Page:Popular Science Monthly Volume 65.djvu/226

222 connection, he does not feel justified in agreeing with those who are ready to accept explanations outside the pale of science. The same naturalist, however, will always be found ready to admit that he is yet exceedingly far from being able to give an 'explanation' of the inner meaning of the real significance of the mutation process.

In order to penetrate into this it is necessary to analyze further the phenomenon of heredity. Concerning this, de Vries has already on a previous occasion published theoretical views which follow in the footsteps of Darwin's celebrated theory of pangenesis. As the chemist operates with molecules and atoms, for the reconstruction of the processes of inorganic nature, so the biologist, when trying to represent to himself living matter, has to take into account the smallest entities, which have received various names from various naturalists, and to which de Vries gives that of 'pangens.'

Pangens are something different from complicated molecules; they can assimilate and they can reproduce themselves. Not only does all living matter, wherever found, consist of them, but those smallest living particles must at the same time be considered, either individually or grouped together, as being bearers of single or of mutually correlated properties of living matter.

An augmentation or a diminution of the number of pangens which represent a certain property will call forth the phenomenon which we have named fluctuating variations; while a modification in the composition of the pangen, for example, by division in two unequal parts, or by substitution—using a term well known in chemistry—will be equivalent to a mutation (progressive mutation), as will also the disappearance of a determined pangen (regressive mutation). Thus, according to these abstract representations which we form of the mysteries of heredity, the fluctuating variation depends on quite a different category of phenomena from those of the chance variation. And we understand directly that the chance variation obeys a more complicated mechanism than fluctuating variation, which depends only on the greater or lesser numerical importance of the preexisting elements, while the chance variation, the formation of species, implies a change of the existent elements.

As to how this change in the pangens periodically takes place, both simultaneously and successively, among a certain number of individuals, or might be aroused or caused; as to how the unequal division or substitution obeys fixed laws in such a way that the mutants, arranged in groups, are alike—all this for the present can not be explained by us.

If we aim to understand the conditions we shall be able to create species, as we can now breed improved races. And as we gradually