PURPOSE AND PLAN 19answers gave the molecular picture the compelling strength of a unifying princi-
ple. Three of these methods are found in Einstein's work of 1905. In March he
counted molecules in his light-quantum paper (19c). In April he made a count
with the help of the flow properties of a solution of sugar molecules in water (5c).
In May he gave a third count in the course of explaining the long-known phe-
nomenon of Brownian motion of small clumps of matter suspended in solution
(5d). The confluence of all these answers is the result of important late nineteenth-
century developments in experimental physics. Einstein's March method could be
worked out only because of a breakthrough in far-infrared spectroscopy (19a).
The April and May methods were a consequence of the discovery by Dr Pfeffer
of a method for making rigid membranes (5c). Einstein's later work (1911) on the
blueness of the sky and on critical opalescence yielded still other counting methods
(5e).
The second problem: the molecular basis of statistical physics. If atoms and
molecules are real things, then how does one express such macroscopic concepts
as pressure, temperature, and entropy in terms of the motion of these submicros-
copic particles? The great masters of the nineteenth century—Maxwell, Boltz-
mann, Kelvin, van der Waals, and others—did not, of course, sit and wait for the
molecular hypothesis to be proved before broaching problem number two. The
most difficult of their tasks was the derivation of the second law of thermodynam-
ics. What is the molecular basis for the property that the entropy of an isolated
system strives toward a maximum as the system moves toward equilibrium? A
survey of the contributions to this problem by Einstein's predecessors as well as
by Einstein himself is presented in (4). In those early days, Einstein was not the
only one to underestimate the mathematical care that this very complex problem
rightfully deserves. When Einstein did this work, his knowledge of the funda-
mental contributions by Boltzmann was fragmentary, his ignorance of Gibbs'
papers complete. This does not make any easier the task of ascertaining the merits
of his contributions.
To Einstein, the second problem was of deeper interest than the first. As he
said later, Brownian motion was important as a method for counting particles, but
far more important because it enables us to demonstrate the reality of those
motions we call heat, simply by looking into a microscope. On the whole, Ein-
stein's work on the second law has proved to be of less lasting value than his
investigations on the verification of the molecular hypothesis. Indeed, in 1911 he
wrote that he would probably not have published his papers of 1903 and 1904
had he been aware of Gibbs' work.
Nevertheless, Einstein's preoccupation with the fundamental questions of sta-
tistical mechanics was extremely vital since it led to his most important contri-
butions to the quantum theory. It is no accident that the term Boltzmann's prin-
ciple, coined by Einstein, appears for the first time in his March 1905 paper on
the light-quantum. In fact the light-quantum postulate itself grew out of a statis-
tical argument concerning the equilibrium properties of radiation (19c). It should