8th. Again, it has been said, that the discussions which are always arising between naturalists as to whether a species should be preserved or regarded as a race, as to the genus, family, order, and sometimes the class in which it should be placed, betray a want of precision in general ideas.—Those who talk in this manner forget the immense number of species and races accepted and classified without discussion. They shut their eyes to all cases except those in which divergences of opinion occur. If, however, facts of this nature prove anything against a science and its fundamental data, then even mathematical theorems must be considered as wanting in precision, for there are disputes among mathematicians. 9th. I have already replied to the arguments drawn from the fertility of certain hybrids by showing to what it is reduced. Writers who insist upon this point invariably forget the lesson taught us by disordered variation and reversion without atavism. I regret being obliged to place among them Darwin, who, in his later writings, has shewn much less reserve than in his earlier publications. In the last edition of his book, he quotes what I have said of the cross between the Bombyx cynthia and the Bombyx arrindia; he speaks of the number of generations obtained, but he forgets to mention that disordered variation appeared in the second generation, and that reversion to one of the parental types was almost complete at the termination of the experiment. X. Species is then a reality. Let us take a group of individuals more or less similar, but always capable of contracting fertile unions, and let us, with M. Chevreul, trace it in imagination to its origin. We shall see it divided into _families_, each of which will have risen either mediately or immediately from one pair of parents. We shall see that the number of these families decrease at _each generation_, and rising still higher we shall at length find the initial term of a _single primitive pair_. Has this really been the case? Has each species indeed arisen from one single pair, or have several pairs, resembling each other perfectly both morphologically and physiologically, appeared simultaneously or successively? These are _questions of fact_ which science neither can nor ought to approach, for neither experiment nor observation is able to furnish us with the smallest data requisite for the solution. But what science may affirm is that _from all appearances_ each species has had, as point of departure, a single primitive pair. CHAPTER IX. CROSSING BETWEEN HUMAN GROUPS.—UNITY OF THE HUMAN SPECIES. I. We now know what are species and race; the phenomena exhibited by mongrels and hybrids furnish us with an experimental means of distinguishing them. We can, therefore, now reply to the question which has necessitated this discussion: Are there one or many human species? Are the human groups races or species?