222 MK. E. J. GAEWOOD & DK. J. W. GEEGOEY [May 1 8985. 



material. Our observations certainly support the views of those 

 who — like Mr. Goodchild, Mr. Upham, and Prof. Crosby — have 

 attributed great importance to it. The distinction between supra- 

 glacial and intraglacial material is no doubt arbitrary, for debris 

 may easily pass from one class to the other : the lowering of a 

 glacier's surface by ablation must increase the supraglacial, at the 

 expense of the intraglacial material. 



The separation of intraglacial from subglacial material appears to 

 us far less definite ; and the six criteria suggested by Mr. ITpham ^ 

 are inapplicable in Spitsbergen. As we have previously remarked 

 (p. 203), there is a very gradual passage from intraglacial to sub- 

 glacial moraine-matter, although a sharp separation between them 

 is often assumed, as, for example, by Upham.^ The melting of an 

 ice-sheet probably takes place on the lower as well as on the upper 

 surface, and this may lead to some intraglacial material becoming 

 subglacial. We agree, therefore, with Crosby ^ when he argues that 

 there is ' no definite distinction between subglacial and englacial 

 till, because, broadly speaking, it has all been englacial.' 



We have especially quoted Mr. Upham as the latest champion of 

 the importance of intraglacial material ; but we have not forgotten 

 that this view was upheld in this country many years previously by 

 Mr. J. G. Goodchild * in his remarkable paper on ' The Glacial Phe- 

 nomena of the Eden Yalley and the Western part of the Yorkshire- 

 Dale District.' We were constantly reminded during our study of 

 the Spitsbergen glaciers of his protest against regarding Glacial 

 Drifts as a ground-moraine, and of his explanation of their deposition 

 by the quiet melting-away of an ice-sheet charged with rock- 

 fragments. This paper was written 23 years ago ; but the views 

 expressed therein as to the transport and deposition of glacial 

 beds agree more nearly with the glacial phenomena seen in Spits- 

 bergen than those of any other contribution that we know to British 

 glacial geology. 



(7) Glacial Gravel-Mils. 



As in the case of recent observers in Greenland,' we saw no sign? 

 of the formation of eskers in either supraglacial or subglacial 

 channels, or by streams flowing through ice-canons. The only 

 eskers that we found must have been formed by torrential action at 



^ W. Upham, ' Criteria of Englacial & Subglacial Drift,' Amer. GeoL- 

 vol. viii (1891) p. 377. 



2 Id., ' Englacial Drift,' Amer. Geol. vol. xii (1893) p. 38. 



3 W. O. Crosby, ' Englacial Drift,' Amer. Geol. vol. xvii (1896) p. 222. 



* Quart. Journ. Geol. Soc. vol. xxxi (1875) pp. 55-99, pi. ii; and 'On Drift,'' 

 Geol. Mag. 1874, pp. 496-510. The former paper is reprinted, with additions 

 and explanations, in Ti'ans. Oumb. & Westmorl. Assoc, no. xii (1887)pp. 111-167. 



° For example, R. D. Salisbury, ' Salient Points concerning the Glacial 

 Geology of North Greenland,' Journ. Geol. vol. iv (1896) p. 809. An esker was 

 described by Kornerup at Arsalik near Holstenborg, ' Geologiske lagttagelser 

 fra Vestkysten af Gronland (66° 55'-68° 15' N. Br.),' Meddel. om Gronl. vol. ii 

 (1881) pp. 192-193, 245. 



