206 F. R. Coioper Reed — On the CJieirurkke. 



to accept the notion of D. Hummel and P. W. Strand that they 

 may have resulted from the action of subglacial streams, and 

 apparently also favours that of Dr, Hoist, who assigns them to 

 streams flowing over the surface of the ice. Now in regard to 

 these theories, it seems to be forgotten by every glacial geologist 

 that a subglacial river or a river flowing on the surface of a glacier 

 differs from other rivers merely in that it flows under a long tunnel 

 or archway of ice, or over a bed of ice instead of a bed of sand or 

 gravel. In every other respect it is a river, and every difficulty 

 which has been already pointed out in regard to the explanation of 

 the asar by river action of any kind is as potent and conclusive 

 against these postulated glacial or glacier rivers as it is against 

 ordinary rivers. In addition they present special difficulties of their 

 own. Let us first look at the theory of Hoist. It is quite true 

 that when the sun beats upon the back of a glacier small streams 

 are sometimes seen on its melting surface, which run for a few 

 yards and then disappear down a crack or a crevasse. Nowhere, 

 not even on the vast ice-plains of Greenland, do these small 

 streams now grow into rivers ; and in order to do so we must 

 suppose that the ice was marked by no cracks or crevasses, and 

 in the particular case of Scandinavia that in a singularly broken 

 and uneven country an ice-mantle could exist without any crevasses 

 or cracks draining its surface. But suppose it could, whence could 

 it derive the materials for making gravel, or the great boulders, when 

 the whole country was, ex hypothesi, blanketed with ice, and no exposed 

 rocks were visible? And having got hold of rocky debris, how were 

 these supraglacial streams- to roll the millions of great boulders of 

 granite, gneiss, and basalt, which form so large a part of the asar, 

 into their rounded and water- worn shapes, and accumulate them in 

 dykes and embankments a hundred miles long and a hundred yards 

 high ? and how is it that the rest of the glacier's back or some 

 part of it was not uniformly strewn with angular and unrolled, or 

 with rolled debris, which should have remained when the glacier 

 melted alongside of the asar? Assuredlj^ the whole idea is 

 incredible, and it is incredible how sober, thoughtful men in our 

 century should have tried to impose it upon science. 

 {To be continued in our next JSTumher.) 



III. — Notes on the Affinities of the Genera of the 



CHEIRURIDiE. 



By F. E. CowFER Reed, M.A., F.G.S. 



IN a former number^ of this Magazine the evolution of the sub- 

 genera of the single genus Cheirurus has been discussed, and 

 it is now proposed to examine the mutual relations of the other 

 genera of the Cheiruridje. Some diversity of opinion has existed 

 as to the genera which may be grouped together to form this 

 family. Barrande^ put only the following five genera into it : 

 Cheirurus, SpJicsrexochus, Flacoparia, StaurocepJialus, and Deiphon. 



1 Reed, Geol. Mag., Dec. IV, Vol. Ill (1896), pp. 117 and 161. 



2 Barrande, Syst. Sil. BoH., toI. i (1852), pp. 336 and 766. 



