But there is another factor to be considered. Those Alpine Lepidoptera, for instance, which have not remained exactly the same in the far north, have formed local varieties in the rest of the area of their distribution also, while species which have remained quite alike in isolated regions, such as the Alps and the north, exhibit no aberrations in other isolated regions, such as the Pyrenees, in Labrador, or in the Altai. Thus one species must have had a tendency in the Glacial period to form local varieties, and the other had not; and I have already attempted to explain this on the hypothesis that the former at the time of their migration and segregation into different colonies were at a period of dominant variability, the latter at a period of relatively great constancy. Leaving aside the question of the causes of this phenomenon, we may take it as certain that there are very variable and very constant species, and it is obvious that colonies which are founded by a very variable species can hardly ever remain exactly identical with the ancestral species; and that several of them will turn out differently, even granting that the conditions of life be exactly the same, for no colony will contain all the variants of the species in the same proportion, but at most only a few of them, and the result of mingling these must ultimately result in the development of a somewhat different constant form in each colonial area.
If we were to try to imitate this 'amixia' artificially we should only require to take at random from the streets of a large town a number of pregnant bitches, and place each of them upon an island not previously inhabited by dogs, and then a different breed of dog would arise upon each of these islands, even if the conditions of life were exactly similar. But if, instead of these variable bitches, the females of a Russian wolf were placed on the islands, the developing wolf-colonies would differ as little from the ancestral species as the various Russian wolves do from one another—similar climate and similar conditions of life being presupposed.
There is thus an evolution of varieties due to amixia alone, and we shall not depreciate the significance of this if we consider that individual variations are the outcome of the fluctuations in the equilibrium of the determinant system of the germ-plasm, to which it is always more or less subject, and that variations of the germ-plasm, whether towards plus or minus, bear within themselves the tendency to go on increasing in the direction in which they have begun, and to become definite variational tendencies. In isolated regions such variational tendencies must continue undisturbed for a long period, because they run less risk of being suppressed by mingling with markedly divergent germ-plasms.
The probability that variational tendencies set up in some ids of the germ-plasm by germinal selection will persist and increase is obviously greater the more the germ-plasms combining in amphimixis resemble each other. For instance, let us call the varying determinants Dv, and assume as a favourable case that these are represented in three-fourths of all the ids in the fertilized eggs of a butterfly-female which has been driven astray on to an island, that is, that they are present in twelve out of sixteen ids; then of 100 offspring of the first generation it is possible that seventy-five or more will contain the determinants Dv, some of them in a smaller number of ids, some in a great number than the mother, according as the reducing division has turned out. If the pairing of the second generation be favourable—and this again is purely a matter of chance—a third generation must arise which would contain the variants Dv throughout, and thus the fixation of this particular variation on this particular island would be begun. In other words, the possibility would arise, that, if individuals with a majority of Dv ids predominated, they would gradually come to be the only ones, since by continual crossing with the minority which possessed only the determinants D, they would mingle the varied ids with those of the descendants of these last, till ultimately germ-plasm with only the old ids would no longer occur.
In following out this process it is not necessary to assume that the first immigrant possessed the variation visibly; if determinants varying in a particular direction occurred in the majority of its ids, these would, as a consequence of persistent germinal selection, go on varying gradually until the externally visible variation appeared. This would not have appeared at all if the animal concerned had remained in the original habitat of its species, for there it would have been surrounded by normal germ-plasms, and its direct descendants, even if they had been as favourably situated for the origin of variations as we have assumed, would not have reproduced only among themselves, and therefore even in the next generation the number of Dv ids would have diminished.
