THE SCHISTOSE ISLES. 30? 



rule may perhaps be applied to many cases where similarly 

 enormous dimensions of the original strata have been 

 imagined to exist. 



This hypothesis may perhaps also in some measure 

 remove a difficulty arising from the present frequent alter- 

 nation of two or more substances, those for example of 

 clay slate and micaceous schist ; since, if the third stratum 

 for example, represented in the diagram as in its original 

 undisturbed state, be called clay slate, and the lowest 

 micaceous schist, it is evident that the sequence, which 

 is originally only the regular superposition of one rock 

 above another, will under the assumed disturbance become 

 a succession of alternations. It is nevertheless evident, 

 that this diminishes the difficulty alluded to without remov- 

 ing it; since it cannot explain the frequent intermixture 

 of these rocks in Lunga and Scarba, or in the other 

 islands ; so that the original argument deduced from those 

 alternations remains in full force. One remark of a general 

 nature still presents itself, and that is, the constancy and 

 regularity of the direction of the strata, when such evident 

 marks of disturbance are found in the dip. It will readily 

 appear, that although we are unable to account for this 

 difference, it is connected with a much more extensive 

 train of phenomena which are indicated in the positions 

 assumed by chains of mountains and the shores of conti- 

 nents in many parts of the world. 



In examining the separations in the chain of the Quartz 

 isles, a perfect correspondence of the opposite sides of 

 the straits may be observed, whether in that of Lunga, 

 in the Coryvrechan, or in the much longer channel that 

 separates Jura from Isla.* No alteration in the elevations 

 or positions of the corresponding strata can be perceived ; 

 the disjunctions appearing to have been produced by 

 causes capable of cutting through the rocks without alter- 

 ing their positions, and the geological continuity remaining 



* Plate XXIII. fig. 3. 



