a perfect mixture of certain parts of A and B,
but this is simply saying that it is identical with what it is now, that
is, with itself, not with something that went before.

Probably no one now believes that water or any other kind of matter is a
continuum, but the bearing of what has been said may be seen when we
remember that this is precisely the present stage of our belief regarding
energy.

No one, so far as I know, has ventured to suggest what may be termed a
molecular theory of energy, a somewhat remarkable fact when we consider
the control now exercised over all thought in physics by molecular
theories of matter. While we now believe, for instance, that a material
body, say a crystal, can by no possibility increase continuously in mass,
but must do so step by step, the minimum mass of matter that can be added
being the molecule, we believe on the contrary that the energy possessed
by the same body can and may increase with absolutely perfect continuity,
being hampered by no such restriction.

It is not the purpose of this paper to discuss whether we have grounds for
belief that there is such a thing as a minimum quantity, or atom, of
energy, that does not separate into smaller parts, no matter what changes
it undergoes. Suffice it to say that there appears to be no _a priori_
absurdity in such an idea. At first sight both matter and energy appear
non-molecular in structure. But we have been forced to look upon the
gradual growth of a crystal as a step-by-step process, and we may some
day, by equally cogent considerations, be forced to regard the gradual
increase of energy of an accelerating body as also a step-by-step process,
although the discontinuity is as invisible to the eye in the latter case
as in the former.

Without following this out any farther, however, the point may be here
emphasized that it is hardly possible for one who, like the majority of
physicists, regards matter as molecular and energy as a continuum, to hold
the same ideas regarding the identity of the two. Efforts to show that
definite portions of energy, like definite portions of matter, retain
their identity have hitherto been made chiefly on the lines of a
demonstration that energy travels by definite and continuous paths in
space just as matter does. This is very well, but it would appear to be
necessary to supplement it with evidence to show that the lines
representing these paths do not form at their intersections continuous
b