(1104 ) 
the one recently obtained for oxygen by E. Marnias and H. KAMERLINGH 
Onnes viz. 3.43. (Comm. N°. 117, March 1911). In a paper that will 
soon be published by C. A. Crommenin and H. Kamuriincn ONNes *) a 
deduction from the isotherms of A, = 3.28 for argon will be given. 
§ 6. Molecular Attraction in Helium. The occurrence of a maxi- 
mum density in a substance of such simple constitution as helium 
gives rise to questions of great import from the point of view of 
molecular theory. With a substance like water it is easy to imagine 
a particular molecular combination by which some of the parts are 
more closely united, while others are separated, the whole leading 
to an inerease of volume as the temperature is lowered, and this 
especially when one considers that the dielectric properties of water 
probably play a part in the phenomenon. But helium atoms we are 
forced to consider as spherical and smooth, and, as appears from 
the Zeeman-effect for helium, of the simplest possible internal con- 
struction; and for their case we seek in the meantime in vain for 
a basis for a similar explanation. Moreover, helium differs from 
ordinary normal substances, but in exactly the opposite way to that 
in which associated substances differ from them. 
A dissociation increasing as the temperature is diminished, leading 
to an increase of the number of molecules (and, therefore of A in 
the equation of state), whieh would account for this deviation in 
the opposite sense, can scarcely be imagined. Should it appear that 
the change was occasioned by an increase in the dimensions of the 
helium atoms (that is, of 4 in the equation of state) as the tempe- 
rature is lowered, then this, too, would be something strikingly 
unusual. The behaviour of helium seems rather to make it clear 
that even in the case of ordinary normal substances two different 
kinds of molecular attraction must be distinguished from each other — 
an attraction of comparatively large sphere of action, and an attrac- 
tion that is local, but more intense, of smaller range, and confined 
to the immediate neighbourhood of the surface of the molecule; 
(his latter attraction causes ordinary normal substances when com- 
pared with helium to resemble rather associative substances; in the 
ease of liquid helium the latter type of action of the attraction 
would, then, be suppressed. 
If it is not, indeed, entirely absent in helium, the sphere of in- 
fluence of this foree must have wholly withdrawn within the space 
occupied by the atom at the lowest temperatures (which is probably 
also to a large extent the case for substances like hydrogen at the 
') Comm. N'. 120a, Proceedings of this Meeting p. 1012, 
