60 



sary, in this particular instance, to place gneiss in 

 the transition class ; while, in the predominant 

 examples, it must necessarily belong to the pri- 

 mary division. No set of internal characters 

 therefore can be assigned, which, either taken 

 singly, or in combination, are capable of forming 

 a distinction between the transition and the pri- 

 mary classes. 



Neither can the former be distinguished by 

 any definite boundaries. With regard to the 

 secondary rocks, its boundary would be the same 

 with that already laid down as the limit of the 

 primary ; should it be present, and should it be 

 held to possess a boundary in this direction. 

 That it has been held to be definite towards the 

 secondary strata, must be presumed ; because the 

 rule of non-conformity, or the absence of the paral- 

 lel consecutive position, and the existence of the 

 rock known by the expressive term old red sand- 

 stone, are by the same geologists held to be gene- 

 rally characteristic of the commencement of the 

 secondary strata. Towards the primary strata, 

 it can have no boundary ; if that must be re- 

 moved as often as a stratum containing organic 



