the colours of certain Lepidoptera. 487 



anterior brown band in each segment and the brown median 

 dorsal stripe are due to true pigment in the epidermic cells, while 

 the broader green band crossing the posterior part of each segment 

 is due to green fat lying beneath the epidermis, which is of a pale 

 yellowish colour over it. The yellow patches on each side of the 

 middle line in the anterior part of each segment are due to part of 

 the more deeply placed yellow fat. 



PLATE XV. 



Figs. 1 and 2. — The dark and light varieties of the larvse of 

 Catocala elocata, obtained in the experiments described on pp. 302, 

 303. The larvae are represented about three-quarters of the natural 

 size, and the difference in shade was far greater than appears from 

 these figures. 



Fig. 3. — The dark and light larviE of Hemevophila abruptaria, 

 obtained in the experiments described on pp. 31G, 317. The larvae 

 are represented about three-quarters of the natural size, and the 

 difference of shade is very well expressed. 



Fig. 4. — A lamp-shade, like that used in Experiment IV. upon 

 Aniphidasis hetularia (1889), see pp. 331, 332. In front of and 

 beside the lamp-shade are represented the five pieces of stick 

 which were used in the experiment. The figure is about one- 

 fourth the real size of the articles. These pieces of stick, placed 

 among the green leaves of nut in the cylinder, turned far more 

 than half the larvae dark. 



Fig. 5. — About one-fifth the real size. The conflicting colour 

 case used chiefly in the experiments on Vanessa urticce and V. io 

 (see pp. 391 — 397 and 420 — 426). A complete description of it is 

 given on p. 393). The difference between the alternate strips of 

 gilt and black paper is not distinct, although it can be made out. 



Fig. 6. — About one-fifth the real size. The conflicting colour 

 case used in the experiments on Vanessa io (see p. 425, where the 

 case is described). The distinct white spots in this and the last 

 figure represent the bosses of silk spun by the larvae. In use, the 

 cover (the upper part of the figure) was placed over the compart- 

 ments (the lower part), so that the white compartments had a 

 black cover, and vice versa, and the dorsal and ventral surfaces of 

 the larvae within were subjected to opposed conditions. 



