224 REPORT ON THE SURFACE DRIFT OF THE ENGLISH CHANNEL 



since it yielded an estimated drift of only 3 miles in a direction W. 24° 

 N. It is obvious that in this case the rapidity of the drift round the 

 Cornish headlands was due to some cause other than the winds, and, as 

 already described in an earlier section of this report, the known peculiar- 

 ities of the tidal streams in this region provide an adequate explanation. 

 The indraught into the Bristol Channel during the flood-tide here causes 

 a marked set to the northward round the Land's End, and this brings 

 about a predominance of the westward tide over the eastward tide in 

 this part of the English Channel. In fact, the English Channel supplies 

 more water to the Bristol Channel during flood-tide than returns to it on 

 the ebb from the same region. 



The last case of all (XL. 1) appears to illustrate the same point. 

 There is an angular error of 45° in my calculated drift, but the estimated 

 distance is approximately correct. The deflecting cause may be regarded 

 as equivalent to a current in a N.N.W. direction having an average 

 velocity of 36 miles in 16 days, i.e., rather over 2 miles per diem. 

 There seems to be no reason for doubting that this deflection was due to 

 the influence of the resultant northward tendency of the tides in the 

 western region of the Channel, which is determined by the indraught 

 into the Bristol Channel during flood-tide in that region. We have seen 

 in the ])receding case (XXX. 1) that the force of this indraught deter- 

 mines a current round the Lizard and Land's End of some 30 miles in 6 

 days, i.e., 5 miles a day ; so that, although this deflecting influence was 

 not fully felt during the whole period of drift in the present case, we 

 are probably correct in attributing the observed deviation to the same 

 cause, since the south-westward drift caused by the winds would bring 

 the bottles nearer to the influence of the indraught on each successive 

 day; and, in order to complete the distance between the eventual 

 estimated and actual positions, it would suffice if the full influence of 

 the indraught (5 miles per diem) were only felt during the last seven 

 days of drift. 



§ 7. Conclusions. — Enough has been said, I think, to show that the 

 method employed here for tracing the actual influence of the winds 

 on the water is sufliciently accurate for practical purposes, and 

 that by its employment, with proper precautions, the influence of 

 the winds may be separated from that of other factors which 

 operate in the production of surface currents. From this point 

 of view the method may be of considerable use in the future for 

 determining the existence of currents not produced by local wind- 

 action. At the same time the method requires to be tested extensively 

 before it can be used as a basis for conclusions. The present report 

 pretends only to show that the relation between wind -action and 

 surface currents is capable of quantitative study, and that the results 



