lz& 



A. W. Jackson — Nomenclature of CrystaUme Rocks. 117 



of nonuniformity and instability. Although it is in a general 

 way true that the theories of one generation become the facts of 

 the next, it is equally true that most of the theories of one 

 generation fail to become the facts of the next; and so long as 

 they remain in the condition of a theory they should be totally 

 ignored in forming rock-names. The value of a fact in this 

 connection is to be measured by its generality; facts of local 

 value should be reserved for local distinctions. 



The kinds of facts which are more immediately accessible to 

 ee: chemical, 'jeofaijiad and mineralogical 

 conceive to be the value of each class for 

 the formation of names. 



1. Chemical. Perhaps the most striking evidence of the im- 

 possibility of basing nomenclature upon chemical composition 

 can be presented in the fact that during all of the years preced- 

 ing the use of the microscope in vestigation, 

 chemistry was unable to discover distinctions that become evi- 

 dent with the first glance into the microscope. Rocks are not 

 minerals and have not the same well-defined stochiometric cora- 

 is have. On the contrary, rocks that are 



■:■:: .■■■:: . ' ' • , ,-■:,•■ , ; 



ical composition, while the same bulk-analysis may be true for 

 rocks of widely different mineralogical composition. It is to be 

 observed from this that knowing the analysis one is notable to 

 predict mineral composition while the converse is not true ; if 

 we know the mineral composition we can predict always quali- 

 tatively, and to an approximation quantitatively, the chemical 

 composition. It would seem then that the fact of mineral 

 composition is of a higher order of utility than the fact of 

 chemical composition. The latter is a function of the former 

 and a nomenclature based upon mineralogical facts will neces- 

 sarily express sufficiently clearly the chemical facts, without 

 the inconvenience of determining the chemical composition in 

 each case before deciding what name shall be applied. 



Again, it is held by some (Scheerer, Jahrbuch der Mineralo- 

 gie etc., 1864, 386, and others) that the chemical composition 

 of a rock is a fact of the first importance and should be made 

 the leading element in classifying (and naming?) because it 

 gives us a clear conception of the nature of the original duid 

 material out of which the rock crystallized; and that it is this 

 original fluid material which constitutes the true "rock-type" 

 about which all of our interest centers. Now. while I fully ad- 

 mit the value of this conception and appreciate the significance 

 of the results to be gained by tracing different rocks back to a 

 common magma, still I hold that a rock is something more than 

 a certain chemical mixture, it is a chemical mixture with a 

 history that is tq\ u , I esting. It has been subjected to various 



