Page:Popular Science Monthly Volume 15.djvu/454

438 Their presence implies an unceasing and more or less violent struggle between the tendency to an untimely death and the natural term of life. The prolonged existence of such variations, especially when they affect the more important organs, is scarcely possible; progressing in acute attacks, or through descent, either toward reversion or toward lineage extinction. Indeed, if abnormities were in their nature as permanent and compatible with the life of the blood as normalities, they would cease to be such, and all distinctions between them would vanish. Some deviations, however, from ordinary development, such as polydactylism, can not be considered morbid, and are in no way incompatible with the continuance of life.

Organic variability is mainly, if not entirely, an effect of changes in condition. The increased use or disuse of any part, a change from one climate to another, and of the quantity or quality of food, have long been noticed to be specially influential. The various species of domesticated plants and animals are far more variable than the same when wild, a difference readily accounted for by the fact that the latter are far less migrant; their nutriment and the culture of parts undergo little if any change in great lapses of time. Some variations, like that of the loss of wool which our sheep undergo when placed within the tropics, are apparently grounded upon adaptation to relation, others upon adaptation to use, and others are dependent on methodical selection.

One and all of these diverging tendencies are opposed by a single conservative principle, that of reversion. The tendency of animals long domesticated when they become feral to revert to the pristine type might also be supposed to arise from adaptation to relation. But the tendency is often seen when there has been no apparent change in the conditions, as in the many and familiar instances of atavism; and more decisively, by the crossing of species. The experiment of the Earl of Powis is one example of the many which evince the marked tendency to reversion through the crossing of blood. He had some domesticated hump-backed cattle crossed by the wild species from India, with the result, not of producing a medium grade of characteristics, but of a marked reversion to the ancient. Here, adaptation to relation can not be regarded as effective; only through the indelibility of types by prolonged transmission is such a prepotency explicable.

The principle of reversion is the touchstone of the fitness for survival of blood variations normal in character. If they withstand for indefinite generations the tendency to be supplanted by older and more deeply fixed characters, their fitness is proved, and at the same time they acquire prepotency for strong maintenance against fresh deviations. The principle is therefore eminently conservative, and not a mere blind tendency of the organic processes. Unlike the coarse and obtrusive struggle for existence between independent organisms, it is