1872.] (47 ) 
V. THE SPECTROSCOPE: ITS IMPERFECTIONS 
AND THEIR REMEDY. 
By Munco PontTon, F.R.S.E. 
y AHAT the spectroscope, as at present constructed, is an 
(es imperfect and fallacious instrument has been already 
: indicated in a previous paper, entitled ‘‘ Molecules, 
Ultimates, Atoms, and Waves.” It is now proposed to 
show in what respect the instrument is objectionable, and 
how it may possibly be improved. 
To understand the subject, it is needful to have some 
acquaintance with the laws of chromatic dispersion. An 
investigation into these laws will be found in the “ Philo- 
sophical Magazine” for 1860, pp. 165, 263, 364. When 
those researches were undertaken, the only existing measure- 
ments of the wave-lengths corresponding to the principal 
fixed lines of the spectrum were those left by Fratinhofer. 
These wave-lengths, however, have been recently re-mea- 
sured with more perfect appliances than Fratinhofer could 
obtain, by M. Angstrom, of Upsal, who has published the 
results in his work on the Normal Spectrum, accompanied by 
an illustrative atlas, in which all the most remarkable lines 
of the spectrum are laid down onan extended scale according 
to their wave-lengths. 
Since that work was published, M. Angstrém’s measure- 
ments have been analysed, and it has been ascertained that 
the whole of the wave-lengths corresponding to the principal 
fixed lines are capable of being calculated from one— 
namely, that corresponding to the more refrangible of the 
two lines marked E. The formule expressing the relations 
on which these calculations are based will be found in a 
memorandum contained in a work recently published.* It 
is there shown, moreover, that there is another curious 
relation connecting the whole of those wave-lengths to- 
gether. While the other seven are all capable of being calcu- 
lated from that of the more refrangible E, the sum of the 
three equations, by which are thus calculated the wave- 
lengths corresponding to the three lines A, B, and C, is equal 
to the sum of the four equations by which are calculated the 
wave-lengths corresponding to the four lines, D, F,G, and H. 
This remarkable relation seems to indicate that the wave- 
* «The Beginning,” &c. Longmans and Co. The memorandum is 
inserted between the notes and the description of the plates. 
