No. 2.

On a Possible Origin of the Rotation of the Solar System.

If two bodies come into collision from rest at the limit of effective attraction, or have a proper motion directly towards each other, the collision may be complete, and there appears to be but little reason to expect a rotation in the nebula produced. But if two bodies have a proper motion in space, and come within one another's influence, a complete collision is very unlikely. If the bodies are email, and the proper motion large, the chances of a collision are infinitesimal; they will almost certainly pass each other. The smaller the proper motion and larger the body, the greater is the chance of a collision. If a collision occurs, it is reasonable to assume that a partial impact will be much more likely than complete impact. Complete impact of two equal bodies is the extreme case. Hence, seeing that all the bodies in the universe have a considerable proper motion, we must expect that almost all collisions will be cases of partial impact.

In every case of partial impact it appears that rotation in the coalesced mass is an absolutely necessary consequence. Let us suppose two equal bodies coming into collision in such a manner that the circumference of each passes through the other's centre, and let us suppose those parts coming into collision to coalesce, and the other parts to pass on in space. I will disregard the question of their escape or return, but will only consider the coalesced mass. The chief bulk of each side of this mass will be made up the one side from the one body, and the other side from the other body, because the section through the centre is both the thickest section and also the densest. Therefore, although the body will have its motion in space destroyed, yet on each side of the mats the momentum will not be balanced, and the resultant momentum will be in different directions on the opposite sides of the mass, owing to the motion of the colliding bodies being opposite. Thus a couple is produced, which will tend to spin the mass on its centre. The two rotating masses will also each attract chiefly one side of the coalesced body, which will tend to produce rotation in the same direction. It thus appears that the result of a partial collision is to produce two retreating pieces and a coalesced spindle-shaped mass, which mass has a tendency to rotation, in a direction such that the ends of the spindle lies in the circumference of the circle, of which the centre is the axis of rotation.