■luxK, 1901.] 



KNOWLEDGE. 



i;u 



whicii was subjected to the analysis of the spectroscope, 

 (lie earliest observations being those of Sir William 

 Mugf'iup and Dr. Miller. The chief facts observetl with 

 regard to its spectrum were that it was compound, being 

 composed of two series of bright and dark lines, super- 

 posed u])on a background of continuous spectrum, and 

 tliat the bright hydrogen lines predominated. jM. Cornu 

 called attention to the matching of the brightest lines in 

 tho stai's spectrum by well-known lines seen bright in 

 (he chromosphere of the sun. With regai-d to the 

 spectrum of Xova Cygni in 187G, Professor Vogcl called 

 attention to the extent of the spectrum which stretched 

 far into the violet regions, the double spectrum of bright 

 and dark lines, especially those of hydrogen, and the 

 gradual changes in the spectrum, winch had become 

 reduced to a single line in the green, very near in position 

 to the chief line characteristic of nebuise, when the star 

 had fallen to the t«nth magnitude. The spectrum of 

 Xova Andromeda in 188.^ was mainly continuous, being 

 similar in this respect to the spectrum of the first days 

 of Nova Persei, ajid, moreover, confined to the green. 

 The spectrum of Nova Aurigje was the first which was 

 permanently recorded for future comparison and study 

 by means of the photographic plate. It was a very 

 wonderful spectrum, with many new details of its own, 

 but still confimiing all the earlier obsei^vations of the 

 spectra of novar. The compound or double spectrum of 

 bright and dark lines was very striking, the more intense 

 bright lines, especially those of hvdrogen, being accom- 

 panied on their more refrangible sides by broad dark 

 companions. The line near the position of the chief 

 nebular line was very marked. The spectrum, too, was 

 of great length, extending from beyond C in the red right 

 down to the further limit of the ultra-violet. It was full 

 of lines, the Stouyhurst photographs, which comprised 

 the region from D in the yellow to H in the violet, show- 

 ing more than 200 lines. Many of these lines, too, were 

 of gi'eat breadth, notably those of hydrogen, extending 

 over several tenth-metres, or units of wave-length. On 

 an average the centres of the dark absorption lines were 

 about 11 such units removed from the most intense points 

 of the bright bands. The match, too, both in character 

 and position of the lines in the spectrum of the star, and 

 in the spectrum of the sun s chromosphere was very 

 evident. Moreover, sharp bright lines appeared on the 

 dark bands after the star had attained its maximum. 

 Such an appearance is not unknown in the spectrum of 

 sun-spots, and on the calcium lines H and K, in the solar 

 spectrum. Lastly, when the star received an accession 

 of light in the August of 1892, the spectrum as examined 

 by Professor Campbell, of the Lick Observatory, was 

 strikingly similar to that of planetary nebulae. 



The beautiful photogi-aphs of Nova Persei by which we 

 are enabled to illustrate this paper, through the kindness 

 of Father Sidgreaves, were taken on March 3rd at 8 p.m., 

 on March 21st at 8.30 p.m., and on ilarch 25th also at 

 8.30 p.m., by means of an objective prism of about 

 4 inches aperture, made by Mr. Thorp, and fixed to the 

 finder, also of the same aperture, made by Cooke & Co., 

 of the 15-inch equatorial. A spectrum of Procyon is 

 placed below the spectra of the nova to show the posi- 

 tions of the hydrogen lines or rather bands Reading from 

 right to left they are H/3 to Ht. the calcium line K, 

 immediately following Ht. Without entering into any 

 details these photogiaphs sufficiently illustrate the chief 

 points to which we have directed attention in the spectra 

 of all novae. There are many other matters worthy of 

 study, but beyond the scope of the present paper. There 

 is. however, one fact with regard to the chanjres in the 



spectrum noted in Nova Persei, which deserves mention, 

 as it has never been observed before. It is that when the 

 spo.-trum was photographed at Harvard, and observed at 

 Kdinburgh on tiic nigiit of February 22nd, it was con- 

 tinuous and crossed by only thin dark lines. By Feb- 

 ruary 24th the bright linos were well in evidence with 

 their dark companions. Tills would seem to point to a 

 sudden raising of the temperature of a dark solid cold 

 body to a white heat. Then came absorption by the 

 vapours thrown off, giving the dark line spectrum so 

 well known in the sun and in the stars of the Orion type, 

 and then the supremacy in radiative power of the gaseous 

 s]/cctrum .suddenly asserted itself, and became the chief 

 feature, as is the case in those storms on the sun which 

 we call spots and prominences. 



How then are we to account for the origin of such 

 stars? At the outset wc must frankly acknowledge that 

 the appearances are extremely puzzling, and that so far 

 no theory has been broached which satisfies all the con- 

 ditions of the pioblem. For any theory of a new star's 

 origin and constitution must needs find an explanation 

 for the sudden outburst of light, for its fluctuations in 

 intensity, for the complicated spectrum, its extent, its 

 multitude of lines, the phenomena of the bright lines and 

 their d.ark companions, together with their e.xceeding 

 great breadth. But the crucial jjoiiit of the whole 

 problem lies in the fact that this shifting is always in the 

 same direction, of great extent, and persists for a long 

 time. I fancy any theories that purport to explain the 

 appearances on the supposition of the motion either of 

 meteors through a nebula, or of a cold dark body through 

 luminous particles, must needs satisfy, first the enormous 

 velocities of the moving bodies in the line of sight, some 

 500 miles a second at the least, if indeed the displace- 

 ments observed are due to motion in the line of sight, 

 the persistence of such velocities for months, and, most 

 difficidt point of all, their invariable occurrence exactly 

 in our line of sight. Should we attribute the great 

 widening of the lines to pressure, after the analogy of the 

 laboratory experiments of Humphreys and Moliler, and 

 of Wilsing, we do not escape the difficulty of the long per- 

 sistence of such pressure, to say nothing of its enormous 

 amount. 



However, we note that the light curve of new stars 

 is very like the curve of sun-spot frequency, with the 

 rapid rise to maximum, and the fluctuations as the ener- 

 gies displayed in the initial stages of the outburst are 

 gradually expended. Next the spectrum of new stars 

 is in the main the spectrum of the solar chromosphere, 

 at least in its earlier stages. Moreover, appearances 



very similar, though on a greatly reduced scale, to the 

 double spectrum of new stars can be observed in the 

 spectra of sun-spots and prominences. And it must be 

 bcrne in mind that a sun-spot is only one of the forms 

 in which the forces in the sun are displayed. For a sun- 

 spot is always accompanied by prominences, as also by 

 faculae at times of vast extent. It is true that in the 

 chromosphere the lines, except at their bases, are very 

 thin, showing that the pressure cannot be very great; but 

 in spots the lines are widened, due no doubt to the 

 density of the matter composing them, and, moreover, 

 frequently shifted, perhaps the partial result of pressure. 

 We have already alluded to the reversals to be seen so 

 frequently in the spectra of spots and prominences, and, 

 moreover, binds have been observed in the spectra of the 

 former. But in spite of all those similarities we cannot 

 claim them as anything more than suggestive of the pos- 

 sible lines on which the problem as to the origin of new 

 stars is to be attacked. 



