an area miglit not extend far enougli seaward to include all areas 

 of mixed-water eddies, it would define the western limits to which 

 bergs might be expected to drift, and locate the major eddies and 

 areas where ice might be expected to threaten traffic on the North 

 Atlantic Track Agreement tracks, and indicate areas for more ex- 

 tensive subsequent surveys. The work of collection of data began at 

 station 3747 located at 43°34' N., 51°30' W., at 1745 on 5 April. At 

 0355 on G April, it w^as necessary to heave-to to await better weather. 

 At 2257 on G April, station work was resumed at station 3751. How- 

 ever, a fathometer failure which had occurred during the heavy 

 weather was found to be beyond possible repair at sea, and upon 

 completion of station 3753 at 1300 on 7 April, the Evergreen laid a 

 course for Argentia to accomplish the necessary repairs, arriving 

 there at 133G on 8 April. 



Advantage was taken of this inport time to replenish exhausted 

 supplies of high-pressure nitrogen for the winches, to stop major 

 leaks in the gas and hydraulic systems of the winches and to clean 

 and adjust their valves to prevent their unloading at wire tensions 

 below about 1,000 pounds. With this done, departure was taken from 

 Argentia at 1130 on 12 April to continue the current survey inter- 

 rupted by the fathometer failure on the first cruise. Work of collec- 

 tion of data began on 13 April at station 3754, located at 42°02.5' N., 

 52°04' W., and progressed northward without major interruptions, 

 speeding up as inexperienced personnel gained facility in the per- 

 formance of their various tasks. Gear was shifted to the port side 

 and that winch used during the occupation of stations 37GG to 3773, 

 while a rough valve (replenishing and servo-pump relief valve) in 

 the starboard winch was polished. It was learned from experience 

 that because of the rolling of the ship, hydraulic oil sump levels had 

 to be maintained at higher than indicated normal levels to prevent 

 air getting into the hydraulic system. One 900-foot bathythermo- 

 graph and one 450-foot bathythermograph were lost through break- 

 ing of the wire under circumstances which indicate the wire had 

 been defective. Two reversing thermometers were broken and two 

 messengers were lost overboard during heavy weather. Several mis- 

 cellaneous Nansen w^ater bottle parts failed. However, few if any 

 essential dalfi were lost and the survey, comprising 57 stations, was 

 completed at station 3810 located at 46°17.5' N., 49°00' ^N., on the 

 afternoon of 23 April and a course laid for Argentia, with arrival 

 there at 1740 on 24 April. 



A third cruise was undertaken Avith departure from Argentia a< 

 OGll on G Ma}'. Previous cruises had indicated little need for extend- 

 ing subsequent survej's westward of tlie Tail of the Grand Banks. 

 This survey was tlierefore intended to cover the area from Flemish 

 Cap to the Tail of the Banks and tlie work of colk'ction of data began 



52 



