30 Hugh Miller — Escarpments and Terraces. 



districts, and can then be attacked chiefly by the dissolving action of 

 soft water — action distributed along the lines of jointing to great 

 depths below the outcrop. The sandstones, on the other hand, are 

 exposed to frost not only along their (more irregular) joints, but 

 grain by grain ; while the calcareous and felspathic elements are 

 liable to be reduced and abstracted by chemical action, with the effect 

 of loosening the texture, rendering it porous, and thus giving frost a 

 better grip. 



For these reasons it is open to doubt whether, bed for bed, the 

 Yoredale limestones are much more rapidly removed than the sand- 

 stones. But the chances of a bed making a good escarpment are 

 directly in proportion to its thickness, which in Yorkshire favours 

 the limestones, in the Tynedale area the sandstones. In this I 

 believe lies the explanation. Just as in the stream section, where 

 sandstone and limestone projected or retired, according to their 

 several thicknesses, so I believe it to be in regions where these 

 rocks crop out in a manner favouring escarpment development. 

 And as the system for liuiestones with grits in the south, and sand- 

 stones with limestones in the north, is a dovetailing one, there is 

 probably some intermediate tract of equal development where both 

 are favoured alike.^ 



In nearly all cases shale is inferior in durability to the other two, 

 and where interlined even with thin courses of sandstone or lime- 

 stone, shows a retreating outline, whether under escarpments or in 

 stream-cut sections ; and this not only where reached by spray, but 

 in every shaly slope not smothered in debris. In certain cases, 

 however, it is apparently durable, as, for instance, where it is backed 

 up by harder rock, and permitted to assume slopes or to sink into 

 hollows favouring preservation and encouraging a conservative 

 covering of vegetable growth. In these cases the denudation of 

 the framework of harder beds by atmospheric action may be said to 

 measure that of the shale. 



The difficulties and objections raised by Mr. Goodchild to the dis- 

 integration theory can perhaps best be treated of under two heads. 



1. Those affecting the character and rate of disintegration of rocks. 



2. Those drawn from the local development of the outcrops. 



1. With the view of proving the incompetency of atmospheric dis- 

 integration to cause great results, Mr. Goodchild gives measurements 

 of blind gorges terminating in waterfalls in which the recession of 

 the fall is contrasted with that of the sides of the gorge. It seems but 

 a fallible test, however, to compare the wearing back of the latter 

 throuo-li tlie crumbling waste of rain and frost with that of a waterfall 

 employing all the tools of subaerial denudation. In comparing gorge 

 with gorge, moreover, the mean annual volume of water would re- 

 quire in all cases to be ascertained; while the nature of the rook, 

 varyino- even in the same material, and apt to be infinitely compli- 

 cated by combinations of several kinds of rock, would present a for- 

 midable obstacle to the attainment of trustworthy results. 



1 That this is the case is, I think, indicated in Phillips's Geology of Yorkshire, 

 Section Physical Geography. 



