280 E. C. ANDREWS. 



to retreat along its old course. If the tributary glacier be 

 small and not constricted along its lower course, while the 

 main glacier is very large and suffers marked constriction 

 at this locality, then undercutting and valley over deepen- 

 ing on the cutting curve will be marked and the retreat 

 upstream of the mouth of the tributary will be correspond- 

 ingly marked. In other words, all other conditions being 

 equal, the weaker the tributary stream action is in such 

 areas the more pronounced will be the hanging valleys. 

 This has been well shown by Davis. 



The following indicates the conditions necessary for the 

 maximum development of Hanging Valleys. 



If the valley constriction be symmetrically arranged 

 to a central axis, then Hanging Valley phenomena may be 

 expected to be fairly equally developed along both sides of 

 the valley. The development of Hanging Valleys shows 

 the stage of land dissection arrived at in such localities 

 during pre-glacial time. For it is evident that valleys 

 hung high up can exist only in perfection when profound 

 channel depths, fairly high channel slopes, resistant wall 

 structures, valley constrictions well marked, and immense 

 stream volume all conspire to work together. In other 

 words such geographical forms cause great local increase 

 of stream velocity, and this involves a still greater local 

 variation in the corrasive strength of streams. This results 

 in marked interruptions of the channel bases, two expres- 

 sions of which are the deep basins below base level and the 

 basined or non-basined "tread" under a heavy declivity, 

 and the immediate corollary of which is the hanging nature 

 of minor tributary valleys. 



But this condition of things could only arise during the 

 early dissection of a high block of land, because a moment's 

 consideration will show that a stage such as the very late 

 maturity of dissection would afford only moderate oppor- 



