496 J. G. Goodchild — On the Origin of Coums. 



so as to leave a bay similar in form to that of the eddy above. 

 Many coums occur in North-Western England whose origin is 

 capable of explanation in the most complete and satisfactory manner 

 by supposing that the greater ice-streams behaved to their tributaries 

 in the same way that a strongly-flowing river does to the weaker 

 tributaries that it is joined by. For instance, the ice of some of the 

 small valleys coming down from the Cross Fell Escarpment into the 

 Eden Valley, must, on meeting with the powerful stream that swept 

 up the valley from the south-west of Scotland over Stainmoor to the 

 North Sea, have been affected by a kind of tangential action, so that 

 much of the tributary ice at high levels was compelled to move 

 backwards in a rudely circular direction until it merged into the 

 direction of the main stream. As a consequence, we find that on 

 nearly every one of these tributaries some of the higher lying scars, 

 where the currents were freest to move, are ground into crater-like 

 recesses that, when viewed from a distance, remind one of nothing 

 so much as of gigantic pot-holes that have had one side cut away. 



Where a stream flows over a rock that is much fissured, so that a 

 great diversity of directions is imparted to the lower currents of the 

 stream, we find that the whirling motion thus produced has caused 

 the rock to be drilled into numerous pot-holes, which vary in shape, 

 position, and size, with the rapidity of the stream, the nature of the 

 rock, and the direction of the larger fissures. The resemblance be- 

 tween the valleys and coums of a mountainous district, as they are 

 seen on a good hill-shaded map, to the wider joint fissures and pot- 

 holes seen in a river's bed, strongly inclines one to the opinion that 

 the analogy between them is complete in nearly every respect but 

 that of size. In the one ease, the pot-holes in the river's bed are due 

 to the eddies caused by the variously directed minor currents result- 

 ing from the inequalities of the river's bed ; in the other, the coums 

 seem to be due to the much larger eddies caused by the variously 

 directed valley streams in the great river, or rather sea, of ice that 

 once overspread perhaps the tops of our highest mountains in the 

 northern parts of our islands. In the one case, the pot-holes have 

 been due to rapid and comparatively intense action in a brief space 

 of time ; in the other, their larger analogues, the coums, have received 

 their form through slowly, and perhaps feebly, acting causes con- 

 tinuing throughout a period of time so long that we can form no 

 conception of its immensity. 



It will therefore be seen that it is here considered that, as the 

 orgin of coums — the so-called Giant's Cauldrons of Sweden — the 

 " round and deep holes with polished sides " that occur on the sides 

 of the Swiss valleys in situations "where the form of the surface 

 will not permit us to suppose that any cascade could ever have 

 existed," 1 — and the existence of similar pits in the rocky bed of 

 lake basins — all admit of the most complete and satisfactory 

 explanation on the assumption that the analogy between the 

 behaviour of a glacier and that of a river, which is known to 

 be almost complete in the puny representatives of the old ice-sheet, 

 1 Lyell, Elements, 5th edition, p. 149. 



