Uevieivs — Dr. J. Petersen on Cheviot Rocks. 233 



rocks under his consideration." I hold that it is not the primary 

 duty of the palaeontologist to determine the age of rocks. His 

 primary duty is to describe the relations of the forms of life with 

 which he has to deal with other forms, both living and extinct, and 

 to define his genera, species, etc., without reference to geological age. 

 It then becomes his duty to ascertain the distribution of his types 

 both in space and time, and in this branch of his work he is led to 

 the conclusion that certain types are limited in their vertical range, 

 and become therefore most valuable for purposes of chronology, 

 whereas other types are persistent and range through several geo- 

 logical periods. He does not fix on some arbitrary period and 

 assert that one and the same type shall receive two different names 

 according as it occurs on the one or other side of his selected line. 

 If a Trilobite were discovered in Secondary or Tertiary formations, 

 or if it were found living at the present day, it would still be called 

 a Trilobite. 



Now I consider that a petrologist should deal with his rocks in 

 precisely the same way. His primary duty is to describe his rocks 

 and define his types in the clearest possible manner, but without 

 reference to age. The introduction of geological age for this purpose 

 is unnecessary if there be a radical distinction between all the Pre- 

 Tertiary and later rocks ; and if in certain cases there be not this 

 distinction, then its introduction obscures natural differences and 

 implies relationships which do not exist. If, after having defined 

 his types, the petrologist discovers that some of them are limited as 

 to their range in time, then these xasiy be utilized for chronological 

 jDurposes. As a matter of fact, there are certain types which appear 

 to be limited in their range, and other types which are persistent. 

 The case is to a certain extent analogous to that of the pala3onto- 

 logist. It must be remembered, however, that the history of 

 geological science shows clearly that petrology is not a very 

 safe guide in the determination of chronological relations. It will 

 be seen then that such cases as the existence of leucite and hauyne 

 only in Tertiary rocks has nothing to do with the question at issue 

 between us. Where there are differences between Pre-Tertiary and 

 later rocks, these differences can be utilized for purposes of classifi- 

 cation. 



As the question is one of so much importance, I will endeavour to 

 put my position from another point of view. 



The whole fabric of science is constructed on the assumption, 

 justified by experience, that the causal relations of natural phe- 

 nomena are independent of situation in space and time. Given 

 similar conditions, similar phenomena occur. Now it seems to me 

 that the refusal to give the same name to two rocks which are 

 similar in structure and composition, merely because they have been 

 produced at two different periods, is equivalent to denying the 

 uniformity on which the whole of science is based. It could only 

 be justified, at least so it seems to me, and it must be distinctly 

 understood that I am only endeavouring, with all diffidence, to make 

 my own position perfectly clear, by the assumption that an entire 



