ON STELLAR SPECTROMETRY. 169 
The most striking object for its singularity which I have met in this ex- 
amination of the heavens, and which is quite unique, with the exception of a 
very faint companion, is y Cassiopeiz. This star showed to me for the first 
time the lines of hydrogen in a luminous state, exactly the reverse of the 
dark lines of the stars of the first type. Thestar 3 Lyre has the same fea- 
ture, but in a very faint degree. 
We have therefore, without doubt, in the heavens a grand fact, the fun- 
damental distinction between the stars according to a small number of 
types ; this opens a field for very many important cosmological speculations. 
2. Another grand fact which was brought out from these researches was, 
that the stars of the same type are occasionally crowded in the same space 
of the heavens. Thus the white stars are thickly gathered in Leo, in Ursa 
Major, in Lyra, Pleiades, &c., while the yellow ones are very frequent in 
Cetus, in Eridanus, Hydra, &c. The region of Orion is very remarkable for 
having all over and in its neighbourhood green stars of the first type, but with 
very narrow lines and with scarcely any red colour. It seems that this par- 
ticular kind of star is seen through the great mass which constitutes the 
great nebula of Orion, whose spectrum may contrast with the primitive 
spectrum of the stars. Sirius is perhaps too near us to be affected by this 
influence. 
This distribution of stars seems to indicate in space a particular distribu- 
tion of matter or of temperature in different regions. 
3. A third very remarkable conclusion arrived at is, that all the spectra 
of the third and fourth type belong to variable stars. The representative of 
these is the wonderful star Mira Ceti. This has been carefully examined ; 
and it is found that even when it is only of the seventh magnitude it has 
the same typical spectrum, only reduced to its few bright columns. a Orionis 
is in the same condition. a Tauri (or Aldebaran) and Arcturus this year 
appeared to be smaller and of a more red hue than in the past year; and in 
the first there appeared traces of columns which were not seen the year 
before ; so that it is evident that the change of these stars depends on a 
periodical change which happens in their atmospheres. 
It is not so, however, with Algol, which has the very same spectrum of the 
first class or type in every stage of magnitude, which induces me to believe 
that there the variation is produced by the passage of an opaque body pass- 
ing between it and the central star, giving thus an example of eclipse of 
a fixed star by its own obscure planet. 
4. Finally, avery delicate question was proposed by myself, to be resolved 
by spectrum analysis; this consists in ascertaining whether the star has a 
proper motion, by the displacement of the lines which ought to take place 
in the spectrum owing to the combined motion of the star and the propaga- 
tion of light. From this new kind of observation it would be easy to ascer- 
tain if a star has a motion whose velocity is five times that of our earth 
around the sun. The star a Lyre, examined in this manner, has not 
given any such displacement; "so that it appears not to have such a motion. 
In some other stars I have found that there is a little displacement, as in 
e Urs Majoris; but this seems especially due to the different breadth of the 
hydrogen line in the star and in the chemical spectrum. TI have used in this 
investigation the comparison of the direct image of a star with its own spec- 
trum, but I have found no appreciable displacement. My first researches 
of this kind commenced with the supposition that in Sirius there was a per- 
fect coincidence of the lines of hydrogen with those of the star. Now Iam 
told that your distinguished countryman, Mr. Huggins, has found a little dif- 
