348 REPORTS ON THE STATE OF SCIENCE.—1912. 
Report on Diffusion in Solids. By Cectu H. Descu, D.Sc., Ph.D. 
[Ordered by the General Committee to be printed in ezxtenso.] 
THE occurrence or non-occurrence of diffusion in solid substances is of 
fundamental importance for the theory of solid solutions and isomor- 
phous mixtures. In order that any mixture of substances in the solid. 
state may be legitimately regarded as a solid solution it must be shown 
that diffusion is capable of taking place within it from regions of high 
to those of low concentration. The ambiguity which at present exists 
in respect to this question is due in part to ambiguity in the definition 
of the solid state, and in part to differences of opinion as to the facts. 
The popular distinction between fluids and solids is traversed by the 
more scientific classification into amorphous and crystalline substances. 
It is evident, however, that such substances as the glasses, although 
connected by all possible gradations with liquids on the one hand, and 
sharply separated from crystalline substances on the other, yet possess 
most of the properties commonly associated with solids, whilst the 
liquid crystals of p-azoxy-anisole, for example, although definitely 
crystalline, cannot be regarded as solid bodies in any sense of the 
word. 
The following classification,! then, represents the possible states 
that may be assumed by a substance, although any one substance is 
not necessarily capable of assuming all of these states :— 
IsorTRopric. ANISOTROPIC, 
Gaseous. Crystalline, including liquid crystals and 
Liquid. one or more solid crystalline modifica- 
— Amorphous or glassy. tions. 
Diffusion takes place in liquid crystals in apparently the same 
manner as in isotropic liquids,? and they are therefore omitted from 
consideration in the present Report. On the other hand, both amor- 
phous and crystalline solids are considered. The former class is restricted 
to the natural and artificial glasses, whilst the jellies, which are 
undoubtedly of a colloidal character, are treated briefly at the close on 
account of the importance of certain periodic phenomena observed in 
them for the general theory of diffusion. The materials in which 
diffusion has been either observed or suspected are treated in the 
following order: Glasses (p. 349); metals (p. 351); minerals (p. 367); 
salts and organic crystalline substances (p. 368); these sections being 
in some cases sub-divided for greater convenience. A short summary 
of the evidence concludes the Report (p. 371). 
1 Modified from that given by G. Tammann, ‘ Krystallisieren und Schmelzen 
(Leipzig, 1903), p. 5. 
? QO. Lehmann, ‘ Fliissige Krystalle’ (Leipzig, 1904), p. 76. 
