*THE POPULAR SCIENCE MONTHLY.*

was appointed to give notice to the Pope, and other dignitaries in various parts of the Christian world, of the time when Easter should be celebrated each year, until a perfectly correct cycle should be established. The most prominent cycle framed for this purpose was one by a mathematician named Victorinus. It consisted of the product of the lunar and solar cycle—i. e., 1928 532. If this calculation had been without defect, any given day would have been the same day of the year, month, moon, and week, that it was five hundred and thirty-two years before or would be five hundred and thirty-two years after.

The Council of Orleans, a. d. 541, decreed that the feast of Easter should be celebrated every year, according to the table of Victorinus. But the tables derived from these data answer only for a limited time on account of the above-mentioned errors in the year and in the lunar cycle. Accordingly, the books which contain tables for finding Easter are good only until the year 1900, when new ones must be made for another period.

As the cycles were fixed by the Latin Church, the era of Christ began in the tenth year of the solar cycle and in the second year of the lunar cycle. Therefore, to find what year of the solar cycle any given year of our Lord is, we add 9 to the number of the year and divide by 28; the remainder, if any, will indicate the number of the year of the cycle. The year of the lunar cycle, i. e., the golden number, is found in a similar manner, by adding 1 to the given year and dividing by 19; the remainder will indicate the year of the lunar cycle.

After the Julian calendar had been used several centuries, the improved state of astronomy disclosed the fact that computed time did not keep pace with actual time, because the year did not consist of three hundred and sixty-five days and six hours but was about eleven minutes ten seconds less. Hence, by inserting an extra day for leap year, we gain upon true time forty-four minutes, forty seconds, which makes an error of a day in about one hundred and thirty-one years; and hence, in 1582, when the correction of the calendar was undertaken by Pope Gregory, the error had amounted to ten days; i. e., instead of counting just 1,582 years it ought to have been 1,582 years and ten days. A correction was accordingly made by taking a leap of ten days and calling October 5th of that year October 15th. This change, as elsewhere stated, brought the vernal equinox to the 21st of March, where it was at the time of the Nicene Council. To prevent the recurrence of the same error in future, it was ordered that every fourth year should be a leap-year as before, but centurial years, though multiples of 4, should not be leap-years unless they were multiples of 400. The loss of 1116 minutes yearly in time, as computed by the Julian method, amounts in 100 years to 18·6 hours. Calling this one hundredth year a common year gives a gain of one day, or 24 hours, which puts computed time ahead of actual time, 24-18·6 5·4 hours, which, in 400 years, equals 21·6 hours, gain. Calling the four hundredth a leap-