For the accuracy of the above statement the authority of Professor Newcomb is an ample guarantee; but there may be modifications required in the data on which it is founded, and these may greatly alter the result. If I do not mistake, the estimate of a hundred million stars is founded on actual counts or estimates of stars of successive magnitudes in different parts of the heavens, and it does not include either those of the denser star clusters nor the countless millions just beyond the reach of telescopes in the Milky Way. Neither does it make allowance for the dark stars supposed by some astronomers to be many times more numerous than the bright ones, nor for the vast number of the nebulæ, great and small, in calculating the total mass of the stellar system.[9] In his latest work Professor Newcomb says, 'The total number of stars is to be counted by hundreds of millions'; and hence the controlling power of the system on bodies within it will be many times greater than that given above, and might even be ample to retain within its bounds such a rapidly moving star as Arcturus, which is believed to be travelling at the rate of more than three hundred miles a second. But there is another very important limitation to the conclusions to be drawn from Professor Newcomb's calculation. It assumes the stars to be nearly uniformly distributed through the whole of the space to which the system extends. But the facts are very different. The existence of clusters, some of which comprise many thousands of stars, is one example of irregularity of distribution, and any one of these larger clusters would probably be able to change the course of even the swiftest stars passing near it. The larger nebulæ might have the same effect, since the late Mr. Ranyard, taking all his data so as to produce a minimum result, calculated the probable mass of the Orion nebula to be four and a half million times that of the sun, and there may be many other nebulæ equally large. But far more important is the fact of the vast ring of the Milky Way, which is now universally held by astronomers to be, not only apparently but really, more densely crowded with stars and also with vast masses of nebulous matter than any other part of the heavens, so that it may possibly comprise within itself a very large proportion of the whole of the matter of the visible universe. This is rendered more probable by the fact that the great majority of star-clusters lie along its course, most of the huge gaseous stars belong to it, while the occurrence there only of 'new stars' is evidence of a superabundance of matter in various forms leading to frequent heat-producing collisions, just as the frequent occurrence of meteoric showers on our earth is evidence of the superabundance of meteoric matter in the solar system.