1 6 A PHYSICAL, STUDY OF THE FIREFLY. 



i meter. When used with a light flint-glass prism the dispersion in the 

 yellow was sufficient to separate the yellow mercury by 0.5 mm., the total 

 length of the spectrum between the green (X = 0.501/*) and the red (X = 0.667/x) 

 helium lines being 37 mm. The firefly light falls within this same region. 

 A small spectrograph, of triple achromatic lenses 6 cm. diameter and 18 

 cm. focal length, which had a much greater light-gathering power, was 

 used in photographing the weak radiations in the red. The same prism (of 

 light flint-glass) was used on both instruments. In the large spectrograph 

 the slit opening was 0.75 mm. and in the small instrument it was 0.4 mm. 

 The collimator slit was covered with a small piece of white cardboard con- 

 taining an opening 5 mm. high and 2 mm. wide, which enabled the operator 

 to center the fireflies over the slit, by means of their own light; for, as already 

 mentioned, the photographing was done in a completely darkened room. 



The method of procedure consisted in holding the insects in the fingers, 

 one or two at a time, over the spectrographie slit. The constant struggle of 

 the insect to get away caused it to flash more frequently than is its accus- 

 tomed rate. If it became quiet and ceased to flash, a movement of the 

 fingers, or allowing the insect to move about, increased the frequency of the 

 flashing. Frequent flashing is an important item when it comes to making 

 exposures to weak sources like the firefly, requiring from 1 to 5 hours to 

 obtain a satisfactory negative. 



In the small spectrograph only from 1 to 60 minutes were required to 

 obtain a uniformly exposed, dense negative. On the small apparatus a 

 single flash from the Photinus pyralis was sufficient to cause a streak across 

 the plate (see Plate 1 , D,V 12; and VII 5, 6, and 8), and in order to obtain a 

 uniformly exposed negative, the slit was covered with a piece of ground 

 glass. In Plate 1 , D, V 2 the negative was obtained in 2 minutes from 50 

 flashes of pyralis through ground glass; III 2 was similarly obtained from 8 

 flashes; III 5 from 20 flashes; III 6 from 35 flashes; III 7 from 60 flashes; 

 III 8 from 5 flashes; and III 9 from 100 flashes or about 4 minutes. On the 

 other hand, the negative directly below this one, III 10, required 30 minutes 

 exposure (without glass) to a male Photuris pennsylvanica which flashed at 

 the rate of about 100 times per minute. 



As indicated elsewhere, the longer time required for exposure to the 

 Photuris is not because of the lack of intensity, but because of the shorter 

 (as compared with pyralis) duration of the flash. Although the luminous 

 organs in the two species are different, those of the Photuris being the smaller, 

 the area covered on the spectrograph slit is not markedly different in the 

 two insects. 



On the spectrometer slit, the Photinus scintilla us and consanguineus could 

 be made to flash about 4 to 5 times per minute. The Photinus pyralis 

 flashed about 20 times per minute when at its best, but the flash lasts longer 

 and therefore (aside from its greater volume) has a greater effect on the plate 

 than the flash of the Photuris pennsylvanica. In the Photuris pennsylvanica 

 the flash, accompanied by a " twinkling effect," often occurs at the rate of 2 

 to 3 times per second, which in some male specimens was sometimes con- 

 tinued for an hour or more with but few interruptions. 



Only a few specimens of Photuris glowed vigorously during the photo- 

 graphing. The Photinus pyralis easilv tires of flashing and then emits a 



