INTENSITY RELATIONS IN THE SPECTRUM OF HELIUM. 
141 
(III.) The Method of Measurement. 
The method adopted for the determination of the intensities of lines in a spectrum 
has been described in a previous communication,* in which it was shown that the 
absolute values of the intensities can be obtained from the “ photographic ” intensities 
by the adoption, as a standard, of the radiation from the positive crater of the carbon 
arc, in which the distribution of intensity along the spectrum can be calculated by 
Planck’s or Wien’s formula. For the purpose of the present investigation, the 
photographic intensities afford all the necessary information, and the results exhibited 
below are accordingly limited to a determination of these values. 
The spectrograph consisted of a large single prism constant-deviation instrument 
by Hilger, with a camera attachment in place of the telescope. Instead of the 
V-shaped slide for reducing the length of the slit, a brass slide with a rectangular 
opening was adopted, and in front of this opening was fixed the neutral glass wedge. 
This consisted of a prism of neutral-tinted glass cemented to a similar prism of 
colourless glass in such a manner that the combination formed a plane-parallel plate. 
When light is allowed to fall on to the slit through this wedge, the resulting 
spectrum is found to consist of lines which are bright at one end, corresponding to 
the thin end of the wedge, and which fade away in the direction corresponding to 
the dense end of the wedge, the length of the line on the plate thus depending on its 
intensity and also on the “ density ” of the wedge for that particular wave-length. 
The spectra under investigation were photographed on Wratten Panchromatic 
plates, and these were developed with a Hydroquinone and Formaline developer 
which gives results showing great contrast. From the negatives thus obtained, 
positives were printed by contact on Paget Half-tone or Paget Slow Lantern plates, 
which were found to give the best results for this stage of the process. These 
positives were then intensified with Mercuric Chloride and Ammonia, and enlarge¬ 
ments were subsequently made on bromide paper using a Zeiss “ Tessar” lens, which, 
under the conditions of use, gave no measurable amount of distortion of the image. 
The enlargements were made with the aid of a ruled process screen, which was placed 
immediately in front of the bromide paper. The resulting enlarged negative image 
was in this way built up from a number of small dots, one-hundredth of an inch apart. 
On the enlargement obtained by this method, it is a matter of no difficulty to pick 
out the last dot visible on each line, and thus to determine with considerable accuracy 
the relative lengths of the lines composing the spectrum. In the absence of the 
process of reproduction of the image in dots, this would be a matter of great 
difficulty, and the results would be subject to considerable personal error. 
The plate-holder of the spectrograph was provided with a rack and pinion motion 
in order to allow of the possibility of photographing a number of spectra on the same 
plate. The spectra under comparison are thus photographed on adjacent portions of 
* Loc. cit. 
