but the inference that the authenticity of the Hebrew computation is thus disproved, does not necessarily follow. From the time of the Septuagint translation it would be read by the Jews of the dispersion, as well as by the other authors who have been referred to; and unless they could refer to the Esdrine copy of the Hebrew, or to some other Hebrew text of authority, the difference of the two calculations would remain unobserved.

But we know that the Jews themselves, on their return from Babylon, had so greatly lost their knowledge of the pure Hebrew, that when Ezra opened the book of the law in the sight of all the people, it was necessary to interpret it. "The Levites read in the book in the law of God "distinctly, and gave the sense, and caused them "to understand the reading." (Neh. viii. 5. 8) Nor can it be supposed that the circumstances of the nation afterwards were such as to revive the knowledge of the more ancient language, or to check the gradually general use of a still more modern dialect; so that the transcription of the law must necessarily have been rare; and, if

b2

transcribed, it must have been still more rarely accessible for any purpose of collation.

Nor is it probable that any doubt arose as to the correctness of the Septuagint; and thus it would necessarily become the sole authority referred to, as being the only one generally understood, or valued. But still, among the learned Jews, previous to the Christian era, traces must be found of the existence of the Hebrew numbers, if they were really according to the original computation: and that such traces are to be observed, I hope hereafter to prove; not only on the authority of that computation made use of in the book of Enoch, which the reader has already seen, but from other sources also.

It will be found that the ancient Jews did not expect the Messiah in the sixth, but in the fourth chiliad; and therefore the supposition that the reckoning of time was altered by them, so as to throw the advent of the Messiah, acknowledged by the Christians, upon the precise time at which his coming had been foretold and expected, will become totally inadmissible; and some other

reason, as well as some other time, for this alteration, will remain yet to be discovered.

Thus, we may discern a possible and not improbable cause for the alteration of the Septuagint, whether we consider it to have taken place during the translation, or shortly before that period.

If, indeed, we give credit to the account of Philo, the translators not only performed their task with exactness; but, although separated, miraculously made use of the very same words. But to such a witness as this I pay no regard; since, of all testimonies, that which proves too much, is the most liable to suspicion.

Nor do I imagine that the pseudo Aristeas can be otherwise regarded than as a work of imagination, in which truth and fiction are so blended, that it is impossible to separate them.

It is, however, to be remarked, that although Philo, within a few pages of the passage above cited, (p. 501 and 496,) speaks of the Hebrews by name, calling Moses also, at p. 823, Eßpatov, and εκ γενους Εβραιων ; and although, in a great number of places, he speaks of the Chaldeans as distinct from the Hebrews, yet repeatedly observes, that the Septuagint was translated, not

from the Hebrew, but from the Chaldean tongue, using the words Χαλδαιος and Χαλδαικος several times*: and by this, as he declares that the copy was sent from Jerusalem, he must have meant the mixed language, which was in use in Judea at that time.

There is thus some evidence in favour of the conjecture of Bp. Horsley, already mentioned; since it would have been much easier to collect a number of interpreters, to whom both that language and the Greek were known, than such as might be qualified to translate the pure Hebrew; and as copies in the vernacular idiom must have been far most easily attainable, the probable circumstances of the case very much coincide with the expression used by Philo.

The testimony of Josephus, depending, as it does, upon the existence of single words, which in the various copies occur with some variation, is of so mixed a character that little reliance can be placed on it, either to support the Septuagint calculation, or to invalidate it; for if there seems cause to conclude that he wrote the sum of years

* De Vit. Mosis, lib. xi. p. 509.

anterior to the flood as 1658, in agreement with the Hebrew; so, on the other hand, the separate numbers ascribed by him to the successive generations, closely approximate to those of the Septuagint.

The variation in the numbers, which are given by the version of Ruffinus, adds to the difficulty of founding any arguments upon the statements of former copies, since the greater proportion of those given by Ruffinus, agree with the Hebrew calculation; while two of them, those of Malaleel and Enoch, neither agree with the Hebrew nor with the Septuagint itself. For the former, the Hebrew has 65 years; the Septuagint, 165; while Josephus gives 162: and for the latter, while there is the same difference of a century between the Hebrew and Septuagint, which have 165 and 65 years, Josephus has 105. Doubtless these numbers originally agreed either with the Hebrew or Greek, but their casual alteration impairs the general credit of the calculation to which they belong, as much as if we were to suppose that they had been wilfully corrupted.

The remark of Fabricius, "That the ancients "seem universally to have paid but little atten