16 J. W. Goldthwait — Glacial Cirques 



or lateral moraines. The lobate margin of the deposit, although 

 suggestive of a convex glacier front, is equally well accounted 

 for by the farther fling of avalanches from the headwall than 

 from' the sides, — indeed, it seems significant that on the head- 

 wall a system of joints which dips steeply, straight down the 

 ravine, is widely exposed, as if large masses had been loosened 

 and had fallen to the floor below. The block mass answers so 

 well the description of the San Juan landslides that one is at 

 once inclined to accept the explanation of post-glacial frost work 

 and avalanches for it. There is, however, an objection to this. 

 Some of the blocks are perched upon others, and in such high 

 places that it seems impossible to account for their positions in 

 any other way than by the melting out of solid ice from under- 

 neath them. The snow and ice which accumulate now in 

 these ravines is neither thick enough nor solid enough to act as 

 a toboggan slide for such heavy material, should it descend 

 from the walls. At the time when the ice sheet was melting 

 away from the range, however, avalanches might easily have 

 descended on remnants of solid ice at the cirque heads ; and the 

 supply of loose blocks at that time must have been very plentiful, 

 both because of the plucking action of the ice sheet, which 

 had become well filled with joint blocks near the crags, and 

 because of more intense frost action then than now. Stagnant 

 masses of lingering ice of this kind, since they had no motion 

 of their own, would leave no distinct moraines, merely a 

 labyrinth of blocks ; and that is what appears to be the case. 



The same peculiar block deposits can be seen in Tuckerman 

 Ravine, Huntington Ravine, the Great Gulf and the Ravine of 

 the Castles. In the first mentioned ravine, around Hermit 

 Lake there is a greater development than elsewhere of ridges 

 of gravelly structure and kame-like habit. One of them curves 

 half way around the lake, retaining an esker-like form for sev- 

 eral hundred feet along the trail. Others near by seem to be 

 composed of unwaterworn and unstratified debris. All are 

 more or less irregular, without persistent or systematic trend 

 or position. They are neither transverse nor longitudinal with 

 respect to the valley, and although when viewed from a distance 

 they seem to assume the form of looped moraines, they invar- 

 iably prove on close examination to wander, split and disappear 

 in a most inconsistent way. They are undoubtedly the kind of 

 ridges described by Tarr in the Ktaadn basins. I am unable to 

 see, however, that they are dependent at all on the presence of 

 valley glaciers in these ravines. All that seems necessary to 

 explain them is the presence for a while, in the closing stage of 

 deglaciation, of stagnant ice of the waning ice sheet, in the 

 shelter of the cirque walls, which received avalanche deposits 

 and esker-stream gravels. 



