30,976 houses would have stood on every square mile; and about 1,239,040 houses, in the whole. This calculation makes no allowance for streets, or vacant spaces, or uninhabited buildings-or for the general inequality of the size of the houses in Rome, one compared with another, which we know to have been very considerable and consequently as to the actual number of dwelling-houses in Rome, it is doubtless prodigiously beyond the truth. That number, at no time, probably amounted to 50,000. The anonymous author above quoted reckons the sum total of Insulæ and Domus at Rome (which together made up the aggregate of its inhabited dwellings) at 46,602 +1,780; that is, 48,382 in all *. The calculation is proposed as one which, under the circumstances of the case, should be received as nearly tantamount to the number of the inhabitants both of Rome and of its suburbs: that is to say, we could scarcely reckon for Rome itself in a calculation of dwelling-houses so made, half an inhabitant to each house: on which principle, the sum total of the population of Rome and of her suburbs in the time of Pliny might be computed at about one million, but not at much more.

* Publius Victor, for the several Regiones, fourteen in num

ber, has 45,795 Insulæ, and 1830 Domus, or 47,625 in all.

VOL. IV.

F

APPENDIX.

DISSERTATION XVI.

On the Jewish and Julian dates of the several years of the Jewish War.

Vide Dissertation xv. vol. ii. page 65. line 16. BEFORE the reader proceeds to the details of the most melancholy septennium, or octennium, which is to be found in the history of the world; it will doubtless be an acceptable service to him, if he is furnished with the means of reducing the Jewish dates, which repeatedly occur in Josephus' account of these times, to their corresponding Julian ones. It is true, we had occasion to do this formerly in part; more especially for the years U. C. 819. and U. C. 823. It may not however be taken amiss, if I exhibit at once, in the following calendar, for all the years in question, from U.C. 819-U.C.826, the two cardinal dates in a Jewish year; the 15th of Nisan, and the 15th of Tisri respectively; by the assistance of which there is no difficulty in ascertaining any of the rest. This calendar, if its correctness can be depended on, must unquestionably be useful to the general student of Josephus; as well as for one of the most interesting of his works, the Bellum Judaïcum

* It is not necessary, by way of preliminary to the calendar in question, that we should enter upon the controversy whether the Jewish year was originally lunar or solar. The proper place for such a discussion, had it been considered requisite,

was Dissertation vii. of volume i. Whatever might be the primitive constitution of the Jewish year, there can be no doubt that for the period which coincides with the duration of the gospel history, and extends to the close of the Jewish war, it was purely

The 15th of Tisri is necessarily to be deduced from

lunar. The testimony of Philo Judæus, and of Josephus, alone is sufficient to place this fact beyond dispute: and to theirs we may add that of the Book of Enoch; (cap. lxxii. lxxiii :) which will be so much the more valuable, if, according to the opinion of the learned translator, the Book of Enoch was really written sometime in the reign of Herod the Great; before the birth of Christ.

I cannot indeed subscribe to this opinion; as I believe it rather to be the production of an Hebrew Christian; though not later than the reign of Hadrian. Upon this particular question, however, its testimony is clear and positive; and it is further supported by the authority of Galen, (Operum ix. 9. A. B.) The ȧkun of Galen coincided with the beginning of the reign of Marcus Aurelius; for he tells us himself he was thirty-seven years of age about the seventh or eighth of that emperor's reign. The Jewish year, the year at least which was observed in Palestine in his time, was lunar; consisting of three hundred and fifty-four days, or twelve months, every two of which contained fifty-nine days in all.

The passages which were cited, in Dissertation vii. vol. i. p. 318, from the Agathobuli and from Aristobulus, prove that the same form of the civil year was in use among the Jews at a much earlier period. The author of the Book of Ecclesiasticus, xliii. 6, 7, 8, bears similar testimony to the state of the case in his time: which was the beginning of the second century before Christ. And according to Anatolius, apud Eusebium, E. H. vii. 32. 287 A. and to Basil, Operum i. 80. C. in Hexaëmeron Homilia vi. the year of the Jews was lunar from the first.

