pies practically 584 days, and consequently five such revolutions are equal to eight years of 365 days. Since 20 when divided into 584 leaves a remainder of 4, it is obvious that the commencement of a Venus-period will fall always on one of five of the twenty day-signs. Further, since 13 when divided into 584 leaves a remainder of 12, it is equally obvious that each successive Venus-period will open with a day of which the numerical sign is one less than that of the preceding period. In several of the MSS. we have the Venus-periods set out in the order in which they occur, viz. on the days cipactli, coatl, atl, acatl and olin; 1. cipactli being followed by 13. coatl, the latter by 12. atl, and so forth. It can be seen that 65 Venus-periods must elapse before the same sign occurs in conjunction with the same number at the commencement of a period, and this amounts to 104 years of 365 days, viz. the longer cycle of the Mexicans. It seems most probable that the Venusperiod was utilized as a means of correcting the 365day year, and I am even inclined to believe that observation of this planet was practised before the institution of the solar calendar. But I shall recur to this point when the subject of the Maya calendar is discussed; meanwhile it is worthy of note that the signs of the two last "suns" are atl and olin, signs which occur as the commencing-days of Venus-periods; that the historical sun was supposed to have been born on 13. acatl; and that the new fire ceremony always took place in the year 2. acatl. Though the morning star played such an important part in the regulation of the calendar, it was not regarded altogether as a beneficent deity, possibly because of its association with war and sacrifice among the hunting peoples. However that may be, its light when it first rose was considered to exercise a baneful effect upon mankind, and chimneys were carefully stopped up to prevent the rays from entering the houses. The cult of Venus was especially
