212 UEl'OKT ON THE SURFACE DRIFT OF THE ENGLISH CHANNEL 



on the shores of Lyme Bay: one on August 22nd on Chisel Beach, 

 the other on August 30th at Charmouth, near Lyme liegis. Yet 

 there is every reason to believe, as will be shown below, that most 

 of the February bottles were in the western part of Lyme Bay at 

 the end of that month. We are conductiug a special series of 

 experiments in this region during the current year. 



It appears accordingly, as a result of the above considerations, that 

 the general eastward drift of the Atlantic is probably not continued 

 through the English Channel as a current independently of the local 

 winds ; and that the deflecting influence of the tides upon the surface 

 drift is inmiaterial in the fairway of the Channel, but may be consider- 

 able in certain well-defined regions of the Channel in proximity to the 

 shore. 



The motion of the surface currents in the Channel will therefore 

 depend principally on the force and direction of the local winds, but will 

 be subject to modification by tidal currents in the regions enumerated. 



§ 4. The Law of Drift. — When discussing the influence of the local 

 winds upon the course of our bottles in § 1 the matter of direction was 

 alone considered, questions of velocity being entirely neglected. 



But as the exact route taken by the bottles is of considerable import- 

 ance, it is desirable that some attempt should be made to determine 

 the ratio between the velocity (or force) of the winds and the velocity 

 of the surface currents set up by their action, even though we can 

 only hope to attain a limited degree of accuracy. 



The only investigation upon this point with which I am acquainted is 

 contained in Mohn's classical memoir on the circulation of the North 

 Ocean, published in the Reports of the Norwegian North Atlantic Expedi- 

 tion (1887, pp. 117-123). According to Mohn, a wind of force 3-9 on the 

 Beaufort scale produces a drift having a velocity of 15 nautical miles 

 per diem. Converting the force of the wind to its velocity according to 

 Scott's table, he gets : 



Wind force 3*9 = wind velocity 22*5 miles per hour = drift 

 velocity 15 miles per diem. 



This result was gained by computing the mean velocities of the equa- 

 torial current and of the trade winds from a large number of cases in 

 which the mean directions of the current and wind approximately 

 coincided. The resultants of the two sets of calculations were then 

 regarded as respectively equivalent. 



The velocity of the drift in this and subsequent cases considered by 

 Mohn was taken to be directly proportional to the velocity of the wind. 



It seems to me to be clear from this account that Mohn's drift equiva- 

 lent is only applicable to cases of permanent currents. In his datum 



