Ray and Tournefort have had from time to time a few imitators, who, in their definition of species, have clung to one of the two ideas. But the immense majority of zoologists have been aware of the impossibility of separating them. To convince ourselves of this fact it is only necessary to read the definitions which they have given. Each one of them, from Buffon and Cuvier to MM. Chevreul and C. Vogt has, so to speak, proposed his own. Now, however they may differ in other respects, they all agree in this. The terms of the definitions vary, each endeavours to represent in the best manner possible the complex idea of species; some extend it still further, and connect with it the ideas of cycle and variation: but in all the fundamental idea is the same. In a case of such difficulty as that of finding a good definition for a combination of ideas, the latest comer always hopes to improve upon his predecessors. For this reason I have also given my formula: “Species is a collection of individuals more or less resembling each other, which may be regarded as having descended from a single primitive pair by an uninterrupted and natural succession of families.” In this definition, as also in that of some of my colleagues, among others of M. Chevreul, the idea of resemblance is made of less importance and subordinate to that of filiation. The fact is that there never is an identity of characters between one individual and another. Putting aside the variations resulting from age and sex, it is at once evident that all representatives of the same specific type differ in some points. Although these differences are very slight, they constitute _individual traits_, _shades_ as Isidore Geoffroy said, which enable us to distinguish between two of the same species. But the differences are not confined within these limits. The specific types are _variable_, that is to say, every kind of physical character is modified in their derivatives and, under the influence of certain conditions, to such an extent as to make it often very difficult to recognise their unity of origin. This, again, is a fact upon which all naturalists agree. Blainville even, who, defined species as “the individual repeated and continued through time and space,” distinctly recognised this _variability_; for the individual is perpetually undergoing modification, and does not retain its similarity during the various stages of life. He admitted, moreover, the existence of distinct races.