—- 
EVOLUTION OF THE UNIVERSE. 583 
To summatise, we see a twofold evolutionary process at work amongst 
the stars, both parts cf the process occurring in this same phenomenon of 
the nova. On the one hand we observe the stars to be collapsing and 
giving birth to white dwarfs, possibly also planetary nebulae and even 
double stars. On the other hand we see nebulous clouds driven off to 
space, there to reinforce the already existing cosmic cloud. Reinforce ? 
Possibly we have struck the real origin of the cosmic cloud '—possibly the 
cosmic cloud is the totality of the débris of the expelled atmospheres of 
novae and other hot stars. The universe is simultaneously condensing 
into hard blocks and evaporating to form a tenuous nebula. I repeat, 
this is what we observe. Sir James Jeans has suggested that the universe 
began as a widely extended nebula, which condensed into stars in virtue 
of gravitational instability. The picture I have attempted to draw is 
that of a concourse of stars on the one hand condensing into denser stars 
in virtue of instability under a waning light-pressure, and on the other 
hand simultaneously expelling their atmospheres and generating a nebula, 
Prof. W. DE SITTER. 
I have been asked to make a short contribution to this discussion 
about the evolution of the universe. I hope you will find it short, although 
its length will, I fear, exceed the diameter of the eatth’s orbit round the 
sun. This way of expressing a time in a unit of length is perhaps some- 
what unusual. To have a distance expressed in a unit of time is, of course, 
very familiar. A lightyear has become an accepted unit of distance. I 
have purposely used the inverse method to call attention to the fact that 
the corresponding units of time and of space are so very widely different 
in relation to actual phenomena. The diameter of the earth’s orbit 
enormously exceeds all lengths that we ever come across in our daily life, 
whilst a quarter of an hour is considered a short interval. Similarly, a 
thousand million lightyears is a long distance, in fact many times greater 
than any actual distance that we have certain cognisance of, but a 
thousand million years is a short time in the evolution of the universe, 
It is only a third or a quarter of the accepted age of the earth, and I do 
not think geophysicists will be ready to take off even one single zero. We 
believe, and, so far as I can see, on good grounds, that the age of the sun 
and the majority of the stars is at least a thousand times greater. The 
time needed for the development of a double star, or even a quadruple or 
sextuple system, and that requited for the evolution of a stellar system 
or spiral nebula are at least of the same order. In the farthest galaxies 
that we can observe things appear to be very much the same as in our 
immediate neighbourhood. There seems to be no indication that these 
far-away systems are in an appreciably earlier state of evolution than 
our own. Although, of course, the trustworthiness of this statement 
decreases very rapidly with the distance, still it seems to show that the 
lapse of time corresponding to the distance of these systems is only an 
entirely inappreciable fraction of their whole lifetime. All these con- 
siderations consistently lead to the conclusion that the time elapsed since 
1 This has been suggested also by Vorontsoy-Velyaminov, Observatory, August 1931, 
and by others. 
