32 A PHYSICAL STUDY OF THE FIREFLY. 
This apparent increase in the cooling effect in the more active (flashing) 
state of the insect is a complex phenomenon requiring further investigation. 
As it stands, it could not be verified by the iron-constantan thermo-couple 
measurements on the temperature of the different parts of the body of the 
insect. ‘The thermo-couple shows, Fig. 13, A, that the temperature of the 
insect is lower than the air (this is in agreement with the radiometric obser- 
vations), and that, used differentially, the temperature of the luminous seg- 
ments is higher than the rest of the body; but it was not possible to show 
with certainty whether the temperature of the luminous segments varied 
(apparently decreased by the radiometric test) during the act of flashing. 
After trying out these heat-conduction effects with the insect close to the 
vacuum thermo-element, the radiometric apparatus was arranged so that a 
mirror, 15 cm. in focal length and 10 cm. diameter, projected an image of an 
opening (2.5 mm. diameter) in a heavy black cardboard upon the active 
thermo-junction. ‘The whole was covered so that no radiation, except that 
which passed through the opening in the cardboard, could fall wpon the 
thermo-element. The firefly was then held with the luminous segments 
over the opening in the cardboard. Both glowing and flashing specimens 
were examined, but in only one instance was any deflection (a kick) observed 
that might be interpreted as possibly being due to the light (flash) from the 
firefly. It was therefore concluded that, since the glass wall of the bismuth- 
platinum thermo-element absorbed the long waves, no radiations less than 
3u were observable. The radiations which are greater than 3u would fall in 
the category of ‘‘animal heat,” which will now be discussed. 
Langley avoided the effect of “animal heat’’ by interposing a glass screen, 
which prevented an interchange between the insect and the radiometer of 
radiations greater than 3 to 4u. It has just been noticed that under certain 
conditions, when the insect was near (or touching) the glass walls of the 
thermopile, the deflection of the galvanometer indicated that the sensitive 
junction was losing heat by radiation; and, from the slowness of the action, 
that this was due to the local cooling of the glass wall surrounding it. Evi- 
dently conditions might be brought about whereby a heating effect, caused 
by a true radiation from the insect, might be counterbalanced by a cooling 
effect (conduction) due to a difference in temperature between the insect 
and the radiometer, the latter being at the higher temperature. Further- 
more, if the light production is a biochemical process, as some believe it to be, 
it is important to know whether this low temperature is due to the fact that 
we are dealing with a “cold-blooded” animal, or due to an endothermic 
reaction. ‘Thermometric measurements were accordingly undertaken to 
determine the nature of the heat distribution in the body of the firefly. The 
first tests were made radiometrically with the vacuum thermo-element just 
discussed. It was found that the glowing luminous segments, when sepa- 
rated from the rest of the insect and placed upon the vacuum thermopile, 
exerted a certain cooling effect which decreased when life was extinct. An 
insect on its back caused a cooling, but it was impossible to determine by the 
vacuum thermo-element whether there was a difference in the temperature 
of the dark and luminous segments. 
‘These experiments were continued by exploring the temperature distri- 
bution in the segments of the insect. For this purpose, thermo-couples of 
