equalized by prolonged hunger. This points to one of those curious relations between sex and alimentation, which are so striking and yet so difficult to fix exactly.
In this short review of Prof. Bachmetjew's results, it is hoped that enough has been said to show that a considerable foundation has been laid down for further researches; but reference should be made to the book itself, which is full of carefully tabulated experiments, and most clearly expressed deductions from them.
It is clear that in dealing with the temperature of insects we have to do with a complex phenomenon dependent on a variety of interacting factors, some of which we have already touched upon. In the remainder of this paper I intend to consider one factor which I believe will have to be taken into account if we wish to gain a complete idea of the temperature relations of the so-called poikilothermic animals, i.e. animals whose temperatures vary with the surrounding medium. This factor is colour. The radiant powers of differently coloured surfaces are notably different; those surfaces which absorb the long-waved colours are better radiators than those which absorb chiefly those colours which lie at the other end of the spectrum. The emissive and absorptive radiant powers of a substance are directly proportional—a good radiator is a good absorber; it must also be remembered that a good reflector of radiant heat is a bad absorber and radiator, and vice versâ. It has for long been pointed out that a dark coloured animal would be able to take advantage of sunshine more readily than a light coloured one, and Lord Walsingham used this fact in explaining certain phenomena of melanism in Lepidoptera. In the controversy which arose on this head nothing conclusive was reached, but a certain amount of evidence was brought forward to show that on the whole, in regions where the sunshine was intermittent, a melanic tendency in the Lepidoptera became the rule rather than the exception. My chief collecting-ground for Lepidoptera abroad has been Haute Savoie, in the neighbourhood of Mont Blanc, and I have been struck there with the fact that the two kinds of butterflies which frequented the highest mountain regions were, on the one hand, the dark brown Erebias, and, on the other, the white Pierids and pale Coliads. This contrast struck me for some time as inexplicable on the theory that the colouration bore any
