500 GENERAL COMPARISON OF 



things will be rendered more obvious by approximating 

 the parts and restoring the supposed continuity; and 

 to aid this purpose an ideal section of that imaginary 

 state is subjoined.* 



For the further purpose of showing the loss of sub- 

 stance exhibited in various parts of this deposit in the 

 vicinity of the islands, a similar section is also added ; 

 including part of Cantyre, where the marks of waste 

 are the most unquestionable .f In the case of the islands, 

 such losses would naturally be attributed to the action 

 of the sea ; but the analogous facts occurring in places 

 where the sea does not reach, show that other causes 

 must be allowed, to have shared at least, in producing 

 those effects to which they probably owe much of their 

 present forms* 



A geological map of Cantyre would have been requisite 

 to illustrate the very peculiar distribution of the sand- 

 stone in that district, but it is here inadmissible : the gene- 

 ral map which accompanies this work will perhaps convey 

 a sufficient idea of it for the present purpose. It must suf- 

 fice, in addition, to remark, that a continuous tract of this 

 rock extends along the eastern shore of that peninsula, from 

 the harbour of Campbeltown southwards ; and that to the 

 north of that spot on the same shore, a few very minute 

 detached portions are also found. On the western shore 

 some independent masses of larger size occur ; all of 

 them presenting the same characters, both in compo- 

 sition, and in their relations to the micaceous schist on 

 which they repose. Indications of their existence are 

 even found extending across the low tract which inter- 

 venes between Campbeltown and Machrianish bay; and 

 here their nature is further proved from their being 

 followed by a portion of the same sandstone and coal 

 that succeed to the red sandstone in the central tract 

 of Scotland which lies between the Clyde and the Forth. 



In describing the alluvia of Arran, it was remarked that 



* Plate XXXIII. fig. 3. f Plate XXXIII. fig. 2. 



