The fact of specificity is supported by the fact of constancy of forms. de Vries has pointed out that regardless of the possible origin of new species by mutation the old species may persevere. Walcott has found fossils of annelids, snails, crustaceans, and algæ in a precambrian formation in British Columbia whose age (estimated on the rate of formation of radium from uranium) may be about two hundred million years and estimated on the basis of sedimentation sixty million years. And yet these invertebrates are so closely related to the forms existing today that the systematists have no difficulty in finding the genus among the modern forms into which each of these organisms belongs. W. M. Wheeler, in his investigations of the ants enclosed in amber, was able to identify some of them with forms living today, though the ants observed in the amber must have been two million years old. The constancy of species, _i. e._, the permanence of specificity may therefore be considered as established as far back as two or possibly two hundred millions of years. The definiteness and constancy of each species must be determined by something equally definite and constant in the egg, since in the latter the species is already fixed irrevocably. We shall show first that species if sufficiently separated are generally incompatible with each other and that any attempt at fusing or mixing them by grafting or cross-fertilizing is futile. In the second part of the chapter we shall take up the facts which seem destined to give a direct answer to the question as to the cause of specificity. It is needless to say that this latter question is of paramount importance for the problem of evolution, as well as for that of the constitution of living matter. _I. The Incompatibility of Species not closely Related_