98 



One of the striking features shown by the surface isotherms south- 

 east of the Banks is that of the tremendous embayments in the 

 boundary between the cold and warm currents. That the warm and 

 cold tongues are especially well developed in the middle part of the 

 ice season is shown by the isotherm chart for the period May 21- 

 June 5, 1928 (fig. 28 in Coast Guard Bulletin 17), and the one for the 

 period June 3-18, 1929 (fig. 29 in this pamphlet). 



Very often where an extra cold tongue from the Labrador Current 

 extends southward on the surface an extra warm tongue will pro- 

 trude north close to it and just to the westw^ard, it would seem almost 

 as a compensation. This sharp contrast of waters at times appears 

 to accelerate the local oceanic circulation along the adjacent intensi- 

 fied temperature walls. The 5-day drift of 160 miles across the forty- 

 fifth parallel of the southeasternmost berg of April, 1928, is not in 

 close agreement with the surface isotherm indications in the area, 

 but it is quite characteristic, and may be typical of the local area and 

 the particular contemporary distribution of surface temperature. 

 The writer believes it to be referrable to the close approximation of 

 tongues of 42° and 54° surface water near 44° 20' N., 45° 50' W., 

 which are plotted on the isotherm chart for the period April 6-21, 

 1928 (fig. 25, Coast Guard Bulletin 17). 



Examples of berg drifts much at variance with the surface isotherm 

 indications are easily found. Off Cape Race, along the eastern edge 

 of the Grand Banks and close to and west of the Tail, such apparently 

 anomalous berg drifts are freci[uent. Possible explanation of this 

 condition can be found in the persistent streaming along of the layers 

 of cold water controlling the bergs. It appears as though the cold 

 water is at times forced under warmer and lighter surface layers, and 

 in these cases the surface isotherms should not be taken as guides for 

 berg drifts. To be able to detect these cases experience in the ice- 

 patrol region or close study of the old ice-patrol records is necessary. 



Before any definite statement can be made it will be necessary to 

 determine more fully under what conditions and how regularly surface 

 isotherm curves can be assumed to mirror the underlying circulation. 

 Present indications are that this is the case and that valuable infor- 

 mation can be. obtained from them in the cases where and when a 

 cold wall is strongly developed. By a cold w^all is meant a region in 

 which the surface isotherms are noticeably packed. It is a narrow 

 belt across which the surface temperature gradient is much greater 

 than it is farther on in either direction along the line drawn at right 

 angles to the different isotherm curves. 



The dynamic investigations have shown that the density wall, 

 which usually controls the circulation, normally lies a variable dis- 

 tance ipshore of the cold wall in the Grand Banks area. When 

 depending on the surface temperatures alone the exact location of 



