106

EFFECTS OF CROSSING.

and but little-specialized forms in the ocean is no argument against these views, since it is probable that, like the lancelet in the sand, they occupy places in the polity of nature in which they do not come into competition with more highly developed forms.

There are many reasons why a lowly organized form should not be exterminated; it may be too prolific and too small to come into competition with larger creatures. Lancelet may compete with lancelet, or minute crustaceans may war with each other or with equally minute forms; but, just as the huge whale is unable to exterminate the prolific Pteropod, highly organized forms rarely completely destroy minute and lowly developed animals.

An objection has been made by several authors that crossing and intercrossing would prevent improvement except under isolation; but it must be remembered that numerous individuals are continually varying in one direction. The circumstances which induce any particular form of variation in one individual, however complex, will tend to produce a similar variation in many; and if these are favoured in any way, they are sure to leave a majority of descendants. Again, one favourably varying individual, although crossed and intercrossed with many having a less favourable form, will undoubtedly tend to improve the whole of

i

their descendants, unless another cross occurs from an individual which has varied in the opposite direction. If the first variation is favourable and the second unfavourable, it is clear that the race will be oftener crossed by individuals having the more favourable variation; and thus the species will eventually inherit the more favourable tendency, which will tend to become more and more marked.

Again, all tendencies towards specialization or the development of new parts will tend towards isolation. The little lancelet-like ancestor of the Vertebrata by the development of fins would become more extended in its distribution; the individuals which were best adapted for swimming would leave the bed of the ocean and occupy a higher level. Thus those capable of swimming would become separated from those incapable of swimming, and would become more or less isolated from their less-active ancestors: whilst the non-evolving race retreated into the sand to escape from enemies increasing on every side, the more highly developed forms would seek the surface of the waves, and thus come under totally new conditions of life. As evolution progressed, the conditions of life would become more and more unlike; certain forms would become isolated in rivers and pools; and these, like the aquatic plants amongst which they passed their

lives, would gradually spread from the water to the damp places around it, and so eventually become habited to the dry land.

We might therefore expect that uniform improvement would go on over the most extended areas, by crossings with the best-adapted variations occurring more frequently from their surviving to the period of maturity more often, or from their obtaining more food and therefore leaving a larger number of vigorous descendants than the lessfavoured varieties. Thus a type form originates, and members of it become isolated by spreading into areas where different conditions appertain, either becoming inhabitants of a different level in the ocean or spreading into colder or warmer latitudes, and becoming altered by the direct effect of conditions, so that they are no longer fertile when crossed with their ancestors. The type form diverges in different directions, and new types occur, each more specialized than the first. These come into competition by migrating beyond their original areas, and mutually react on each other. If they are mutually strong, each may become modified, and new forms may arise under these new conditions, so that each type becomes modified and innumerable forms arise from the highly specialized types; these continually prevent any considerable deviation, in the direction of specialization, amongst

the older and simpler forms, which cannot compete with them.

Thus we should expect to find exactly what we do find-type forms with groups of allied forms around them. Transitional forms would occur rarely, as they do, but would not have innumerable forms descended from them, except in certain definite directions. The study of classification and the affinities of living beings is entirely confirmatory of the views which have been adopted here.

Mr. Darwin treats this subject briefly in his 'Origin of Species.' He says:-"As natural selection acts solely by the preservation of profitable modifications, each new form will tend, in a fully stocked country, to take the place of and finally to exterminate its own less-improved parent form and other less-favoured forms with which it comes into

competition." The expansion of this pregnant sentence includes most of the views expressed in the foregoing chapter.

110

CHAPTER XIV.

ON THE ORIGIN OF THE HIGHER
VERTEBRATES.

In order to bring the argument into a narrower compass, the application of the preceding views will be attempted to the case of certain special forms. The three great terrestrial types of vertebrates (reptiles, birds, and mammals) have been selected for this purpose. These have all attained a high degree of specialization; and they have been chosen because they present especial difficulties.

As we have already seen, the facts of geographic botany have led M. De Candolle to believe that aquatic plants are the most ancient; zoology and comparative anatomy tend to the same view amongst the higher vertebrates; and an endeavour will be made to show that the marine and aquatic forms of reptiles, birds, and mammals tend to unite those forms with each other and with the lower amphibian and fish or ichthyopsid types in a very remarkable manner.

It is universally admitted that the Amphibia form the connecting link between the fishes and

1