236 AKNUAL REPORT SMITHSONIAN INSTITUTION, 19 31 



In N. G. C. 3115 we saw the raw material of the process — a gaseous 

 mass of extreme tenuity, already molded, as a result of shrinkage and 

 consequent increase of rotation, to the stage at which disintegration 

 is about to commence. Further shrinkage takes place, and in N. G. C. 

 5866 and 4594 we see the ejection of the jets of matter from which 

 the future stars will in due course be made. In N. G. C. 891 and 

 M. 51 individual stars are beginning to form, although at present 

 only as vague condensations in what is still a continuous nebula 

 mass. Finally, each condensation forms a separate star, until the 

 whole nebula is transformed into a star cloud. Thus the great 

 nebulae prove to be the birth places of the stars. 



Long before this complete evolutionary sequence was known, I 

 had taken a preliminary step in the reverse direction, and had shown 

 that the stars had in all probability been born out of a uniform mass 

 of tenuous gas by a process which I designated " gravitational in- 

 stability." If all the matter of our own system of stars were uni- 

 formly sjDread throughout the space occupied by the system, it would 

 form a gas of density about 10"-^. 



I showed that such a medium would be unstable, and that its in- 

 stability would cause it to break up into condensations whose dis- 

 tances apart could be calculated mathematically, which calculation 

 showed that these distances would be about equal to the actual aver- 

 age distance of the stars. Thus the single supposition that the stars 

 had been born out of a uniformly spread mass of gas was found to 

 explain at a single stroke why the stars all have approximately the 

 same mass, and why these masses are what they are. 



A similar situation has recently arisen with respect to the nebulae. 

 In a telescope they appear to differ widely in shape, size, and bright- 

 ness. But Doctor Hubble has shown that differences in size and 

 brightness between nebulae of the same shape are almost entirely 

 due to a distance effect. If all the nebulae were put in a row at the 

 same distance from us, nebulae of the same shape would be found 

 to have approximately the same dimensions and luminosity, while 

 even nebulae of different shapes would exhibit only comparatively 

 small ranges of dimensions and luminosity, especially the latter. 



Because of this, it is possible to estimate the distances of all nebu- 

 lae, even the very faintest, with fair accuracy; their faintness gives 

 a measure of their distance. The faintest which can be observed 

 photographically in the 100-inch telescope prove to be at the amaz- 

 ing distance of about 140,000,000 light-years. Some 2,000,000 nebu- 

 lae lie within this distance. 



Doctor Hubble finds that these are fairly uniformly spaced at an 

 average distance of about 1,800,000 light-years apart. To construct 

 a model, we may take 300 tons of apples and space them at about 10 

 yards part, thus filling a sphere of about a mile diameter. This 



