246 Professor Forbes on the Leading I'/ienomenu of Glaciers. 



maintain that this is precisely what a semifluid body might 

 be expected to do. For the structure near the centre is 

 always imperfectly developed, exactly because there the dif- 

 ferential motion is least ; I mean, that there is least discon- 

 tinuity of parts, because the velocity is nearly the same through- 

 out a considerable space ; and if two glaciers unite, and move 

 tolerably uniformly together, they will preserve, for a long way, 

 the structure which they had already acquired, before the 

 new one (representing a single united stream) is superinduced 

 upon it. Now this is exactly what takes place at the union 

 of the glaciers of Lechaud and Geant,— of the two branches of 

 the Glacier of Talefre, and of the Glacier of La Noire and Le 

 Geant, all of which, originally double in structure, finally be- 

 come single, and cut the separating moraine at an angle. 

 But I appealed here also to experiment, and found, that by 

 pouring double streams of viscid plaster down a single channel, 

 the separate forms were veri/ slorvly worn out indeed, and per- 

 petuated far beyond the point of union of the streams. Thus 

 the proposed objection became a strong confirmation of my 

 theory. One of these models, also shewn to the Royal Society, 

 is represented in the annexed figure. 



FieiK 0/ a model, shewing the effect of th& union of two Mtreams on the motion of a viscid 

 fluid. 



The illustrations now given will, it is hoped, shew that there 

 is a striking conformity between the facts of motion and the 

 facts of structure in a glacier, and that the two, mutually sup- 



