All analogous interpretations are absolutely arbitrary. In fact, superiority between human groups depends essentially upon intellectual and social development; it passes from one to another. The Chinese and Egyptians were already civilized, when all Europeans were true savages. If the latter had judged our ancestors as we too frequently judge foreign races, they would have found many signs of inferiority in them, commencing with the white skin of which we are so proud, and which they would have been able to regard as betraying an irremediable degeneration. Is the fundamental superiority of one race really betrayed outwardly by some material sign? We are still in ignorance upon this point. But when we examine it more closely, we are led to think that it is not so. In expressing myself thus, I know that I am separating myself from the opinions which are generally admitted, and am at variance with men whose works I value most highly. But I hope to give decisive proofs in my favour further on. Differences of every kind nevertheless exist between one human group and another. These must be taken for what they are, for _characters_ of _race_, for _ethnical characters_. It is the duty of the anthropologist especially to recognise these differences, to make use of them for defining the groups, then to connect or separate, according to their affinities, the races thus characterised. In other terms, his work is the same as that of the botanist or geologist describing and classifying plants and animals. Men of an impatient or venturous disposition will perhaps reproach me with making anthropology too _descriptive_. I shall only make a partial defence against the accusation. Provided that the description embraces the entire being, it enables us to become acquainted with it. If we take our stand on this point of view, we remain on the ground of positive knowledge, and run less risk of losing ourselves in hypotheses. I still consider it the right and almost the duty of the anthropologist, to investigate the causes which may have given rise to the appearance of the features which characterise races. The study of the actions of the conditions of life sometimes gives valuable indications on this subject. The evolution of the human being from his appearance in the embryonic state to the adult state, especially furnishes facts of great interest. A simple _arrest_, a slight _excess_ in the evolutive phenomena, are, it appears to me, the causes of the principal differences which separate races, and particularly the two extremes, the Negro and the White.