very different schemes of coloration.
The Barbary sheep, again, which inhabits the dry hills bordering the deserts of Northern Africa, where the vegetation is parched and scanty at all seasons of the year, and the rocks of a red brown colour, is itself of a uniform reddish brown which harmonises exactly with its surroundings, and makes it very difficult to detect when lying at rest amongst rocks. This perfect harmony of coloration with its surroundings in the Barbary sheep may have been brought about by the need of protection from enemies, but seems to me far more likely to have been caused by the influence of the colour of its environment, for its four-footed foes hunt by scent and by night far more than by sight during the daytime.
The male moufflon of Sardinia, which lives in a temperate climate where the colours of its surroundings are much brighter and more diversified than is the case in the habitat of the Barbary sheep, is a much more conspicuously coloured animal than the latter, or than the females of its own kind. As the females and young of the Sardinian moufflon, which are of a uniform brown colour, are more difficult to see than the males in their somewhat conspicuous autumn and winter coats, the latter cannot be said to be protectively coloured. Either through the influence of sexual selection or that of an environment the general colour of which varies very greatly at different seasons of the year, the male of the Sardinian moufflon becomes during autumn and winter conspicuously coloured compared with the female, without detriment, however, to the well-being of the species.
During my long sojourn in the interior of South Africa, I made large collections of butterflies. There was one species (Precis artaxia, Hewits) which always puzzled me. This handsome insect is only found in shady forests, is seldom seen flying until disturbed, and always sits on the ground amongst dead leaves. Though handsomely coloured on the upper side, when its wings are closed it closely resembles a dead leaf. It has a little tail on the lower wing which looks exactly like the stalk of a leaf, and from this tail a dark brown line runs through both wings (which on the under sides are light brown) to the apex of the upper wing. One would naturally be inclined to look upon this wonderful resemblance to a dead leaf in a butterfly sitting with closed wings on the ground amongst real dead leaves as a remarkable instance of protective form and coloration. And of course it may be that this is the correct explanation. But what enemy is this butterfly protected against? Upon hundreds of different occasions I have ridden and walked through the forests where Precis artaxia was numerous, and I have caught and preserved many specimens of these butterflies, but never once did I see a bird attempting to catch one of them. Indeed, birds of all kinds were scarce in the forests where these insects were to be found. I now think that the form and colour of the under wings of Precis artaxia have more probably been produced by the influence of its environment than by the need for protection.
During the rainy season in South Africa, the open glades in the forests bordering the rivers are gay with multitudes of brightly coloured butterflies of many different species, and after a night's rain butterflies of various kinds may often be seen settling in masses round pools of water along waggon roads. Most of these butterflies are conspicuously coloured, though they are in perfect harmony with the sunlit flowers which spring up at the time of year when they appear. I cannot, however, believe that the need for protection against birds or other enemies has had anything whatever to do with the determination of their various colours, as in all my experience (and I have been all my life a close observer of nature) I have never once seen a bird feeding upon butterflies in Africa.
The coloration of certain animals in the Arctic and sub-Arctic Regions is somewhat remarkable, as at certain seasons it is conspicuously out of harmony with its surroundings, and cannot therefore be protective. The musk ox retains its dark brown coat the whole year round, although it lives almost constantly amidst a snowy environment. Mr. Wallace tells us that the reason why the musk ox does not turn white is because it has no enemies to fear, and therefore has no need of a protective coloration. He says: "Then we have that thoroughly Arctic animal the musk sheep, which is brown and conspicuous; but this animal is gregarious, and its safety depends on its association in small herds. It is therefore of more importance for it to be able to recognise its kind at a distance than to be concealed from its enemies, against which it can well protect itself so long as it keeps together in a compact body." As, however, according to the experience of Arctic travellers, large numbers of young musk oxen are annually killed by wolves, this explanation of a case in which an animal is manifestly not protectively coloured does not seem altogether satisfactory. Mr. Wallace, it may be noted, calls special attention to the coloration of the giraffe, which he considers to be protective; yet nothing, I think, is more certain than that a far smaller percentage of giraffes are killed annually by lions in Africa than of musk oxen by wolves in Arctic America. If this is so, the musk ox has more need of protective coloration than the giraffe. The musk ox is, I think, the only one amongst the few truly Arctic mammals which does not turn white during the winter months, for, unlike the barren ground caribou, it does not migrate southwards in the autumn to the dark spruce forests, which change of habitat no doubt has had an influence on the colour of the latter animals; since Peary's caribou, the most northerly form of the genus, whose habitat lies far within the Arctic Circle, where trees of any kind are non-existent, is almost absolutely white in colour. In spite, however, of the fact that the caribou inhabiting Ellesmere Land and the adjacent land masses are white, and therefore harmonise well in colour with the snowy wastes amongst which they live, they form the principal food of the white wolves inhabiting the same regions, which hunt them by scent and run them down just as easily as the grey and black wolves of Alaska capture the dark-coloured and very conspicuous caribou which frequent the mountain ranges of that country. It appears to me that the colour of a caribou's coat, whether it be white, black, or brown, cannot afford it any protection against wolves, which probably possess as keen a sense of scent as any animals in the world, and must surely hunt entirely by scent during the long dark months of the Arctic winter. If this is so, then the great diversity in the coloration of the various species of caribou inhabiting the North American Continent must be due to some other cause than the necessity for protection against wolves, practically their only four-footed enemies.
Speaking of other Arctic animals, Mr. Wallace believes that the Arctic fox of necessity turns white in winter in order to enable it to capture the white Arctic hares upon which it chiefly lives. Very little, however, is known as to the life-history of these two animals. But if the Arctic foxes hunt by scent, as they almost certainly do, during the constant darkness of the long Arctic winter, and the hares burrow beneath the snow, and are caught as a rule when completely hidden from sight below its surface, I think it is arguable that the influence of environment has been at least as potent a factor in bringing about the white coloration of these animals in winter as the necessity for protective coloration. At any rate, in Alaska and the Yukon Territory of Canada, where the country is covered with snow for more than half the year, and where the hares are white throughout the long winter, the foxes are red, black, or a mixture of these two colours, all the year round, and the lynxes grey; yet these two species of carnivorous animals depend almost entirely on the hares for their food supply. It is somewhat remarkable that in the sub-Arctic forests of Alaska and the Yukon Territory, where the cold is intense and the ground covered with snow for so many months of every year, only the hares and the stoats amongst mammals turn white in winter. But in these countries the land is covered for the most part with dark spruce forests, the influence of which—if there is anything in the influence of environment—may have been greater in determining the coloration of the mammals of this district than that of the snow-covered ground.
During winter in the Yukon Territory, moose turn very dark in colour on the under parts of the body, and at this season of the year leave the thick forests and live in the comparatively open valleys amongst willow and birch scrub, where they are said to stand out like haystacks amidst their snowy surroundings. The local race of caribou (Rangifer osborni), which live all the year round on the treeless mountain plateaus, are very dark in colour (with the exception of their necks), and, as I myself can testify, stand out very plainly when the open ground they frequent is covered with snow. Of the various races of wild sheep inhabiting the mountains of Alaska, the Yukon Territory, and Northern British Columbia, some are white all the year round, and therefore very conspicuous in summer when there is no snow on the ground, though difficult to detect in the winter; some are grey, with white heads, necks, and rumps; whilst others are nearly black,