MUSEUM OF COMPARATIVE ZOuLOGY. 277 



No natural distinction can be drawn between rocks of the Tertiary 

 and Pre-Tertiary ages, since the glass and fluidal inclusions, crystalline 

 texture, and the various other characters fail, exactly where they are 

 most needed, to divide the roclis into older and younger, as is done 

 by the majority of lithologists. 



The writer believes that rocks should be studied, by beginning with 

 their most compact or glassy state, and by then tracing them through 

 to the most crystalline form, following every alteration, whether it be 

 chemical or mechanical. Every rock that can be traced in this way 

 forms a distinct species, whatever may be its state, — whether amorphous, 

 glassy, crystalline, fragmental, tufaceous, or otherwise, — and whatever 

 may be its age. The modifications, if of sufficient importance, form 

 varieties simply, which shoidd be included under the specific name. A 

 natural classification of rocks must be empirical, and must be based on 

 the rock as a whole, while a natural mineralogical classification is an 

 impossibility, as it is based on part of the characters only. 



If we except the veinstones and the majority of those rocks that are 

 composed of one mineral, the species of rock forming the crust of the globe 

 are very few. Believing that this earth is a cooling globe, all mani- 

 festations of internal heat giving rise to rocks (the only thing with which 

 we are at present concerned) are here termed volcanic, and all such pro- 

 ducts are styled volcanic rocks. The testimony of the rocks is that all 

 sedimentary forms came primarily from volcanic ones, volcanic energy 

 having been more active than now in the past ages of the globe. This 

 derivation is consonant with that which we see taking place at the 

 present time, and agrees with the law of dissipation of energy ; while the 

 reverse view, at present popular, — that eruptive rocks were derived 

 from sedimentary ones, — is contrary to the positive testimony of the 

 rocks themselves, to the facts that are observed in nature, and to physi- 

 cal laws. 



Taking the consolidation of any rock as its initial point, the minerals 

 and rock fragments contained therein fall naturally into three classes : 

 1. Minerals and fragments of prior origin ; 2. The products of that con- 

 solidation ; 3. The products of alteration and infiltration. 



These thi'ee classes are most marked in the volcanic rocks, as is 

 natural ; the first two predominating in the younger and least altered, 

 the latter in the older and more altered ones, while the first and third 

 classes predominate in sedimentary rocks. These alterations apparently 

 take place thiough the agency of the ordinary percolating waters, which 

 are not uccc-ssarily hot. Tlie minerals aud fragments of the first class, 



