fathoms. Most of the observations were in water which was deeper than 

 this and the 5 percent difference from the vahie used would affect only 

 the absolute and average values of (A-B) as in the case of the field in- 

 tensity. Another possible source of error is that no correction has 

 been made for lee-way and it is possible that the ship's head was oriented 

 more into the weather than was the electrode cable. More recent obser- 

 vations by von Arx and others, however, in which the orientation of 

 the electrode cable was determined by means of a remote-reading 

 Magnesyn compass system discovmts this soiu'ce of error as of little 

 probable significance in the results of the observations reported here. 



Another possible source of error which has not yet been exhaustively 

 studied is that involved in taking the product of the average of half- 

 hourly values of the normal component on any run between stations 

 and the station interval instead of the integral of the products of ele- 

 mental increments of distance and corresponding instantaneous values 

 of normal component. Work is continuing on the development of the 

 current meter and the prospect is good that solutions to the various 

 problems involved will permit the ultimate application of the method 

 to ice patrol current mapping with a reduction in the elapsed time 

 required for the production of a useful current map of a given area in 

 the Grand Banks region. 



While the problems of current mapping have to do with the com- 

 plete velocity vector another application of the von Arx current meter 

 involving only the direction of the current seems assured with the 

 instrument and methods as presently existing. This is the use of the 

 current meter to assist a patrol cutter so equipped in locating bergs 

 which have been reported in areas not covered by a recent current map. 

 In such an application the cutter would proceed to the reported position 

 of the berg and thence down-stream as indicated by the current meter, 

 instead of the present procedure of box- or ladder-searches. If the 

 down-stream direction indicated by the current meter is in error by 5°, 

 the cutter would be 12 miles off course after steaming about 140 miles. 

 It is expected that the berg would be located either visually or with 

 radar assistance before such a distance had been traveled. If wind- 

 currents affected the berg's drift significantly during the interval between 

 the time the berg was reported and the time the cutter began its down- 

 stream search, presimiably such currents would have been sufficiently 

 well established as to persist until the time of the cutter's search. The 

 major anticipated source of discrepancy is the circumstance in which 

 the current meter would indicate the direction of transitory wind cur- 

 rents different in direction from that of the deeper currents and too 

 shallow to have a significant effect on the berg's drift. Should tests 

 result in unsuccessful searches, a possible remedy would be the towing 

 of a su})sin'face pair of electrodes in addition to the surface electrodes. 

 Such practical test searches still await the combination of simultaneously 

 available ships, current meter equipments and bergs. 



96 



