40 Dr. Mac Culloch on 



that character is defective. I have shown further, that these 

 have heen found under gneiss ; whence, according to the ar- 

 rangement, that gneiss ought to be a transition rock ; and thus 

 every rock found above any one that is discovered to possess 

 such an organized body, must necessarily fall into the same 

 class. The consequences are here sufficiently obvious ; and 

 it is equally plain that there are no characters, either singly 

 taken or combined, by which transition rocks can be distin- 

 guished from the primitive. Neither is there any boundary 

 by which the transition class can be separated from the first. 

 None has been assigned ; and it is evident, from the last re- 

 mark, that whatever boundary may be assumed, it must be 

 removed as often as any organic substances, or even any 

 imbedded fragment of a rock, shall be found in a position, or 

 in a rock, inferior to that which has last been fixed on as its 

 lowest member. Lastly, the transition rocks are frequently 

 absent altogether ; a fact which, when the primitive and 

 secondary are present, implies a contradiction in terms. This 

 class, therefore, as a natural division, has every possible 

 defect ; as it is not always present, does not form a transition, 

 has no permanent or certain characters, and has no assignable 

 boundary. 



Such at least is the judgment which must be pronounced 

 upon it, as far as our knowledge of the early rocks yet goes. 

 It is not impossible, as I formerly suggested, that there may 

 be a division of rocks intermediate between the primary and 

 secondary, and it is possible that its characters and boundaries 

 are assignable. It is even possible that some of those which 

 have been named as transition rocks, may belong to such a 

 division ; but it is very certain that, at present, we have no 

 definite notions of it, and cannot pronounce even on its exist- 

 ence, much less on its characters and boundaries. As an 

 artificial arrangement, it serves no useful purpose, as it does 

 not facilitate the examination or classification of rocks. On 

 the contrary, it is pernicious, as it leads to a vicious kind of 

 reasoning in a circle ; certain rocks being first called by the 

 name of transition, and then used to determine the class. 

 To a certain extent, it is true, the same incorrect mode of 

 reasoning may be applied to the primitive and secondary 



