We cannot share Bather’s confidence in the security of a bridge composed of even linked species. Let such a series be never so perfect, let the gradation be never so minute, as it might conceivably be made, when not merely distinct species, but also hybrids, mutants and fluctuants are available as stopgaps, the bare fact of such intergradation tells nothing whatever concerning the problem of genetical origin and specific relationship. The species-by-species method does, however, represent the very minimum of requirement imposed upon the palæontologist, who professes to construct a fossil pedigree. But, when all is said and done, such a method, even at its best, falls considerably short of the mark. However perfectly intergradent a series of fossils may be, the fact remains that these petrified remnants of former life cannot be subjected to breeding tests, and that, in the consequent absence of genetical experimentation, we have no means of determining the real bearing of these facts upon the problem of interspecific relationship. Only the _somatic_ characters of extinct floras and faunas have been conserved in the rock record of the past, and even these are often rendered dubious, as we shall see presently, by their imperfect state of preservation. Now, it is solely in conjunction with breeding experiments, that somatic characters can give us any insight into the nature of the _germinal constitution_ of an organism, which, after all, is the cardinal consideration upon which the whole question of interspecific relationship hinges. All inferences, therefore, regarding the descent of fossil forms are irremediably speculative and conjectural. When we are dealing with living forms, we can always check up the inferences based on somatic characteristics by means of genetical experiments, and in so doing we have found that it is as unsafe to judge of an organism from the exclusive standpoint of its external characters as it is to judge of a book by the cover; for, apart from the check of breeding tests, it is impossible to say just which somatic characters are genetically significant, and which are not. Forms externally alike may be so unlike in germinal constitution as to be sexually incompatible; forms externally unlike may be readily crossed without any discernible diminution of fertility. “Who could have foreseen,” exclaims Bateson, “that the apple and the pear—so like each other that their botanical differences are evasive—could not be crossed together, though species of _Antirrhinum_ (Snapdragon) so totally unlike each other as _majus_ and _molle_ can be hybridized, as Baur has shown, without a sign of impaired fertility?” (_Heredity_, Smithson. Inst. Rpt. for 1915, p. 370.) We cannot distinguish between alleged specific, and merely mutational (varietal), change, nor between hybridizations and factorial, chromosomal, or pseudo-, mutations, solely on the basis of such external characters as are preserved for us in fossils. It