ning at 0145 on 17 July. Tlie rectangular components of the fluctu- 

 ating currents were again determined, this time for all observations, 

 and are shown by small circles in figure 26. 



It would seem from figure 26 that the surface current did have a 

 fluctuating component that was about one-tliird of a knot, approxi- 

 mately circular and had a period of about 16 hours. The difference 

 between 16 lunar hours and 16 solar hours (about 32 mmutes) is too 

 small to distinguish in this series of observations considering their 

 short duration, scatter of the plotted points, and the fact that indi- 

 vidual measurements are not considered to be better than 0.05 knot. 

 If the periodic current is tidal one should expect to find its constituents 

 related to the constituents of the tides in period, although not neces- 

 sarily following the same ratio of amplitudes. Amongst the many 

 tidal constituents considered by the United States Coast and Geodetic 

 Survey •" in predicting tides one which is well down the list in impor- 

 tance as a tidal constituent, the M3 tide, is related to periods of 8 

 lunar hours and might conceivably form the basis of this observed 

 current period of 16 hours. The connection seems unsatisfactorily 

 remote. For the convenience of those who may wish to use these data 

 in connection with other studies of tidal currents it is noted here that a 

 new moon occurred on 15 July and that the local civil time of its 

 upper transit on 17 July was 1412. 



It has been suggested that the periodic current may be an inertia 

 current with its period that of a half pendulum day. For latitude 

 45°27' the half pendulum day would amount to 16.84 hours. This 

 would be close enough to be indistinguishable from the observed period. 



A rough check on the surface currents as measured by the current 

 meter is possible by comparison with the positions of the berg if it 

 is assumed that the berg's motion was the same as the surface current. 

 These positions were determined by celestial or Loran methods or 

 combinations of the two as available. Beginning at the position for 

 1745 on 16 July the berg was 2.1 miles 105° from the position projected 

 by the current measurements at 0800 on 17 July. Interpolating 

 between the positions for 0800 on 17 July and 0800 on 18 July, the 

 berg was 7.5 miles 108° from the position projected by the current 

 measurements at the end of the series of measurements at 2345 on 17 

 July. After 0930 on 22 July the berg was reported as a growler and 

 because of its shallow draft might have been more directly affected by 

 wind after that time. Between 1745 on 16 July and 0930 on 22 July 

 the drift of the berg averaged 0.37 knot in a direction 1 14° whereas the 

 average current deduced from the current meter observations was 0.25 

 knot 129°. This left a berg movement, unaccounted for by the surface 

 current, of 0.14 knot 087°. This may be regarded as being made up of 

 the discrepancy between the surface current and the average current 



9 Schuroinan, Paul, "Manual of harmonic analysis and prediction of tides." U. S. Coast and Geodetic 

 Survey Special Pub. No. 98, 1940 edition (1941), Washington. 



84 



