1878.] Past Changes in the Universe. 361 
recognised atflion of physical causation be as desirable as in 
the case of geological changes ? 
One insuperable barrier in the way of explaining past 
changes in the Universe, by causes at present in operation, 
has been the supposition — in the absence of any explanation 
of the mechanism of gravity — that the range of gravity ex- 
tends to indefinite distances, and consequently that every 
portion of the Universe tends to agglomerate or collet 
together into one mass, whereby all conditions of stability 
or permanence are excluded. Another difficulty has been to 
conceive how the energy (in the form of heat) dissipated by 
matter in the ether of space could ever again be made 
available or re-transferred to matter. Now, in the first 
place, it has never been proved that the range of gravity 
extends to indefinite distances, and, on examination, con- 
clusions so incongruous are involved in this assumption 
that they themselves render the assumption unlikely. Thus 
imagine the incongruity of the idea of an unstable Universe 
tending to agglomerate into one (perhaps infinite) mass, 
gravity itself becoming (in the limit) infinite, so that the 
conditions of life would be rendered utterly impossible. 
Can it appear reasonable to conclude that physical causa- 
tion should be so adapted as inevitably (in time) to produce 
such a state of things as this, or to involve instability and 
confusion ? 
According to the only physical theory of gravitation which 
has yet withstood the test of criticism (viz., that of 
Le Sage*), and which has been received with some favour 
by some of the most eminent physicists, it is certain that 
the range of gravity cannot extend to indefinite distances. 
It is therefore quite possible that the range of gravity may 
not extend to the stellar distances so that the stars do not 
gravitate towards each other.f This would evidently be the 
condition required to satisfy the very important point of 
stability in the Universe. The range through which gravity 
has been actually observed to prevail is known to be 
infinitesimal compared with the stellar distances, so that 
the range of gravity may perfectly well lie within these 
limits. Thus the condition for stability is the first important 
* In three papers published in the Philosophical Magazine (September and 
November, 1877, and February, 1878) it has been my objedl to show how 
Le Sage’s original theory may be modified, and certain difficulties in the way 
of its acceptance (in its original form) removed by the light of the modern 
investigations conne&ed with the kinetic theory of gases. A fourth paper 
(after the MS. of the present paper was written) has also been published in 
the Philosophical Magazine for April last. 
f Of course we do not allude to double stars, in close range. 
