On page 61 of his Principles of Biology, he further assures us "that arrest of coördination is death, and that imperfect coördination is disease."
A superficial view of Mr. Spencer's definition would involve the inference that, upon the evolution hypothesis, only one of two things is possible. Either there is an ever-continuing imperfect coördination, or there is an always perfect coördination. As parts subserve actions, the perfect coördination of the latter must be dependent upon the perfect coördination of the former. Now, evolution implies a constant change. In fact, according to the hypothesis, constant change is the only normal state. The variation of parts, then, would entail their imperfect coördination, and, consequently, the imperfect coördination of their actions; for the only conceivable way in which the imperfect coördination of actions is possible, is by a change in the parts subserving those actions. As variations, then, are ever occurring, imperfect coördination must always exist.
The following is the alternative view. The evolutionist might assume an ability in each organism to effect, on the occurrence of each variation, a re-coördination. This view manifestly admits only of perfect coördination. But the advocate of evolution may avoid these absurd conclusions by affirming, as he has tacitly done, that, while the organism is capable of coördinating any number of characters, imperfect coördination may ensue by a too sudden change in any part or parts. This is the issue which we desired to produce, the decision of which will, we conceive, legitimately preclude further argument. The question is, Is the organism capable of coördinating any number of characters? or, are all the characters of the species alone susceptible of coördination? The reader will perceive that the latter is a mere recurrence of our proposition that the proportionate development of all the parts is necessary to perfection, and that the absence of any part is deleterious to the organism. If we prove this, we shall have completely disproved the evolution hypothesis.
There is a fact adduced by Darwin which places the validity of our theory beyond all doubt, and which is, at the same time, grossly at variance with the conception of evolution. The fact to which we allude is, that good results from crossing. Observing this result, Darwin propounds a general law of nature, that all organic beings are benefited by an occasional cross. This law he employs as a somewhat important factor of evolution, and essays to harmonize it with his theory. In this attempt he succeeds. But mere congruity with a law is no proof of the validity of a theory, where that law is only an empirical one. Of this every person conversant with science is aware. It is equally well known, however, that when a theory is shown to accord with a law; to furnish an explanation of it; and to resolve it into a higher law, thus changing it from an empirical into a derivative law; proof conclusive and incontrovertible has been adduced. If the reader has not already mentally anticipated our argument, it remains for us to prove that the theory of reversion fulfils these requirements.
Our theory manifestly implies that the more proportionate the development, the greater is the approach to perfection. It also implies that the more characters of the species there are in each variety, the nearer is the approximation to perfect coördination. It is apparent at a glance, then, that crossing furnishes a crucial test of the truth of our views. For most varieties are distinguished from each other by the possession of positive features. The presence of the peculiar character of one variety, of course, implies its absence in the others. Each variety possesses a character or characters which the others lack, and lacks what the others peculiarly possess. When, then, two such varieties cross, good must of necessity accrue to their offspring. For, in the formation of the latter, each variety supplies a deficiency of the other. Could a reason be more obvious? or could proof of a view be more conclusive? So conclusive is it, we conceive, that were any other result consequent on crossing, such a circumstance would be at variance with our theory.
Of the fact that good results from crossing, not a doubt can reasonably be entertained. Darwin, so far from questioning the fact, is its most strenuous advocate. But upon his conception, it is crossing per se which produces the favorable effects. In other words, this is another of Darwin's ultimate laws. Being purely empirical, the general law of nature which he assumes, fails utterly to explain the cause of the variations in the quantity of the effects. The crossing of pigeons, for instance, is attended by the greatest gain in constitutional vigor, while comparatively little good results from the crossing of the varieties of the horse, sheep, or cow. On our doctrine, the explanation is clear. The many widely distinct varieties of the pigeon necessarily imply great disproportionate development of each. They are, then, extremely susceptible of improvement. The races of the horse, sheep, and cow, on the other hand, approximate, as we have seen, to proportionate development. There is, therefore, much less room for improvement. Strikingly in harmony with this interpretation is the fact that, with pigeons, the more highly bred the crossed varieties are, the greater is the gain from a cross. Equally congruous is the fact that the more highly bred the breeds of the horse, cow, and sheep are, the less is the gain. The reason is, careful and select breeding produces increased divergence of character with pigeons; but with horses, sheep, and cattle it induces increased convergence. The former become widely distinct, while the latter converge in character. All the characters are developed in each variety of the latter; but in the former different characters are developed in different varieties. While, then, coördination in the horse, sheep, and cow advances toward perfection, coördination in the pigeon is rendered more imperfect by careful breeding. Each variety of the pigeon possesses a character which, when joined with those of another variety, will entail a great advance toward due coördination. This concurrence is effected by crossing, and the result is, as one would be led to expect upon our doctrine, great beneficial effects. With the horse, sheep, and cow the effects of a cross between varieties are less marked, owing to less imperfect previous coördination.
In noting the advantage accruing to crossed offspring, we have particularly referred to gain in constitutional vigor. We have occasion now to speak of gain in fertility. Seeing that hybrids—the product of a cross between species—are invariably sterile, it is clear that, if the conception that varieties are incipient species is a valid one, we are bound to expect that the more marked, distinct, and widely divergent varieties are, the greater will be their sterility. The mere circumstance that such an effect is not observable, goes far to invalidate the conception. What, then, must the inference be when an effect diametrically opposite to that necessitated by the conception is shown to result—when increased fertility is seen to follow crossing, and when this increased fertility is observed to be directly proportionate to divergence of character? Such results would, we apprehend, negative completely the hypothesis of evolution, and would conclusively confirm our view, that the beneficial effects are owing to the disproportionate development which a multiplicity of widely distinct varieties necessarily implies. These results we have, and they are indisputable. For the fact that crossing induces increased fertility, and that this increased fertility is directly proportionate to divergence of character, is so well known that it is scarcely necessary to adduce proofs from Darwin in support of it. But that the least shadow of a doubt may not remain, we will quote a few of Darwin's remarks on the subject.
Constant reference to crossing may be found in any portion of his late work. But a somewhat lengthy chapter is devoted exclusively to this subject and to close interbreeding. In the conclusion of this chapter (p. 142, vol. ii.) he says:
"In the early part of this chapter it was shown that the crossing of distinct forms, whether closely or distantly allied, gives increased size and constitutional vigor, and, except in the case of crossed species, increased fertility to the offspring. The evidence rests on the universal testimony of breeders.... Although animals of pure blood will obviously be deteriorated by crossing, as far as their characteristic qualities are concerned, there seems to be no exception to the rule that advantages of the kind just mentioned are thus gained even when there has not been any previous close interbreeding. The rule applies to all animals, even to cattle and sheep, which can long resist breeding in-and-in between the nearest blood relations. It applies to individuals of the same sub-variety, but of distinct families, to varieties or races, to sub-species, as well as to quite distinct species.
"In this latter case, however, while size, vigor, precocity, and hardiness are, with rare exceptions, gained, fertility, in a greater or less degree, is lost; but the gain cannot be exclusively attributed to the principle of compensation; for there is no close parallelism between the increased size and vigor of the offspring and their sterility. Moreover, it has been clearly proved that mongrels which are perfectly fertile gain these same advantages, as well as sterile hybrids."
On page 174, he reiterates these statements, which place the subject of increased fertility beyond all doubt.