(October, 1912. 



KNOWLKDGE. 



iiiiinher of stars, each mo\iti,i,' onward in a straight 

 line and in a direction wiiich. with rare exceptions, 

 has nothing in common with the motion of an\- 

 other star." 



Suppose that we substitute for this the following 

 statement : (Supposition h) There exist scattered 

 through space, in every direction, millions of stellar 

 aggregations, each of which is a galaxy, liach 

 galaxy is composed of many millions of stars niov ing 

 for the most part in definite star-streams with 

 velocities which increase with their age, .\s viewed 

 from a given point within our Galaxy, the motions 

 of the nearer stars may be grouped in a small 

 number of prevalent drifts which may he under the 

 gravitational control of great central masses, or 

 highly-condensed clusters of stars, but the local 

 drifts do not necessarily coincide with the main 

 drifts of the dominant streams. 



This supposition puts a very different interpre- 

 tation on the facts of observation, and ma\- 

 reconcile difficulties which seem, at first sight, 

 almost insuperable, 



Newcomb gives in his work on " The Stars " a 

 very accurate anahsis of probabilities as applied to 

 the nearer stars on supposition a. If we replace 

 this criterion by supposition b, we have a less rigid 

 rule, and some of the former conclusions are abro- 

 gated, or can be explained differentls'. For example, 

 the percentage of negative apical motions, derived 

 from the Bradley-.Auwers stars, is found to increase 

 as the cross-motion diminishes. Newcomb concluded 

 that " this arises from the fact that in the case of 

 the nearer stars the apical motions are necessarilj- 

 larger, whether positive or negative." If, however, 

 we suppose that not much beyond the general sphere 

 of the nearer stars there are great star-streams 

 moving nearly parallel to the Sun's way, there ought 

 to be a large increase of negative apical motions on 

 approaching the confines of the streams, together 

 with a cessation of cross-motion. 



The determination of the general motions of 

 the Galaxy requires a knowledge of the trend of 

 myriads of stars composing its major streams. The 

 movements of a few stars, or even of a few thousand 

 stars in several local clusters, or subordinate streams, 

 are obviously inadequate to characterize the grand 

 trend of the formation as a whole ; but the greater 

 the number of available stars the nearer we shall 

 come to the ideal, especially if the phenomena of 

 star-drift are studied. Professor Boss has made a 

 slight move in this direction. In his first paper on 

 " Precession and Solar Motion,"* he says of certain 

 " closely accordant proper-motions : Each of these 

 was condensed into one mean star representing the 

 whole, except in the case of the moving cluster in 

 Taurus, where four stars taken at random were 

 employed as representative of the entire forty-one 

 stars therein." To be entirely consistent, he should 

 have classified the remaining stars hv drifts, substi- 



tuting for each drift one, or at most a very few mean 

 representative stars. This he has not done, but has 

 preferred the ordinary hypothesis that the peculiar 

 motions of the stars are at random. In this way he 

 has reached the conclusion that the mean motion 

 of three thousand five hundred and fort\-nine stars 

 of sixth magnitude, or brighter, having annual proper 

 motions less than 0"-20, isMo=t)"-0566, and that the 

 parallactic motion at 90° from the solar apex is 

 M=0"-0399; while, for five hundred and fiftv-nine 

 stars having annual motions between 0"-17 and 

 0"-80, Mo = 0"-319andM = 0"-2158. That the ratio 

 fjio/M is nearly the same in either case testifies to a 

 similar distribution of the stars of either selection 

 among the several drifts, but does not disprove the 

 existence of diverse drifts. Consequenth', the variation 

 of average proper motion in the two groups presum- 

 ably signifies a real variation in the mean distances 

 in the proportion of six to one, which is another 

 testimony to the great e.xtent of the local drifts. 



When Professor Boss grouped his grand total of 

 five thousand four hundred and thirteen stars 

 according to their galactic latitude, a very different 

 result was obtained. The ratio /ji-JM was found to 

 vary as follows : — 



Here we have a demonstration that the motions 

 of the lucid stars are in some way associated with 

 the direction of the galactic plane. It may be that 

 most of the stars are moving in planes parallel to the 

 galactic plane, and that the stars in low galactic 

 latitudes have their motions foreshortened, as 

 Professor Boss suggests : or it ma}' turn out that 

 the actual linear velocities diminish as the Galaxy 

 is approached : but the main point is that the 

 Galax}- is not so far away as to be out of touch 

 with these stars. 



The transformation of the proper motions 

 into linear velocities will depend upon the value 

 attached to measures of parallax, and thus upon the 

 computer's point of view. The value of the mean 

 motion for three thousand five hundred and forty- 

 nine stars of mean magnitude 5-2 was |Uo=: 0"-0566, 

 and that for five thousand four hundred and thirteen 

 stars of mean magnitude 5-7 was /*„ ^ 0"- 05.58, or 

 a change of barely 0"-003 for a variation of half a 

 magnitude. This, like other analyses of considerable 

 numbers of stars, evidences only a slight connection 

 between stellar brightness and proper motion. 



One other piece of evidence will complete the 

 sources from which we can draw our conclusion. 

 It is known that the radial inotions of the helium 

 stars are exceptionally small. The fact was brought 

 out by Frost and Adams, ^ and by Kapte\n and 



■■'■ Astronomical Journal. No. 612, page 99, .April 25th, 1910. 

 i '■ Radial Velocities of Twenty Stars." Publications of the Yerkes Observatory. Vol. 11, pages 143 to 250. 



