each species of graduation of structure so insensible that not a link in the chain of evidence is wanting."
On the other hand, the writer has recently completed a microscopic study of a group of ancient lamp-shells—animals which looked somewhat like molluscs, but which were very different internally—with altogether different results. The particular changes involved were minor matters of surface markings, which could have had no conceivable importance to the animals. Selection, therefore, may be virtually ruled out: indeed, many of the different forms lived close together, with apparently equal success. But in the small markings on the shells there appear, as one follows the series from bottom to top, very decided changes, and those changes are, in some cases, abrupt and complete.
In others the variations are very small—indeed they could be distinguished only with the microscope—but so far as could be told, were distinct. This, therefore, points to a course of evolution that was clearly a matter of mutation, without any apparent governing by the process of natural selection.
The conclusion which we may reach, therefore, is that both natural selection and mutation operate in the development of new forms from old. The variations, for which Darwin was at a complete loss to account, are in many cases the mutations emphasized by de Vries and his followers. But to what extent climate, food; habits, and multitudinous other environmental factors, coupled with such in-
