To make clear the relations of these two factors to one another and to heredity, let us take a case in which the operations of all three may be severally identified and distinguished.

Here is one of those persons, occasionally met with, who has an additional finger on each hand, and who, we will suppose, is a blacksmith. He is neither aided nor much hindered by these additional fingers; but, by constant use, he has greatly developed the muscles of his right arm. To avoid a perturbing factor, we will assume that his wife, too, exercises her arms in an unusual degree: keeps a mangle, and has all the custom of the neighbourhood. Such being the circumstances, let us ask what are the established facts, and what are the beliefs and disbeliefs of biologists.

The first fact is that this six-fingered blacksmith will be likely to transmit his peculiarity to some of his children; and some of these, again, to theirs. It is proved that, even in the absence of a like peculiarity in the other parent, this strange variation of structure (which we must ascribe to some fortuitous combination of causes) is often inherited for more than one generation. Now the causes which produce this persistent six-fingeredness are unquestionably causes which "operate through inheritance." The Duke of Argyll says that "natural selection includes and covers all the causes which can possibly operate through inheritance." How does it cover the causes which operate here? Natural selection never comes into play at all. There is no fostering of this peculiarity, since it does not help in the struggle for existence; and there is no reason to suppose it is such a hindrance in the struggle that those who have it disappear in consequence. It simply gets cancelled in the course of generations by the adverse influences of other stocks.

While biologists admit, or rather assert, that the peculiarity in the blacksmith's arm which was born with him is transmissible, they deny, or rather do not admit, that the other peculiarities of his arm, induced by daily labour—its large muscles and strengthened bones—are transmissible. They say that there is no proof. The Duke of Argyll thinks that the inheritance of organs enfeebled by disuse is "not generally disputed;" and he thinks there is clear proof that the converse change—increase of size consequent on use—is also inherited. But biologists dispute both of these alleged kinds of inheritance. If proof is wanted, it will be found in the proceedings at the last meeting of the British Association, in a paper entitled "Are Acquired Characters Hereditary?" by Professor Ray Lankester, and in the discussion raised by that paper. Had this form of inheritance been, as the Duke of Argyll says, "not generally disputed," I should not have written the first of the two articles he criticizes.

But supposing it proved, as it may hereafter be, that such a functionally-produced change of structure as the blacksmith's arm shows us, is transmissible, the persistent inheritance is again of a kind with which natural selection has nothing to do. If the greatly strengthened arm enabled the blacksmith and his descendants, having like strengthened arms, to carry on the battle of life in a much more successful way than it was carried on by other men, survival of the fittest would ensure the maintenance and increase of this trait in successive generations. But the skill of the carpenter enables him to earn quite as much as his stronger neighbour. By the various arts he has been taught, the plumber gets as large a weekly wage. The small shopkeeper by his foresight in buying and prudence in selling, the village-schoolmaster by his knowledge, the farm-bailiff by his diligence and care, succeed in the struggle for existence equally well. The advantage of a strong arm does not predominate over the advantages which other men gain by their innate or acquired powers of other kinds; and therefore natural selection cannot operate so as to increase the trait. Before it can be increased, it is neutralized by the unions of those who have it with those who have other traits. To whatever extent, therefore, inheritance of this functionally-produced modification operates, it operates independently of natural selection.

One other point has to be noted—the relative importance of this factor. If additional developments of muscles and bones may be transmitted—if, as Mr. Darwin held, there are various other structural modifications caused by use and disuse which imply inheritance of this kind—if acquired characters are hereditary, as the Duke of Argyll believes; then the area over which this factor of organic evolution operates is enormous. Not every muscle only, but every nerve and nerve-centre, every blood-vessel, every viscus, and nearly every bone, may be increased or decreased by its influence. Excepting parts which have passive functions, such as dermal appendages and the bones which form the skull, the implication is that nearly every organ in the body may be modified in successive generations by the augmented or diminished activity required of it; and, save in the few cases where the change caused is one which conduces to survival in a pre-eminent degree, it will be thus modified independently of natural selection. Though this factor can operate but little in the vegetal world, and can play but a subordinate part in the lowest animal world; yet, seeing that all the active organs of all animals are subject to its influence, it has an immense sphere. The Duke of Argyll compares the claim made for this factor to "some bit of Bumbledom setting up for Home Rule—some parochial vestry claiming independence of a universal empire." But, far from this, the claim made for it is to an empire, less indeed than that of natural selection, and over a small part of which natural selection exercises concurrent power; but of which the independent part has an area that is immense.

It seems to me, then, that the Duke of Argyll is mistaken in four of the propositions contained in the passages I have quoted. The inheritance of acquired characters is disputed by biologists, though he thinks it is not. It is not true that "heredity is the central idea of natural selection." The statement that natural selection includes and covers all the causes which can possibly operate through inheritance, is quite erroneous. And if the inheritance of acquired characters is a factor at all, the dominion it rules over is not insignificant but vast.

Here I must break off, after dealing with a page and a half of the Duke of Argyll's article. A state of health which has prevented me from publishing anything since "The Factors of Organic Evolution," now nearly two years ago, prevents me from carrying the matter further. Could I have pursued the argument it would, I believe, have been practicable to show that various other positions taken up by the Duke of Argyll do not admit of effectual defence. But whether or not this is probable, the reader must be left to judge for himself. On one further point only will I say a word; and this chiefly because, if I pass it by, a mistaken impression of a serious kind may be diffused. The Duke of Argyll represents me as "giving up" the "famous phrase" "survival of the fittest," and wishing "to abandon it." He does this because I have pointed out that its words have connotations against which we must be on our guard, if we would avoid certain distortions of thought. With equal propriety he might say that an astronomer abandons the statement that the planets move in elliptic orbits, because he warns his readers that in the heavens there exist no such things as orbits, but that the planets sweep on through a pathless void, in directions perpetually changed by gravitation.

I regret that I should have had thus to dissent so entirely from various of the statements made, and conclusions drawn, by the Duke of Argyll, because, as I have already implied, I think he has done good service by raising afresh the question he has dealt with. Though the advantages which he hopes may result from the discussion are widely unlike the advantages which I hope may result from it, yet we agree in the belief that advantages may be looked for.