426 Mr. E. B. Poulton. Experimental Proof that the Colours 



where the tough cuticle could not be easily penetrated by their small 

 weak mandibles. The larger larvae do not experience the same 

 difficulty. 



It will be well, however, to repeat the experiment with other larvee, 

 some of which may be expected to have greater powers of endurance. 

 I would suggest Mamestra brassicce and Phlogophora meticulosa as 

 suitable for the purpose. Freshly cut mid-ribs might be offered 

 every day or perhaps twice a day. 



Conclusions from the Experiments. Assuming that the results ob- 

 tained in Experiment (2) are not pathological, and I believe that 

 this assumption is justified, it follows that etiolin (1), no less than 

 chlorophyll (3), can be transformed into a larval colouring matter, 

 which may be either green or brown, and is so disposed as to form a 

 ground colour. 



The fact that brown pigments may be thus formed is new. In my 

 previous paper (* Roy. Soc. Proc.,' 1885, pp. 269 et seq.) I gave 

 reasons for the conclusion that the green pigments are derived from 

 plants, but argued that brown pigments are proper to the larva. 

 This still remains true in. many cases. Thus the green larvae of 

 Amphidasis betularia, investigated in 1892, are coloured by derived 

 pigments contained in the superficial fat, while the brown larvae are 

 coloured by true pigment contained in the epidermic cells (' Trans. 

 Ent. Soc. Lond.,' 1892, pp. 357359), so that the green fat which lies 

 beneath is concealed. The intensely opaque and dark larvae of many 

 other Geometrae are probably similarly coloured by true pigments in 

 the cuticle or epidermis. But the brown ground colour of many 

 Nbctua larvae will probably be found to be due, like that of T. pronuba, 

 to modified plant pigments. 



A comparison of the larvae fed on pigmentless food (Plate 3, fig. 2, 

 Plate 4, figs. 1 3) with those fed upon etiolated leaves (Plate 3, 

 fig. 1) and the similar larvae fed upon green leaves, proves that both 

 green and brown ground colours are modified plant pigments. When 

 the larvae fed on etiolin were being compared on October 10, one of 

 them became irritated and expelled a drop of fluid from its mouth. 

 This fluid was of a faintly bluish-green colour. This observation 

 suggests that the change of etiolin into a soluble green pigment takes 

 place in the digestive tract. Chlorophyll similarly becomes soluble 

 and forms a green solution (turning brown on exposure) in the di- 

 gestive tract of larvae. It is possible that the brown ground colour 

 of the larvae is also a result of oxidation : at any rate, it is a change 

 in the direction of greater stability ; for I have shown that the colours 

 of certain brown larvae, evidently coloured like those of T. pronuba, 

 are far more persistent after preservation than those of the green 

 varieties of the same species ('Roy. Soc. Proc.,' 1885. pp. 275, 276). 



Although the brown ground colour, probably situated in this 



