In order to understand the following it is necessary to consider more precisely the nature of coloured light. In this two factors are always to be distinguished, the brightness of the light, in physical terms, the intensity, and the colour of the light, or in physical terms its wave-length. In studying the colour-sense of animals we are concerned only with the question whether they are able to distinguish light of different wave-lengths, and we have to take the greatest care, therefore, that the reactions of the animals are not connected with differences in the intensity of the light. v. Frisch made use of the "chess-board" method. On a square board were fixed a large number of pieces of cardboard of all shades of grey, from the lightest white-grey to the darkest black-grey. Among these was placed a piece of cardboard of the colour on which the bees had been trained (blue). If the bees had no sense of colour, but distinguished between the different pieces of cardboard only by their degree of brightness, they would confuse the blue with one or other of the greys, one of which would certainly be of the same tone. If, on the other hand, they had a true colour-sense, such confusion would not occur. The result of the experiment was in favour of the possession of such a sense by the bees. The trained bees, having been accustomed to find their food on a blue background, flew to the blue pieces of cardboard only, and paid no attention to any of the others.
Since these fundamental researches were made by v. Frisch, the colour-sense of bees has been studied with much finer tech nique, particularly by Kuehn. Above all, we have learnt to train these insects to distinguish the pure colours of the spectrum. We
are now able, therefore, to form a fairly precise estimation of the colour-sense of these insects. Bees perceive all wave-lengths between 650-313 ,u,u. Within this range they distinguish f our different colours. The first comprises wave-lengths from 65o to about soo 12/2, and includes our red of shorter wave-length, and also yellow and green. Seemingly, all these appear to them as one colour. On the other hand, the red of longer wave-length, which we use in our photographic dark-rooms, bees do not distin guish at all; they confuse it with black. The second comprises wave-lengths from 500-480 ALA and corresponds to our blue-green, the third lies between 48o and 400 (our blue and violet). Lastly, bees perceive a fourth visual region in the ultra-violet, which is beyond our powers of sight.
It lies between 400-3101.44. Bees may therefore be trained to distinguish ultra-violet just as well as blue or yellow.
A sense of colour has been proved to exist also in Lepidoptera, dragonflies and flies; perhaps it is universal among insects.
We have gone to work in quite a different manner to prove the existence of a colour-sense in Crustacea. Some of these possess the remarkable power of matching their colouration to that of the surface on which they live. On light backgrounds they are light, on dark ones they are dark. This is brought about by means of particular pigment cells (chromatophors) which are present everywhere beneath the transparent cuticle. The majority From these experiments the fact of Adaptation may be proved as a logical inference. The sensitiveness of our eyes is consider ably greater in weak light than in strong. The eye thus adapts itself completely to the light conditions present. Starting with an original intensity of i,000 light units, Balanus first reacts when this is diminished by 33 units. With an original intensity of 25o units the animal first notices a diminution by io units. On pursu ing the subject further, we find that with an original intensity of 25 units, diminution of this by only one unit is noticed by the animals. The eye, therefore, becomes more sensitive as darkness increases.