This principle is carried into colour in most of our useful arts. The wheelwright instinctively lines out the rim and spokes and does not cross them, feeling that the effect would be to suggest weakness. Moreover, in all our handicraft work, the points and tips are emphasized with colour. This principle seems to hold good throughout nature. It is not suggested that the colouration is applied to important parts _in order to_ emphasize them, but rather that being important parts, they have become naturally the seats of most vivid colour. How this comes about we cannot here discuss, but shall refer to it further on. It is owing to this pervading natural principle, that we find the extreme points of quadrupeds so universally decorated. The tips of the nose, ears and tail, and the feet also proclaim the fact, and the decoration of the sense organs, even down to the dark spots around each hair of a cat's feelers, are additional proofs. Look, for instance, at a caterpillar with its breathing holes or spiracles along the sides, and see how these points are selected as the seats of specialized colour, eye-spots and stripes in every variety will be seen, all centred around these important air-holes. This leads us to our second principle, that of repetition, which simply illustrates the tendency to repeat similar markings in like areas. Thus the spiracular marks are of the same character on each segment. The principle of repetition, however, goes further than this, and tends to repeat the style of decoration upon allied parts. We see this strongly in many caterpillars in which spiracular markings are continued over the segments which lack spiracles; and it is probably owing to this tendency that the rib-like markings on so many mammals are continued beyond the ribs into the dorsal region. Upon these two principles the whole of the colouration of nature seems to depend. But the plan is infinitely modified by natural selection, otherwise the result would have been so patent as to need no elucidation. Natural selection acts by suppressing, or developing, structurally distributed colour. So far as our researches have gone, it seems most probable that the fundamental or primitive colouration is arranged in spots. These spots may expand into regular or irregular patches, or run into stripes, of which many cases will be given in the sequel. Now, natural selection may suppress certain spots, or lines, or expand them into wide, uniform masses, or it may suppress some and repeat others. On these simple principles the whole scheme of natural colouration can be explained; and to do this is the object of the following pages.