Some words should perhaps be said on the geographical works of the Arabs. One of the most important of these, by Yacut, is in the form of a huge Gazetteer, arranged in alphabetical order; but the greatest geographical work of the Arabs is by Edrisi, geographer to King Roger of Sicily, 1154, who describes the world somewhat after the manner of Ptolemy, but with modifications of some interest. He divides the world into seven horizontal strips, known as "climates," and ranging from the equator to the British Isles. These strips are subdivided into eleven sections, so that the world, in Edrisi's conception, is like a chess-board, divided into seventy-seven squares, and his work consists of an elaborate description of each of these squares taken one by one, each climate being worked through regularly, so that you might get parts of France in the eighth and ninth squares, and other parts in the sixteenth and seventeenth. Such a method was not adapted to give a clear conception of separate countries, but this was scarcely Edrisi's object. When the Arabs—or, indeed, any of the ancient or mediæval writers—wanted to describe a land, they wrote about the tribe or nation inhabiting it, and not about the position of the towns in it; in other words, they drew no marked distinction between ethnology and geography.
But the geography of the Arabs had little or no influence upon that of Europe, which, so far as maps went, continued to be based on fancy instead of fact almost up to the time of Columbus.
Meanwhile another movement had been going on during the eighth and ninth centuries, which helped to make Europe what it is, and extended
