Makch 1, 1892.] 



KNOWLEDGE 



49 



list comprises mainly stars of the second magnitude, 

 which, for some unluiown reason, possess exceptionally 

 slow visible motions : and it excludes those faintly-luminous, 

 swift objects so frequently chosen as the subjects of 

 experiments for parallax. Hence, when spectrographic 

 measurements are extended to stars of the sixth and 

 seventh magnitudes, radial motions of hundreds of miles 

 a second may be brought to light. There seems no good 

 reason why they should not exist : for in our present 

 ignorance as to the dynamical relations of the stellar 

 crowd, we cannot with any confidence anticipate the 

 display by its members of preferences in the shaping of 

 their courses. Arcturus, then, which chances to have 

 nearly the whole of its tremendous speed directed ncmns 

 the line of sight, may prove to have opposite counterparts 

 in stars telescopically fixed, but rushing with scarcely less 

 rapid I'ml-on motions towards, or away from the earth. 

 The discovery, however, of such hypothetical objects will 

 demand perseverance or good fortune ; since no suspicion 

 can be raised beforehand of a peculiarity betrayed by no 

 visible symptom. 



If indeed there were reason to believe the stars combined 

 into a stable system, and wheeling unanimously round a 

 common centre, then their tangential might be expected to 

 prevail very greatly over their radial velocities, as measured 

 from a point anywhere in the neighbourhood of tliat centre. 

 Thus, the planetary movements would lie nearly altogether 

 across the line of sight of an observer on ^lercury ; and 

 the movements in the Saturnian system would be similarly 

 related to one stationed on Mimas. But the plan of stellar 

 organization bears no traceable resemblance to that 

 followed in the planetary and Saturnian systems ; so the 

 argument from analogy falls to the ground. Still, it is 

 imoortant to recall that the postulate of approximate 

 equality between the average values of different components 

 of stellar velocity, besides being flatly contradicted by our 

 present, certainly most imperfect, experience, really involves 

 some kind of theory, or negation of a theory, regarding 

 sidereal construction. 



This postulate has lately served for the foundation of 

 an attempt to solve the problem of mean stellar distance. 

 Dr. Kleiber (A.str. Xnclt., Xo. 3037) laid it down as a 

 principle not likely to be disputed, that, in a sufficient 

 number of casually distributed velocities, the sum of 

 movement projected in any given diiection cannot differ 

 very appreciably from the sum of movement projected in 

 any other ; and proceeded to compare the radial speed of 

 22 stars measured by Professor \'ogel with their proper 

 motions as determined by Dr. Auwers. The sum of these 

 last proved to be, in right ascension 4-83" (after due 

 multiplication' by the cosine of the declination), in declina- 

 tion 5'5f"; while the sum of the corresponding end-on 

 movements was 363 kilometres per second. Hence, taking 

 an arc of 4-83" to stand for so many kilometres as result 

 from multiplying 363 by the number of seconds in a year 

 (31,556,929), the mean parallax of the 22 stars comes out 

 0'065", implying a light journey of 50 years ; the proper 

 movements in declination giving similarly a parallax of 

 O'07-l" (light journey=44 years). The mean magnitude 

 of the stars made the subjects of this interesting, if not 

 entirely con viucmg, experiment is 1'8 ; so that its upshot is 

 in plausible agreement with Dr. Elkin's directly measured 

 mean parallax of 0-089" for 10 stars of the first magnitude. 

 Tlie real significance of stellar movements has yet to be 

 penetrated. Setting aside the accordant impress stamped 

 upon them by the sun's onward march, they seem almost 

 purely erratic. Certainly they obey no obvious stream of 

 tendency. They are executed indifferently in all planes, 

 and show no methodical graduation of velocities. They 



disregard the lie of the Milky Way, and refuse to allow a 

 ■• rotation-component ' with reference to it to be elicited 

 from them. Shall we then give ear to a "counsel of 

 despair," and assert that the "mighty maze'' is "without a 

 plan"? This would surely be rash ; despair always is rash. 

 Millenniums indeed are but as moments in the development 

 of the harmonizing idea which we expect to lay hold of and 

 apprehend in the course of a few generations of ephemeral 

 existence. But that such an idea is present — that the 

 stars do not cruise at random and rudderless, drifting at 

 the mercy of wind and tide, so to speak, through space — a 

 very little consideration suffices to show. 



The case, on a general view, seems to stand thus. The 

 recorded observations of stellar movements all, or nearly 

 all — for radial measurements are independent of distance — 

 refer to stars comparatively near the sun. These are 

 promiscuously distributed throughout the vast region 

 enclosed by the annulus of the Milky Way, yet after a 

 maimer not wholly irrespective of its structure. They 

 show, on the contrary, decided condensation towards the 

 plane of that stupendous collection, and form with it, 

 undoubtedly, one highly complex aggregation. No aggre- 

 gation of moving bodies can, however, continue to exist 

 apart from their subjection to some governing law. " Bees 

 in a swarm, ' for instance — to adopt Professor Young's 

 chosen example — would very quickly cease from that con- 

 dition imless maintained in it by an efficacious " clustering 

 power." Their courses must be more or less pendulous, if 

 not centrally deflected. A few seconds of persistent flight 

 along straight hues would bring about the disintegration of 

 the little community. And so, just because the sidereal 

 aggregation is a fundamental reality, and not the casual 

 product of an eternal and meaningless flux of things, star- 

 movements must he controlled by some primary and 

 overmastering force, acting vmiformly, universally, and 

 irresistibly. Did they truly possess the lawless and in- 

 subordinate character sometimes imputed to them, we 

 should be compelled to regard the constellated suns of our 

 midnight skies as mere adventurers from the void brought 

 into temporary companionship within the circuit of the 

 galaxy, and obeying only King Arthur's law of hfe — 



" From till" gri'.it deep to the great deep they go." 



But this is simply incredible. 



[.\ general aggregation of the brighter stars towards 

 the zone of the Milky Way has long been recognized. Dr. 

 Gould has carefully studied the symmetry in the arrange- 

 ment of the brighter stars, and points out (in the I'rano- 

 iiuiriii Ai-(i,')iUna, p. 301), that the brighter stars are more 

 uniformly distributed with respect to the medial plane of 

 the zone of great stars (described in the article on the 

 Pleiades cluster in Knowledge for May, 1891) than with 

 respect to the medial plane of the Milky Way. The zone 

 of large stars follows a great circle round the heavens, 

 making an angle of about 20° with the medial plane of 

 the MUky Way. 



It does not follow, from the general symmetry in the 

 arrangement of the brighter stars which is observable, that 

 all the brighter stars belong to the Galactic system, or that 

 they are coimected with the great zone above referred to. 

 A third or even a half of their number may be distributed 

 at random over the sphere. The chance distribution of a 

 certain number of stars would not obliterate any symmetry 

 that might be evident in the arrangement of the stars of 

 the Galactic cluster. 



Some twelve years ago Professor Simon Newcomb showed 

 that stars ha%dng a velocity in space of 25 miles a second 

 could not be controlled by the gravitation of the J\Iilky Way 

 system, and that such swift stars could not therefore be 



