196 Frerp MusEum or NaturaAt History — Geotoecy, Vot. III. 
whole no serious error need be involved. To confine reported analyses 
to those which were only stated in the mass form would reduce the 
number materially and fail to represent our true knowledge of the 
chemical composition of meteorites. 
The second purpose for which the grouping of the analyses has been 
made was, as has been stated, to propose a quantitative classification. 
The principles of this classification are the same as those for terrestrial 
rocks proposed by Cross, Iddings, Pirsson, and Washington.* It was 
suggested by Washington in his publication on the Chemical Analyses 
of Igneous Rocks and their Classificationt that such a classification of 
meteorites be made, and the writer held a brief conference with 
Dr. Washington on the subject. The need of such a classification of 
meteorites is, perhaps, even more acute than was the case with terres- 
trial rocks. Of the various classifications of meteorites which have 
been proposed none can be considered quantitative. The classification 
chiefly used for stone meteorites at the present time is that which has 
been gradually evolved through the labors of Rose, Tschermak, Cohen, 
and Brezina. It is presented in its most complete form by Brezina 
in the Catalogue of the Ward-Coonley Collection of Meteorites.t As 
is well known, the groups of this classification are based primarily upon 
structure but also upon mineralogical characters. The stones are 
first subdivided into achondrites, chondrites, and siderolites. The 
achondrites are divided into a number of groups distinguished by 
mineralogical composition. These include the eukrites, chladnites, 
howardites, etc. Among the chondrites the subdivisions are based 
chiefly on color, the groups being designated as white, gray, black, 
intermediate, carbonaceous, etc., with additional divisions according 
to structure giving spherulitic and crystalline. Other subdivisions are 
based upon the presence or absence of veins and breccia-like structure. 
Of these divisions, that according to color cannot be regarded as resting 
upon any important or fundamental character, although it finds some 
sight justification in the fact that the lighter-colored meteorites are 
likely to contain more enstatite than the darker ones. Another weak 
feature of the classification in the view.of the present writer is its failure 
to take account, in any definite way, of the metallic content of me- 
teorites. The metal of meteorites is an important feature which 
should serve as a distinguishing mark. 
So far as the iron meteorites are concerned the present system of 
* Quantitative Classification of Igneous Rocks, Chicago, 1903. 
} U. S. Geological Survey, 1903, Prof. Pap. No. 14, pp. 9 and 61. 
{t Henry A. Ward, Chicago, 1904, pp. 97-101. 
