ENTROPY AND PROBABILITY 57temperature derivative. Using known data, he could check his hypothesis, which,
he found, actually worked fairly well for a limited range of carbon compounds
(with molecular weights mainly of the order of 100) but not for lighter molecules,
such as water.
Einstein's hypothesis is, of course, incorrect. As is now well known, even in the
simplest semiphenomenological models (such as the Lennard-Jones potential),
the intermolecular forces not only have a characteristic strength constant but also
depend on the molecular size. This first paper by Einstein is of interest only in
that it shows how from the start he was groping for universal principles, in the
present case for a relation between molecular forces and gravitation. 'It should be
noted,' he remarked, 'that the constants c increase in general but not always with
increasing weight; however, this increase is not linear. Therefore the question if
and how our forces are related to gravitational forces must for the time being be
kept completely open' [E5]. The purpose of his second paper [E6] was likewise to
obtain information on his conjectured force law. Here, no comparison data were
available. The paper concludes with an apology by Einstein for not being in a
position to contribute personally to the experimental clarification of his theoretical
ideas.
That Einstein was quite taken with the concept of a universal molecular force
is seen from a letter to Grossmann in 1901. 'I am certain now that my theory of
the attractive forces ... can be extended to gases ... Then the decision about the
question of the close relation of molecular forces with the Newtonian forces acting
at a distance will come a big step nearer' [E7]. Then follows a lyrical passage: 'It
is a wonderful feeling to recognize the unifying features of a complex of phenom-
ena which present themselves as quite unconnected to the direct experience of the
senses.'
In December 1907, Einstein wrote to Stark: 'I am sending you ... all my pub-
lications except for my worthless first two papers [E8]. And so we meet for the
first time a trait typical of Einstein throughout his life. He could be very enthu-
siastic about his own ideas and then, when necessary, drop them some time later,
without any pain, as being of no consequence.
I have dwelt at disproportionate length on these first two papers simply because
by doing so I shall have no need to return to them. Two final comments about
them: (1) one thermodynamic relation contained in the first paper did survive;*
and (2) in 1911 Einstein briefly returned one more time to the molecular theory
of liquid surface phenomena.**
'Let / be the heat capacity at constant pressure p of a liquid held in a container, <o the liquid surface,
and a the surface tension. Einstein derived the relation [E5]This result is discussed by Schottky [S3].
**In a short note on the Eotvbs relation between surface tension, specific volume, and temperature
[E9].