254 



NA TURE 



[July ii, 189; 



more rigid than any other, and therefore any irregularity under 

 it would lie almost impossible. 



Closely associated with this reference to double swarms in the 

 case of \-ariable stars are the phenomena of so-called " new 

 stars." Indeetl. the whole conception of the mcteoritic 

 h}"pothesis arose from a consideration of thi>se IxKlies which 

 sometimes ijuite suddenly make their appearance in the heavens. 

 We have had during the last thirty years five of these new stars. 

 and it was during the appearance of one in the constellation 

 Cygnus in l S76 thai I was led to the views w hich I still hold 

 with regard to their origin. 



One of the most remarkable features of the.se new stars is the 

 rapidity with which they lose their lirilliancy, and it wa-s this 



Kic 35. — Thc_ region'in the heavens where Nov.-i .Auri^j.-u 

 (l)aflcr iLi dls.ippear.-ince ; (2) when Imghtly \isihle (ne.irly 



w.x-i observed 

 in the centre). 



which led me in 1877 to write, in connection with Nova Cygni 

 (NATt:RE, vol. xvi. p. 413, 1877): "We seem driven, then, 

 from the idea that the.se phenomena arc proiluced by the in- 

 candescence of large ma.sses of matter, because if they were so 

 produceil, the running down of brilliancy would l>e exceedingly 

 slow. 



" Let us crjnsidcr the ca.se, then, on the sup|K)sition of small 

 masses of matter. Where are we to find Ihem ? The answer is 

 easy : in those small meteoric ma.s.ses which, an ever-increasing 

 ma.ss of evidence tends to show, occupy all the realms of 

 space. . . . The Nova now exists as a nebula, so fiir .is its 

 s|5ectnim goes, and the fact not only goes far to sup|Mirt the view 

 I ha%-c suggested, as again.st that of Ziillner, but it affords 

 collateral evidence of the truth of Thom.son and Tail's 

 hy|)othesis of the true nature of nebul.x-." 



we get the greater light formed at the moment when two swarms, 

 one revolving round the other, are nearest together. 



Kortimateiy for science, one of these new stars appeareii in 

 1S92 ; it is known as Nova .-Xurii^.e, and two photographs will 

 give us an idea of the sort of thing w hich an astronomer sees in 

 the heavens when the discovery of a new star is announced. 

 The photograjjhs show a [xirlion of the constellation of .Auriga, 

 and a star which is verj- clearly seen in the photograph taken 

 very soon after this star had burst ujwn us, is absent from one 

 taken a few months later. 



Since the spectroscope w.as first applied to the stars, five new 

 stars have l>een observed and spectroscopically examined. One 

 ap|)earecl in Corona Borealis in 1866, one in Cygnus in 1876, and 

 one in .\ndromeda in 1S85 : then came the one in .\uriga in 

 1892, to which reference has already been made, and last of all 

 was one in the .southern hemisphere, discovered in 1893. The 

 first three of these were observed by eye only, but in the two 

 recent ones we have the immense benefit of photographic 

 records. 



It was therefore a ver)' interesting point w hen a new star came 

 along, to see whether there was any additional light thrown by 

 it upon the ])roblem of two bodies ; and further, upon one of the 

 points in which, if the meleoritic hypothesis failed, it was worth 

 absolutely nothing at all. If there was any truth in the idea of 

 the light of these l>odies being produced by the clash of meteor- 

 swarms, w hen the clash was over the swarms should go back 

 into their native oi)scurity, or cont!iti<in of low tcmpemlure, and 

 should, if they were seen at all, put on the spectriini of sparse 

 swarms in other parts of the sky ; that is, they should put on the 

 spectrum of a nebula. 



That, you see, was a ver)' crucial point : it was a ptiint which 

 could be settled by the spectroscope, (irovided always we had 

 one of these marvellous bodies at such a distance from us that 

 we could still observe it spectroscopically, and see what the 

 ilifferent changes really amotinted to. 

 I -Vlready in the case of Nova Cygni, the s|>ectrum had l)een 

 obser\ed to change from a rather complicated one of liright 

 lines and flutings to a very simple one, similar to that of a 

 planetary nebula. The observations <li<l not, however, furnish 

 any direct evidence that more than a single bo<ly was concerned 

 in the outburst. 



The apjiearance of Nova .\uriga-, however, furnished a s])lendiil 

 opportunity of testing the many theories wliidi have been at 

 various times ad\"anced to account for the phenomena. This 

 No\a was discovered at Kdinburgh by Or. .\nderson, who was 

 modest enough to announce his discovery by sending an anonymous 

 post-card to Or. Copeland, the .Astronomer Royal for .Scotland, on 

 Kebruar)' I, 1892. It was then a star of the fifth magniliule, 

 and on confirming the true nature of the newly-discovered star by 

 means of the spectroscope, Dr. Copeland m.ide the news public. 

 Information was received at nu)st observatories (Hi l*"ebruary 3, 

 and on the same evening two photographs of the spectrum were 

 taken at South Kensington. During the next two or three weeks 

 the star fluctuated consideralily in brightness, though being 

 generally ()n the down grade : and by .\pril 26 had fallen to the 



FrnMimminMrTi 



K I. . (, 1' 



Flo. 36. — IMiolugr.iph of the Ajicclrum of Nuva Aurig.t, taken at .Nouth Kensington, February 7, 1892. 



Returning to the subject of new stars in 1887, in a general 

 discnssirm of the mcteoritic hy|Kilhesis, I «iw no reitson to change 

 my views, and an incjuiry into the siwctroscopic phenomena leil 

 me to state that -" New stars, whether seen in cimneclion with 

 neliuLx' or not. arc produce<I by the clash of meteor swarms, 

 Ihc bright lines seen having low tem|>eraturc lines of elemen(s, 

 the ■" ' ' ; K-.st brilliant at a low stage of heat." 



A iiinn iif all the new stars which had 



I*' firmed the subject of a communication 



lo ' 1 w :ls shown that the hy|xilhesis would 



e»|'i iighl, the changes of colour, and the 



ftpcctriKriiopii. .ip|H.-.tr.i)i«.c>. 



To make a vcr)- long story short, it «a.s suggested that the 

 phenomena of new stars were prinluccd by exactly the same 

 cause as that which wa.s at work in the variable stars in which 



NO 1341. vol. 52] 



l6th magnitude, .so that it could only be picked up at all in the 

 very largest tele.sco|)cs. Thanks to the photographic records ol 

 the stars, it was pos.sible to learn something of ihe earlier history 

 of the new star. It had really been pholograplieil l)y I'rol. 

 Pickering two immlhs before its existence was known. 



Kig. 36 shows usa photograph of the spectrum of this wonderful 

 star itself, and it will be .seen that In the case of all the chief 

 lines we get a bright lint and a dark line side by side. There 

 arc the hydr<igen lines ; that is, in the spectrum of that body wo 

 were ilealing with the giving out of hydrogen, and the absoip- 

 lion of hydrogen. Now, the same .set of iiarticles cannot be 

 producing bright and dark linesal Ihe same time. We were then 

 obviously dealing with two sets, and the first photograph, there- 

 fore, which was taken of the spcclriim of this strange body, put: 

 beyond all ipiestion the fact that we were really dealing with two 



