confirm what has been said on this point. The conquest of France, Spain, England, and other portions of Europe, by the Romans took place about two thousand years ago. The evidences and results of this conquest, in the languages, institutions, features, and character of the conquered nations, are everywhere apparent at the present day. No one dreams of doubting them, and this simply because we have written and monumental evidence of the facts. Why should we doubt that the results of a conquest made two thousand years earlier may survive in equal vigor, though in the nature of things no written or monumental record of it can have come down to our time? The memorials of it which remain are of a different, but, to an ethnologist, not less convincing character. One of these may be noted in the other fact to which reference has been made. A very high authority in comparative philology. Dr. Friedrich Müller, in his great work on Linguistic Science (Grundriss der Sprachwissenschaft), after remarking that the numerical system of the Indo-Germanic languages rests on the decimal system, adds that the Celtic alone shows traces of the vigesimal system, which are to be referred to the influence of the Basque language. The Iberian Basques reckon by digits to twenty, which is, in their language, a distinct word, hogei (or oguei); forty is berrogei, "two hogeis"; eighty is "four hogeis"; and ninety-seven would be "four hogeis and ten-seven." The Celtic has a double system. Twenty is fiche, a corruption of the Aryan term; for forty the Celt can say either cethor cha, an Aryan contraction of "four-tens," or dá fichit, "two twenties." Ninety-seven is either "nine tens and seven," or, as in the Basque, "four twenties and ten-seven." Now, the French language, as is well known, adopts both methods, in different parts of its ascending scale. As far as sixty it proceeds by the decimal system; then it abruptly changes to the vigesimal. The Frenchman, when for ninety-seven he says "four-twenties-ten-seven" (quatre-vingt-dix-sept), has no idea—unless he is a philologist—that he is translating an ancient Iberian idiom into a corrupt form of Aryan speech. If we consider what this fact really signifies, we shall see that the whole ethnological history of France is embodied in it. This French system of enumeration, now in actual use, tells us that the people who employ it are mainly of Iberian origin; that an Aryan language in its most corrupt and disintegrated form, the Celtic, was once imposed upon them; that this has again given place to the Latin form, which has been further mangled and debased by the influence of a still later Teutonic conquest; and that through the whole of these overlying strata, caste imposed upon caste, the vigorous Iberian element has forced its way to the light, and governs to this day, in this composite population, that most striking manifestation
