26 A PHYSICAL STUDY OF THE FIREFLY. 
VIII. LUMINOUS EFFICIENCY AND CANDLE-POWER 
MEASUREMENTS. 
In the experiments on the Cuban firefly, Pyrophorus noctilucus, Langley* 
and his assistants found that to the eye the insect gave one-eighth as much 
light as an equal area of the candle flame, and that the actual candle-power 
of the insect was 1/1600 candle. ‘The light from this insect appears to be 
emitted continuously, but to fluctuate somewhat in intensity. 
Accurate measurements ona fluctuating source, such as the flashes emitted 
by the local species of fireflies, is, of course, impossible; but it is highly 
desirable to have some idea of the magnitude of the light emission. A 
simple ‘grease spot” photometer was therefore provided, the comparison 
source being a standard sperm candle. 
In the measurements on the glow of the firefly, the photometer screen was 
placed at 5 meters from the comparison source and the insect, with the 
luminous segments turned toward the screen, was moved back and forth 
until the neutral point was reached. This distance varied from 1.0, 1.5, 
to 2.2 cm., depending upon the intensity of the glow and the size of the 
insect. Some specimens for a short time glowed very strongly, 1/50,000 
candle, then suddenly decreased to 1/150,000 candle or Jess. ‘The weakest 
measurements were 1/250,000 candle; but these were strong as compared 
with the faint glow observed while photographing the spectrum of the 
Photinus scintillans. 
Measurements were made of 8 to 10 different specimens, the average 
intensity of the glow being of the order of 1/50,000 candle. Because of the 
inability to cause the insect to flash rapidly for any length of time, the 
measurements were more difficult, and hence uncertain. In this case the 
insect was held over a glass plate at a fixed distance (about 7.5 cm. from the 
photometer screen), and the latter was moved back and forth before the 
comparison source. The main difficulty encountered was the variation in 
intensity of the flash. A fresh specimen sometimes emitted several strong 
flashes which would suddenly become weak. 
As in the measurements on the glow, the specimens examined were of two 
kinds, viz, fireflies that had been in captivity for over a week and fireflies 
that had just been caught in the fields. Several other simple tests were 
tried, but no satisfactory measurements were made of the flash, due princi- 
pally to the great fluctuations in intensity. The measurements on the 
candle-power of the flash indicated a variation from 1/50 candle to 1/400 
candle, the predominating values being 1/400 candle. In all, about a dozen 
measurements of the intenstity of the flash were made on three healthy 
specimens of Photinus pyralis. It would have been interesting to make 
similar measurements of the Photuris, but it was too late in the season. As 
mentioned elsewhere, male specimens of the latter are frequently obtained 
which emit a “twinkling” flash 3 times per second, and will sometimes con- 
tinue to do so for 1 to 13 hours without serious interruption. 
In the previous investigation the radiant efficiencyf of the Photinus pyralis 
was found by multiplying the spectral energy curve by factors representing 
*Annals Astrophys. Obs., 2, p. 5, 1902. 
tIves and Coblentz, Bull. Bur. Standards, 6, p. 330, 1909. 
