lurs that not only forbid any practical attempt at identification on
emergence, but make it doubtful whether we can in any true sense call the
issuing path identical with the entering one. Otherwise the identity of
energy can be admitted to be only that kind of identity that could be
preserved by matter if its molecular structure did not exist. One who can
admit that this sort of identity is the same sort that can be preserved by
molecular matter may be able to hold the identity of energy in the present
state of the evidence, but the present attitude of physicists would seem
to show that, whether they realize the connection of the two subjects or
not, they cannot take this view. In other words, modern views of the
identity of matter seem closely connected with modern views of its
structure, and the same connection will doubtless hold good for energy.

Regarding the probable success of an attempt to prove that energy has a
"structure" analogous to the molecular structure of matter, any prediction
would doubtless be rash just now. The writer has been unable, up to the
present time, to disprove the proposition, but the subject is one of
corresponding importance to that of the whole molecular theory of matter
and should not be entered upon lightly.

* * * * *

The writer freely acknowledges at present that the illustrations in the
foregoing are badly chosen and some of the statements are too strong, but
it still represents essentially his ideas on the subject. No reputable
scientific journal would undertake to publish it. The paper was then sent
to Prof. J. Willard Gibbs of Yale, and elicited the following letter from
him:

"NEW HAVEN, JUNE 2, 1897.

"MY DEAR MR. BOSTWICK:

"I regret that I have allowed your letter to lie so long
unanswered. It was in fact not very easy to answer, and when one
lays a letter aside to answer, the weeks slip away very fast.

"I do not think that you state the matter quite right in regard to
the mixture of fluids if they were continuous. The mixing of water
as I regard it would be like this, if it were continuous and not
molecular. Suppose you should take strips of white and red glass
and heat them until soft and twist them together. Keep on drawing
them out and doubling them up and twisting them together. It would
soon require a microscope to distinguish the red and white glass,
which would be drawn out i