406 PROPER MOTIONS OF THE STARS. SECT. XXXVI. 



directed towards a point in a line joining the two stars p> and 

 * Herculis at a quarter of the apparent distance of these two 

 stars, reckoning from * Herculis. This investigation was founded 

 upon no law assumed or observed, such as the circulation of all 

 the stars of our firmament about a common centre, though 

 philosophers have speculated as to the probability of such a 

 motion in the sun and stars in the plane of the Milky Way. 

 Should the sun and his stellar companions be moving in a nearly 

 circular orbit, the centre of motion would be in the plane passing 

 through the sun perpendicular to the direction of his motion. 

 The constellations through which that great circle would pass 

 are Pisces, Australis, Pegasus, Andromeda, Perseus, &c. M. 

 Argelander is of opinion that the sun's orbit is nearly in the plane 

 of the Milky Way, and, therefore, that its centre must probably be 

 in Perseus, while M. Madler places it in the Pleiades, which seems 

 to be inadmissible; but the data are too uncertain at present to 

 admit of any absolute conclusion as to the sun's orbit and the 

 general motion of the stellar firmament : for though the stars in 

 every region of the sky tend towards a point in Hercules, it is not 

 yet known whether their motions are uniform or variable, whether 

 the sun's motion be gradually changing, and whether the stars 

 form different independent systems, each having its own centre 

 of attraction, or if all obey one powerful controlling force which 

 pervades the whole universe. Accurate observations of the places 

 of a select number of stars of all dimensions in the Milky Way 

 continued for a series of years would no doubt decide this point. 

 The proper motion of a star combined with the progressive 

 velocity of light alters the apparent periodic time of the revolving 

 star of a binary system. If the orbit of a double star be at right 

 angles to the visual ray, and both the sun and the star at rest, the 

 periodic time of the revolving star, say of 10,000 days, would 

 always be the same. But if the centre of gravity of the star 

 were to recede in a direct line from the sun with the velocity of 

 one tenth of the radius of the earth's orbit in a day, then at the 

 end of 10,000 days it would be more remote from us by 1000 of 

 such radii a space light would take 57 days to traverse : 

 hence, although the periodic time of the star would really be the 

 same, the completion of its period would only be known to us 57 

 days after it had taken place, so that the periodic time would 



