g6 DR. J. A. SMYTHE ON 



height for a considerable distance. This gives place to a 

 winding ridge at the south end. The material of this ridge is, 

 in places, rather morainic than gravelly. 



The flat terraced deposits of sands and gravels occur most 

 extensively in the larger drift-filled valleys, though not con- 

 fined exclusively to these. They are well developed in the 

 valleys of the Till and Aln, the Coquet below Felton, where 

 they cover a triangular patch of about five square miles 

 flanked on the east by the Crowden Hill Ridge (v. post., p. 97); 

 in the Font at Netherwitton and at the junction of the Font 

 and Wansbeck. Sections of the last near Mitford show these 

 deposits resting on a great thickness (over 80 feet) of boulder 

 clay, and composed of lenticular patches of sand, pebbles, 

 rounded and subangular, boulders with striae in all stages of 

 demolition, and thin bands of clay containing scratched 

 boulders. Traced southwards, from the exposures on the Font 

 to those on the Wansbeck, these beds pass into a clean, 

 compact, cemented conglomerate (Photograph 2.). At Nether- 

 witton the size of the terraces is out of all proportion to that 

 of the modern haughs. Near Rothbury, as noted by Topley, 

 the contents of these deposits are chiefly sedimentary rocks, 

 in marked contrast to the modern porphyrite-bearing haughs 

 of the Coquet. 



These facts point to rapid erosion of glacial (and other) 

 materials and deposition in torrential waters at the close of 

 the ice-age, when the modern streams, greatly swollen by 

 melting ice and the bursting of ice-dams, were coming into 

 being and their valleys were yet uncut. When foreign ice 

 held the catchment, as in the case of the Coquet, the com- 

 position of the detrital matter differed from that transported 

 after the ice had melted and new species of rock had become 

 exposed within the drainage area. 



Though these various deposits have been described 

 separately, there is much evidence to show that they are 

 connected in part and belong to the same order of phenomena. 

 This has been emphasised by Hugh Miller [6] who noted the 



