452 
CAPTAIN ABNEY AND MAJOR-GENERAL FESTING 
intended by the makers to be worked at about this intensity. The light represented 
by curve A was brilliant, and would certainly be supposed to be whiter than that of 
gas burnt in a good fishtail burner. On plotting curve D, however, which is that of 
gas burnt in a fishtail burner, an image of the flame having been thrown on the slit, 
we were somewhat astonished to find that it indicated a whiter light than our incan¬ 
descence lamp would give with 34 boxes. On comparing the shadows of the two 
lights in an ordinary Rumfgrd photometer, we found a confirmation of the truth of 
our curves, for the shadow illumined by the gas was distinctly blue-grey as compared 
to the yellow shadow illumined by the lamp. Our unaided eyes, therefore, had failed 
to appreciate the colour of the light. We found that with 34 boxes of the secondary 
battery the lamp gave a light of almost precisely the same colour as that of the 
candles which we used in our measurements. 
We next determined the luminosity-curve of a candle, and found that it lay very 
closely on that of the lamp with 34 boxes. There was, however, a slightly larger 
amount of blue-green in the former than in the latter, which is a point of some 
interest. 
Reverting to the curves of the incandescence lamps, it will be remarked that at 
comparatively low temperatures there is marked deficiency in the yellow-green of the 
spectrum about X 5G30, as shown by a decided bend in the curve, which, however, 
diminishes with increase of temperature, and in curve C is scarcely discernible, 
showing that the emissivity of carbon for that part of the spectrum increases propor¬ 
tionally more than for the adjacent parts. It would appear, then, that cold carbon 
should probably be less absorbent of these rays than of others, a point which we have 
not determined as yet. The scale of A must be multiplied 6'67 tunes, and that of 
B 1 ’8 times to arrive at the actually observed values compared with C. 
Luminosity of Incandescence Light and Spectra. 
Intensity. 
26 boxes. 
Scale 
reading. 
C 
2S boxes. 
Scale 
readme. 
B “ 
34 boxes. 
Scale 
reading. 
A 
Gas. 
Scale 
reading. 
D 
Intensity. 
26 boxes. 
Scale 
reading. 
C 
23 boxes. 
Scale 
reading. 
B ~ 
34 boxes. 
Scale 
reading. 
A 
Gas. 
Scale 
reading. 
D 
100 
46-2 
46-25 
46-3 
46-35 
100 
46-2 
46-25 
46-3 
46 - 35 
96 
46-45 
46-45 
46-5 
46-75 
96 
45"65 
45-65 
45-8 
46-00 
90 
46-72 
46-75 
46-83 
47-28 
90 
45-35 
45-35 
45-55 
45"65 
* 80 
46-92 
47-08 
47-3 
47-9 
80 
45-05 
45-05 
45-10 
45"25 
70 
47-10 
47-35 
47-70 
48-4 
70 
44-88 
44'85 
44-9 
45-00 
60 
47-28 
47-50 
48-10 
48-75 
60 
44-7 
44-7 
44-7 
44"85 
52 
47-50 
47-72 
48-50 
49-00 
50 
47-60 
47-85 
48-60 
49-05 
50 
44-60 
44-6 
44-6 
44-78 
46 
47-80 
48-10 
48-75 
49-15 
40 
48-20 
48-50 
49-00 
49-4 
40 
44-45 
44'45 
44-45 
44-60 
30 
48-90 
49-20 
49-50 
49-90 
30 
44-25 
44-25 
44-28 
44-35 
20 
49-70 
50-00 
50-20 
50-5 
20 
44-00 
44-00 
44-00 
44-10 
15 
50-10 
50-45 
50 65 
50-95 
10 
50-80 
51-00 
51-30 
5T60 
10 
43-6 
43-6 
43-6 
43-6 
6 
51-5 
51-8 
52-2 
52-6 
6 
43-4 
43-4 
43-4 
43"4 
4 
5200 
52-5 
53-2 
54-00 
4 
43-2 
43'2 
43-2 
43-2 
