546 Proceedings of Royal Society of Edinburgh. [skss. 
spectrum. It may be specially noticed that, from the theoretical 
point of view, the absorption-liim must under all circumstances he 
on the violet side of the emission-line. 
(3) The expanding atmosphere, formed from the volatilised sub- 
stances of the star matter, and being at temperatures comparable to 
those prevalent in star-atmospheres, will spectroscopically resemble 
the chromosphere. This conclusion is confirmed by the table on 
pp. 286-7 of Scheiner-Frost’s Spectroscopy , in which a comparison 
is made between the lines seen in Nova Aurigae and those most 
frequently and most intensely noticed in the solar chromosphere. 
(4) Besides the expanding atmosphere, account must be taken 
of the revolving ring of nebular matter which, after the collision, 
has been brought under the permanent gravitational sway of the 
star. The presence of motions of this character explains uot only 
the enormous width of the bright bands, but also the appearance 
of symmetrically grouped maxima within them. We are further 
enabled to understand the strong displacement of the bright band 
towards the red in one case (Nova Aurigae), and the absence of 
such a shift in another (Nova Persei). 
(5) When once this gyrating ring of matter has been established, 
further direct impact of meteoric matter upon the star will be im- 
peded, since a considerable number of nebular particles may collide 
already inside the ring without reaching the surface of the star. 
This enhances, on the one hand, the luminosity of the ring, and 
reduces, on the other hand, the incandescence of the nucleus. 
Consequently we notice a decrease of the continuous spectrum 
coupled with an increase of those gaseoul radiations which are caused 
by the incandescence or luminosity of the gyrating nebular matter. 
The expanding atmosphere having gradually faded away, the 
chromospheric spectrum has also disappeared, and has been super- 
seded by those lines which are peculiar to the spectrum of gaseous 
nebulae.* This is, briefly, the course of events which theory would 
lead us to expect. At the same time, it is also in many respects the 
sequence of phenomena shown by observation. The gradual dis- 
appearance of the continuous spectrum together with the lines 
which belong to the chromospheric radiations, and the simultaneous 
intensification of the nebular lines, — the peculiar process of 
* See note at end of paper. 
