AND NEIGHBOURING SEAS DURING 1897. 221 



during the gales, and partly to the narrowing of the Channel in this 

 region. This result accords with the experience of navigators, for, 

 according to the Pilot's Handbook (12th ed., 1893, p. 142), "strong W. 

 gales prolong the intermediate stream running E., and retard the stream 

 running S.W. At such times the streams at the Eidge shoal have been 

 found to run 8 hours to the N.E., and only 4 hours to the S.W." 



But there is no reason to assign to any of the Calais bottles the 

 remarkable velocity attained by this Eastbourne-Terschelling bottle 

 (XL 3) during the latter half of March. The latter bottle would begin 

 to experience the acceleration caused by the narrowing of the English 

 Channel almost immediately after the commencement of the westerly 

 gales in mid-March, as Beachy Head marks the %vestern boundary 

 of the funnel-shaped extremity of the Channel ; but the Calais bottles 

 at this time (March 16th) were 50 miles to the westward, off the east 

 coast of the Isle of Wight, in one of the widest parts of the Channel. 

 During the third week of March, therefore, although the direction 

 of drift of the Calais bottles would be deflected to a course parallel with 

 the coast, their velocity would scarcely differ from that directly due to 

 the pressure of the wind, which we have already seen was calculated to 

 be 120 miles for the fortnight, or 60 miles for the week, which would 

 bring them slightly to the eastward of Beachy Head. From this point 

 they would begin to experience an accelerated velocity ; and, if we 

 assume that they travelled at the same average rate as the Eastbourne- 

 Terschelling bottle (18 miles a day), their position at the end of March 

 would be 125 miles to the north-eastward of Eastbourne, or somewhere 

 in a line between Harwich and the Hook of Holland, and probably 

 on the westward side of the middle of this line. Such a position would 

 be about 60 miles N.i^.E. or N. by E. of Calais. Now the estimated 

 direction of drift in this region from April 1st to 30th, based on tlie 

 North Foreland winds, was W. i S., 29 miles, which would bring the 

 bottles to a position slightly south of Harwich, off the Naze, by 

 the end of that month. The Naze is rather under 60 miles N.N.W. 

 of Calais. The estimated drift from May 1st to 13th, based on N. 

 Foreland winds, was approximately E.S.E. (actually S.E. by E. |^ E.), 56 

 miles ; or, based on the winds of Cape Gris Nez, S.E. | S., 59 miles ; so 

 that, obeying this drift, the bottles on May 13th would be within a mile 

 or two of the French coast between Calais and Dunkerc^ue, a result 

 which coincides remarkably with the actual position of the bottles 

 about that time, and demonstrates the general accuracy of the method 

 employed in this report for deducing the course of drift from the 

 direction and pressure of the winds. 



Apart from the employment of this quantitative method, a mere 

 survey of the records contained in the table showing the recovery 



V 2 



