It is quite natural that polygenists should have neglected facts of this nature. In their opinion a population with _mixed_ characters is a _species_ as much as any other, which is intermediate between two given specific types. But the indifference or the mistake of monogenists is less easily explained. They are evidently ignorant of the phenomena of crossings among plants and animals. When they meet with a race of undecided characters, and which presents more or less distant analogies with two different types, they have generally felt embarrassed, and have put the question on one side, or have at most invoked the action of conditions of life in a vague manner. It is quite true that the latter, when effecting a resemblance between foreign races and the local race, leads to results analogous to those which result from crossing. We have seen an example of it in the United States. Yet crossing has its peculiar phenomena, which are persistent even after several generations. Moreover, to the indications drawn from physical and physiological characters we may add others borrowed from very different orders of facts, and which, in many cases, permit us to draw a conclusion with remarkable certainty. The mixture of beliefs, customs, and manners often furnishes valuable information. But the comparison of languages generally throws an unexpected light upon problems apparently most difficult. From time to time legends and history confirm inductions drawn from the orders of facts which I have just pointed out, and testify to the correctness of views which, at first sight, might appear conjectural. As an example I will quote the Zulu Kaffirs. They are one of the groups of which some polygenists make a distinct species. They are in fact distinguished from other negro races by several characters. But by these characters they are brought nearer to the white type. Moreover, various travellers inform us that they present a great variability of feature. Missionaries who have lived among them add that, in the same family, and under conditions which render all crossing impossible, individuals are met with who have the hair and colour of a Negro, and others whose hair is smooth and whose colour is brown. These facts alone would authorise the conclusion that the Zulus are a mixed race. Philology confirms this conclusion. Philologists agree in placing the Kaffir languages in the group of Zimbian languages, whose grammar and vocabulary are fundamentally negro, but which also include arab, nilotic, and malgach elements. Thus language, as well as physical characters, points to a mixture of blood.