seconds in a year, and you will find that Sirius is more than 40 billion miles away using billion in our English way for a million million. (In France a billion means much less: they use it to mean only a thousand million.)
Now this is far enough in all conscience; but what are we to say of Canopus and Rigel and Spica? They are so far away that Sir David Gill could not find any "squint" or parallax at all, and had to write zero in the third column. If the "squint" had been as large as that for β Centauri, or even half as large, he could have measured it. These stars must be at least twice as far off as β Centauri, which is 100 "light years," or, say, 440 billion miles from us.
In the last column of the table I have calculated the actual sizes of the stars compared with α2 Centauri, assuming that their surfaces are all just as bright as that of α2 Centauri; which again is not a justifiable assumption, but we may venture to make it for the sake of illustration, and in order to bring out the next point. You will see that some are very big and some very small; they vary in size at least as much as dolls, I think. I never saw a doll more than four or five feet high say fifty inches; and I never saw one less than about half-an-inch; so that the biggest would be about 100 times the smallest. We may fairly say that Canopus must be at least 100 times the size of α2 Centauri.
Before we go further, let us think for a moment how astonishing it is that light should reach us from those vast distances at all. If the stars sent out light in only one direction as a searchlight does, we could understand it better; the rays of a search-
