182 Proceedings of the Royal Physical Society. 



part, 18 inches imbedded, was surrounded by motionless till. 

 And if this reasoning holds good in general, the thickness of 

 the fluxion-layer cannot have exceeded a few inches. This 

 line of inference, however, is a very insecure one, depending 

 primarily upon an unascertained fact, viz., the direct proximity 

 of the glaciating agent to the upper surface of the boulder. I 

 have frequently seen little bits of shale, not more than ^V^h 

 of an inch deep, of which the upper surface was well rubbed, 

 while the lower surface was almost untouched. The sole of the 

 loidders, as Eobert Chambers used to term their flatter side, 

 seems generally, in fact, to have travelled uppermost. And 

 it may be remarked that the fact noticed by Dr James 

 Geikie, that there is " some curious connection " between the 

 size of the stone and the coarseness of its striae, seems to 

 imply that every boulder was liable to be striated by just 

 such materials as were more mobile than itself. But, quitting 

 these obscurities, it seems at least quite certain that in sections 

 such as that of Fig. 6, where different boulder-clays overlap, 

 the movement extended no considerable distance downward. 

 I am aware that it is usual to suppose the pressure and 

 motion of the ice to have been transmitted through thick 

 masses of drift and down to the surface of the rock below, so 

 that the till, like a rolling snowball, licked up its additions 

 from beneath. Mr Tiddiman, for instance, takes this view of 

 the drift of the Lancashire uplands, and represents the local 

 drift as actually forming beneath an older and further-derived 

 drift superposed. It appears more probable, however, that 

 the local drift was manufactured whilst the foreign import 

 was still on the way. This, at least, is the only intelligible 

 explanation of either the structure or the order of the drifts 

 of which I have any practical knowledge. It seems too much 

 to assume that " the whole mass of debris " was in motion at 

 once. 



Distribution of the Fluxion- Structures. 



Fluxion-structure, so far as I am aware, is generally char- 

 acteristic of well-kneaded till upon tolerably open ground. 

 My first attempts to discover it; about seven years ago, met 



