becoming the favourite game of the educated classes. This is proved by the fact that in a book of sermons published in the latter half of the thirteenth century one Jacobus de Cessoles, a Dominican, attempts to set forth a system of rules for right living founded on the rules of the _ludus scaccorum_.[395] The game has naturally undergone many changes in the course of time; for instance, the Arabians originally had elephants in the place of our bishops; but it has always preserved the character of a battle, and is so represented in old Arabian manuscripts.[396] [Illustration:] Our third group of board plays is comprised of those which add the attraction of chance to intellectual enjoyment of the contest. It is true that to a certain degree chance is an element in the purest games of reason, since the most skilful player can not foresee all the consequences of a move, and various uncontrollable influences[397] may interfere with the best-laid plans; but in the games which we are now considering there is a blending of risk with calculation, which has a peculiar charm. Perhaps the most familiar game of this kind is backgammon, which was certainly known to the Greeks and Romans, and possibly to the Egyptians and Phœnicians. In this game and kindred ones the object is to throw away men whose value is determined chiefly by chance, while the advance to advantageous points is a matter of calculation, thus affording a combination of direct and indirect fighting. Backgammon is of peculiar ethnologic interest because of the prominent part it plays in the controversy as to whether Asiatic influence is traceable in primitive American civilization. E. B. Tylor has stated in several passages[398] that a kind of backgammon played on a cruciform board is a favourite amusement of the East Indians, and is called by them Patschisi (in Burmah: Patschit), and a very similar form of the game was known to the pre-Columbian Mexicans under the name of Patolli. Tylor considers the complicated nature of the game as a sufficient disproof of its independent origin, and from this, and a certain kinship to chess which is apparent in it, he concludes that the whole group of games furnishes an important argument in favour of Asiatic influence on American life before the time of Columbus.