It has been proved by experiment that when this animal is blinded the change does not occur.

In all these cases, therefore, we have some form of reflex or sense action by which the change is produced, probably by means of pigment cells beneath the skin as in the chameleon.
The second class consists of certain larvae, and pupae, which undergo changes of colour when exposed to differently coloured surroundings.
This subject has been carefully investigated by Mr.E.B.Poulton, who has communicated the results of his experiments to the Royal Society.[65] It had been noticed that some species of larvae which fed on several different plants had colours more or less corresponding to the particular plant the individual fed on.

Numerous cases are given in Professor Meldola's article on "Variable Protective Colouring" (_Proc.
Zool.

Soc._, 1873, p.

153), and while the general green coloration was attributed to the presence of chlorophyll beneath the skin, the particular change in correspondence to each food-plant was attributed to a special function which had been developed by natural selection.