302 B. Bmits — On the Mechanics of Glaciers, 



I wish it to be understood ttiat I have no fault to find with Mr. 

 Goodchild's theory as such. I have no doubt but that it is the true 

 explanation of the phenomena in question. Indeed, I accept it as 

 such, and regard it as a beautiful proof of the erroneous character of 

 Mr. Croll's theory. It is certainly most creditable to Mr. Good- 

 child's powers of observation that he should have seen in the field 

 the cause of these hollows, even when misled by accepted theories. 

 It is not to be wondered at that logic should have suffered a little to 

 effect a compromise. 



The physical cause of the motion of glaciers is a problem so im- 

 portant and interesting, and is so far from being solved, that any 

 suggestion or theory that pretends, or is likely, to help to a solution, 

 should be welcomed. I have done my best to show what I consider 

 faults in Mr. Croll's theory. I have done so because I consider it 

 has been misleading geologists and lulling physicists to inaction by 

 the notion getting abroad that the whole question was settled. If 

 Mr. Croll had been as dissatisfied with his theory as I have been, 

 and, as I trust, I have shown I have had very good reason for being, 

 I am sure he would have returned to the subject before this to the 

 profit of science. Nor would I have this theory too readily dis- 

 carded. The motion of glaciers may be a very complex afi'air, and 

 this molecular theory may yet be useful to supplement other theories 

 so as to enable them to cover all the phenomena of glaciers. 



To further the investigation as far as I can, I would here raise the 

 consideration of a question I hinted at when speaking of '' cre- 

 vasses." Though surface molecules melt most frequently, it is part 

 of Mr. Croll's theory to suppose that each and every molecule of the 

 glacier melts many times over. Now suppose a molecule to melt 

 somewhere in the interior of the glacier, and suppose that under it 

 there is just room for it to move in a downward direction. The 

 question I wa^t to put is, would it move at all ? We are dealing 

 with very minute particles of matter, at weicy close quarters, and 

 must be careful. What says the doctrine of Capillary Attraction on 

 this point ? As I write among the wilds, and have no books to 

 refer to, I do not wish to commit myself to any direct statement on 

 the subject at present ; but, after observing the behaviour of mole- 

 cules of water among molecules (or crystals rather) of sugar, it will 

 not astonish me, supposing Mr. Croll, on returning to the subject, 

 makes the glaciers go up-hill. 



There is another consideration, which I mention the more cheer- 

 fully as it helps the molecular theory. If Mr. Croll returns to the 

 subject, and still adheres to his theory, I am sure he will adopt my 

 suggestion to regard the heat and molecules as moving in the same 

 direction. Indeed, it may have been his intention to do so, but 

 that he was unlucky in his illustration. Suppose a cubical stone 

 somewhere in a glacier, with faces up-hill and down-hill. Further, 

 suppose heat to be travelling down the glacier, and let us consider 

 its effect on rows of molecules, whose lines pass through the stone 

 normal to these faces. First, consider the molecules below the 

 stone. Those next the stone move down-hill a little, and again 



