TRANSACTIONS OF SECTION C. o09 
due to the deep cutting of its valley by the Sheaf, and its resulting predominant 
power in capturing consequent streams north and south. 
The northward bend of the Don, after its entrance into the central plain, is due 
to river-capture by a feeder of the Aire. The course of the old Don river from 
Thorne, along the north side of Hatfield Chase to Adlingfleet, on the Trent, is 
clearly traceable, and may have been one of the previous channels of the whole 
river. 
4, On the Elephant Trench at Dewlish, Dorset: Was it a Pitfall? 
By Rev. O. Fisner, JA. 
The author refers to papers upon the trench at Dewlish, by the late Mr. Mansel- 
Pleydell,! Mr. Clement Reid,’ and himeelf.* After describing the course of the 
excavations made by Mr. Pleydell in 1889, he points out the apparent impossi- 
bility of accounting for the formation of such a trench by natural agency, and for 
its containing the remains of so many elephants. He then refers to the present 
practice of taking elephants in pitfalls by the natives of Africa, and suggests that 
the trench at Dewlish was artificially dug for a similar purpose. 
The species of eiephant found here being Z. meridionalis, the author refers to 
flints, considered to have been worked, being procured from the ‘forest bed’ of 
Cromer, where they are associated with this early species of elephant, and alludes 
briefly to some of the geological questions involved. 
5. Wotes on the Glaciation of Holyhead Mountain. 
By Epvwarp GREENLY. 
The glaciation of Holyhead Mountain is of considerable interest on account of 
its position far out in the Irish Sea Basin. Its northern and eastern slopes are 
strongly rubbed and rounded, and striated in a general N.E.-S.W. direction, 
with local deflections. Striz cross the very summit, 721 ft. above the sea, running 
8. 40° W. This direction agrees with that of the general glaciation of Anglesey. 
A strong feature traverses the mountain from N. to 8. facing W., and forms 
also the line of great sea cliffs near the North Stack. The edge of the crag is 
polished, and, in spite of being sheltered by a steep rocky brow rising some 50 or 
100 ft. behind it to the east, is traversed by striz from N.E. to S.W. Examples 
are also given of undercut furrows, and of glaciation of overhanging surfaces. 
The mountain is very bare of drift, but a little till occurs in hollows. At the 
summit are abundant fragments of the green mica schists of the neighbourhood of 
the town, which do not occur 7m sttw at a higher level than 200 or 220 ft. These 
have, therefore, been raised some 500 ft. above their source. 
The phenomena, when all are considered, appear to the author to be ascrib- 
able, with probability, to land ice. 
Finally, there are some ill-defined mounds, composed of local débris, which 
appear to be moraines, as if small local glaciers had gathered here for some time, 
in spite of the moderate elevation. 
8 Report on the Brratic Blocks of the British sles —See Reports, p. 237 
' broc. Lorset Nut. Hist. and Antiy. Field Club, 1889, vil. ; 1893, xiv. 
* Geol. Survey : Mem. on the Country round Dorchester, 1899. 
* Q.S.G@US., 1888, xliv. 818, 
