210 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1911. 
chanics of the universe. The influence of atomic compressibilities 
may be perceived everywhere, and in most cases each fact seems to fit 
easily and without constraint into its place in the hypothesis. Even 
apparent exceptions, such as the abnormal bulk of ice, may be ascribed 
in a reasonable fashion to superposed effects. A detailed discussion 
of many applications of the theory is impossible here, but a few may 
be suggested; in order to make clearer its possibilities. 
The satisfying of each valence of an atom would cause a depression 
on the atomic surface, owing to the pressure exerted by the affinity 
in that spot. The stronger the affinity, the greater should be this 
distortion. Evidently this conception gives a new picture of the 
asymmetric carbon atom, which, combined with four other different 
atoms, would have upon its surface depressions of four unequal 
magnitudes, and be twisted into an unsymmetrical tetrahedron. The 
combining atoms would be held on the faces of the tetrahedron 
thus formed, instead of impossibly perching upon the several peaks. 
According to this hypothesis, the carbon atom need not be imagined 
as a tetrahedron in the first place; it would assume the tetrahedral 
shape when combined with the other four atoms. One can easily 
imagine that the development of each new valence would change 
the affinities previously exercised, somewhat as a second depression 
in the side of a rubber ball will modify a forcibly caused dimple in 
some other part. Thus a part of the effect which each new atom 
has on the affinities of the other atoms already present may be 
explained. 
Many other physico-chemical phenomena assume a new aspect 
when viewed from the standpoint of this idea. New notions of the 
mechanism of the critical phenomena, surface tension, ductility, 
malleability, tenacity, and coefficient of expansion are gained. The 
peculiar relations of material and light, such as magnetic rotation, 
fluorescence, partial absorption, and so forth, may be referred to 
the modified vibrations of distorted atoms. The deviations from 
the exact fulfillment of many older generalizations concerning volume 
(such as the equation of van der Waals already cited, the comparative 
volumes of aqueous solutions, especially of electrolytically disso- 
ciated substances, and the variations in the crystal forms of isomor- 
phous substances) are seen to be a foregone conclusion. Moreover, 
the theory, although not necessarily dependent on the modern belief 
that atoms are built up of numbers of much smaller corpuscles, is 
consistent with that belief, for would not such an entity be 
compressible ? . 
The more closely the actual data are studied, the more plausible 
the hypothesis of compressible atoms appears. Ten years’ experience 
1 Baxter has very recently discussed this matter from the point of view of the theory of compressible 
atoms, (Journal American Chemical Society, 1911, vol. 33, p. 922.) 
