may thus suggest the rustling of the leaves and the piping of early birds, while the more guttural accents of a Flemish name remind us of the war-cry of wild hordes and the beating of "frameas." The Frankish invasions of the fifth century may be considered the most important event of Belgium's early history. Whether the unity of the Belgian nation is questioned or upheld, we must inevitably go back to the cause of its real or apparent division. If such division, from being racial and linguistic, had become political or economic--that is to say, if the language boundary had coincided with some of the boundaries which divided the country at a later stage--the idea that Belgium was born in 1830 and constituted an "artificial creation of European diplomacy" might not be groundless. Here, as in many other countries of Europe, nationality would have been determined mostly by race and language. This, however, is not the case. At no period of Belgian history did any division follow the linguistic frontier. On the contrary, most of the political and ecclesiastical units created during the Middle Ages included both elements of the population, and, through frequent intercourse and common interests, these two people, speaking different languages, became gradually welded into one. When in the fifteenth century the various duchies and counties came under the sway of the dukes of Burgundy, national unity was realized, as it was realized in England or in France at the same time, through the increasing power and centralizing action of modern princes. A few prejudiced writers have vainly endeavoured to exaggerate the racial or linguistic factor, and contended that, in the eyes of science, Belgian nationality could not exist. The duty of a scientist is not to distort the manifestations of natural phenomena in the light of some more or less popular idea. His duty is to explain facts. The development and permanence of Belgian nationality, in spite of the most adverse conditions, is one of these facts. The existence of the Swiss nation, far more deeply divided than the Belgian, shows that it is not unique. But even if it were unique, it ought to be accounted for. It is far easier to indulge in broad generalizations than to devote oneself to a close study of nature or man. It is not the rules, it is the exceptions which ought to retain our attention, for only exceptions will teach us how imperfect are our rules. CHAPTER II FROM SAINT AMAND TO CHARLEMAGNE