AN" EVOLVING UNIVERSE — JEANS 233 



It is natural, then, to interpret our sequence of nebulae as one of 

 bodies which are rotating at different speeds. And as we know that 

 the speed of rotation of a body increases as it shrinks, we may rea- 

 sonably conjecture that this sequence of nebulae corresponds to dif- 

 ferent stages of development. At the one end, we have the globular 

 fuzzy mass of gas with little or no rotation; at the other end, we 

 have the flat cart-wheel shape in which rotation predominates and 

 governs the structure of the whole mass. A satisfactory confirma- 

 tion of this is to be found in the fact that a number of these flat 

 nebulae have been observed to be in a state of rapid rotation. 



Now before Doctor Hubble had arranged the nebulae in sequence 

 in the way I have described, I had tried to work out, as a problem of 

 abstract mathematics, the sequence of configurations which a mass 

 of rotating gas would assume as it cooled and shrank and, as a con- 

 sequence, increased its speed of rotation. I arrived at a sequence 

 of shapes which agreed almost exactly with that which Doctor Hub- 

 ble subsequently found when he arranged the observed nebulae in 

 sequence guided solely by the facts of observation, and deliberately 

 putting theoretical considerations out of his mind. This leaves little 

 room for doubt that the nebulae we see in the sky are members of 

 this theoretical sequence, that they began as rotating masses of gas, 

 and that we see them in various stages of development. 



If a rotating mass consists of water or some entirely incompressi- 

 ble substance, an increase in the speed of its rotation merely in- 

 creases its flatness. But compressibility of substance, such as comes 

 into play with a gaseous nebula, introduces new features in addition 

 to flattening. 



At first the spinning mass simpl}^ flattens and assumes the shape 

 of an orange. After a time a new feature appears — a pronounced 

 bulge all round its equator. Finally this becomes so marked that 

 the equator is merely a sharp edge; the rotating mass has assumed 

 the shape of a double-convex lens as in N. G. C. 3115. 



This configuration forms a noteworthy landmark in the evolu- 

 tionary path of a nebula. Until it is reached, the effects of shrink- 

 age can be adjusted, and are adjusted, by a mere change of shape — in 

 spite of its reduced size, the rotating mass carries the same angular 

 momentum as before by the simple expedient of rotating more rap- 

 idly and bulging out its equator. But we find that this is no longer 

 possible when once this landmark has been passed. 



Further shrinkage now involves an actual break-up of the nebula. 

 This can no longer carry all its angular momentum as a single body ; 

 it is in the state of a fly-wheel which is rotating too fast for safety, 

 and it relieves the situation by the ejection of matter from its equa- 

 tor. This brings us to the type of configuration shown in N. G. C. 

 5866, 4594, and 891. 



