143 



gravel, which itself would eventually contribute to make up some 

 or other of the various forms of drift. 



Under the action of the various agents I have briefly noticed, 

 an infinite variety of combinations of mud, clay, sand, gravel, and 

 boulders of all kinds would be accumulated. At one place the 

 materials might be deposited, from top to bottom, in the form of 

 a stiff paste of clay stuck full of stones of all sizes, dropped, or 

 melted, into the clay without the slightest regard to either their 

 shape or their size. At another point, glacial streams resulting 

 from the melting of the ice might deposit sand, gravel, or water- 

 worn material, in any proportion, and these deposits were just as 

 likely to be left in a stratified form as in any other. Far away 

 from the margins of the hydrographical basins, at the points where 

 much ice-water was flowing seawards, the whole of the material 

 liberated by the ice might be expected to be more or less water- 

 worn in character. And here it occurs to me to make a remark 

 suggested years ago by a criticism on my views on this subject put 

 forward in the Geological Magazine by my colleague the late Mr. 

 E. T. Hardman. I have already pointed out that the character of 

 the erratics in a glacial deposit at any given spot was determined 

 by a very complicated set of causes. The results may be briefly 

 summarized as the fiett results of all the transportals the several 

 erratics have undergone since being detached from the parent 

 rock. Now when the Ice Sheet melted, the subglacial streams, 

 being influenced in their directions by an entirely-different set of 

 causes from those affecting the movements of the ice, must neces- 

 sarily have transported water-worn material in different directions 

 from what those materials had ever taken when they were in the 

 ice. And so it must often have happened that the water-worn 

 detritus characteristic of the drift of one area has been rolled to a 

 distance, and mingled there with drift totally different as regards 

 the parent sources of their constituent erratics. 



The explanation above given of the origin of the drift mounds 

 or drumlins, will serve also to account for the fact that the large 

 axes of these drumlins lie approximately in the same direction as 

 those of the preponderating directions of movement of the ice 



