56 DOMAIN I. 3IDEROU8. 



species; there would be bitt one and the same 

 name for all the mineral kingdom. Thus I dis- 

 tinguish granite from porphyry, porphyry from 

 trap, this from petrosilex, roches de corne, and 

 argillolites, because the well-characterised indi- 

 viduals of these different kinds are evidently 

 distinct; and I do not embarrass myself because 

 there are transitions or intermediate varieties, 

 which I do not exactly know to which kind I 

 ought to refer. 



" I have in this only to regret a deviation 

 from the acceptation that M. Dolomieu has 

 given to the name of trap, in the excellent work 

 he published, Journal de Physique, An 2. Part I. 

 page 257. He had given this name to the cor- 

 neus trapezius of Wallerius, which is a simple 

 stone of the genus corneennes, with a fine and 

 Compact fracture. But I have already observed 

 in another place, that the genus of simple cor- 

 neennes does not require this subdivision, whilst 

 the class of composites or rocks appears inca- 

 pable of avoiding it; and of them the cele- 

 brated Werner has even formed such a class, 

 where, under the head of trapp formazion, he 

 includes grunstein, the amygdaloide, the por- 

 phyrschiefer, and bazalt. 



" I shall also observe, that the Swedes give 

 the name of trap, not only to a simple and com- 



