432 NATURAL SCIENCE. Dec. 



original find on Moel Tryfaen, On the same shore, near to where I live, 

 my sons and I have found no less than 70 species, though, probably, 

 not more than 20 live in the immediate locality. On gravelly beaches 

 the fragmentary shells far outnumber those that are unbroken. The 

 glacial she!ly fragments are also generally much rounded and watej"- 

 worn. If we add to the ordinary effect of wave action that of shore- 

 ice, which would prevent, except under very peculiar circumstances, 

 the preservation of any deposit undisturbed, the mystery with which 

 these glacial shell-beds has been surrounded is largely dissipated. 

 Of course, it would, in some respects, be more satisfactory to meet 

 with the shells as perfect as in a museum, but in that case I make no 

 doubt that their glacial origin would not be admitted. The explana- 

 tion of the fragmentary condition of the shells also explains why 

 barnacles are not found on the stones. The mixture of northern and 

 southern species takes place at the present day at Cape Cod. 



4. — The Rocks of the Drift are never found north of their origin. 



This is another sweeping statement which wants more proof 

 than that vouchsafed us. Very little attention has been paid by 

 observers to the point, though, doubtless, the great majority of the 

 stones are from positions north of where they are found. The tracking 

 of erratics to their origin is a very difficult and laborious process 

 in most cases, as I know from personal experience. It is only where 

 the masses of rock yielding them are in considerable force, and their 

 character marked like Shap and Eskdale granite, that this can be 

 satisfactorily done. It can be done, in other cases, by a system of 

 tracking and exclusion. To differentiate Dalbeattie from Criffel 

 granite when found in the Drift, as has been attempted, is well-nigh 

 impossible. Eskdale granite is to be found in the Drift between 

 Ravenglass and St. Bees, north of its origin. The rocks yielding 

 erratics in the greatest abundance must have been the sites of 

 glaciers, and these were principally in the Lake District and in 

 Wales. How far north of their origin Welsh rocks are to be found 

 cannot be known, because the sea occupies the space to the north. 

 Charnwood Forest rocks are found in the Boulder Clay of Nottingham, 

 and therefore must have travelled 15 miles northward {Q.J.G.S., 

 vol. xlii., p. 480). Other examples could be quoted of varying value. A 

 great deal of gypsum is found in the Drift by the Estuarj^ of the Dee, and 

 this, there is every probability, came from further south, in Cheshire. 

 There is, however, an absence of characteristic rocks south of Lanca- 

 shire and Cheshire by which to trace the flow. In the high-level 

 shelly drift the rocks, though from all levels, mostly come from 

 localities where they may have been derived from high levels. 



In my opinion, the distribution of the erratics is quite explicable 

 by the prevalence of north-westerly winds, and by the tides. It is 

 not probable, with a tidal current running through the " Severn 



