368 Past Changes in the Universe . [July? 
These considerations would point to the general con- 
clusion that recurring changes take place in the Universe 
whereby the continuance of useful activity and life is 
insured, and the deadlock of useless inadbion and uniform 
repose prevented. In other words, they would lead to the 
inference that physical causation is not so constituted as to 
defeat itself and bring the operations of the Universe to a 
standstill. 
It should be noted that, as a matter of theoretic principle, 
the problem of recurring changes in the Universe undoubt- 
edly admits of solution, since it is an admitted dynamical 
principle that masses immersed in media whose particles 
are in a state of translatory motion must themselves inevi- 
tably acquire some translatory motion, and the stellar 
masses moving freely in straight lines in the gravific medium 
represent (as it were) a larger scale gas immersed in a 
smaller scale gas (viz., the gravific medium). It is a mere 
question of scale, and dynamical principles are (admittedly) 
independent of scale. The difficulty, if any, would there- 
fore be one of degree, not of principle. There might be a 
difficulty in accounting for the high value of the translatory 
motion of the stellar masses that observation appears to 
point to. But it should be noted that it does not follow 
that the value of the translatory motion we observe is the 
true mean value ; for it might be exceptionally above it. 
For it is known that the translatory motion of the molecules 
of gases (or of any bodies moving freely among each other 
according to kinetic theory) varies from zero towards infinity. 
It might therefore well be that the translatory motion we 
observe in our immediate neighbourhood might be exception- 
ally above the mean value, and we happen to observe this 
particular value because it is suited to the conditions of life 
(or the low translatory motion, on account of the feeble 
heat developed, would not be so suited).* 
* Possibly there may be some supplementary cause tending to produce trans- 
latory motion in the stars. The spectroscope shows the molecules of matter 
to possess a considerable complexity, or their parts have a considerable capa- 
city for taking up motion. It would appear reasonable to conclude that the 
molecules of matter of the Universe, immersed in a medium constituted 
according to the kinetic theory, must acquire also in their parts a certain 
degree of motion, owing to the dynamic aCtion of the impinging particles of 
the medium in which they are immersed. The intensity of motion thus 
acquired would (as is known) be greater in proportion as the parts of the 
molecule which are capable of motion are smaller; and this motion of the 
parts of the molecules would apparently tend to produce translatory motion 
in the molecules as wholes — much as the development of motion in the parts 
of the molecules of a gas (as occurs when they are exposed to the pulsations 
of waves of heat) is known to produce translatory motion in the molecules as 
wholes, thereby producing expansion in the volume of gas. Indeed a balance 
