January 6, 1923] 
NATURE 19 

There can be little doubt that large radiation-pressure, 
tending to overcome gravity, conduces to instability, 
so that larger masses have small chance of survival. 
Somewhere about one-half the sun’s mass the radiation- 
pressure no longer counts seriously, so that there is no 
tendency for the primitive material to break into 
smaller units. 
The existence of radioactive minerals on the earth 
seems to supply another reason for believing that its 
material was originally subjected to high temperature 
or to physical conditions of a different order from 
those now prevailing. In radioactivity we see a 
mechanism running down which must at some time 
have been wound up. Without entering into any 
details, it would seem clear that the winding-up process 
must have occurred under physical conditions vastly 
different from those in which we now observe only a 
running-down. The only possible guess seems to be 
that the winding-up is part of the general brewing of 
material which occurs under the intense heat in the 
interior of the stars. 
The trend of this argument has been against the 
Chamberlin-Moulton hypothesis and in favour of some 
form of nebular origin of the solar system. It is, of 
course, accepted that the details of the original nebular 
hypothesis of Laplace require modification. Also the 
word nebula is meant to signify diffuse gaseous material 
in general, and has no immediate connexion with 
those objects which we see in the sky, and call nebula 
more particularly. There is still controversy as to 
what process of evolution is represented by the spiral 
nebule which are seen in such numbers—what they 
will ultimately turn into; but the controversy is 
whether the spiral nebula will give rise to a cluster 
of a few hundred stars, or whether it will turn into a 
stellar universe on the same scale as the great system 
of some thousands of millions of stars which forms 
our galactic system. There is now no suggestion that 
it has anything to do with the formation of so in- 
significant a system as the solar system. But in 
preferring the nebular hypothesis to that of Chamberlin 
and Moulton, it is necessary to make a certain reserva- . 
tion. We have hitherto taken it for granted that the 
formation of a system of planets is a normal feature 
of the evolution of a star. Most of my arguments 
have referred to the development of stars in general, 
and would become irrelevant if it could be admitted 
that the solar system were an exceptional formation 
violating ordinary expectation. 
We know that at least a third of the stars are double 
stars, and I do not think there is any reason to think 
that planetary systems would be formed when the 
evolution takes that course; but until recently it 
was taken for granted that the remaining single stars 
would generally (or at least frequently) i the rulers 
of systems of planets. Jeans has recently pitched a 
bombshell into the camp, suggesting that the solar 
system is a freak system—the result of a rare accident, 
which could only happen to one star out of a very 
large number. He found no way of accounting for it 
as a normal process. I have not the specialist know- 
ledge necessary to criticise the details of the working 
of the nebular or of the planetismal theory of develop- 
ment, but before regarding Jeans’s argument as con- 
clusive (he himself makes reservations) I should be 
NO. 2775, VOL. 111] 
more satisfied if the effect of radiation-pressure had 
been taken into account. It is fairly clear that radia- 
tion-pressure plays a great part in the separation of 
nebulous matter into stars, and although I have no 
definite reason to think that it can account for the 
separation of planets from the sun, I do not feel satisfied 
that we have got at the whole truth until that point 
has been duly examined. : 
Supposing, however, that we are forced to accept 
Jeans’s suggestion that the solar system is a freak 
system, some of my objections to the Chamberlin- 
Moulton hypothesis are removed. I cannot admit that 
the conditions of collision which that hypothesis 
requires are normal features in the formation of stars ; 
but they might have happened occasionally in the 
history of the universe, and produced the solar system, 
the sun being thus as an exceptional star born out of 
due time. But if my arguments against Chamberlin’s 
hypothesis fall to the ground, there are probably 
other astronomers prepared to attack it in other 
directions. 
The new views as to the age of the earth are now 
pretty well known to geologists. I may sum them up 
briefly in the statement that Lord Kelvin’s estimate 
of the extent of geological time need not now be taken 
any more seriously than Archbishop Ussher’s, and that 
the geologist may claim anything up to 10,000 million 
years without provoking a murmur from astronomers. 
Although there may still be some difficulties about 
the exact source from which ‘the vast heat-energy 
the stars pour out into space is derived, it is now 
clear that the Helmholtz contraction-theory is inade- 
quate to give the necessary supply. The astronomer 
has no such precise means of measuring geological 
time as the physicist has now discovered by the analysis 
of radioactive minerals; but he can add his contri- 
butory evidence that the sun, and presumably there- 
fore the earth, is much older than Lord Kelvin allowed. 
In the Cepheid variable stars it seems possible to 
measure the actual rate at which evolution is pro- 
ceeding—the rate at which the star is condensing from 
a diffused state to a denser state. The star is believed 
to be pulsating, and as it expands and contracts the 
light varies in quantity and character. In a pulsating 
gravitating mass the period is proportional to the 
inverse square root of the density, so that by observing 
the rate at which the period is changing we can deduce 
the rate at which the density is changing. I may add 
that the law that the period depends on the inverse 
square root of the density is very closely confirmed by 
comparing the values for the various Cepheids. In 
this way we find that for the best observed of these 
stars, 6 Cephei, the density is changing 500 times 
slower than the contraction hypothesis assumes. It 
would, of course, be risky to assume that the same 
proportion holds at all stages of the evolution of a 
star; but it suggests that Lord Kelvin’s estimate of 
20 million years for the age of the sun might well be 
multiplied by 500 to give 10,000 million years. At 
any rate, the Cepheid observations show that the stars 
must have some other source of energy besides con- 
traction. 
I suppose it must be a matter of interest to geo- 
logists whether the intensity of the sun’s heat has been 
constant or whether it was at one time hotter than 
