three times greater. According to the best judgment we can form,
our solar system is situate near the central region of the girdle,
so that the latter must be distant from us by half its diameter.
It follows that if we can imagine a gigantic pair of compasses, of
which the points extend from us to Alpha Centauri, the nearest
star, we should have to measure out at least five hundred spaces
with the compass, and perhaps even one thousand or more, to reach
the region of the Milky Way.

With this we have to connect another curious fact. Of eighteen new
stars which have been observed to blaze forth during the last four
hundred years, all are in the region of the Milky Way. This seems
to show that, as a rule, they belong to the Milky Way. Accepting
this very plausible conclusion, the new star in Perseus must have
been more than five hundred times as far as the nearest fixed
star. We know that it takes light four years to reach us from
Alpha Centauri. It follows that the new star was at a distance
through which light would require more than two thousand years to
travel, and quite likely a time two or three times this. It
requires only the most elementary ideas of geometry to see that if
we suppose a ray of light to shoot from a star at such a distance
in a direction perpendicular to the line of sight from us to the
star, we can compute how fast the ray would seem to us to travel.
Granting the distance to be only two thousand light years, the
apparent size of the sphere around the star which the light would
fill at the end of one year after the explosion would be that of a
coin seen at a distance of two thousand times its radius, or one
thousand times its diameter--say, a five-cent piece at the
distance of sixty feet. But, as a matter of fact, the nebulous
illumination expanded with a velocity from ten to twenty times as