86 



KNOWLEDGE 



[May 'I, 1892. 



all that was known of functions dependent on masses 

 derived its origin from Galileo and Newton, and appeared 

 to indicate that such functions always either increase or 

 decrease with the increase of mass, as in the case of the 

 attraction of celestial bodies. The numerical expression 

 of the phenomena was always found to be proportional to 

 the mass, and in no case was an increase in mass followed 

 by a recurrence of properties such as is disclosed by the 

 Periodic law. This is thought to indicate that the true 

 conception of mass, and perhaps also of the mode of action 

 of gravitation, must be sought in the study of the chemical 

 atoms, and of the connection between them by a periodic 

 function of their mass. 



THE SYSTEM OF ALGOL. 



By Miss A. M. Clerke, Aiit/ioresx of " The Si/steiii of t/w 



Sturs," ami '■ 21ic Hinturij of Astronomij durimi tlir 19th 



CenUtry," d'c, dv. 



THE steady advance of exploratory research in the 

 system of Algol promises to furnish one of the 

 most curious and instructive episodes in the 

 history of science. Vague hypothesis, determinate 

 theory, and triumphant verification, have already 

 played their logically sequent parts in the discovery of the 

 eclipsing satellite. Goodricke's conjecture, however, had 

 to wait nearly a century for Pickering's fonuulization, 

 while this was ratified within a decade by Vogel's dis- 

 closure of the anticipated tell-tale spectroscopic effects. 

 Progress has, indeed, of late notably quickened its pace ; 

 and we may therefore hope for a prompt and effective 

 application of the Ithuriel-spear of adapted observation to 

 the latest creation of speculative intelligence in the lately 

 organized department of " dark stars." 



Since Argelander's time it has been tolerably evident 

 that Algol had other attendants besides the agent in 

 producing its periodical eclipses. For their recurrence was 

 shown by him to be subject to minute irregularities in 

 point of time, and these irregularities are of such a nature 

 as to demand for their explanation the presence of at least 

 one disturbmg mass. A highly complex piece of mechanism 

 could plainly be seen to be at work ; yet the penetration of 

 its intricacies presented a task so formidable that 

 astronomers of, at any rate, the present generation might 

 well have despaired of its accomplishment. It has, never- 

 theless, been undertaken by Dr. Chandler ; and his labours 

 have been rewarded with an encouraging measure of 

 success.* They have been necessarily of a more or less 

 tentative character, and their result must be looked upon 

 as merely provisional ; but there is much reason to 

 suppose that it at least approximates to the truth. 

 It is, moreover, perfectly plain and straightforward ; 

 there is nothing of the (ilmnrum per ohxcurius about it ; 

 the consequences it involve;5 are definite, and admit of 

 definite verification. 



The new and enticing hypothesis now presented for the 

 consideration of astronomers is mainly founded upon certain 

 well-ascertained inequalities in Algol's period of variation. 

 These were shown by Dr. Chandler's discussion some little 

 time since ' to be slowly compensatory. They are 

 oscillatory, not progressive. Consistently in advance of 

 their due time down to about the year 1804, the obscura- 

 tions of the star then began to fall behind it, and the delay 

 had accumulated in 1843 to 165 minutes. A gradual pro- 

 cess of restoration thereupon set in, and the normal 

 epoch was reached near the beginning of 1873. It 



* Astronomical Joiiriinl. Xos. 255, 256. 

 t Ibid, vol. vii,, pp. 165-183. 



was quickly, however, transcended, for acceleration is 

 still going forward, and is likely to continue operative 

 during some years to come. These irregularities are 

 evidently comprised in a cycle considerably exceeding one 

 hundred years ; and for that very reason, it is difficult to 

 account for them on gravitational principles ; since a 

 third body, exterior to the close pair, should, in order to 

 produce any marked perturbational eff'ects, revolve much 

 nearer to them than would be consistent with so long a 

 period. Another mode of explanation is, accordingly, 

 resorted to by Dr. Chandler. The varying intervals needed 

 for the transmission of light from difl'erent parts of a large 

 orbit described by Algol and its dark satellite round a 

 remote primary, are, in his view, the fundamental cause of 

 the alternate anticipations and retardations in the occur- 

 rence of Algol's eclipses. They are, in fact, apparently 

 shifted backwards and forwards in time, just in the same 

 way as are the eclipses of .Jupiter's satellites through the 

 orbital movement of the earth. 



Algol may, then, be regarded as the solitary luminous 

 member of a multiple combination of opaque masses. The 

 common centre of gravity, round which the pair hitherto 

 known revolves in a period of about 131 jears, lies by the 

 present hypothesis at a distance from it just equal to that 

 of Uranus from the sun. The path thus traced out is, 

 we are further informed, sensibly circular, and its plane is 

 inclined 20^ to our line of vision. Obviously, however, 

 during the whole time occupied in travelling over its remoter 

 half, the light-minima of the star must be recorded somewhat 

 later than if we saw them in the precise order of their actual 

 occurrence ; and this remoter half was swept over between 

 the years 1804 and 1869, when the observed phases 

 were always in arrear of calculation. Now, on the 

 other hand, that the star is on the hither side of its orbit, 

 the epochs of its eclipses are apparently anticipated, 

 and will not coincide with their true times imtil the 

 passage of the " ascending node " about 1934. The 

 dimensions of Algol's orbit, together with its inclination, 

 of course prescribe the amplitude of the oscillations by 

 which its periodicity appears to be disturbed ; and this 

 " light-equation," as we may call it, proves to be 149 

 minutes. This should be the maximum extent, whether 

 of acceleration or of retardation ; but in point of fact, as 

 we have seen, delay mounted up in 1843 to 165 minutes. 

 Hence the theory cannot be said to represent the observa- 

 tions as satisfactorily as could be dcFiicd. The deviations, 

 indeed, are large enough to suggest to Dr. Chandler further 

 complications, the onravelmeut of which may challenge 

 the utmost skill and patience of investigators. Meantime, 

 a touchstone of the general truth of his hypothesis will 

 soon be at hand ; for it involves a cessation within the 

 next ten or twelve years, and a subsequent reversal, of the 

 shortening process at present afl'ecting the star's period of 

 luminous change ; and the fulfilment of this prediction 

 will serve as a hall-mark of its genuine ijuality. An 

 additional test may be derived from spectrographic evidence. 

 The velocity of Algol in the large orbit attributed to it is 

 2-7 miles a second ; but of this, less than one half, or 

 about one mile per second, is at present directed towards 

 the earth. It constitutes, however, a goodly proportion 

 of the 23 miles of continuous approach determined from 

 the Potsdam plates ; but which should in the course of a 

 score of years, if the new theory be true, completely dis- 

 appear, neutralized by the altered direction of the star's 

 orbital motion. It remains, indeed, to be seen whether 

 the whole of its supposed translatory speed may not really 

 be of a circulatory character. 



Dr. Chandler's theory does not rest wholly on the 

 cyclical inequalities of Algol's light-changes. He alleges 



