who knows nothing of a Keltic speech at all. The Welshman in turn is physically allied to the Irish and distinct from many of the Gaelic-speaking Scotch, although these last two speak even the same subtype of the Keltic language. Such racial affinity as obtains between certain of these people is in utter defiance of the bonds of speech. The Breton should be more at home among his own folk in the high Alps in respect of race, even although he could hold no converse with the Swiss people in their own tongue. If these examples be not enough, turn to other parts of
Europe. The Walloons and the Flemish, component parts of the Belgian nation, are indeed quite distinct in race and in language alike. It is only an accident. For if we turn to Switzerland, seeking for physical differences along the boundary of the French-and German-speaking cantons, they are not to be found. In northern Italy to-day there are considerable communities still bearing the German speech and customs, evidence of the Teutonic invasions of historic times. These people have become so completely absorbed that they are not distinguishable physically from their Italian neighbors. There are indeed spots in Italy where German racial traits survive, but they are quite remote from these islets of Teutonic language.
Nor in eastern Europe is the picture less confusing. The Bulgarian language, spoken by people so outlandish to Europeans that they gave the word "bogie" to our nurses wherewith to frighten children, went first. Now it is the Roumanian speech which, in its turn, is disappearing before the Slavic tongue. Magyar, the language of the Hungarians, spreading toward the east, displaced by German, which is forcing its way in from the northwest, is also on the move. Beneath all this hurry-skurry of speech the racial lines remain as fixed as ever. Language, in short, as a great philologist has put it, "is not a test of race. It is a test of social contact." Waves of language have swept over Europe, leaving its racial foundations as undisturbed as are the
- Annales de Demographie, iv, p. 224; or more fully, with maps by the author, in Publications of the American Statistical Association, v, p. 28.
- The French and German portions of Switzerland are shown in Forschungen zur Deutschen Landes- und Volkskunde, viii, No. 5, 1894; and Rundschau für Geographic und Statistik, xiii, p. 337, appendix map.
- Map in Petermann's Geographische Mittheilungen, 1877, plate 17; vide also Globus, lxvi, p. 165; and Journal of the Anthropological Institute, ii, p. 108.
- Revue de Geographie, xxxvii, p. 321. Perhaps the best compilation of references to ethnological maps extant, bringing them down to 1885, is given by Dr. Andrée in Mittheilungen des Vereins für Erdkunde zu Leipzig, 1885, pp. 173 seq.; less fully to 1879 in Archiv für Anthropologie, xi, p. 454. The editor confesses that nearly all of them are indeed not ethnological, but merely "speech" maps.