INFRA-RED ABSORPTION MEASUREMENTS. 41 
above is a sample. The size of the specimen examined was about 1.5 X 
7mm. In this same illustration, Fig. 14, curve 6 gives the transmission 
through one of the eye-like fenestrate membranes on the thorax of the Cuban 
firefly, and through which is emitted the so-called ‘thoracic light.”” The 
specimen was only 1.52 mm., and hence very difficult toexamine. After 
removing all the photogenic cell ligaments which adhered to it this specimen 
was fairly translucent. In the region from 1p to 2u the transmissivity is 
probably higher than observed, the low values being due to scattering of the 
radiation in passing through the specimen. 
Curve c, Fig. 14, gives the transmission through the outer integument 
covering the photogenic cells of Photinus pyralis. ‘The specimen examined 
was a single dried abdominal segment, 1.54 mm., which had been moist- 
ened in water and freed from all ligaments and photogenic material. The 
specimen was not very smooth and it was only semi-translucent. No doubt 
a better sample could have been prepared by dissecting a fresh undried 
insect. In both curves, 6 and c, the opacity at 1 to 2u and at 5u is greatly 
increased by scattering of the transmitted radiation which did not reach 
the bolometer. However, as was anticipated in previous communications, 
these curves have the characteristic absorption spectrum of complex carbo- 
hydrates, in which there is great (and usually complete) opacity from 2.8y 
to 3.8u, and beyond 6, with a fairly transparent region at 4.54 to 5u. It 
may be noticed that these specimens were dry. If they had been fresh, 
then the water present would have absorbed considerable of the radiation at 
4 to Su. 
During the present summer an examination was made of the infra-red 
absorption of the fresh photogenic substance of Photinus pyralis, excised and 
placed between fluorite plates. The material was an inhomogeneous white 
mass, which in the preliminary examination showed complete opacity from 
3u to 3.8u, a more transparent region at 4.5 to 5.5u, followed by complete 
opacity beyond 5.8u. This is the same sort of absorption as found in the 
chitinous layer, Fig. 14. A thin layer of the white fluorescent fluid which 
exudes freely from the wings of the pyralis was also examined between fluo- 
rite plates. Complete opacity was observed from 2.9 to 3.24 and beyond 
5.84; and an appreciable absorption band was observed at 4.74. These 
bands were considered to be due to water, and since no others were observed 
a more thorough examination seemed unnecessary. Apparently the fluor- 
escent material has no appreciable absorption bands in this region of the 
spectrum. 
The experiments of Ives have bridged the gap from the visible to 1.5y. 
The data then published by the writer showed that but little if any radiation 
of wave-lengths greater than 2.5 would pass outinto spaceeven if generated 
inthe photogenic cells. Thatwhichcan beemitted at 4.5u to 5.5u will beso 
weak in intensity that it will hardly be distinguished from the emission spec- 
trum due to the ‘‘animal heat’? (maximum emission at 8 to 1ou) radiated 
from the surface of the body. By exclusion, we have, therefore, reduced the 
unexplored region to that part of the spectrum lying between the wave- 
lengths 1.5uand 2.5u. There are no types of radiators known which would 
lead one to expect emission bands in this region of the spectrum of the light 
of the firefly, and while it is important to be on the alert for the unexpected, 
