as the one to be accepted until further investigation shall prove its unsoundness.
The second limitation has been the possible structure of an infinite universe. The mathematical reader will easily see that the conclusion that an infinite universe of stars would fill the heavens with a blaze of light, rests upon the hypothesis that every region of space of some great but finite extent is, on the average, occupied by at least one star. In other words, the hypothesis is that if we divide the total number of the stars by the number of cubic miles of space, we shall have a finite quotient. But an infinite universe can be imagined which does not fill this condition. Such will be the case with one constructed on the celebrated hypothesis of Lambert, propounded in the latter part of the last century. This author was an eminent mathematician, who seems to have been nearly unique in combining the mathematical and the speculative sides of astronomy. He assumed a universe constructed on an extension of the plan of the Solar System. The smallest system of bodies is composed of a planet with its satellites. We see a number of such systems, designated as the Terrestrial, the Martian (Mars and its satellites), the Jovian (Jupiter and its satellites), etc., all revolving round the Sun, and thus forming one greater system, the Solar System. Lambert extended the idea by supposing that a number of solar systems, each formed of a star with its revolving planets and satellites, were grouped into a yet greater system. A number of such groups form the great system which we call the galaxy, and which comprises all the stars we can see with the telescope. The more distant clusters may be other galaxies. All these systems again may revolve around some distant center, and so on to an indefinite extent. Such a universe, how far so ever it might extend, would fill the heavens with a blaze of light, and the more distant galaxies might remain forever invisible to us. But modern developments show that there is no scientific basis for this conception, attractive though it is by its grandeur.
So far as our present light goes, we must conclude that although we are unable to set absolute bounds to the universe, yet the great mass of stars is included within a limited space of whose extent we have as yet no evidence. Outside of this space there may be scattered stars or invisible systems. But if these systems exist, they are distinct from our own.
The second question, that of the arrangement of the stars in space, is one on which it is equally difficult to propound a definite general conclusion. So far, we have only a large mass of faint indications, based on researches which cannot be satisfactorily completed until great additions are made to our fund of knowledge.
A century ago Sir William Herschel reached the conclusion that our universe was composed of a comparatively thin but widely ex-
