1904 - 5 .] On Prof. Seeligers Theory of Temporary Stars. 545 
I cannot believe that these results, combined with the cor- 
roborating evidence of Professor Pecker’s observations, leave any 
doubt as to the fact that the displacements in the spectra of new 
stars depend exclusively on the wave-length, and are not caused 
by agencies which depend on the atomic structure of the emitting 
substances. This fact must be considered as the touch-stone 
of theories on temporary stars; so much so, indeed, that we 
may at once dismiss any explanation, however plausible in other 
respects, which is not in entire accordance with it. 
We are now in a position to form, step by step, a mental picture 
of the evolution of a new star, and to compare our deductive 
conclusions with the observed facts. The more important events 
in the star’s history as a radiating body may be thus summarised : — 
(1) The immediate consequence of the impact between star and 
cosmic cloud is a more or less one-sided incandescence of the star’s 
surface, causing a purely continuous spectrum. This stage was 
noticed here in Edinburgh about sixteen hours after the outburst. 
(2) In consequence of the sudden and enormous heating a gaseous 
envelope is formed, which expands very rapidly in radial directions. 
The velocity of expansion may be assumed to exceed that noticed 
in solar eruptions. The expanding gases now begin to influence 
the spectrum. At first absorption predominates, and is shown by 
broad absorption-lines, displaced towards the more refrangible side. 
The lines must be broad and hazy, on account of the density and 
the divergent motions of the gases. This stage w T as observed 
here in Edinburgh about twenty hours after the discovery, when 
the visual spectrum was strongly continuous, but interrupted at 
various places by faint broad absorption-bands. As the density 
must have decreased while the atmosphere was more and more 
expanding, the broad and hazy absorption-bands in course of 
time reduced to sharp dark lines, which ultimately thinned out 
and faded away. This peculiar feature, too, has been noticed by 
observers (see F. Sidgreaves’ note quoted above). At the same 
time the star’s atmosphere becoming more extensive, its radiation 
outside the star’s disc grows more and more prominent, giving rise 
to broad emission-bands in normal positions. Hence, after a time, 
the spectrum shows bright bands, bordered on their violet edges 
by absorption-bands. This constitutes the typical new star 
PROC. ROY. SOC. EDIN. — YOL. XXV. 35 
