almost invisible as they glide among the foliage, by a similar coloration.

Thus we see that protective resemblance realizes the talisman of the fairy tale, and gives its possessor the power of rendering himself invisible. It requires no argument to show that those who possess it in the highest degree will oftenest escape their foes and catch their own prey, and that therefore every variation that tends to perfect it will be preserved, while variations. which are not useful will be neglected and suffered to disappear.

Mimicry. We now come to creatures whose colours neither conceal them nor make them to look like vegetable or mineral substances; who, on the contrary, are conspicuous enough, but so completely resemble creatures of a quite different group, that ordinarily they are mistaken for them. They remind us of masqueraders dressed up and painted for amusement, or of swindlers passing themselves off for respectable members of society. If this were done with conscious intention, it would be mimicry, strictly so called; but as we do not suppose that butterflies and birds put on the dress of others by voluntary imitation, the word is used only in a figurative sense. The term was adopted by Mr Bates, who first discovered the object of these curious imitations, and explained them as a result of natural selection.1 In the Brazilian forests, there are great numbers of birds which catch insects on the wing, and destroy, among others, many butterflies; but among the wings of butterflies found on the ground where the bodies have been devoured, you will not detect any of the Heliconidæ― butterflies, which are sandy in colour and slow in flight 1 Transactions, Linn. Soc., vol. xxiii. 1862, p. 495.

—though you find some of the large Nymphalidæ, which fly more swiftly. Or if you watch the birds bringing butterflies to their nests to feed their young, you will not find Heliconidæ among them, though they are flying lazily about in great numbers, and could easily be caught. The secret is that these butterflies have a strong and disagreeable odour, which renders them distasteful to the birds; and any young bird, after a few trials, would give up catching them. Under these circumstances, it is evident that any other butterfly of a group which birds were accustomed to devour would be almost as well protected by resembling a Heliconia as by resembling a leaf or a twig, always supposing that there were only a few of them among a great number of the Heliconias. Now, it is actually the case that among the white butterflies, forming the family Pieridæ, is a genus of rather small size (Leptalis), some species of which are white like their allies, while the greater number exactly resemble the Heliconidæ, not only in the colouring of the wings, but also in their form. In structural characters, the Pieridæ are as easily distinguishable from the Heliconidæ as buffaloes are from bears; but externally there is this resemblance, fitted to deceive the eye of a bird; nay, every length and shape of wing, every stripe and shade of colour, every band and tint and spot, are so much the same that only the close examination of a naturalist can make out the difference. It is intelligible, then, that when variations occur in the white butterflies, a slight approach in colour and form to the Heliconidæ will be some slight advantage, tending to deceive a bird at a long distance, and so secure a longer lease of life; which, though it be but one additional day, may suffice in many cases for

ORIGIN OF SPECIES.

81

laying eggs and leaving offspring. Among the offspring many will inherit the peculiarity which has been the safeguard of their parent, and if any exhibit it to a greater extent they will benefit by it all the more.

One of the most remarkable cases of mimicry is that of Charis melipona, a South American longicorn beetle, which has been so named from its resemblance to a small bee of the genus Melipona. The thorax and body are densely hairy, like those of the bee, and the legs are tufted in a manner most unusual in the order Coleoptera. Another longicorn (Odontocera odyneroides) has the abdomen banded with yellow, and constricted at the base, and is altogether so exactly like a small common wasp of the genus Odynerus, that Mr Bates informs us he was afraid to take it out of his net with his fingers, lest he should be stung. No doubt many a hungry bird had declined the morsel, being similarly deceived by the beetle's disguise. Nor is mimicry confined to insects. In the island of Bouro, and again in the adjacent island of Ceram, Mr Wallace discovered birds which he constantly mistook for each other, though they belonged to distinct families-orioles mimicking honeysuckers, and no doubt benefiting by the hypocrisy, since the orioles are weak birds, with small feet and claws, while the honeysuckers they resemble are strong, wellarmed, pugnacious, and able to defy the hawk.

The Origin of New Species through the Accumulation of Variations. It is abundantly clear that, in every species of animal and plant, variations are constantly occurring, and that through the agency of "selection " they are sometimes made to accumulate in definite directions till they constitute new breeds or varieties;

F

but now we have to inquire whether the difference ever becomes so great that the progenitor and the descendant must be classed as distinct species? To our minds, two species of plant or animal are separate pictures, as distinct from one another as a picture of St Peter's at Rome and St Paul's in London: Can the one pass into the other like the slow gradation of a dissolving view? Our authors say it can, and that in several ways.

First, by Artificial Selection, both conscious and unconscious. Of pigeons-the breeds of which are methodically produced-at least a score might be chosen, which, if shown to an ornithologist, and he were told that they were wild birds, would certainly be ranked by him as well-defined species. "Moreover," says Mr Darwin, "I do not believe that any ornithologist would place the English carrier, the short-faced tumbler, the runt, the barb, the pouter, and fantail in the same genus; more especially as in each of these breeds several truly-inherited sub-breeds, or species, as he might have called them, could be shown him." Yet, as before stated, it is the common opinion of naturalists that all have descended from the rock-pigeon; and if we could collect all the pigeons which have ever lived, from before the time of the Romans to the present day, we should be able to group them in several lines, diverging from the parent bird. Each line would consist of almost insensible steps, occasionally broken by some slightly greater variation or sport, and each would culminate in one of our present highly-modified forms.

Amongst dogs the selection exercised by man-sometimes purposely, sometimes unconsciously-has produced changes of form, of size, and of speed greater than those which distinguish different species. The greyhound

STERILITY OF ARTIFICIAL SPECIES.

83

differs much more from the wolf or the dingo than the racer does from the wild Arabian. The smallest dog is probably no bigger than the head of the largest. No wild dog, fox, or wolf is so small as the tiniest spaniel, or so large as the biggest Newfoundland dog; and no two wild animals of the family differ so widely as the bull-dog and the greyhound. The known range of variation, therefore, is more than enough for the derivation of all the forms of dogs, wolves, and foxes, from a common ancestor.

:

There is one great difference, indeed, between these apparent species produced artificially, and species not so produced the hybrids of the former are fertile with one another, while those of the latter are sterile. That is to say, while, if the horse is crossed with the ass, the resulting mules produce no offspring when paired together; the case is different with dogs and pigeons, for the carrier and fantail, or any other two pigeons, will produce mongrels which are as fertile between themselves as carriers are with carriers. This is certainly a great difference, but it is not fatal to the theory; because, although sterile hybrids have not yet been produced artificially, nobody has proved that such production is impossible, and future experiments, when the causes of sterility are better known, may be successful.1

1 Mr G. H. Lewes (Fortnightly Review, April 1868) states, that the European guinea-pig is believed to have descended from the American, but that the two will not couple together.

It occurs to the writer that sterile hybrids might perhaps be produced if the breeder directed his sharp eye to the food preferences of an animal, instead of selecting peculiarities of external structure. Sterility must depend on peculiarities of constitution not manifest externally, but correlated with the animal's diet. Let a number of pigeons be taken, let every variety of food be