If the Hindûs possessed in 1491 a knowledge of the heavenly motions sufficiently accurate to enable them to calculate backwards for 4592 years, it follows that they could only have obtained this knowledge from very ancient observations. To grant them such knowledge, while refusing them the observations from which it is derived, is to suppose an impossibility; it would be equivalent to assuming that at the outset of their career they had already reaped the harvest of time and experience. While on the other hand, if their epoch of 3102 is assumed to be real, it would follow that the Hindûs had simply kept pace with successive centuries down to the year 1491 of our era. Thus, time itself was their teacher; they knew the motions of the heavenly bodies during these periods, because they had seen them; and the duration of the Hindû people on earth is the cause of the fidelity of its records and the accuracy of its calculations.

It would seem that the problem as to which of the two epochs of 3102 and 1491 is the real one ought to be solved by one consideration, viz., that the ancients in general, and particularly the Hindûs, as we may see by the arrangement of their Tables, calculated, and therefore observed, eclipses only. Now, there was no eclipse of the sun at the moment of the epoch 1491; and no eclipse of the moon either fourteen days before or after that moment. Therefore the epoch 1491 is not based on an observation. As regards the epoch 3102, the Brâhmans of Tirvaloor place it at sunrise on February 18th. The sun was then in the first point of the Zodiac according to its true longitude. The other Tables show that at the preceding midnight the moon was in the same place, but according to its average longitude. The Brâhmans tell us also that this first point, the origin of their Zodiac, was, in the year 3102, 54 degrees behind the equinox. It follows that the origin—the first point of their Zodiac—was therefore in the sixth degree of Aquarius.

There occurred, therefore, about this time and place an average conjunction; and indeed this conjunction is given in our best Tables: La Caille's for the sun and Maier's for the moon. There was no eclipse of the sun, the moon being too distant from her node; but fourteen days later, the moon having approached the node, must have been eclipsed. Maier's tables, used without correction for acceleration, give this eclipse; but they place it during the day when it could not have been observed in India. Cassini's tables give it as occurring at night, which shows that Maier's motions are too rapid for distant centuries, when the acceleration is not allowed for; and which also proves that in spite of the improvement of our knowledge we can still be uncertain as to the actual aspect of the heavens in past times.

Therefore we believe that, as between the two Hindû epochs, the real one is the year 3102, because it was accompanied by an eclipse which could be observed, and which must have served to determine it. This is a first proof of the truth of the longitude assigned by the Hindûs to the sun and the moon at this instant; and this proof would perhaps be sufficient, were it not that this ancient determination becomes of the greatest importance for the verification of the motions of these bodies, and must therefore be borne out by every possible proof of its authenticity.

We notice, 1st, that the Hindûs seem to have combined two epochs together into the year 3102. The Tirvaloor Brâhmans reckon primarily from the first moment of the Kali Yuga; but they have a second epoch placed 2d. 3h. 32m. 30s. later. The latter is the true astronomical epoch, while the former seems to be a civil era. But if this epoch of the Kali Yuga had no reality, and was the mere result of a calculation, why should it be thus divided? Their calculated astronomical epoch would have become that of the Kali Yuga, which would have been placed at the conjunction of the sun and the moon, as is the case with the epochs of the three other Tables. They must have had some reason for distinguishing between the two; and this reason can only be due to the circumstances and the time of the epoch; which therefore could not be the result of calculation. This is not all; starting from the solar epoch determined by the rising of the sun on February 18th, 3102, and tracing back events 2d. 3h. 32m. 30s., we come to 2h. 27m. 30s. a.m. of February 16th, which is the instant of the beginning of Kali Yuga. It is curious that this age has not been made to commence at one of the four great divisions of the day. It might be suspected that the epoch should be midnight, and that the 2h. 27m. 30s. are a meridian correction. But whatever may have been the reason for fixing on this moment, it is plain that were this epoch the result of calculation, it would have been just as easy to carry it back to midnight, so as to make the epoch correspond to one of the chief divisions of the day, instead of placing it at a moment fixed by the fraction of a day.

2nd. The Hindûs assert that at the first moment of Kali Yuga there was a conjunction of all the planets; and their Tables show this conjunction while ours indicate that it might actually have occurred. Jupiter and Mercury were in exactly the same degree of the ecliptic; Mars being 8° and Saturn 17° distant from it. It follows that about this time, or some fifteen days after the commencement of Kali Yuga, and as the sun advanced in the Zodiac, the Hindûs saw four planets emerge successively from the Sun's rays; first Saturn, then Mars, then [pg 726]Jupiter and Mercury, and these planets appeared united in a somewhat small space. Although Venus was not among them, the taste for the marvellous caused it to be called a general conjunction of all the planets. The testimony of the Brâhmans here coïncides with that of our Tables; and this evidence, the result of a tradition, must be founded on actual observation.

3rd. We may remark that this phenomenon was visible about a fortnight after the epoch, and exactly at the time when the eclipse of the moon must have been observed, which served to fix the epoch. The two observations mutually confirm each other; and whoever made the one must have made the other also.

4th. We may believe also that the Hindûs made at the same time a determination of the place of the moon's node; this seems indicated by their calculation. They give the longitude of this point of the lunar orbit for the time of their epoch, and to this they add as a constant 40m., which is the node's motion in 12d. 14h. It is as if they stated that this determination was made thirteen days after their epoch, and that to make it correspond to that epoch, we must add the 40m. through which the node has retrograded in the interval. This observation is, therefore, of the same date as that of the lunar eclipse; thus giving three observations, which are mutually confirmatory.

5th. It appears from the description of the Hindû Zodiac given by M. C. Gentil, that on it the places of the stars named the Eye of Taurus and the Wheat-ear of Virgo, can be determined for the commencement of the Kali Yuga. Now, comparing these places with the actual positions, reduced by our precession of the equinoxes to the moment in question, we see that the point of origin of the Hindû Zodiac must lie between the fifth and sixth degree of Aquarius. The Brâhmans, therefore, were right in placing it in the sixth degree of that sign, the more so since this small difference may be due to the proper motion of the stars, which is unknown. Thus it was yet another observation which guided the Hindûs in this fairly accurate determination of the first point of their movable Zodiac.

It does not seem possible to doubt the existence in antiquity of observations of this date. The Persians say that four beautiful stars were placed as guardians at the four corners of the world. Now it so happens that at the commencement of Kali Yuga, 3,000 or 3,100 years before our era, the Eye of the Bull and the Heart of the Scorpion were exactly at the equinoctial points, while the Heart of the Lion and the Southern Fish were pretty near the solstitial points. An observation of the rising of the Pleiades in the evening, seven days before the autumnal equinox, also belongs to the year 3000 before our era. This and similar observations are collected in Ptolemy's calendars, though he does not give their authors; and these, which are older than those of the Chaldeans, may well be the work of the Hindûs. They are well acquainted with the constellation of the Pleiades, and while we call it vulgarly the “Poussinière,” they name it Pillaloo-codi—the “Hen and chickens.” This name has, therefore, passed from people to people, and comes to us from the most ancient nations of Asia. We see that the Hindûs must have observed the rising of the Pleiades, and have made use of it to regulate their years and their months; for this constellation is also called Krittikâ. Now they have a month of the same name, and this coïncidence can only be due to the [pg 727]fact that this month was announced by the rising or setting of the constellation in question.