Page:Life and death (1911).djvu/297

 trillion cells ought each to find its own predetermined place.

In crystallography also we understand by form the organization which crystals present. The grouping of the elements of crystals is, perhaps, more simple. They are none the less organized, in the same sense that living bodies are.

Their organization, while more uniform than that of living bodies, still shows a considerable amount of variation. It should not be assumed that the area of a crystal is completely filled, with contiguous parts applied one to the other by plane faces, as might be supposed from the phenomenon of cleavage which dissociates the parts of the crystalline body into solids of this kind. In reality, the constituent parts are separated from each other by spaces. They are arranged in a quincunx, as Haüy put it, or along the lines of a network, to use the terms of Delafosse and Bravais. The intervals left between them are incomparably larger than their diameters. So that in the organization of a crystal it is necessary to take into account two quite different things:—An element, the crystalline particle, which is a certain aggregate of chemical molecules having a determinate geometrical form; and a more or less regular, parallelopipedic network, along the edges of which are arranged in a constant and definite manner the aforesaid particles. The external form of the crystal indicates the existence of the network. Its optical properties depend upon the action of the particles, as Wallerant has shown: Thus we must distinguish in a crystal between two kinds of geometrical figures—that of the network and that of the particle—and their characters of symmetry may be either concordant or discordant.