408 MR. E. B. POULTON ON THE COLOUR-RELATION BETWEEN EXPOSED 
healthiest larvae, while the remaining three were small, and of these only one lived to 
pupate. Hence an unhealthy condition, or even a stunted size, might become 
secondarily associated with one of the two varieties of a dimorphic species after the 
power of being influenced by the surroundings had been lost. Nevertheless, it is not 
necessary for the growth of some such association that the loss of susceptibility should 
have taken place, for it has already been shown that the gilded appearance and cor¬ 
responding absence of pigment colours are associated with the presence of parasites in 
the pupae of the Vanessidae. It is clear from my experiments, and the previouslv 
quoted experience of others, that the susceptibility to corresponding colour influences 
has been lost in P. machaon, and Dr. Fritz Muller shows that it has also been lost 
in P. polydamus. On the other hand, the striking observations of Mrs. Barber upon 
P, nireus, and of Mr. Trimen upon P. demoleus, prove conclusively that these species 
are highly susceptible to the influence of certain colours ; and when it is remembered 
that both of the former non-sensitive species are dimorphic, and furthermore present 
in each case the two varieties, green and brown, which harmonise best with their 
surroundings, it appears probable that such dimorphism is the remnant of a former 
susceptibility which has, at any rate to a great extent, disappeared. Future experi¬ 
ments must finally decide whether the relative amounts of illumination produce any 
effect upon P. macliaon, as Mr. Harwood believes, or whether either form of pupa 
exhibits a more or less constant relation to a healthy or unhealthy condition, or finally, 
whether the formation of either variety is the spontaneous result of individual 
variability. 
Experiments upon Pieris brassiere and P. rap®. 
Having read of Mr. T. W. Wood’s observations, and those of other naturalists, I 
was extremely anxious to obtain the larvae of these species to investigate in the 
manner already described under V. urticce, &c. I could only obtain single specimens 
in Oxford, and I thought that the experiments would have to be delayed for another 
season. However, just when the experiments on V. urticce came to an end, on 
September 8, I went to Seaview, in the Isle of Wight, and there found the kitchen 
gardens ravaged by the larvae of P. brassicce, while those of P. rapee were very 
abundant on mignonette. Accordingly, I made the experiments described below, and, 
although they were not as accurately or minutely conducted as those on F. urticce, they 
yield some valuable results, and entirely confirm the previously adopted conclusion 
that the colour-effects are due to larval and not to pupal sensitiveness. Being away 
from my laboratory, and not expecting such an opportunity, it was, of course, impossible 
to carry out the experiments in the most satisfactory manner. The results obtained 
with the two species are described together, because they were, in nearly all cases, 
kept under similar conditions and were, in fact, often placed in the same cylinders. 
Furthermore, the results were remarkably uniform. When no locality is mentioned 
in any experiment it is understood that the included larvae were captured at Seaview. 