In opposition to this weight of evidence, the authority of Syncellus, who certainly supposes the Jewish months to have been solar, consisting some of thirty, and others of thirty-one days, may justly be considered as good for nothing.

I shall

barely subjoin then the following statement of the names and order of the Syro-Macedonian months, which Josephus employs as appellations for the lunar ones in use among his countrymen: with references to the passages, where their Jewish names occur, but to those only.

Ant. Jud. i. iii. 3: ii. xiv. 6: iii. x. 5: xi. iv. 8.

viii. iii. I.

5 Lous (Hecatombæon) Ab

6 Gorpiæus

Elul

Tisri...

Marchesvan..
(Tebeth) Chasleu..

Tebeth

Sebat

Adar..

the 15th of Nisan: since from the nature of months, which consisted alternately of twenty-nine and of thirty days each, or vice versa, and were six in number, the 177th day inclusive, from the 15th of Nisan exclusive, must fall on the 15th of Tisri. The 15th of Nisan, for the years in question, I obtain from eclipses calculated in the Art de vérifier les dates; and eclipses in every instance so near to the paschal terms, March 18, and April 16, that the times of the mean full moons as thence deduced must represent, with very little error either of excess or of defect, the actual times of the true. I have added also the Dominical letter, and the day of the week; observing only, that though the former is actually the letter for the corresponding year of the solar cycle; the latter is two days in advance of the corresponding day of the week.

U. C. 819. A. D. 66. D. Let. E. 15 Nisan. March 30. Tuesday. 15 Tisri Sept. 23. Thursday. D. 15 Nisan. March 19. Saturday.

U. C. 820. A. D. 67.

U. C. 821. A. D. 68.

U. C. 822. A. D. 69.

Thursday.

U. C. 823. A. D. 70.

Monday.

This calendar expires with the recapture of Masada, the last act of the war in Judæa, U. C. 826, on the 15th of Xanthicus or Nisan; that is, Tuesday, April 11: as, if we compute the duration of the war from

one Jewish passover to another, it may be considered to have begun on the fifteenth of the same month, U. C. 819, Tuesday, March 30.

The correctness of the above calculations for the years U.C. 819, U. C. 820, and U. C. 823, I think was abundantly proved in the Dissertation already referred to a Its correctness for the year U. C. 821. may be further illustrated as follows.

If the 15th of Xanthicus or Nisan fell that year on April 6, but the year before on March 19, this is enough to prove that the year was intercalated. Hence we might naturally expect, about the period of the passover, an allusion to the fruits of the ground as ripe. Such an allusion occurs Bell. Jud. iv. vii. 2.

Again, if the 15th of Xanthicus fell on April 6, and the year was intercalated, the 4th of Dystrus would answer to the 4th of Veadar, and the 4th of Veadar to the 25th of February. About this time then the spring rains would naturally be at their height; and the Jordan might well be so much encreased by them as to be impassable. Accordingly, Bell. iv. vii. 3. 5. this appears to have been the case*.

But perhaps the clearest proof that we have rightly assigned the 15th of Nisan, in these several years, to its corresponding Julian date, is supplied by what admits of being established concerning the 15th of Nisan, U. C. 822, in particular. It will follow as a necessary consequence, that this 15th of Nisan is justly supposed to coincide with March 26, if the 15th of Tisri, corresponding to it, can be proved to have coincided with September 19.

* Bell. iv. viii. 1. mention occurs of the second of Dæsius. Nisan 15 coinciding with April 6, Dæsius I coincided with May 21.

a Vide vol. i.

And this, I think, ad

The next year, U. C. 822. (when Vespasian again took the field, (Bell. iv. ix. 9.) Dæsius 5 coincided with May 14.

p. 412-434.