350 GLACIAL EROSION IN NORTH WALES. [Aug. I909, 



was not convincing : not because the process of river-capture was 

 unreal or ineffective, but because the various consequences that 

 must accompany such a capture were not clearly deduced and 

 compared with the appropriate facts. Until such deduction and 

 comparison was made the explanation remained uncertain. The 

 difficulty that Prof. Garwood found in accounting for small branch- 

 valleys that hang over hanging lateral valleys was not a difficulty 

 in the theory of glacial erosion ; it was, indeed, an expectable 

 and essential consequence of that theory, for the relation of the 

 branch-valleys to a lateral valley was the same as the relation of the 

 lateral valleys to their main valley. It was to be noted that, in 

 Prof. Garwood's favourable consideration of the protective action 

 of glaciers as a means of accounting for hanging lateral valleys, 

 he had taken no account of the various difficulties which stood 

 in the way of accepting that explanation. Until these difficulties 

 were considered and set aside, the Author found it difficult to 

 accept the theory that glaciers were protective agents. The 

 explanation of cirques by protective ice and snow involved 

 assumptions as to the action of weathering in the steepening of 

 the cirque-head cliffs : and these assumptions had not been 

 sufficiently examined and justified. 



