idea a scientific basis for chemistry, the progress of that
subject has been wonderful beyond any conception that could
previously have been entertained; and the atomic theory in
some form appears to be an indispensable part of the framework
of physical science. Now this doctrine of material atoms is
an almost necessary corollary to the doctrine of a universal
aether. For if we held that matter is continuous, one of two
alternatives would be open. We might consider that matter and
aether can coexist in the same space; this would involve the
co-existence and interaction of a double set of properties,
introducing great complication, which would place any coherent
scheme of physical action probably beyond the powers of human
analysis. Or we might consider that aether exists only
where matter is not, thus making it a very rare and subtle
and elastic kind of matter; then we should have to assign
these very properties to the matter itself where it replaces
aether, in addition to its more familiar properties, and the
complication would remain. The other course is to consider
matter as formed of ultimate atoms, each the nucleus or core
of an intrinsic modification impressed on the siurounding
region of the aether; this might conceivably be of the
nature of vortical motion of a liquid round a ring-core,
thus giving a vortex atom, or of an intrinsic strain of some
sort radiating from a core, which would give an electric
atom. We recognize an atom only through its physical
activities, as manifested in its interactions with other atoms
at a distance from it; this field of physical activity would
be identical with the surrounding field of aethereal motion
or strain that is inseparably associated with the nucleus,
and is carried on along with it as it moves. Here then we
have the basis of a view in which there are not two media
to be considered, but one medium, homogeneous in essence and
differentiated as regards its parts only by the presence of
nuclei of intrinsic strain or motion---in which the physical
activities of matter are identified with those arising from
the atmospheres of modified aether which thus belong to its
atoms. As regards laws of general physical interactions,
the atom is fully represented by the constitution of this
atmosphere, and its nucleus may be left out of our discussions;
but in the problems of biology great tracts of invariable
correlations have to be dealt with, which seem hopelessly
more complex than any known or humanly possible physical
scheme. To make room for these we have to remember that the
atomic nucleus has remained entirely undefined and beyond
our problem; so that what may occur, say when two molecules
come into close relations, is outside physical science---not,
however, altogether outside, for we know that when the vital
nexus in any portion of matter is dissolved, the atoms
will remain, in their number, and their atmospheres, and all
inorganic relations, as they were before vitality supervened.
Nature of Properties of Material Bodies.---It thus appears
that the doctrine of atomic material constitution and the
doctrine of a universal aether stand to each other in a
relation of mutual support; if the scheme of physical laws
is to be as precise as observation and measurement appear to
make it, both doctrines are required in our efforts towards
synthesis. Our direct knowledge of matter can, however,
never be more than a rough knowledge of the general average
behaviour of its molecules; for the smallest material speck
that is sensible to our coarse perceptions contains myriads of
atoms. The properties of the most minute portion of matter
which we can examine are thus of the nature of averages.
We may gradually invent means of tracing more and more
closely the average drifts of translation or orientation,
or of changes of arrangement, of the atoms; but there will
always remain an unaveraged residue devoid of any recognized
regularity, which we can only estimate by its total amount.
Thus, if we are treating of energy, we can separate out
mechanical and electric and other constituents in it; and
there will be a residue of which we know nothing except its
quantity, and which we call thermal. This merely thermal
energy--which is gradually but very slowly being restricted
in amount as new subsidiary organized types become recognized
in it--though transmutable in equivalent quantities with the
other kinds, yet is so only to a limited extent; the tracing
out of the laws of this limitation belongs to the science of
thermodynamics. It is the business of that science to find
out what is the greatest amount of thermal energy that can
possibly be recoverable into organized kinds under given
circumstances. The discovery of definite laws in this
region might at first sight seem hopeless; but the argument
rests on an implied postulate of stability and continuity of
constitution of material substances, so that after a cycle
of transformations we expect to recover them again as they
were originally---on the postulate, in fact, that we do
not expect them to melt out of organized existence in our
hands. The laws of thermodynamics, including the fundamental
principle that a physical property, called temperature, can be
defined, which tends towards uniformity, are thus relations
between the properties of types of material bodies that
can exist permanently in presence of each other; why they
so maintain themselves remains unknown, but the fact gives
the point d'appui. The fundamental character of energy
in material systems here comes into view; if there were any
other independent scalar entity, besides mass and energy,
that pervaded them with relations of equivalence, we should
expect the existence of yet another set of pualities analogous
to those connected with temperature. (See ENERGETICS.)
