II1O General Notes. [Dec. 
obtain some knowledge as to the origin of rocks, their relations 
to each other, and the causes of the great variation in structure 
observed in them. The paper is full of interesting facts and 
valuable suggestions. It begins with a discussion of the theory 
proposed by Rosenbusch to account for the successive crystal- 
lization of rock-constituents. It has been recognized for a long 
time that the minerals in a rock did not crystallize in the inverse 
order of their fusibility, as might at first thought be expected. 
The order in which the separation takes place bears no relation 
to the temperature of solidification. Rosenbusch regards the 
acidity of the minerals as the important element governing their 
crystallization. He states, as a general law, that the order of 
the separation of the minerals from a rock-magma is inversely 
as their acidity. Roth’ has called attention to the very many 
exceptions to this rule, and in its place gives seven empirical 
laws of association. Lagorio refuses to accept Rosenbusch’s 
law, and proceeds to investigate the subject by chemical methods. 
He analyses the glassy ground-mass and crystallized portions . 
(Ausscheidungen) of artificial and natural glasses, and thoroughly . 
discusses the figures thus obtained. He finds that the sodium- 
silicates possess a stronger tendency to crystallize from a molten 
magma than do the corresponding isomorphous potassium com- 
pounds. Sodium is concentrated in the crystallized portions, 
and potassium in the residual ground-mass. With this discovery — 
as his governing principle, Lagorio examined rocks varying 
that in which all the other silicates are dissolved, or, more ab 
cisely, the last substance to crystallize from a molten rock © 
SiO, This 
the different amounts of the elements in solution and their wes n 
ity for each other, the sodium-bearing minerals always 8 era 
lizing before the corresponding potassium-bearing compou i 
) The order of solubility of various compounds in o ae ze 
magnesium-, iron-salts; and, finally, oxides of the bay aie 
the latter being the most soluble. The greatest amount 0 am Me 
saturation obtains in the case of the most soluble compoun 
From such a supersaturated solution—a rock-magm | 
separate in the order of the excess of saturation,—V!4+ 
of the heavy metals first, then the silicates of the heavy 
then sodium-silicates, and, finally, quartz and the po 
1 J. Roth, Chem. Geologie, ii. pp. 49 and 69. 
E 
