What is true of language is obviously even more true of culture. In other words, if a certain type of man migrated over a considerable area before its language assumed the form which can now be traced in related linguistic groups, and before its culture assumed the definite type the further development of which can now be recognized, there would be no possibility of ever discovering a correlation of type, language, and culture, even if it had ever existed; but it is quite possible that such correlation has really never occurred. It is quite conceivable that a certain racial type may have scattered over a considerable area during a formative period of speech, and that the languages which developed among the various groups of this racial type came to be so different that it is now impossible to prove them to be genetically related. In the same way, new developments of culture may have taken place which are so entirely disconnected with older types that the older genetic relationships, even if they existed, can no longer be discovered. If we adopt this point of view, and thus eliminate the hypothetical assumption of correlation between primitive type, primitive language, and primitive culture, we recognize that any attempt at classification which includes more than one of these traits cannot be consistent. It may be added that the general term “culture,” which has been used here, may be subdivided from a considerable number of points of view; and different results again might be expected when we consider the inventions, the types of social organization, or beliefs, as leading points of view in our classification. After we have thus shown that language, culture, and type cannot be considered as constantly associated, and after we have recognized that the same type of man has developed distinct languages, the question still remains open, whether the languages developed by any one stock bear marks of superiority or inferiority. It has been claimed, for instance, that the highly developed inflected languages of Europe are much superior to the cumbersome agglutinative or polysynthetic languages of northern Asia and of America (Gabelentz). We have also been told that lack of phonetic discrimination, lack of power of abstraction, are characteristics of primitive languages. It is important to show whether these traits are really associated with any languages of primitive man. In a way this consideration leads us back to the study of alleged mental characteristics of distinct human types.