yards, and three quarters of an inch per second, as that it should possess that particular characteristic implied in the absence of all movement whatsoever. It thus appears that, even in the absence of direct testimony on the subject, there is only one chance that the sun should be at rest, while the chances that it is not at rest are absolutely infinite. Under these circumstances rational beings will conclude that the sun is not at rest, and once we have admitted that the system is in motion, our next duty will be to discover the characteristics of that movement.
There are several different methods by which this problem has been investigated. They all lead to results which are in such substantial accordance that there can be no reasonable doubt that the difficult problem of the motion of the sun has been solved with considerable approximation.
It will be noted that in this inquiry there are two different problems which have to be considered. The first of these relates to the direction of the sun's movement, and the second to its velocity. These investigations have hitherto generally been conducted simultaneously. It is, however, now apparent that the most satisfactory solution of the problem is to be obtained by employing one of the methods for the determination of the direction of the sun's motion, and a quite different method for the determination of its velocity. I shall deal with the two branches of the subject consecutively.
The only method of learning the actual displacement of the solar system in space must be founded on the observation of bodies external to that system; at all events to the extent of not participating in the motion with which the system is animated. Of course the only such
