Variation is the departure in any direction from the mean character of the species. When the variation in a large number of individuals, generally more or less isolated in space, is of a marked and constant type, the group of individuals which exhibit such variation is termed a variety. On the Darwinian Theory (q.v.) the existence of variation in the individuals of a species under nature is essential to natural selection, which proceeds by the elimination of those individuals varying in such manner as to render them unsuited to the conditions of their existence; just as the similar variation of domesticated animals and plants is the basis of artificial selection, which proceeds by the intentional choosing out of those individuals varying in any desired or desirable direction.
A question which is to-day much discussed among biologists is whether variation is determinate or indeterminate in direction; that is to say, whether organic forms have or have not any tendency to vary in particular ways. According to the extreme advocates of the Darwinian Theory there is primitively no tendency to any special mode of variation, any existing tendency being the result of the selection of those individuals which chanced to vary along these particular lines. According to other biological observers and thinkers there is, apart from the guidance of natural selection, an inherent bias, differing in different groups of organisms towards variation in determinate lines. This may be due to the inheritance of characters individually acquired under the stress of surrounding conditions (direct environmental determinism); or to constitutional tendencies inherent in the individuals of each species, and analogous to the inherent tendencies of inorganic substances, such as calcite or aragonite, hornblende or augite, to assume definite crystalline forms (innate specific determinism); or to the deep-seated anti-thesis of nutrition and reproduction as conditioned by the organic nature of the species (protoplasmic or physiological determinism).
It has been claimed by certain American biologists that palaeontological evidence establishes the existence of determinate variation. The teeth and the limb-bones of more than one series of fossil ungulates are found to exhibit variation along definite and determinate lines. The facts may be admitted; but the reasoning based thereon is inconclusive. The variation adduced is confessedly along lines that are advantageous to the individuals in which it occurs. It would, therefore, on the Darwinian Theory, escape that elimination which would be the fate of non-advantageous and neutral variations. This may be made clearer by an illustration. Suppose a pendulum free to swing in any direction. Let a number of slight impulses indeterminate in direction be applied to it, but let all those which are in any but one particular direction be damped down and thus eliminated. Let this occur many times in succession, and the result will be that the pendulum will swing with considerable amplitude in the determinate direction through the checking of the oscillations in all other directions. So, too, if the teeth of mammals varied indeterminately, and if all variations save those along one line (or several correlated lines) were neutral or non-adaptive, these latter would be eliminated through inter-crossing, while the adaptive variation would become evident. In the fossil forms the variations along non-adaptive lines would be so slight as to escape detection, while those in a plus or minus direction along adaptive lines would be assigned to different stages in the evolution of the variation in question.
We seem forced, therefore, to the observation of existing forms if we would settle the question of the determinate or indeterminate nature of variation. Even here the difficulties are great if not insuperable. For if we take the individuals descended from a particular pair of parental organisms, it is scarcely possible to prove that determinate variation, should it be found to occur, is not the result of hereditary idiosyncrasy peculiar to those parents. And if we take the adult individuals of the species indiscriminately, it is scarcely possible to prove that determinate variation, should it be found to occur, is not the result of the elimination, at an early stage of life, of those individuals which were found wanting in this adaptive variation.
A. R. Wallace and others have tabulated some results of the observation of variation in the state of nature; and Wallace has shown that variations in size or length of particular parts are considerable, 'usually reaching 10 or 20, and sometimes even 25 per cent. of the varying part,' and occurring in 5 to 10 per cent. of the specimens examined. These results incidentally show that in the species under examination there was no very rigid elimination, and that inter-crossing did not suppress variations from the mean to such an extent as is sometimes supposed.
As a concrete example of variation we may take that which is found to occur in the 'wing'-bones of bats as observed and tabulated by the present writer. The wing of the bat was selected (1) because the bones are readily measured even in dried specimens; (2) because they form mutually related parts of a single organ; and (3) because they offer facilities for the comparison of variations, not only among the individuals of a single species, but also among several distinct species. The bones chiefly concerned are those of the arm and forearm, the metacarpals, and the phalanges of the digits. A glance at the tables dealing with the noctule (Vesperugo noctula), the long-eared bat (Plecotus auritus), the pipistrelle (Vesperugo pipistrellus), and the whiskered bat (Vespertilio mystacinus), together with the measurements of the greater and lesser horseshoe bats (Rhinolophus ferri-equinum and R. hipposideros), show (1) that variations of not inconsiderable amount (10 to 12 per cent., or even in some cases twice this amount) occur among the related bones of the bat's wing, and (2) that these variations are to a considerable extent independent of each other. Comparing now not individuals of the same species, but the mean results for different British bats, it is shown that the different types of wing may have resulted from emphasising certain variations and suppressing others. For example, the long and narrow wing of the noctule would result from the accentuated development of the metacarpals of the second and third digits and the correspondingly reduced development of the metacarpals and phalanges of the fifth digit. On the other hand, the broad and ample wing of the horseshoe bats would result from the reduction in length of the second and third metacarpals, and the relative increase in the length of the phalanges of the fourth and fifth digits and of the tibia of the hind-limb to which the hinder edge of the wing-membrane is attached.
On the hypothesis of indeterminate variation it must be confessed that we are to-day not much in advance of Charles Darwin, who said: 'Our ignorance of the laws of variation is profound. Not in one case out of a hundred can we pretend to assign any reason why this or that part has varied.' It may fairly be surmised that the variation in size of parts may result from variation in the vascular and nervous supply of such parts. But how these variations in vascular and nervous supply are occasioned we do not know. When we say that they are the result of chance and are indeterminate, we merely mean that their causes are unknown, and that the mode of their determination is at present undiscovered.
Weismann has endeavoured to elucidate the problem of the origin of variations by the study of the germinal cells from which the individual is developed. A portion of the nuclear substance of the ovum or egg is lost before or slightly after the act of fertilisation by the extrusion of the polar cells. It may be that by this process, and possibly analogous processes in connection with the sperm, material is got rid of, the loss of which by disturbing the hereditary equilibrium gives rise to variation. Weismann maintains that amphimixis, or the mingling of the peculiarities of two individuals in fertilisation, 'has arisen from the necessity of providing the process of natural selection with a continually changing material, by the combinations of individual characters.' The suggestion has hardly yet, however, got beyond the speculative stage.
Much careful work is being done by Ponton and others on the influence of the environment on the individual with the production of acquired variations not necessarily inherited. Caterpillars of the pepper-moth (Amphidasis betularia) reared from eggs produced by the same mother and fed on the same food-stuff were found to vary from nearly black through brown and gray to green (due to the colour of the blood showing through the transparent skin) according as the twigs or other objects upon which they habitually rested were dark or light. The nature and proximate causes of individual variation, and the experimental inquiry whether such variations are in any cases inherited, offer a wide field for future research.
For references, see article HEREDITY; cf. also Darwin's Origin of Species (1859), Variation of Animals and Plants under Domestication (1868), &c.; Spencer's Principles of Biology (1866), Factors of Organic Evolution (1886); E. D. Cope, Origin of the Fittest (New York, 1887); P. Geddes, article 'Variation and Selection,' Ency. Brit.; A. R. Wallace, Darwinism (Lond. 1889); the present writer's Animal Life and Intelligence (1890), 'Nature and Origin of Variation' (Proc. Bristol Nat. Soc., vi. 1890-91); G. J. Romanes, Darwin, and After Darwin (1892); H. de Varigny, Experimental Evolution (Lond. 1892); and A. Weismann, Essays upon Heredity and Kindred Biological Problems (Oxford, 1889-92).