DEPARTMENT OF aiERIDIAN ASTROMETRY. 1 59 



right-ascension 90° in the Milky Way) and in the opposite direction have a 

 greater velocity than those at right angles to this direction in the approxi- 

 mate ratio of 7 to 4. This result explains and makes more definite the 

 phenomenon indicated in the researches by Kapteyn some years ago, on the 

 basis of which he propounded the hypothesis that the universe of stars con- 

 sists essentially of two intermingled streams moving in opposite directions, 

 each containing substantially like mixtures of stars of the various types. 

 Our researches brought out the fact that stars of the B type seem to have no 

 special tendency toward larger mean motions in the preferential directions 

 already indicated in the foregoing. This second branch of our recent re- 

 searches, owing to unavoidable delays, has not yet been published, but the 

 details have been prepared in full and have been preserved on our records 

 since May 191 1. From these it seems as if the preference of the larger 

 motions for a given direction does not exist for the younger type of stars, B, 

 but that it has gradually been impressed upon the stars with increasing age. 



I have advanced the hypothesis that the acceleration of the motions, acting 

 in a manner which seems to be selective on the basis of the physical condi- 

 tion of the stars, may have been due to repulsive rather than gravitational 

 forces, and that the preference for a given direction may have been due to a 

 kind of polarity in the action of these forces making them stronger in the 

 given direction to which there is a tendency of the larger motions. It must 

 be admitted that this explanation is speculative, but something like this is 

 apparently rendered necessary by the phenomena which have been already 

 demonstrated as having an existence, and they seem to be reasonable from 

 the analogy of repulsive action upon the tails of comets. 



In working out these results, my assistants. H. Raymond and Benjamin 

 Boss, have taken a part that justifies their admission, in some measure, as 

 joint authors, to whatever of credit attaches to results so important. 



The work has suggested extension of these researches in various direc- 

 tions, and strongly indicated the advantages which are likely to result from 

 a large accession of results for proper-motion, such as the projected works 

 of the Department of Meridian Astrometry are likely to supply. But a great 

 increase in the measurements of radial velocities is likewise demanded, and 

 a further extension of the classification of spectral types covering stars of 

 the seventh and even some stars of fainter magnitudes is alm.ost equally 

 indispensable for a full elucidation of the peculiarities of systematic motion 

 already indicated. 



The return of the stafif of San Luis was preceded by the dispatch to Al- 

 bany of the instrumental record of the meridian-observations, of which there 

 were 87,000 in round numbers. This has afforded the opportunity to begin 

 the systematic reduction of the observations. The constants for reduction 

 in right-ascension except azimuth are completed and tabulated throughout 

 from April 1909 to January 191 1. Much progress has been made in their 

 application to transits of standard stars. 



