no information with regard to the movement of the solar system. But when there is so large a number of stars distributed in such a way that we are entitled to assume that on the whole there is about as much motion in one direction as there is in exactly the opposite direction, then it becomes possible to eliminate the disturbing effects introduced by the fact that the stars are not themselves at rest. When we employ stars enough, we may proceed in the inquiry exactly as if the stars were individually at rest, and as if all the motion perceived could be attributed to the movement of the solar system. No doubt there would be a fallacy in this proceeding if it happened that there was a general drift of the stellar motions in one direction. If this was the case, then the methods now employed would attribute, and would falsely attribute, a movement to the solar system equal and opposite to that with which the sidereal system was animated. It is, however, found that by a judicious selection of the stars, we are able to preclude the possibility of such a consentaneous sidereal movement as is here contemplated.
It is obvious that a movement of the solar system towards one region in the heavens must be accompanied by a general diminution of the distances between the stars in that region and our solar system. On the other hand, the distances from the solar system to the stars at the opposite region of the celestial sphere will be generally increased by such a movement. If the stars were at rest, or if a sufficient number of stars were taken to eliminate the consequences of their individual motions, then the spectroscopic search for the movements of the solar system becomes very simple. It is only necessary to measure the rate at which the solar system alters its
