developing into the salient the axis of Avhich is near the -fTth meridian 

 in figure 16. As the two surveys were made about 2i days apart 

 and as the salients as thus identified diti'er in position by about 150 

 miles, their rate of progress would have averaged about G miles a day. 



If bergs had been present the region of potential ice hazard would 

 have moved from that southeastward of 42°30' N., 47°00' W., to that 

 southward of 4:4:°0(y N., 45°30' W. These conditions, shown in figure 

 16, are such as accompany the ''break-through" of bergs southeasterly 

 from the area between the Grand Banks and Flemish Cap directly 

 toward the U. S. -European steamer lanes. This break-through of 

 bergs has occurred with sufficient frequency to be considered a sea- 

 sonal characteristic of early May. 



It has been assumed that the formation of the eddies or salients 

 mentioned above is conditioned by factors peculiar to this region, 

 such as bottom topography and the junction of the Labrador and 

 Atlantic Currents.- It has been assumed further, that the position 

 of the boundary betAveen the Atlantic Current and the mixed water 

 along its outer edge is controlled by the volumes of flow of the Lab- 

 rador Current and the Atlantic Current, and that fluctuations in the 

 size, shape, and degree of incursion of the salients are associated with 

 fluctuations in the volume of flow of these two currents. An interest- 

 ing corollary of such an hypothesis in the light of the apparently 

 seasonal break-through is that there are seasonal fluctuations in the 

 volume of flow of the Labrador Current or the Atlantic Current or 

 both. Aside from its fundamental importance, it is important that 

 the interrelationship of the elements stated above be determined, since 

 it is regarded as a prerequisite to the practical forecasting of the 

 geographical extent of the ice hazard from week to week. 



While demonstration of the validity of the assumptions is not ade- 

 quate, some supporting evidence has accumulated. In an earlier Bul- 

 letin of this series,"* the location of the outer boundary of Atlantic 

 Current water found during the ice seasons of 193-1—41 was presented 

 and discussed with relation to observed fluctuations in volume of 

 flow of the Labrador Current and volume of flow of the Atlantic 

 Current inferred from differences in sea level across the Charleston- 

 Bermuda section. The criterion used for the boundary of Atlantic 



' See B. Haurwitz and II. A. Paiioft^ki, "Stability and Meandering of tlie Gulf Stream." 

 Trans. Am. Cieophy.s. Union. Vol. 31, Xo. 5, pp. 723-731 (Oct. lO-lO), Washington. In thi.s 

 paper its authors conclude that unstable waves may develop in the Gulf Stream after it 

 leaves the continental shelf and that these unstable wave.s form the eddies or meanders along 

 the outer edge of the Gulf Stream. The conditions are similar to those existing in that part 

 of the Atlantic Current iu the vicinity of the Grand Banks where the observed speed of prog- 

 ress of Atlantic Current salients is of the same order of magnitude as the speed of propaga- 

 tion of unstable waves computed from "realistic values" by Haurwitz and Panofski. 



' Soule, Floyd M., and C. A. Barnes, "International Ice Observation and Ice Patrol Service 

 in the North Atlantic Ocean — Season of 1941." U. S. Coast CJuard Bull. Xo. 31, pp. l.j-24 

 (1950), Washington. 



61 



