This Table is formed from one given by Petavius, Vol. i. p. 161, corrected and improved. The Table of Jackson, Vol. ii. p. 85, is imperfect and unsatisfactory; it omits the reduction of the several months of the year of confusion to their corresponding beginnings in the Julian, and thereby leaves the reader at a loss to account for the interval between the intercalation of the first month after February, and of the two others after November *.

Thus did this admirable correction, the skill and ingenuity of which will be more admired the more closely it is considered, by its judicious intercalations, bring the beginnings of the new months nearly to a conformity with those of the old, and thereby preserve the Calends nearly in their old places; while, by inserting the additional days at the ends of the months, the Nones and Ides were unaltered: Cæsar thereby respecting the prejudices of the vulgar for these holy days.

* This perhaps will explain a passage in Cicero's epistle to Ligarius, Lib. vi. Epist. 14, which otherwise seems unintelligible: "Ego tamen ad quint. calend. intercalares priores, mane venissem ad Cæsarem."

JULIAN YEAR.

The reformed year, which succeeded the year of confusion, was called Julian, from Julius Caesar, with whose fourth consulate it began, the first of January, U. C. 709, or B. C. 45.

In modelling the months of it, he first restored all the even months of 30 days, instituted by Romulus, which Numa had altered for the worse, to 29; he next retained the four months of 31 days, used by Romulus and Numa, and added thereto three more, namely, January, Sextilis, and December. February he retained from Numa, according to the following Table, including the twelve Dii Majores, to whom the months were dedicated. And from which it is remarkable, that Janus, the ancient god of time, was excluded, to make way for Juno Matrona.

To compensate for the additional quarter of a day, every year above the 365 days, Julius Cæsar inserted a whole day every fourth year; which, because it was inserted after the old terminalia, or 23d day of February, and was therefore reckoned

By this inaccuracy of substituting "the 21st day of March," instead of " the day of the vernal equinox," as the decree ought to have been worded, in the time of Pope Gregory XIII. A.D. 1582, or 1257 years after the Council of Nice, the equinox was found to be on the 11th of March, having fallen back ten days. In order, therefore, to bring it forward to its former place of the 21st, he left out ten days in October, calling the 5th, the 15th day of that month. Whence, in that year of confusion, the 22d day of December became the first of January A.D. 1583, which was the first year of the Gregorian Era. In making this correction, he was principally assisted by the celebrated mathematician Clavius.

1. But to prevent the repetition of this error in future, a further reformation of the Julian Calendar was wanting. Because the vernal equinox fell backwards three days in the course of 390 years, Gregory, chiefly by the assistance of Aloysius Lillius, decreed that three days should be omitted in every four centuries: namely, that every first, second, and third centurial year, which would otherwise be bissextile, should be a common year; but that every fourth centurial year should remain bissextile. Thus, the years A.D. 1700, 1800, 1900, and 2100, 2200, 2300, were to be common years; but A.D. 1600, 2000, 2400, to remain leap years.

By this ingenious reform, the Julian Calendar is rendered sufficiently accurate for all the purposes of Chronology, and even of Astronomy, for 6000 years to come; because the tropical year itself is subject to a periodical inequality.

2. At the end of 6000 years, a further correction will be necessary. For since the deficiency of the Tropical from the Julian year, 11 min. 3 sec. or 663 seconds, will amount to 265,200 seconds, in the course of 400 years; while three entire days, amounting only to 259,200 seconds, are deficient therefrom 6000 seconds, or one hour and 40 minutes, every 400 years; this deficiency from the correct deduction, will in fifteen such periods, or 15 × 400 = 6000 years, amount to 25 hours; or to one day and one hour. Therefore the omission of one day in every 6000 years (by making the 6000th year common, which would otherwise be bissextile) will leave only one hour to be corrected.

3. And this may be done by omitting one day more at the end of 144,000 years, which is the product of 24 × 6000; in which time the hour will amount to 24, or to one day. And this will

bring every thing right to a second of time; if the world shall last so long *.

The Gregorian, or reformed Julian year, was not adopted in England until A.D. 1751, when, the deficiency from the time of the Council of Nice then amounting to eleven days, this number was struck out of the month of September, by Act of Parliament; and the 3d day was counted the 14th, in that year of confusion. The next year, A.D. 1752, was the first of the New Style +, beginning January 1, instead of March 25.

The following Tables, selected from the most approved, will be found useful in such Chronological calculations as depend upon Astronomy.

• From the analogy of the 6000 years in the second correction, to the six millenary ages of the world; and of the 144,000 years, in the third, to the sealed number, in the Apocalypse, Rev. vii. 4; xiv. 1–3; the late Edward King, Esq. formed a very ingenious, but rather fanciful scheme, of a grand Sabbatic Era of 144,000 years, in his Morsels of Criticism, Vol. III. p. 188.

By the Act of 24 Geo. II. A.D. 1752, relative to the Alteration of the Style, it was enacted,

"That Easter-day, on which the rest of the moveable feasts depend, is always the first Sunday after the full moon which happens upon, or next after, the 21st of March: and if the full moon happens upon a Sunday, Easter-day is the Sunday after."

Here our Legislators seemed to have considered the 21st of March as the earliest Paschal full moon; and in consequence of this, the earliest Easter-day, in the Tables annexed to the Calendar in the Book of Common Prayer, is fixed to the 22d of March; and the latest, to April 25, including a range of 35 days; on account of the variations of the time of full moon during a synodical month of 291 days, added to 7, the variations of Sunday, in that time = 36; from which subducting 1, for the difference between the full moon and Easter-day, the remainder is 35.

But at present, the vernal equinox, which they supposed could not be earlier than the 21st of March, now falls on the 20th; and may fall back to the 19th, before the expiration of 130 years. And when this shall be the case, if the mean Paschal new moon shall happen on the 4th of March upon a Friday; then to March 4, add 15 days; and the Paschal full moon will fall on the 19th of March, on a Saturday; and will coincide with the day of the vernal equinox: and consequently the 20th of March will be Easter Sunday, according to the spirit, but not the letter of the Act.

Hence, the Paschal astronomical rule might have been worded more correctly, thus: “Easter-day is always the first Sunday after the full moon which happens either upon, or next after, the vernal equinox; which, in the New Style, will ever fall either on the 19th, or on the 20th day of March; reckoning the hours from midnight."

For this judicious correction, we are indebted to Kennedy's Astronomy, p. 362. By the foregoing Act of Parliament, the beginning of the year was changed from March 25, to the first of January. Russia is the only civilized state of Europe that retains the Old Style.

*The rernal Equinox in A. D. 325, fell before March 21; or more correctly, March 20; 8 h. 21 min. according to Kennedy's Astronomy, p. 360. See a very ingenious method of finding the times of the Equinoxes and Solstices, arithmetically, in Beverege,

Lib. ii. cap. 2. De Equinoctiis et Solstitiis, p. 145-154. Edit. 3.

+ Add one day after this, when it is Leap year.