200 GENERAL REMARKS [Ch. X. 



various binary combinations, more particularly some of those 

 which have been classed under the head of quartz-rock by 

 Macculloch, and which are considered as stratified rocks, 

 must be placed among the granitic, since they not only form 

 an integrant part of the so called unstratified rocks, but, 

 when in the slates, have the same relations thereto as the 

 eurites, syenites, shorl-rock, porphyries, and other members 

 of the granitic group. Farther evidence in favour of this 

 arrangement will be immediately brought forward, under the 

 head of the primary slates, the consideration of which will 

 now occupy our attention. 



Of all the members of the schistose group, gneiss is one of 

 the most important : it is not, as once supposed, more inti- 

 mately connected with granite by mineral transitions than 

 other slates, nor is it necessarily always in sequence to this 

 rock, though it commonly occurs in this position. It is 

 formed of the same crystalline minerals as those which enter 

 into the composition of the granitic rocks; though these 

 commonly assume a different arrangement, being parallel to 

 each other, whence arises the laminated or schistose struc- 

 ture, which is very often the only distinction between gneiss 

 and granite. The component minerals of gneiss, like those 

 of the granitic group, are united together in every possible 

 proportion, and in granules of various sizes ; and the resem- 

 blance of constitution is still greater, for each district of 

 gneiss, and even different parts of the same district, possess a 

 characteristic mineral; and this, by insensible gradations, 

 passes into others : thus, hornblende, mica, and talc in Scot- 

 land and Norway, mutually graduate into each other, whence 

 result different kinds of gneiss which alternate together. 

 These continual fluctuations in the composition of the pri- 

 mary rocks are exceedingly interesting; at first sight they 

 appear to be devoid of all order, but a careful examination 

 has shown that this is not the case, several examples of 

 which have been quoted from the works of Von Buch and 

 Macculloch, in the fifth chapter, showing that the dif- 



