158 Mr. S. T. Preston on the Continuance of 



of volume), like the mean temperature, is the same through- 

 out — though indefinite fluctuations of dimensions would occur 

 from one mass to another, in harmony with the fluctuations 

 of velocity. 



It would further follow from the principle that molecules of 

 different densities (molecular weights) tend forcibly to become 

 uniformly diffused, that by an adequate past duration of the 

 universe the different kinds of matter must be uniformly dif- 

 fused (per unit of volume) by the continued interchange of 

 motion, though considerable fluctuations of mixture within 

 ranges less than a unit of volume would be possible in harmony 

 with the kinetic theory. 



It may be observed that in principle, in order to account 

 for the continuance of change in the universe, the existence 

 of some process of recurrence is absolutely essential. The 

 cooled down material of extinct suns must in some way be 

 made available for the development of fresh suns or centres of 

 heat. For if this were not the fact, there would be a continual 

 accumulation of the material of extinct or useless suns in the 

 universe, and processes of renewal and maintenance of the 

 activity of the universe would come to a deadlock in the ab- 

 sence of matter to operate upon. It seems inconceivable how 

 this end could be effected under recognized dynamical prin- 

 ciples, excepting through an exchange of motion going on 

 among the matter of the universe, involving collisions and an 

 alternating renewal and loss of heat : or this seems on broad 

 principle the only conceivable way in which there should be 

 recurrence, under the condition that the same matter should be 

 used again. There would appear to be a simple grandeur (not 

 out of harmony with the recognized characteristics of nature) 

 in this great result being brought about by the mere move- 

 ment of the stellar masses according to the kinetic theory. 

 Moreover the kinetic theory has been mathematically proved 

 (when a large number of masses are concerned) to produce a 

 system of order and symmetry (or mean similarity of the con- 

 ditions in all parts of the system) which is rigidly and auto- 

 matically maintained by a process of self-correction under 

 dynamical principles — a self-acting adjustment of the motions 

 continually taking place, whereby a system of harmony and 

 order is maintained everywhere, a perfect state of mobile equi- 

 librium existing in all parts. 



Those who are inclined to view the physical causation of 

 the past in the light of the physical causation of the present, 

 or who look upon the principle of the conservation of energy 

 as a truth as necessary in the past as in the present (or who 

 are disposed to regard physical truths as independent of time), 



