If the idioplasm really possessed the power of spontaneous variability ascribed to it by Nägeli; if, as a result of its own growth, it were compelled to undergo gradual changes, and thus to produce new species, we should expect that the duration of species, genera, orders, &c. would be of approximately equal length respectively, at least in forms of equal structural complexity. The time required by the idioplasm to undergo such changes as would constitute transformation into a new species ought to be always the same at equal heights in the scale of organization, that is, with equal complexity in the molecular structure of the idioplasm. It appears to me to be a necessary consequence of Nägeli’s theory that the causes of transformation lie solely in this molecular structure of the idioplasm. If nothing more than a certain amount of growth, and consequently a certain period of time during which the organism reproduces itself with a certain intensity, is required to produce a change in the idioplasm, then we must conclude that the alteration in the latter must take place when this certain amount of growth has been reached, or after this certain period has elapsed. In other words, the time during which a species exists—from its origin as a modification of some older species, until its own transformation into a new one—must be the same in species with the same degree of organization. But the facts are very far from supporting this consequence of Nägeli’s theory. The duration of species is excessively variable: many arise and perish within the limits of a single geological formation, while others may be restricted to a very small part of a formation; others again may last through several formations. It must be admitted that we cannot estimate the exact position of extinct species in the scale of organization, and the differences may therefore depend upon differences of organization: or they may be explained by the supposition that certain species may have become incapable of transformation, and might, under favourable conditions, continue to exist for an indefinite period. But this reply would introduce a new hypothesis in direct antagonism to Nägeli’s theory, which assumes that the variability of idioplasm takes place as the consequence of mere growth, and necessarily depends upon molecular structure. Nägeli himself asserts that the essential substance (idioplasm) of the descendants of the earliest forms of life is in a state of perpetual change, which would continue even if the series of successive generations were indefinitely prolonged[201]. Hence there can be no rest in the process of change which the idioplasm must undergo; and this is as true of each single species as it is of the organic world taken as a whole. We could, perhaps, find shelter in the insufficiency of our geological knowledge, but the number of ascertained facts is too great for this to be possible. Thus it is well