It is well to consider for a moment what these facts mean. Professor S. Newcomb, one of the highest authorities on these problems, tells us that the long series of measurements to discover the parallax of Canopus, the brightest star in the southern hemisphere, would have shown a parallax of one-hundredth of a second, had such existed. Yet the results always seemed to converge to a mean of 0".000! Suppose, then, we assume the parallax of this star to be somewhat less than the hundredth of a second--let us say 1/125 of a second. At the distance this gives, light would take almost exactly 400 years to reach us, so that if we suppose this very brilliant star to be situated a little on this side of the Galaxy, we must give to that great luminous circle of stars a distance of about 500 light years. We shall now perceive the advantage of being able to realise what a million really is. A person who had once seen a wall-space more than 100 feet long and 20 feet high completely covered with quarter-inch spots a quarter of an inch apart; and then tried to imagine every spot to be a mile long and to be placed end to end in one row, would form a very different conception of a million miles than those who almost daily _read_ of millions, but are quite unable to visualise even one of them. Having really seen one million, we can partially realise the velocity of light, which travels over this million miles in a little less than 5-1/2 seconds; and yet light takes more than 4-1/3 years at this inconceivable speed to come to us from the very _nearest_ of the stars. To realise this still more impressively, let us take the _distance_ of this nearest star, which is 26 _millions_ of _millions_ of miles. Let us look in imagination at this large and lofty hall covered from floor to ceiling with quarter-inch spots--only _one_ million. Let all these be imagined as miles. Then repeat this number of miles in a straight line, one after the other, as many times as there are spots in this hall; and even then you have reached only one twenty-sixth part of the distance to the nearest fixed star! This _million_ times a _million_ miles has to be repeated twenty-six times to reach the _nearest_ fixed star; and it seems probable that this gives us a good indication of the distance from each other of at least all the stars down to the sixth magnitude, perhaps even of a large number of the telescopic stars. But as we have found that the bright stars of the Milky Way must be at least one hundred times farther from us than these nearest stars, we have found what may be termed a minimum distance for that vast star-ring. It may be immensely farther, but it is hardly possible that it should be anything less. THE PROBABLE SIZE OF THE STARS