512 
all mixtures of different substances to rapidly 
change into the same substance. There is in fact 
nothing in Sir J. Larmor’s theory to account for 
the stability of atoms which according to any such 
theory would be broken up by collisions and could 
not preserve their identity in compound molecules. 
This theory fails to explain why free positive 
electrons can not be obtained while negative elec- 
trons are easily isolated. On the other two theories 
the stability of the atoms can be ascribed to the 
stability of the positive sphere or nucleus. When 
negative electrons are knocked out of the atom the 
excess of positive charge can only be neutralized 
by an equal number being replaced, which restores 
the atom to its original condition. Important evi- 
dence as to the nature of atoms is provided by the 
properties of gases interpreted according to the 
kinetic theory. The viscosity, rate of diffusion and 
other properties of gases seem to be best explained 
if the molecules are regarded as having a definite 
volume of radius about 10“ em. This is usually 
termed the radius of molecular action and two 
molecules which are separated by a distance 
greater than about 10 cm. can be regarded as 
having no appreciable action on each other. Sim- 
ilar conclusions ean be derived from the theory of 
surface tension and many other phenomena. The 
energy necessary to remove a negative electron 
from an atom, for example, indicates that the 
positive and negative charges are about 10° cm. 
apart in the atom. 
The most important evidence from the kinetic 
theory is derived from the ratio of the specific 
heat at constant pressure to that at constant vol- 
ume. This ratio y is equal, according to the 
theory, to 1+ 2/n where n is the number of de- 
grees of freedom per molecule. For helium, neon, 
argon, krypton, xenon, mercury vapor and other 
gases y==1.66, which requires that n==3. This 
means that when such gases are heated all the 
energy goes into the three degrees of freedom of 
translational motion of the molecules, so that the 
molecules acquire no energy of rotation or vibra- 
tion. To explain this it appears necessary to sup- 
pose that the molecules behave like rigid smooth 
spheres. Such gases are therefore believed to be 
monatomic, each atom behaving like a smooth 
sphere. These gases, however, give spectra con- 
taining many lines so that it is certain that their 
atoms contain electrons which can vibrate. It is 
necessary to suppose that collisions between these 
atoms do not set their electrons in vibration, which 
seems to require the electrons to be protected in 
some way. This seems to be strongly in favor of 
SCIENCE 
[N.S. Vou. XXXV. No. 900 
Sir J. J. Thomson’s theory and against the other 
two theories, for if the electrons were describing 
orbits outside it is hard to see how they could 
escape violent disturbance during a collision. 
Gases like nitrogen, hydrogen and carbonmon- 
oxide have y= 1.4, which gives n=5. This 
means that their molecules behave like smooth 
rigid solids of revolution. These gases are di- 
atomic, so that it appears that the two atoms in 
the molecule are firmly fixed together. This can 
be explained on Sir J. J. Thomson’s theory by 
supposing that the two positive spheres stick to- 
gether and it can not be explained on the other 
two theories. 
The theory of the constitution of chemical com- 
pounds seems to require the atoms in compound 
molecules to be firmly fixed together in definite 
relative positions. The numerous cases of stereo- 
isomerism and optically active isomers seem con- 
elusive as to this. Similar conclusions follow from 
the properties of crystals which can be explained 
in many cases by supposing the atoms combined 
in a definite way throughout the crystal so that 
the whole crystal is, so to speak, one large mole- 
culet The rigidity of many crystals requires the 
atoms to be firmly fixed together. 
On Sir J. J. Thomson’s theory two atoms can 
stick together if one or more electrons are trans- 
ferred from one to the other. In this case they 
would attract each other with great force and we 
may suppose the positive spheres to be flattened 
up against each other by the pressure so as to 
form a rigid combination. 
Sir J. Larmor’s theory and Rutherford’s planet- 
ary theory are difficult to reconcile with the idea 
that atoms become firmly fixed together in com- 
pounds and rigid solids. On such theories we 
should expect to have nothing but gases and 
liquids and only very simple compounds. 
Another important property of solids is their 
impenetrability by gases. This seems to require 
the atoms to occupy nearly the whole volume. The 
compressibility of solids is not great and does not 
diminish much at low temperatures. If atoms 
were made up merely of electrons of very minute 
volume we should expect solids at very low tem- 
peratures to contract to a very small volume or at 
any rate to become easily compressible. This 
again is in fayor of atoms with a definite volume 
of radius about 10“ em. as on Sir J. J. Thomson’s 
theory. This argument is taken from the Faraday 
lecture for 1911 by Professor Richards, who puts 
1See A. E. H. Tutton, ‘‘Crystallography,’’ Lon- 
don, 1911. 
