412 Prof. J. P. Cooke on the Variation of Constitution in a 
nite crystalline form is compatible with quite a wide variation of 
composition, and has in this way pointed out an explanation of 
the variation observed in the mineral kingdom. But more than 
this, the investigation has also indicated a method by which, 
amidst all this variation, the true constitution of the mynimeral can 
be determined. 
In the compounds of zine and antimony, although the definite 
crystalline form was compatible with a wide variation in the 
proportions of the constituent elements, yet the point corre- 
sponding to the typical composition was marked by several un- 
mistakeable properties, which clearly enough indicated the true 
formule of the compounds. These properties are discussed at 
length in my original memoir, and need therefore only to be 
alluded to in this connexion. 
It has already been stated that the crystals, both of Sb Zn® 
and Sb Zn?, having the theoretical composition are, as a rule, 
larger and more generally aggregated than those containing an 
excess of either metal. Moreover, in Sb Zn? the general character 
of the crystals appears to be modified by the change of compo- 
sition, although the crystallographic elements remain the same. 
Thus in the crystals having the theoretical composition, the 
octahedral planes are greatly developed, giving to the crystals 
the general appearance of a truncated octahedron*. But as the 
crystals take up an excess either of antimony or zinc, the basal 
planes become more and more dominant, and the crystals are at 
last reduced to thin plates. In fact, so marked are these changes, 
that, after a little experience, a person could tell the approxi- 
mate composition of the erystals from their general appearance. 
Similar changes in the appearance of many minerals are familiar 
to the mineralogist. They are seen in calcite, heavy spar, Angle- 
site, and others, and may serve as guides in tracing variations of 
composition. 
Again, the specific gravity of the crystals, both of Sb Zn? and 
Sb Zn°, was taken with great care through the whole series, and 
the results are tabulated below. The union of the two elements 
is attended with an increase of volume, and this increase is at a 
maximum at the points corresponding to the theoretical composi- 
tion. These points would therefore be marked ina set of crystals 
by being points of minimum specific gravity ; and they could be 
determined with great accuracy by means of this property, even 
in a series of alloys of the two metals which had not been cry- 
stallized. This fact is illustrated by the following Table, reprinted 
from the original memoir. 
* See figure accompanying my original memoir. 
