suggestion of a nebular constitution, but within a month or two
characteristic nebular lines began to appear, and in less than six
months the whole spectrum had been transformed to the nebular type. In
the mean time the shifting of the spectral lines indicated a
complication of rapid motions in several directions simultaneously.
These motions were estimated to amount to from one hundred to five
hundred miles per second.

The human mind is so constituted that it feels forced to seek an
explanation of so marvelous a phenomenon as this, even in the absence
of the data needed for a sound conclusion. The most natural
hypothesis, perhaps, is that of a collision. Such a catastrophe could
certainly happen. It has been shown, for instance, that in infinity of
time the earth is sure to be hit by a comet; in the same way it may be
asserted that, if no time limit is fixed, the sun is certain to run
against some obstacle in space, either another star, or a dense meteor
swarm, or one of the dark bodies which there is every reason to
believe abound around us. The consequences of such a collision are
easy to foretell, provided that we know the masses and the velocities
of the colliding bodies. In a preceding chapter we have discussed the
motions of the sun and stars, and have seen that they are so swift
that an encounter between any two of them could not but be disastrous.
But this is not all; for as soon as two stars approached within a few
million miles their speed would be enormously increased by their
reciprocal attractions and, if their motion was directed radially with
respect to their centers, they would come together with a crash that
would reduce them both to nebulous clouds. It is true that the chances
of such a ``head-on'' collision are relatively very small; two stars
approaching each other would most probably fall into closed orbits
around their common center of gravity. If there were a collision it