e yellowish-green when on a branch of the bottle-brush tree
whose half-dried leaves were of this colour, and yellowish when attached
to the wooden frame of a box. A few other observers noted similar
phenomena, but nothing more was done till Mr. Poulton's elaborate series
of experiments with the larvae of several of our common butterflies were
the means of clearing up several important points. He showed that the
action of the coloured light did not affect the pupa itself but the
larva, and that only for a limited period of time. After a caterpillar
has done feeding it wanders about seeking a suitable place to undergo
its transformation. When this is found it rests quietly for a day or
two, spinning the web from which it is to suspend itself; and it is
during this period of quiescence, and perhaps also the first hour or two
after its suspension, that the action of the surrounding coloured
surfaces determines, to a considerable extent, the colour of the pupa.
By the application of various surrounding colours during this period,
Mr. Poulton was able to modify the colour of the pupa of the common
tortoise-shell butterfly from nearly black to pale, or to a brilliant
golden; and that of Pieris rapae from dusky through pinkish to pale
green. It is interesting to note, that the colours produced were in all
cases such only as assimilated with the surroundings usually occupied by
the species, and also, that colours which did not occur in such
surroundings, as dark red or blue, only produced the same effects as
dusky or black.

Careful experiments were made to ascertain whether the effect was
produced through the sight of the caterpillar. The ocelli were covered
with black varnish, but neither this, nor cutting off the spines of the
tortoise-shell larva to ascertain whether they might be sense-organs,
produced any effect on the resulting colour. Mr. Poulton concludes,
therefore, that the colour-action probably occurs over the whole surface
of the body, setting up physiological processes which result in the
corresponding colour-change of the pupa. Such changes are, however, by
no means universal, or even common, in protectively coloured pupae,
since in Papilio machaon and some others which have been experimented
on, both in this country and abroad, no change can be produced on the
pupa by any amount of exposure to differently coloured surroundings. It
is a curious point that, with the small tortoise-shell larva, exposure
to light fr