Part III — Fifteenth Annual Report 



359 



time from September 1894 to December 1896, the prevailing winds on 

 the east coast of Scotland blew from between the two points W. and W. by 

 S. If the direction of the prevailing winds at each of the three Scotch 

 stations be marked on a chart (see pi. x.), it will be found that at 

 North Unst, in both years, it points towards the north of Norway 

 parallel to the coast ; at Buchan Ness, Aberdeenshire, it points in both 

 years to the southern part of the west coast of Norway; in 1895 to the 

 vicinity of Skudesnaes ; in 1896 to the neighbourhood of the Sogne 

 Fjord (with Bergen midway between) ; at the Bell Rock it points in 

 both years to the coast of Jutland, in the vicinity of the Limfjord. 

 And Dr Buchan informs me that the prevailing wind in Scotland, as a 

 whole, is, as a general rule, W.S.W. 



It is, I think, evident from these facts that the movement of the 

 floats on the east coast of Scotland is not directly and immediately due 

 to the action of the prevailing winds ; it is also clear that the drifters are 

 not, as has been sometimes contended, driven by the wind independently 

 of the movement of the water in which they are suspended. If they, or 

 the surface water, were merely driven before the wind from the east coast 

 of Scotland, the general course would be north-east and easterly, whereas 

 we have seen that the almost invariable coarse is southerly, at right angles 

 to the direction of the prevailing wind. That is, I think, an important 

 point to establish. 



The resultant wind at the English and Continental stations does not 

 show the same agreement in the two years, a circumstance probably due 

 to the fact tint the calculation was based only on the 8 a.m. observations. 

 At Shields, the prevailing wind, on the basis stated, blew, in 1895, from 

 N.N.W. § N., and, in 1896, from S.W. by W. \ W. ; at Yarmouth, in 

 1895, it blew from E. § S., and, in 1896, from N. h W. ; at Helder, in 



1895, it blew from W. \ N., and, in 1896, from W.N.W. \ W. ; at the 

 Scaw it blew, in 1895, from N., and, in 1896, from W.S.W. \ W. The 

 variation is therefore considerable. By grouping together the three 

 southern stations, Shields, Yarmouth, and Helder, the resultant for 1895 

 is N.W. by W. \ W. (3-3), and for 1896 N.W. by W. \ W. (50'6), 

 practically the same direction, but much stronger ; and the four months, 

 September to December 1894, give for the group N.W. \ W. 



If the two English stations, Shields and Yarmouth, be taken together, 

 the combined resultants are: — In 1895, N.E. by N. .1 E. (3'3), and, in 



1896, N.W. \ W. (5*6). The two Continental stations, Helder and the 

 Scaw combined, give for 1895 W. by N. \ W. (Ill), and for 1896 W. 

 by N. \ W. (139). In the four months of 1894, September to Decem- 

 ber, the resultant for the two English stations is N. \ E. (30), and for 

 the two Continental stations W. \ S. (48). We thus see that while the 

 yearly resultant at the different stations — each founded on a possible 

 maximum of 365 observations — varies somewhat in the two years, this 

 variation is reduced when the stations are combined, and the number of 

 observations thereby doubled or trebled. It must be remembered that 

 the observations exclude light airs and gentle breezes ; and an examina- 

 tion of the daily weather reports shows that when a fresh or strong breeze 

 is blowing it usually blows from about the same quarter at the stations 

 in the same areas. While the combination of the observations at two 

 stations reduces the variation shown in the two years, it also approxi- 

 mates the resultants to those obtained from the more extensive observa- 

 tions at the Scottish stations. Thus, while in each of the years the 

 prevailing wind is westerly, the southerly element is also well marked 

 at the most northern station (North Unst, Shetlaud) and at Buchan Ness, 

 Aberdeenshire, and a northerly element comes in at the Bell Rock. The 