Obviously it is to a certain extent a matter of chance whether in the isolated descendants the variation or the normal form remains the victor, for it depends on the number of Dv ids originally present in the fertilized eggs, then on the chances of reducing divisions, and finally on the chance which brings together for pairing individuals in which the similarly varied ids preponderate. The probability of the conquest of the variation will depend in the main on the strength of the majority of the varied ids in the fertilized eggs of the parents; if this be an overwhelming majority, then the chances of favourable reducing divisions and pairings will also be great. The origin of a pure amixia variety will thus depend upon the fact that the same variational tendency Dv was present in a large number of the ids of the ancestral germ-plasm. We need not wonder therefore that of the numerous diurnal butterflies of Corsica and Sardinia only eight have developed into endemic, probably 'amiktic,' varieties.
But since we know that so many species in oceanic islands and other isolated regions are endemic or autochthonous, i.e. of local origin, there must obviously be some other factor in their evolution in addition to the mere prevention of crossing with unvaried individuals of the same species. The variational tendencies which have arisen in the germ-plasm through germinal selection may—as we have already seen—gain the ascendancy in various ways; first, by being favoured by the climatic influences, then by being taken under the protection of personal selection, whether in the form of natural or of sexual selection.
As the inhabitants of insular areas are not infrequently subject to special climatic conditions, we may assume at the outset that many of the 'endemic' species are climatic varieties, but in many cases this explanation is insufficient. For instance, special local forms of mocking-bird live on several of the Galapagos Islands, but this cannot depend upon differences of climate, for the islands are only a few kilometres apart, and resemble one another as regards the conditions of life which they present. But as the differences between these local forms show themselves especially in the male sex, as colour variations of certain parts of the plumage, we must take account of sexual selection, which, though with its basis in germinal selection, has in many islands followed a path of its own. Sexual selection operates especially in the case of sporadically occurring characters which are in any way conspicuous. But it is just such variations as these that are called into existence by germinal selection, whenever it is allowed to continue its course undisturbed through a long series of generations. Characters of this kind, such, for instance, as feathers of abnormal structure or colour in a bird, or new colour spots in a butterfly, make their appearance when a group of determinants has been able to go on varying in the same direction for a long time unimpeded, that is, without being eliminated as injurious by natural selection or obliterated by crossing. This is very likely to happen in the case of an isolated area, and as soon as the conspicuous character thus brought about makes its appearance, sexual selection takes control of it, and ensures that all the individuals, that is, all the germ-plasms which possess it, have the preference in reproduction.
I believe, therefore, that a large number of the endemic species of birds and butterflies in isolated regions result from amixia based upon germinal selection, whose results have been emphasized by sexual selection. Experience corroborates this, as far as I can see, for many of the endemic species of birds in the Galapagos and other islands differ from one another solely or mainly in their colouring, and in many it is especially the males which differ greatly.
As to the humming-birds we may say, without going into details regarding their sexual characters and their distribution, that the many endemic species which inhabit the Alpine regions of isolated South American volcanic mountains differ from one another chiefly in the males and in the secondary sexual characters of these. The family of humming-birds is characteristically Neotropical, that is, it has its centre in the Tropics of the New World, and by far the greater number of humming-bird species—there are about a hundred and fifty—occur there only, while a few occur as migrants north of the Tropical zone, and visit the United States as far north as Washington and New York. We know that many of the most beautiful species have quite a small area of distribution, that many are restricted to a single volcanic mountain, living in the forests which clothe its sides. These species are isolated there, for they do not migrate; apparently they cannot endure the climate of the plains, but remain always in their mountain forests. Without doubt they originated there, chiefly, I am inclined to think, through the variation of the males due to sexual selection. Any one who has seen Gould's magnificent collection of humming-birds in the British Museum in London knows what a surprising diversity of red, green, and blue metallic brilliance these birds display, what contrasts are to be found in the diverse colour-schemes, and what differences they exhibit in the length and form of the feathers of the head, of the neck, of the breast, and especially of the tail. There are wedge-shaped, evenly truncate, and deeply forked tails, some with single long, barbless feathers, and so on. All these characters are confined to the males, and are at most only hinted at in the female; in no species does the female even remotely approach the male in brilliance or decorativeness of plumage.