But the surest method of arriving at conclusive evidence in anthropology is necessarily _travels_. Doubtless the study alone of the materials collected from afar is of the greatest possible use. But we repeat concerning the study of mankind what we said about the study of animals; the anthropologist must leave his library and go into the great continents, in order to study by means of his own eye-sight. “We can only arrive at the distinction of species,” says M. Flourens, “by _direct and complete personal observation_.” That it must be complete, we have endeavoured to show; but the only condition for its being complete is its being direct. Had we even the genius of Buffon,[358] we should see but poorly by means of others; facts reach us distorted and altered, because they have not always been observed by competent men; they are not comparable, resulting as they do from diverse individual impressions. It will be especially necessary to control with care travellers’ tales as regards the study of intellectual tendencies, since they are too often influenced by their own ideas on the subject. Let us say this before concluding: among the _à priori_ proofs which polygenists can bring forward on their side, there is one which is of some importance; it is this, that while contrary ideas have been sustained and defended by men who never go beyond their own studies, the former have been generally brought forward by travellers and sailors, by those indeed who have been best able to put in practice this _direct observation_, which is generally conclusive and decisive. It is these whom we find the most ready to separate mankind into distinct groups, and to recognise in the inferior species a manifest tendency to approach nearer the nature of the anthropomorphous apes. A valuable source of information, from which anthropologists must not neglect to borrow, are the accounts of those who landed for the first time on certain islands and continents. If they have even conceived any erroneous ideas, it must usually be acknowledged that they are most likely to be able to give us a tolerably faithful portrait of the nations with whom they have met, even more important in certain points of view than the accounts afterwards given of them, since at that time these people have not been submitted to the various influences which necessarily result from contact with Europeans.