Su(>plement to ''Nature," March 13, 1902. 



accompanied by dark lines on their more refrangible 

 sides." Substantially the same announcement was made 

 by Prof. Pickering about the same time. These are the 

 facts, but they have been overlooked by Prof. Newcomb, 

 who does not mention Sir Norman Lockyer or Prof. 

 Pickering in connection with the Nova, but gives Profs. 

 Campbell and Vogel the credit of the discovery of the 

 double spectrum, and ascribes to the latter the demon- 

 stration of the existence of hydrogen a fortnight after the 

 element had been detected in England. The question of 

 priority is a small matter, but the principle involved is 

 important, namely, that care should be taken in the 

 selection of significant observations by whomsoever they 

 are made. It can scarcely be said that Prof. Newcomb 

 has e.xercised impartial judgment in this case. 



Early observations of the occurrence of bright hydrogen 

 lines in the spectrum of Mira Ceti are overlooked in 

 much the same way as with Nova Auriga?. Bright 

 hydrogen lines were photographed in the spectrum of this 

 variable star so long ago as 1886, and were observed by 

 Mr. Espin about the same time ; yet we read " Campbel 

 found that near the time of maximum, the bright hydrogen 

 line Hy was very strong and overexposed on all the plates." 

 The disregard of observation of Beta Lyrre is even more 

 unsatisfactory. Prof. iVIyers's mathematical theory of the 

 constitution of the star's system is described, while the 

 numerous spectroscopic researches relating to the star are 

 scarcely mentioned. The .South Kensington photographs 

 indicated long ago the presence of two bodies giving dark- 

 line spectra — one like Rigel and the other like Bellatrix. 

 The shifting bright lines are superposed upon the double- 

 absorption spectrum, and the spectroscopic variations 

 which occur bear a constant relation to the period of the 

 stars changes of lucidity. Prof. Belopolsky's measures 

 of the velocity of the star in the line of sight, and his 

 proof that the radial velocity is zero when the star is 

 passing through the primary and-secondary minima, have 

 practically proved that the chief cause of variability is 

 eclipse by such bodies as those mentioned. Why, then, 

 should these established facts be left out of consideration 

 while space is given to mathematical inquiries which have 

 yet to receive spectroscopic confirmation? 



The study of these binary and multiple spectroscopic 

 systems has already led to results of deep significance, 

 and Prof. Ncivcomb devotes some attention to ihem. In 

 connection with the account of Prof. Vogel's investiga- 

 tions of Algol, however, it is worth remark that the eclipse 

 theory of the variability of the star was practically estab- 

 lished by Prof. Pickering by observations of spectro- 

 scopic changes before precise measures of the radial 

 velocity were made at Potsdam. Algol is a case of 

 May wedded to December — a fervid body linked to one 

 apparently decaying or dead. In true spectroscopic 

 binaries like Zeta Ursas Majoris (not Xi Ursa: Majoris, 

 as is stated on p. 167), we have the component stars in 

 more or less the same stage of development. From close 

 binaries of this kind with periods of a day or so it is 

 now possible to pass almost without a gap to the double 

 stars with periods reckoned in centuries. Prof. Newcomb 

 gives eleven years as the shortest period of a telescopic 

 binary, but the star ji 883 has a period of only half this 

 length, which brings the two classes of binary systems 

 closer together than he supposes. 

 NO. 16 9, VOL. 65] 



Breaches of continuity are usually only apparent, and 

 become filled as knowledge increases. We have passed 

 the days of special creations, and are beginning to see 

 a course of evolution in all natural processes. The 

 biologist has adapted himself to this idea more readily 

 than the astronomer, who looks askance at all attempts 

 to discover a spectroscopic order of evolution in celestial 

 bodies. Prof. Newcomb brings together a number of 

 interesting facts concerning the structure of our universe, 

 but he does not show much relationship between them. 

 Why is it that nebuls which give bright-line spectra, 

 bright-line stars and new stars are almost exclusively 

 confined to the Milky Way ? For though it is true that 

 diffuse nebuUp " are least numerous in the Milky Way 

 and increase in number as we go from it in either direc- 

 tion," it is also true that there are more planetary nebula; 

 near the Milky Way than in any other part of the 

 heavens. It is reasonable to suppose that there is a 

 large amount of dark meteoritic matter in the region 

 of the Galaxy, as well as a greater depth of stars. Sir 

 Norman Lockyer has suggested this in explanation of 

 the overwhelming number of very faint stars which occur 

 in the Milky Way. Such stars may only be faint be- 

 cause of the partial extinction of light caused by fine 

 particles of matter ; so that the same stars would appear 

 brighter if they happened to be nearer the galactic poles. 

 Upon this hypothesis it is possible to explain phenomena 

 which otherwise appear inexplicable ; at any rate, it is as 

 worthy of analysis as the "grindstone" theory of the 

 structure of the universe. 



It is too much to assert that no extinction or modifi- 

 cation of light occurs in space. We can only analyse 

 what reaches us, and it is just as logical to assume the 

 existence of an absorbing medium as it is to neglect a 

 possible cause which, while not contrary to any observed 

 facts, facilitates the explanation of related phenomena. 

 The test of any theory is not so much how it explains 

 one group of facts or phenomena, but how far it admits 

 of general application ; and it is because the meteoritic 

 hypothesis satisfies this condition that it claims atten- 

 tion. Most of us will agree with Prof Newcomb in the 

 remark, " So far as we can judge from the enumeration 

 of the stars in all directions, and from the aspect of the 

 Milky Way, our system is near the centre of the stellar 

 universe." But the argument is nevertheless unsound. 

 It would be just as reasonable for a horse tethered to a 

 stake to conclude that the grass he was able to get at 

 represented all there was in the world. We can only 

 sound the depths of space so far as our instruments 

 permit us, and the facts as to the distribution of stars or 

 other celestial bodies must be considered in relation to 

 others of a spectroscopic character before we are justified 

 in assuming that the sun is situated towards the centre 

 of our universe. 



There are many other points in Prof. Newcomb's book 

 which invite discussion, but they cannot be dealt with 

 here. It is worth mention that no account is taken of 

 Prof. Perry's reasons for believing that the sun may have 

 radiated heat for much more than twenty million years, and 

 that the results of studies of photographs of stellar spectra 

 taken at South Kensington are entirely neglected. The 

 misprints we have noticed are Sirs A. A. Common and 

 Isaac Roberts (p. iv.), telescopic for spectroscopic (p. 12) 



