We may rest assured of one thing, at all events, which is that the architectural remains in France are as sufficient for the construction of an ethnographic map of that country as the rocks are for the compilation of a geological survey. If the one opens out to the student an immense expanse of scientific knowledge, the other is hardly of less interest, though in a less extended field. There are few studies more pleasing than that of tracing the history of man through his works, and none bring the former condition of humanity so vividly back to us as those records which have been built into the walls of their temples or their palaces by those who were thus unconsciously recording their feelings for the instruction of their posterity. [Illustration: 543. Diagram of the Architectural Divisions of France.[16]] The first thing that strikes the student in examining architecturally the map of France is the recurrence of the same phenomenon as was remarked in that of Italy, a division into two nearly equal halves by a boundary line running east and west. In both countries, to the southward of this line the land was occupied by a Romanesque people who, though conquered, were never colonised by the Barbarians to such an extent as to alter their blood or consequently the ethnographic relations of the people. North of the line the Goths and Lombards in Italy, and the Franks in Gaul, settled in such numbers as to influence very considerably the status of the races, in some instances almost to the obliteration of their leading characteristics. In France the boundary line follows the valley of the Loire near its northern edge till it passes behind Tours; it crosses that river between that city and Orleans, follows a somewhat devious course to Lyons, and up the valley of the Rhone to Geneva. In the Middle Ages the two races were roughly designated as those speaking the Langue d’oc and the Langue d’œil—somewhat more correctly those to the south were called Romance,[18] those to the north Frankish; but the truth is, the distinction is too broad to be now clearly defined, and we must descend much more into detail before any satisfactory conclusion can be arrived at.