straight line approximations for only a fraction of the whole year are 

 still on a rather shaky foundation, one should be careful not to draw 

 from them conclusions that are statistically unwarranted. It is, 

 nevertheless, interesting to note that throughout the summer season 

 covered by the tentative normals shown in figure 25 the volume trans- 

 port leaving the Bonavista triangle across sections SW. and SE. exceed 

 the volume transport entering the triangle across section NW. by 0.3 

 to 0.4 million cubic meters per second, and that this discrepancy 

 increases from April through July. Normal barometric charts for 

 these months indicate in this vicinity a geostrophic \nnd vector which 

 swings from being directed toward about 097° T with a weak gradient 

 in April to approximately 058° T wdth a much stronger gradient in 

 July. Thus a possibility exists that normally during this part of the 

 year there is an upwelling in the vicinity of the Bonavista triangle. 

 A qualitative confirmation may be deduced from considering that the 

 temperature at the 1,000-decibar surface is warmer than that at 

 shallower levels and that the excess of heat transport leaving the tri- 

 angle over that entering the triangle is greater than the product of 

 excess volume transport multiplied by normal mean temperature. 



The seasonal change in the normal barometric pressure distribution 

 has been called upon to explain the seasonal change in the number of 

 bergs making the southward passage between the Grand Banks and 

 Flemish Cap (see p. 66 of Bulletin No. 39 of this series). As the 

 season advances, more and more bergs are diverted eastward and 

 northeastward instead of making the southward passage. At both 

 the South Wolf Island section and at the Bonavista triangle the volume 

 transport normally increases as the season advances. For sections 

 T, U, and W, located south of the latitude of Flemish Cap, the volume 

 transport normally decreases as the season advances. At section H, 

 figure 26 shows a seasonal increase in volume transport. At sections 

 G, F2, and F, however, the sign of the seasonal change in volume 

 transport is not well established as may be seen from the plot of individ- 

 ual points in figures 26 and 27. While the montlily rates of change 

 have been computed as —.18, +.13, and +.08, respectively, for these 

 sections, additional observations could very well change both their 

 signs and magnitudes. 



When it is remembered that the bulk of the transport is below the 

 depth where seasonal warming can directly affect the Labrador Cur- 

 rent, the consistent seasonal increase in its mean temperature during 

 the summer season must find its explanation in seasonal changes in 

 the amounts of water along its margins which move as part of the 

 current, and in changes in velocity distribution between the cold 

 inshore part and the warmer offshore part. This principle was em- 

 ployed in Bulletin No. 35 (pp. 83-85) of this series to explain the steep 

 seasonal increase in mean temperature of the West Greenland Current 

 off Cape Farewell. It was assumed that at this section the West 



49 



