96 
SIR W. DE W. ABNEY AND PROF. W. WATSON ON 
our unit is such that when the intensity of illumination by the D light on the disc, I, 
is one lux the energy per square centimetre is equal to 100,000 such units. 
A measurement of the light diffused from the magnesium oxide surface showed 
that when the illumination of this surface was one lux the radiation it sends out 
normally per square centimetre is equal to 0'000026 candles, so that if the cosine law 
be assumed to hold the coefficient of reflection is 0'83. 
In order to obtain the energy distribution in the spectrum, a linear thermopile was 
attached to the slide, FG, fig. 1, and the deflections of a sensitive galvanometer were 
read as the pile moved through the spectrum. In order to reduce the effect of stray 
heat a pair of slits were placed one in front of the other before the pile and a water 
cell was interposed between the Nernst glower and the collimator lens. As a test to 
see whether stray radiation had been eliminated, the water cell was replaced by one 
containing a solution of iodine in carbon bisulphide. This cell, while entirely cutting 
off the visible spectrum, allowed most of the infra red to pass. While there was a 
large deflection when the pile was placed beyond the red, there was no observable 
deflection while the pile moved throughout the region ordinarily occupied by the 
visible spectrum. Since a similar spectrum apparatus to that shown in fig. 1, except 
that the crater of an electric arc is the source of light, has been used for some of the 
experiments described in this paper, and further that it is the one used throughout 
the series of researches conducted by one of us for many years, we have determined 
the energy distribution of this source. In Table I. and fig. 3 we give the results 
Table I.—Energy Distribution in Spectrum. Sources of Light : the Nernst 
Glower, and Positive Crater of Arc. 
SSN. 
A (/*/*). 
Energy. 
SSN. 
A (/x/x). . 
Energy. 
Nernst. 
Arc. 
Nernst. 
Arc. 
64 
721-7 
396 
182 
35 . 
504-3 
27-0 
42-3 
62 
695-7 
325 
166 
34 
500-2 
25-2 
40-0 
60 
672-8 
261 
151 
32 
492-4 
21 * 2 
35-2 
58 
652-1 
205 
137 
30 
484-8 
17-9 
30-7 
56 
633 • 0 
165 
126 
28 
477-6 
J 5 • 3 
26-9 
55 
624-2 
151 
121 
26 
470-7 
12-9 
23-8 
54 
615-2 
137 
116 
25 
467-0 
12-0 
22-3 
52 
599 • 6 
114 
106 
24 
463-9 
11-2 
20-8 
50 
585 • 0 
96 
97 
22 
. 457-8 
9-7 
18-2 
48 
572-0 
80 
89 
20 
451-7 
8-4 
16-0 
46 
559-6 
68 
81 
18 
445 ■ 9 
7-2 
13-8 
45 
553-8 
62-5 
77 
16 
440-4 
6-1 
11-9 
44 
548-1 
57 
73 
15 
438-1 
5-7 
11-1 
42 
537 • 3 
46-5 
65-5 
14 
434-9 
5-2 
10-4 
40 
527-0 
39-6 
58 
12 
429-6 
4-2 
9-0 
38 
517-2 
34-3 
51-5 
10 
424-5 
3-3 
8-2 
36 
508-5 
29-6 
45-5 
