39 



thermometers were computed for each group. These probable 

 differences varied between 0.003° and 0.014° C. and the probable 

 difference for the entire 1,434 comparisons was 0.008° C. It is 

 therefore considered that the observed temperatures are accurate 

 to 0.01° C. 



The current maps prepared during the season were made following 

 the procedure used in 1935, the season's maps being referred to the 

 1,000 decibar surface. The cmTent map representing conditions 

 found on the post-season cruise is referred to the 1,500 decibar 

 surface. 



The dynamic ciu-rent map for April, figure 25, shows the Labrador 

 Current to be somewhat subnormal in surface velocity in the northern 

 part of the map, but having unusual width. In latitudes 44° to 45° 

 its flow is augmented by the Grand Banks whorl wliich is frequently 

 a characteristic of the circulation in tliis region. Another feature 

 less frequently encountered is the slow but definite clockwise whorl 

 near the eastern edge of the Banks in about latitude 47° north. Its 

 effect is to set icebergs in onto the Banks; a natural tendency here 

 where the shoaling bottom below the Labrador Current probably 

 exerts some influence in a deflection toward the right. A part of the 

 Labrador Current ^\^ll be seen to recurve northeastward north of 

 latitude 45°. Although the measurements do not extend to the 

 westward of about longitude 50° west it is evident that south of the 

 Tail of the Banks the Labrador Current continues west and south 

 before recurving eastward. Gulf Stream water is to be seen in the more 

 rapidly flowing currents at the southern edge of the map. 



From figure 26 it will be seen that the Labrador Current decreased 

 in width and increased in velocity from April to May. The clock- 

 wise whorl noted in the April map near the eastern edge of the Banks 

 in about latitude 47° north joined the Banks whorl centered near 

 44°30' N., 49°30' W. This Banks whorl contributed water off the 

 Grand Banks to what would seem at first glance to be the Labrador 

 Current. However, it will be seen that that portion westward of 

 about the 971 isobath was part of the Banks whorl and cannot be 

 considered part of the ice-bearing Labrador Current. Much of the 

 latter continued to recurve northeastward between latitudes 47° 

 and 44° north although a part still continued to the Tail of the Banks.. 

 The Gulf Stream margin wealdy evidenced on the eastern edge of 

 the April map between latitudes 44° and 45° north had intruded 

 with strength and began the pinching oft' of the Labrador Current 

 and resulted in closing off and partly assimilating the cold-water pool 

 near the Tail of the Grand Banks. 



The June current map, figure 27, shows that the Labrador Current 

 was much weakened in velocity, while the intruding Gulf Stream 

 salient had made no further westward jDrogress from the situation shown 



