72 THE CONSTITUTION OF THE SILICATES. [bull. 125. 
becomes Mg=Si 2 7 =H 2 (MgOH) 2 , which may be derived either from 
2Mg 2 Si0 4 or 2MgSi0 3 , with loss of magnesium in one case and gain in 
the other. On the orthosilica.te basis it is simply derivable from the 
polymei Mg 4 (Si0 4 ) 2 , and is related to the intermediate alteration prod- 
uct villarsite, as follows: 
Mg 4 (Si0 4 ) a . Villarsite. Serpentine. 
Mg Mg H 2 
H II II 
„ / Si °4 X /Si0 4X Si0 4X 
Mg/ \Mg Mg< >Mg Mg/ >Mg 
II / \ / \ 
Mg H Mg— OH H Mg— OH 
On this scheme the formula for serpentine corresponds with that of 
chondrodite; and the fact that the latter mineral alters readily into 
serpentine is strong evidence in its favor. If pyroxene is written 
structurally 
O 
O— Si— O x 
Mg< \Mg 
O 
the derivation of serpentine from it by action of infiltrating magnesian 
solutions becomes easy to understand, and the orthosilicate formula for 
serpentine is rendered still more probable. In short, that formula best 
indicates the genetic relationships of serpentine, and on these grounds 
is preferable to the alternative diorthosilicate expression. The latter 
is not disproved ; it is simply rendered iess advantageous as regards 
existing evidence, which is the evidence now to be interpreted. 
In some former investigations, carried on jointly with Dr. Schneider,* 
I sought to obtain experimental data in support of the orthosilicate 
formula here assigned to serpentine. By acting on serpentine with 
dry gaseous hydrochloric acid we found that a part of the magnesium 
could be removed as chloride, while olivine and the magnesian micas 
were not attacked. .A t first it seemed probable that the reaction would 
give a quantitative measure of the magnesium combined as MgOH; 
but our later experiments and those of Lindner f have shown that the 
expectation was not well founded. I still believe, however, that the 
reaction discriminates between those magnesian silicates which con- 
tain MgOH and those which contain Mg and H combined otherwise, 
for only the members which must belong to the first class are acted 
* Bull. U. S. Geol. Survey No. 78, p. 11 ; Bull. No. 90, p. 11, and Bull. No. 113, pp. 27 and 34. 
f A. Lindner, Inaugural Dissertation, Breslau, 1893. 
