622 REPORT—1899. 
Perhaps I may illustrate the danger in the use of the conception of the ether 
by considering the common way of describing the electro-magnetic waves, which 
are all about us here, as ether waves. Now in all cases with which we are 
acquainted, these waves start from matter; their energy before starting was, as far 
as we can guess, energy of the matter between the different parts of the source, 
and they manifest themselves in the receiver as energy of matter. As they travel 
through the air, I believe that it is quite possible that the electric energy can be 
expressed in terms of the molecules of air in their path, that they are effecting 
atomic separations as they go. if so, then the air is quite as much concerned in 
their propagation as the ether between its molecules. In any case, to term them 
ether waves is to prejudge the question before we have sufficient evidence. 
Unless we bear in mind the hypothetical character of our mechanical concep- 
tion of things, we may run some risk of another danger—the danger of supposing 
that we have something more real in mechanical than in other measurements, 
For instance, there is some risk that the work measure of specific heat should be 
regarded as more fundamental than the heat measure, in that heat is truly a ‘ mode 
of motion.’ On the molecular hypothesis, heat is no doubt a mixture of kinetic 
energy and potential energy of the molecules and their constituents, and may even 
be entirely kinetic energy; and we may conceivably in the future make the 
hypothesis so definite that, when we heat a gramme of water 1°, we can assign 
such a fraction of an erg to each atom. But look how much pure hypothesis is 
here, The real superiority of the work measure of specific heat lies in the fact 
that it is independent of any particular substance, and there is nothing whatever 
hypothetical about it. 
Another illustration of the illegitimate use of our. hypothesis, as it appears to 
me, is in the attempt to find in the ether a iixed datum for the measurement of 
material velocities and accelerations, a something in which we can draw our co- 
ordinate axes so that they will never turn or bend. But this is as if, discontented 
with the movement of the earth’s pole, we should seek to find our zero lines of 
latitude and longitude in the Atlantic Ocean. Leaving out of sight the possibility 
of ethereal currents which we cannot detect, and the motions due to every ray of light 
which traverses space, we could only fix positions and directions in the ether by 
buoying them with matter. We know nothing of the ether, except by its effects 
on matter, and, after all, it would be the material buoys which would fix the 
positions and not the ether in which they float. 
The discussion of the physical method, with its descriptive laws and explana- 
tions, and its hypothetical extension of description, leads us on to the consideration 
of the limitation of its range. The method was developed in the study of matter 
which we describe as non-living, and with non-living matter the method has sufficed 
for the particular purposes of the physicist. Of course only a little corner of the 
universe has been explored, but in the study of non-living matter we have come to 
no impassable gulfs, no chasms across which we cannot throw bridges of 
hypothesis. Does the method equally suffice when it is applied to living matter ? 
Can we give a purely physical account of such matter, likening its motions and 
changes to other motions and changes already observed, and so explaining them ? 
1 This risk of imagining one particular kind of measure more real than another, 
more in accordance with the truth of things, may be further illustrated by the 
common idea that mass-acceleration is the only way to measure a force. We stand 
apart from our mechanical system and watch the motions and the accelerations of 
the various parts, and we find that mass-accelerations have a certain significance in 
our system. If we keep ourselves outside the system and only use our sense of 
sight, then mass-acceleration is the only way of describing that behaviour of one 
body in the presence of others which we term force on it. But if we go about in 
the system and pull and push bodies, we find that there is another conception of 
force, in which another sense than sight is concerned—another mode of measurement 
much more ancient and still far more extensively used—the measurement by weight 
supported. Each method has its own range; each is fundamental in that range. It 
is one of the great practical problems in physics to make the pendulum give us the 
exact ratio of the units in the two systems. 
a 
