338 Supposed Connection of 



the other hand, rocks of which the mineralogical and geo- 

 logical characters are dissimilar, may furnish a soil essentially 

 the same. Thus, the following branches of the primitive 

 series, granite, gneiss, mica slate, and clay slate, being all 

 composed principally of silex and alumine, usually give rise, 

 when converted into soil, to a sandy clay ; while, with respect 

 to the secondary rocks, it may be said of many of them that 

 they run into one another by insensible gradations, and that 

 each will produce a soil similar to that from any one of the 

 rest. There are, indeed, various obvious facts which confirm 

 the foregoing remarks. Every one knows that, in an exten- 

 sive tract composed of a single rock, the soil is very far from 

 being uniform, but may consist of not a few kinds. On the 

 other hand, it is almost equally well known that the leading 

 soils are much the same every where, however different the 

 rocks may be. It is even true that the soil of secondary tracts 

 is frequently not dissimilar to that met with in primitive dis- 

 tricts ; nor is this much to be wondered at, when we consider 

 that, in point of the principal chemical or mineral ingredients, 

 the one series of rocks is by no means very different from 

 the other. 



It thus seems, a priori, highly improbable that the native 

 plants are regulated by the rocks ; as it appears that, gene- 

 rally speaking, the roots of vegetables are not in contact with 

 the rocks ; and that these frequently do not furnish the soil 

 which lies over them. It is also true, that, frequently, the 

 varied forms of a particular rock differ from one another with 

 regard to the circumstances most likely to influence vege- 

 tables, more than some rocks do from others which are of a 

 different species. 



It is, however, only by facts that the question can be set- 

 tled, though' the preceding observations may not be unworthy 

 of notice. As the best means, within my reach, of putting 

 the matter at issue to the test of experience, I propose to 

 compare, in some measure, Aberdeenshire with the tract 

 around Edinburgh, and with Devonshire. It is not in my 

 power to offer a minute account of the rocks of these dif- 

 ferent regions ; but, indeed, for the present purpose, nothing 

 is necessary beyond a short notice of their solid surfaces. 



Devonshire appears to be composed of granite and slates, 

 together with sandstone and limestone. The country around 

 Edinburgh is a secondary tract, chiefly consisting of trap, 

 sandstone, and mountain limestone. In Aberdeenshire, the 

 surface is almost entirely composed of granite or gneiss ; but 

 I cannot form any estimate of the relative quantities of these 

 rocks, as it is not always easy to trace the distinction between 



