30 A PHYSICAL STUDY OF THE FIREFLY. 
no appreciable difference to expose the two junctions or only the dark june- 
tion to radiation. ‘This is due in part to thesmallness of the bright junction. 
The radiating of such a thermopile by the candle test has but little meaning 
because of the absorption by the glass walls, which absorb perhaps 40 to 50 
per cent of the incident radiation. The test was applied, however, giving 
a deflection of 2.5 cm. for a sperm candle at 2 meters and galvanometer 
sensitivity of 7=3 107? amp., or 10 cm. for a candle at 1 meter. 
In actual work on fireflies the galvanometer sensitivity was increased to 
i=1.5X10- amp. to 7X10~ amp. or 5 times as sensitive as for the 
candle test. In his measurements Langley found that: 
A portion of the flame of a standard sperm candle, equal in area to the bright part of 
the insect, gave, under the same circumstances, a bolometric effect of such magnitude 
that had the heat of the insect between 1/80,000 been as great as this from the candle, it 
would certainly have been recognized. 
‘The sensitivity of the present apparatus, under conditions similar to 
those just quoted, was such that radiation 1/60,000 as great as that of the 
candle would have caused a deflection of 1 mm. The thermopile in vacuum 
A 
C 
Cu OCT 
% To Bi 
sx Galv. 
Cu \ 2008 
B 
Pt. ON Cu i SS et 
a iy N E 
BI Cu a } 2202. 
Pt. i, 
Fic. 13.—Thermopiles for measuring the temperature and radiation of fireflies. 
is far more steady than a bolometer in air and covered with a glass window, so 
that while the present sensitivity appears to have been less than that used 
by Langley and his assistants, the efficiency or reliability was no doubt con- 
siderably greater. ‘That the sensitivity of the instrument was very high is 
indicated by the fact that when a firefly was placed on the glass walls over 
the sensitive junction, or within 5 to 10 mm. of the same, the galvanometer 
indicated a cooling of this junction which attained its maximum in about 
I minute. 
It will be shown presently that the temperature of the firefly is some- 
what less than the surrounding air; hence in the heat interchange the pile 
would radiate to the insect. However, the radiation from the pile could not 
pass through the glass, and the explanation offered is that the glass was 
cooled (principally by conduction) by the insect and in turn the dark junc- 
tion radiated to the glass; whence the slowness of the operation. This 
observation of a cooling effect when the insect was near the thermopile came 
as a surprise, and for a while it appeared that there might be an endothermic 
action taking place in the photogenic cells instead of the long-sought-for 