Returning now to the aether, on our present point of view
no such complications there arise; it must be regarded as a
continuous uniform medium free from any complexities of atomic
aggregation, whose function is confined to the transmission of
the various types of physical effect between the portions of
matter. The problem of its constitution is thus one which
can be attacked and continually approximated to, and which
may possibly be definitely resolved. It has to be competent
to transmit the transverse waves of light and electricity,
and the other known radiant and electric actions; the way
in which this is done is now in the main known, though
there are still questions as to the mode of expression and
formulation of our knowledge, and also as regards points of
detail. This great advance, which is the result of the gradual
focussing of a century's work in the minute exploration of
the exact laws of optical and electric phenomena, clearly
carries with it deeper insight into the physical nature
of matter itself and its modes of inanimate interaction.
If we rest on the synthesis here described, the energy of the
matter, even the thermal part, appears largely as potential
energy of strain in the aether which interacts with the kinetic
energy associated with disturbances involving finite velocity of
matter. It may, however, be maintained that an ultimate
analysis would go deeper, and resolve all phenomena of elastic
resilience into consequences of the kinetic stability of steady
motional states, so that only motions, but not strains, would
remain. On such a view the aether might conceivably be a
perfect fluid, its fundamental property of elastic reaction
arising (as at one time suggested by Kelvin and G. F.
Fitzgerald) from a structure of tangled or interlaced vortex
filaments pervading its substance, which might conceivably
arrange themselves into a stable configuration and so resist
deformation. This raises the further question as to whether
the transmission of gravitation can be definitely recognized
among the properties of an ultimate medium; if so, we know
that it must be associated with some feature, perhaps very
deep-seated, or on the other hand perhaps depending simply on
incompressibility, which is not sensibly implicated in the
electric and optical activities. With reference to all such
further refinements of theory, it is to be borne in mind that
the perfect fluid of hydrodynamic analysis is not a merely
passive inert plenum; it is also a continuum with the
property that no finite internal slip or discontinuity of
motion can ever arise in it through any kind of disturbance;
and this property must be postulated, as it cannot be explained.
Motion of Material Atoms through the Aether.--An important
question arises whether, when a material body is moved through
the aether, the nucleus of each atom carries some of the
surrounding aether along with it; or whether it practically
only carries on its strain-form or physical atmosphere, which
is transferred from one portion of aether to another after
the manner of a shadow, or rather like a loose knot which can
slip along a rope without the rope being required to go with
it. We can obtain a pertinent illustration from the motion
of a vortex ring in a fluid; if the circular core of the ring
is thin compared with its diameter, and the vorticity is not
very great, it is the vortical state of motion that travels
across the fluid without transporting the latter bodily with
it except to a slight extent very close to the core. We might
thus examine a structure formed of an aggregation of very
thin vortex rings, which would move across the fluid without
sensibly disturbing it; on the other hand, if formed of stronger
vortices, it may transport the portion of the fluid that is
within, or adjacent to, its own structure along with it as
if it were a solid mass, and therefore also push aside the
surrounding fluid as it passes. The motion of the well-known
steady spherical vortex is an example of the latter case.
Convection of Optical Waves.--The nature of the motion,
if any, that is produced in the surrounding regions of
the aether by the translation of matter through it can be
investigated by optical experiment. The obvious body to take
in the first instance is the earth itself, which on account
of its annual orbital motion is travelling through space at
the rate of about 18 miles per second. If the surrounding
aether is thereby disturbed, the waves of light arriving
from the stars will partake of its movement; the ascertained
phenomena of the astronomical aberration of light show that
the rays travel to the observer, across this disturbed aether
near the earth, in straight lines. Again, we may split a
narrow beam of light by partial reflexion from a transparent
plate, and recombine the constituent beams after they have
traversed different circuits of nearly equivalent lengths,
so as to obtain interference fringes. The position of these
fringes will depend on the total retardation in time of the
one beam with respect to the other; and thus it might be
expected to vary with the direction of the earth's motion
relative to the apparatus. But it is found not to vary at
all, even up to the second order of the ratio of the earth's
velocity to that of light. It has in fact been found, with
the very great precision of which optical experiment is
capable, that all terrestrial optical phenomena--reflexion,
refraction, polarization linear and circular, diffraction
--are entirely unaffected by the direction of the earth's
motion, while the same result has recently been extended to
electrostatic forces; and this is our main experimental clue.
We pass on now to the theory. We shall make the natural
supposition that motion of the aether, say with velocity
