TWO objections, however, have been made to these hypothetical two swarms. It has been urged that the secondary swarm which we saw moving in a closed orbit round the primary one would soon spread out into a line along the orbit, so that there would always be some parts of it mixed up with the constituents of the parent swarm. That is a perfectly fair objection, supposing we are dealing with millions and billions of years, but I think that those who have made it do not know the history of astronomy. Let us take, for instance, the history of the November swarm which cuts the earth's orbit, so that in certain Novembers, generally about thirty-three years apart, we get this swarm of meteorites passing through our atmosphere, getting burnt out in that passage, and giving us one of the most magnificent sights which it is possible for mortals to see—a whole hemisphere of sky filled with shooting stars. Some of you may remember such a phenomenon as that in the year 1866, some of us are hoping to see the recurrence of it in 1899, for which we have not long to wait. But the fact that we only get this appearance every thirty-three years shows that, at all events, that swarm of meteorites to which the phenomena are due has not changed during our life-time—nay, it has not changed during the last thousand years, for man has known of that November swarm for more than a thousand years, and we have only known of the variability of Mira for 300 years; so that you see such an objection as that is entirely out of court, because it lacks the historical touch.