ARRAN. GEOLOGY. SCHIST. 359 



are however undefinable ; as this valley is deeply covered 

 with alluvial matter so as to prevent any accurate investi- 

 gation. 



In the preceding topographic account, it has been 

 unavoidably necessary to consider all the schistose 

 rocks of this island under one general term ; the 

 confusion attending them, no less than their geogra- 

 phical discontinuity of position, and the difficulty of 

 obtaining a free access, from the incumbrance of the 

 investing soil, preventing all attempts to arrange them 

 in that order which has on other occasions been adopted. 

 That confusion and irregularity are however in them- 

 selves interesting, since they bespeak circumstances 

 in the constitution of Arran which are of the first 

 importance in its general history ; the consideration of 

 which will be advantageously deferred to the con- 

 clusion of this subject. 



Notwithstanding this intermixture of the various schists, 

 it may be inferred, partly from observations on this island 

 itself, partly from the analogy of the neighbouring island 

 Bute, and of the corresponding parts of the mainland, that 

 there are here portions of two different series of primary 

 strata. One of these is the micaceous schist which 

 forms the adjoining districts of Cantyre and Cowal; 

 the other is the argillaceous schist which follows that 

 rock in the order of superposition and is immediately 

 succeeded by the secondary strata. This series will 

 hereafter be shown to occupy a very extensive tract in 

 Scotland. 



As the general topography of the whole has been 

 laid down in the preceding sketch, it would also be 

 desirable to trace that of each of these series ; but 

 this is impossible, for the reasons already assigned, 

 and from the following causes. Even where those 

 two sets of strata occur in a linear and undisturbed 

 direction, as they do in Bute and nearly throughout the 

 whole of the mainland, their relative limits are unde- 



