[ 187 ] 
III. On the Ultra-Violet Spectra of the Elements. —Part I. Iron (with a map). 
By G. D. Liveing, M.A., F.R.S., Professor of Chemistry , 
and J. Dewar, M.A ., F.R.S., Jacksonian Professor, University of Cambridge. 
Received June 8,—Read June 15, 1882. 
[Plates 1-5.] 
o 
Angstrom’s “normal solar spectrum” has served, most spectroscopists as a standard 
of reference for wave-lengths in the visible part of the spectrum. Cornu’s con¬ 
tinuation of it, and particularly the map of the iron lines which he used in the 
construction of it, serves very well for such a standard up to the limit of the solar 
spectrum, i.e., to a wave-length 2948 (U). For the region above this we have had to 
use Mascart’s and Cornu’s wave-lengths of the cadmium lines, and Cornu’s wave¬ 
lengths of some magnesium lines. The intervals between those lines are, however, 
too great for any close approximation to the wave-lengths of intervening lines by 
interpolation, and, moreover, the wave-lengths did not appear to be determined with 
sufficient precision to serve as a standard, and the lines are ill adapted to that purpose 
by reason of their diffuse character. For the determination of the wave-lengths of lines 
in this higher region we have, therefore, been driven to form a standard for ourselves. 
For this purpose we have chosen the iron spectrum which had been employed by 
Cornu in the region which he mapped, and seemed to answer the purpose well, both 
from the number of lines which it presents and their characteristic grouping by which 
they may easily be recognised by anyone wdio has once become familiar with them. 
The wave-lengths of the most prominent lines were determined by means of a 
Kutherfurd diffraction grating, as detailed below, between the wave-lengths 2948 
and 2327 ; but beyond this there is a remarkable falling off in the intensity of the 
iron lines, and between wave-lengths 2327 and 2135 (which is near the limit of 
transparency of Iceland spar) we have preferred to determine the wave-lengths of 
the prominent copper lines which are numerous and strong in that region. The wave¬ 
lengths of a series of lines at short intervals having thus been determined, those of 
the intervening lines were obtained by interpolation, and the result is shown in the 
map of the iron spectrum above U which accompanies this paper. 
2 b 2 
