3 3 6 D O C . 3 3 2 O N T H E E T H E R
this can be easily concluded out of the so-called Lorentz contraction. Consequent-
ly, as in dynamics, the geometry of bodies is co-determined by the ether.
The general theory of relativity removes a trouble spot in classical dynamics:
According to the latter, inertia and gravity appear as entirely different, mutually in-
dependent phenomena, despite both being defined by the same body-constant, the
mass. Relativity theory overcomes this deficiency in that it establishes the dynam-
ical behavior of the electrically neutral mass-point by the law governing the ge-
odetic line, in which the inertial and gravitational effects are no longer kept apart.
It thereby adds characteristics to the ether that are variable from point to point and
determine the metric and the dynamic behavior of material points. They, for their
part, are defined by physical factors, namely, by the distribution of mass or energy.
The ether of the general theory of relativity, consequently, differs from the one of
classical mechanics, i.e., the special theory of relativity, in that it is not “absolute”;
its local variable properties are rather determined by the ponderable matter. This
determination is complete if the world is spatially finite and closed. It is more
characteristic[14]
of the mathematical form of this theory than of its physical con-
tent that, in the general theory of relativity, there are no preferred space-time coor-
dinates uniquely linked with the metric.
In implementing the formal apparatus of the general theory of relativity, it was
not possible to attribute all the mass inertia to electromagnetic fields or even to
fields generally. Neither, in my view, have we until now gone beyond a superficial
inclusion of electromagnetic forces into the scheme of the general theory of rela-
tivity. The metric tensor codetermining the effects of gravitation and inertia, on one
hand, and the tensor of the electromagnetic field, on the other hand, seem now, as
before, to be essentially different expressions for the ether’s state, whose logical in-
dependence one would surely be far more readily inclined to attribute to the imper-
fection of our theoretical framework than to a complex structure of reality.
Although Weyl and
Eddington[15]
did find by a generalization of Riemannian
geometry a mathematical system that appears to unite both types of fields under a
unified aspect, nevertheless, the simplest field laws that this theory yields seem to
me not to lead to advances in physical knowledge. Overall, it seems today that we
are much farther away from an understanding of the electromagnetic elementary
laws than seemed to be the case at the beginning of this century. As a basis for this
opinion, I would like to point out here briefly the problem of the magnetic, terres-
trial, and solar
fields[16]
as well as the problem of light quanta, problems which to
some extent concern the coarse structure and the fine structure of the electromag-
netic field, respectively.
The Earth and the Sun possess magnetic fields whose orientation and sense are
in approximate relation to the rotational axis of these celestial bodies. According
[p. 91]