the counterclockwise eddy is wanting; displaced by 

 the Gulf Stream encroaching toward the Tail of the bank. An 

 earlier series of current maps (Smith 1927b) shows the development 

 of such a phenomenon in the first week in May. 1927. As long as 

 the Gulf Stream maintains this position ice is blocked from pursu- 

 ing a course west of the Tail. (See lower right-hand sketch, fig. 

 101, also overlay figs. 105 and 106.) 



(c) The cold and warm currents after meeting off the Tail of the 

 Grand Bank proceed southeastAvard in a parallel set to the vicinity 

 of latitude 41°. longitude 47°. where the streams bend sharply to the 

 left. The existence of this cul-de-sac of the current, and occasionally 

 for bergs, 150 miles soutlieast of the Tail shows evidently that the 

 drift is affected by the bottom configuration, even to as great a depth 

 as 4.000 meters. ' (See Smith 1922. Chart H and lower left-luind 

 sketch, fig. 100.) 



(d) The ice patrol suspected for several years that the Gulf 

 Stream after turning sharply northward on the forty-seventh merid- 

 ian pressed in toward the bank up the submarine embayment l)etween 



