FORMATION OF CRYSTALS—GAUBERT. 273 
there takes place what Vogelsang and his predecessors have observed 
with sulphur or other bodies, but who worked with supersaturated 
drops or amorphous particles, or little spherulites of unstable form, 
which later underwent modifications into more stable forms, and 
the normal development of which can then be followed. 
Nevertheless, in spite of the observations of Frankenheim, O. 
Lehmann, and others, the idea of the embryonic state of the crystal 
has not disappeared from science, and the hypothesis of Vogelsang, 
supported by De Schoen, Cartaud, and others, resting on misinter- 
preted observations, still finds some credit. 
108 
When the crystal is once formed—that is, becomes visible under the 
microscope—how does it grow? Several cases may be presented: 
First, the mother liquid is in a state of rest, the cooling or evapora- 
tion is extremely slow, and the crystalline particles are built up by 
diffusion alone. In this case the growth is too slow to be constantly 
followed under the microscope. In the second case the lquid is 
cooled or evaporated with such rapidity that the quantity of matter 
deposited on the crystal produces an enlargement microscopically 
visible. Movements in the liquid are thereby set up. It is an estab- 
lished fact that currents called “ currents of concentration” passing 
over a crystal, deposit a thin coating of substance, followed by a 
second, and so on, until, for example, one can see on a crystal of lead 
nitrate, having a diameter of half a millimeter, as many as twelve 
of these successive layers deposited. If the process were suddenly 
interrupted and the crystal examined any observed face would not be 
a plane, but would show a sort of step arrangement, of which the 
highest step would indicate the point of contact of the current of 
deposition. 
These successive deposits have no interspaces and the crystal 
may be perfectly transparent. If the crystal of lead nitrate is, 
however, subjected to the influence of two or of several currents of 
concentration, the corresponding coatings laid upon it start from 
different points in the periphery and may not be of the same thick- 
ness. Ordinarily they do not join exactly at their point of meeting. 
In this way are then produced inclusions and the crystal is no longer 
transparent, but becomes milky. On a glass plate it is easy to 
produce at will a transparent or milky crystal of lead nitrate. In the 
experiment it is necessary to agitate the crystal very slightly with a 
needle in order to subject it to the influence of one or several currents. 
These concentration currents produce other peculiarities (vicinal 
faces, etc.), which it would take too long to describe in detail. I 
shall confine myself to calling attention to the influence they may 
