CHAPTER XVIII.

ON THE CALENDAR.

327. AMONG the different divisions of time, the civil year is one of the most important. The solar year, or the interval elapsed between two successive returns of the sun to the same equinox, includes all the varieties of seasons.

The civil year must necessarily consist of an exact number of days. But the solar year consists of a certain number of days and of a part of a day, (art. 214.) Hence an artifice is necessary to keep the commencement of the different seasons, as nearly as possible, in the same place of the civil year: that is, if the sun enter the equator on the 20th of March in one year, that it may always enter it on the same day, or nearly on the same day, and that the solstices may be always as nearly as possible on the same day.

The common civil year consists of 365 days. The solar year of 365 days, 5 hours, 48 minutes, and 50 seconds, or 365 days, 6 hours nearly.

It is evident that if each civil year were to consist of only 365 days, the seasons would be later and later every year, and in process of time change through every part of the year.

328. In the infancy of astronomy, it was not to be expected that the exact length of the solar year could be obtained with much accuracy, and we find the Egyptians and other nations availing themselves of another method, by which they regulated the times of their agricultural labours. They observed when

Sirius or Arcturus, or some other bright star, after it had been obscured by the splendor of the solar rays, first became visible in the east, before sun-rise. This is called the heliacal rising of a star. From this time they reckoned a certain number of days to the commencement of the respective seasons of ploughing, of sowing, and of other labours in husbandry.

In this manner they dispensed with an exact knowledge of the length of the year. They were ignorant of the precession of the equinoxes, which in a few centuries would have occasioned their rules to fail, or rather to change.

329. The first useful and tolerably exact regulation of the civil year, by help of the solar, took place in the time of Julius Cæsar. It was then provided that every fourth civil year should consist of 366 days, and the addition of the day should be made, "die sexto calendas martias," whence the term bissextile applied to the year that consists of 366 days: we usually call it leap year, and the additional day is called the 29th of February.

The Calendar so ordered was called the Julian Calendar. 330. By the council of Nice, held in the year 325, it was fixed that the feast of Easter, by which the moveable fasts and festivals of the church are regulated, should be the first Sunday after the first full moon, which happened on or after the 21st of March. At that time the equinox happened on the 21st of March. Thus the festival of Easter was intended to be regulated by the spring equinox.

At that time it must have been known that the excess of the solar year above 365 days was not quite six hours, and that therefore, in using the Julian Calendar, the equinox would happen sooner every year. There however seems to have been no provision made on that account.

The true length of the solar year being less than 365 days, 6 hours, by 11 minutes nearly, the equinox every fourth year was nearly 44 minutes earlier, and in course of time the 21st

of March, instead of being the day of the equinox, might have

been the day of the summer solstice. festival of Easter might have been

summer.

Thus the fast of Lent and observed in the middle of

This inconvenience was foreseen before any material alteration had taken place. In the time of Pope Gregory, in 1577, the equinox happened on the 11th of March, or ten days before the 21st. It was then determined to remedy the error that had already taken place, and to provide against a future accumulation.

It must he generally allowed, that it was right to guard against an increase of the error, but it may be doubted whether a greater inconvenience did not take place to the people in general by correcting the error of the ten days, than if it had remained.

331. The 5th of October, 1582, was called the 15th, and thus the equinox was restored to the 21st of March.

A recurrence of error was prevented in the following manner. The true length of the solar year, as far as it was then known from the best tables, founded on the observations of Copernicus, Ptolemy, and Hipparchus, was 365 days, 5 hours, 49 minutes, and 16 seconds. By adding a day every fourth year, in 4 years the addition was 4 × (10m 44s) too much, or the accumulation of error in 400 years = 400 × (10m 44$) = 2 days, 23 hours, and 33 minutes nearly. Hence if, instead of making every fourth year leap year, every hundredth year for three centuries successively be made a common year, and the fourth hundred year be a leap year, the error in 400 years will be only about 27 minutes, and therefore the error in 20000 years would not be more than a day.

Hence the correction adopted by Pope Gregory, that the years 1700, 1800, 1900, 2100, 2200, 2300, 2500, &c., which, by the Julian Calendar, are leap years, should be common years,

and that the years 2000, 2400, &c. should remain leap years, is quite sufficient. The more correct length of the solar year, as now determined, proves the Gregorian correction less exact, but not materially so.

332. The Gregorian, or the new style, was not adopted in Protestant countries, till a considerable time had elapsed. When it was adopted in England in the year 1752, the error amounted to 11 days. This was remedied by calling the 2nd of September, 1752, the 13th.

The effect of thus putting, as it were, the seasons backward by 11 days, must at that time have been disagreeable. That our mode of reckoning time was made the same as that of other nations, was doubtless a convenience. But it might have been more conformable to our climate and the original notions of the festival of Easter, which regulates the other moveable fasts and festivals of the church, if the error that had already accumulated from the Julian Calendar had remained, and the Gregorian correction against future error had been only adopted.

The early climate of Italy might have principally induced Pope Gregory to bring back Easter to the regulations of the equinox: and it may have been a powerful motive in Russia for not adopting the Gregorian alteration in the style, that by retaining and suffering the errors of the Julian Calendar to accumulate further, the fast of Lent and festival of Easter will fall at times more convenient in respect to their seasons.

The year 1800 having been by the Julian Calendar a leap year, and by the Gregorian a common year, the Russian date is now 12 days behind that of the other countries of Europe.

333. The time of the festival of Easter depends on the first full moon on or after the 21st of March, and therefore, strictly, recourse should be had to astronomical calculation to ascertain the time of Easter for each year. But it is sufficient for this

purpose to use the Metonic Cycle, (art. 137,) the numbers of which are called Golden numbers.

Short rules and brief tables are given in the Act of Parliament for changing the style, and are usually prefixed to the Book of Common Prayer, by which the times of Easter may be found for any number of years to come. The computation so made, must sometimes differ from what a more exact calculation would give, and the time of Easter, if exactly computed, may vary considerably from the computations founded on the Metonic Cycle. However, as the latter mode of calculation is prescribed by the Act of Parliament, no inconvenience, from uncertainty as to the time in which the festival of Easter is to be observed, can arise.

By exact computation the 1st of April, 1798, should have been Easter Sunday, whereas by the Calendar prescribed it was not celebrated till the Sunday after. Also the 29th of March, 1818, should have been Easter Sunday, instead of the 22nd of March, as found by the prescribed mode of calculation.