ed. Heiberg: ἀπὸ μὲν τῆς Ναβονασσάρου βασιλείας μέχρι τῆς ̓Αλεξάνδρου τελευτῆς ἔτη συνάγεται κατ ̓ Αἰγυπτίους υκδ, ἀπὸ δὲ τῆς ̓Αλεξάνδρου τελευτῆς μέχρι τῆς Αὐγούστου βασιλείας ἔτη Σφὸ ἀπὸ δὲ τοῦ ᾶ ἔτους Αὐγούστου κατ ̓ Αἰγυπτίους τῆς ἐν τῷ Θωθ ά μεσημβρίας, ἐπειδὴ τὰς ἐποχὰς ἀπὸ μεσημβρίας συνιστάμεθα, μέχρι τοῦ ιζ ἔτους ̔Αδριανοῦ ̔Αθὰρ ζ μετὰ δύο ἰσημερινὰς ὥρας τῆς μεσημβρίας ἔτη γίνεται ρξα καὶ ἡμέραι ξε καὶ ὧραι σημερινα β. Hence it follows that Ptolemy counted the first year of Augustus from Thoth 1 = August 31 in 30 B.C. Again we learn from the πρόχειροι κανόνες (Ptolemaei et Theonis opera, ed. Halma, vol. vi.) that the fixed and variable calendars coincided at the beginning of the fifth year of Augustus, and that four years later the difference amounted to one day : -p. 30. γέγονε δὲ ἡ εἰρημένη διὰ ̓Αυξ ἐτῶν ἀποκατάστασις ἀπό τινος ἀρχῆς χρόνου ἐν τῷ πέμπτῳ ἔτει τῆς Αὐγούστου βασιλείας, ὡς ἐκ τούτου πάλιν τοῦ χρόνου τὴν ἀρχὴν εἰληφέναι τοὺς Αἰγυπτίους προλαμβάνειν καθ ̓ ἕκαστον ἐνιαυτὸν τῷ τετάρτῳ μέρει τῆς ἡμέρας. From the beginning of the fifth year of Augustus (26/25 B.C.) the Egyptian calendar began to anticipate the Alexandrine at the rate of a quarter of a day each year; this difference amounted to one day in 22 B.C., which was the first year in which a day was intercalated. The examples of double dates in the πρόχειροι κανόνες agree with this; we have first 1. c. p. 30, 31, ἀπὸ τῆς ἀρχῆς τῆς Διοκλητιανοῦ βασιλείας ἔτη οζ Θωθ κβ. εὕρομεν δὲ τὸν κατ ̓ Αἰγυπτίους μῆνα Χοιὰκ τὴν δὲ ἡμέραν κη. This is the year 360/1 A.D., and 360 = 4 + (4 × 89); the difference is made up of 8 days Thouth + 30 Phaophi + 30 Hathyr + 28 Choiak = 96 = 89 + 7. Again the date of the total eclipse +7. of the sun which took place on June 16, 364 A.D., is thus given:-1. c., p. 77. τῷ ὀγδοηκοστῷ ἔτει Διοκλητιανοῦ μηνὶ καθ' Ελληνας Παυνί. p. 81. ἔσχομεν καὶ τὸν κατὰ τὴν ἀκριβῆ τῶν φώτων συζυγίαν κατ' Αἰγυπτίους χρόνον, μεθ ̓ ὥρας βεδ' κ' τῆς ἐν τῇ εἰκοστῇ τετάρτῃ τοῦ Θωθ μεσημβρίας : and p. 82. εὕρομεν τὸν κατ' ̓Αλεξανδρέας μῆνα μὲν Παυνὶ ἡμέρας δὲ κβ, ἧς μετά τε β δ' ὥρας καιρικὰς τῆς μεσημβρίας φαμεν τὴν σύνοδον ἀποτελεῖσθαι. άTоTEλETOOα. The year is 363/4 A.D. and 363 = 3 + (4 × 90): the difference consists of 8 days Pauni + 30 Epeiph + 30 Mesore + 5 Epagom. + 24 Thouth 97 90 + 7. = = Thus the evidence of the papyri and the statements of Ptolemy and Theon are in complete accord; it may be regarded as certain that the fixed Alexandrine year was instituted in 26 B.C., and that 22 B.C. was the first year in which a day was intercalated. This method may now be employed to determine the limits within which the horoscope, Ox. Pap. II. ccxxxv. 5, must lie the date, in which the number of the year is lost, is given thus: : κατὰ [τὸ ἔτος Τιβερίου μηνὶ Φαΐφι ἄ, κατὰ δὲ τοὺς ὥρᾳ τετάρτῃ τῆς νυκτός. The solution depends on whether the noon of Phaophi 1 corresponded to the noon of Phaophi 11 or of Phaophi 12: in the former case the difference is 10 days, and the earliest date is Sept. 29, 15 A.D., the latest Sept. 28, 18 A.D.; in the latter, the difference being 11 days, the earliest date is Sept. 29, 19 A.D., and the latest Sept. 28, 22 A.D. It seems most probable that, in the case of dates given according to the Alexandrine calendar, the civil day was counted from midnight to midnight, in accordance with the Roman style; the authority of Pliny may be quoted in support of this inference:-ipsum diem alii aliter observavere, Umbri a meridie ad meridiem, vulgus omne a luce ad tenebras, sacerdotes Romani et qui diem finiere civilem, item Aegyptii et Hipparchus a media nocte in mediam. Pliny, Hist. Nat., II. 188. This statement is open to doubt in the case of Hipparchus, and can be true of the Egyptians only after the introduction of the Alexandrine calendar; but we are justified in supposing that, when the calendar was revised, the beginning of the civil day was fixed, according to the Roman style, at midnight. The form in which dates-κατ ̓ Αἰγυπτίους, οι κατ ̓ ἀρχαίους—of events which took place at night, are given (e. g. Pawpi ia riç iẞ), implies that in former times the Egyptians, like the vulgus omne of Pliny, regarded the day as extending from sunrise to sunset, and that the nights were dateless; hence, when they found it necessary to refer to the night, they were forced to speak of it as the night leading from one day to the next. Ptolemy always gives his own dates in the old Egyptian style, and various reasons have been assigned by Pétau, Ideler, and Boeckh (see Boeckh, Sonnenkreise, p. 301) to account for the manner in which he refers to the night: these reasons all depend on the assumption that the method was invented by Ptolemy himself; but it is now known, from instances in the unpublished Petri papyri, that this was not so, and that he merely adopted a system which had been in popular use for centuries before his time in Egypt. In dating events at night by the Alexandrine calendar this system is abandoned, and the time is given by the number of a single day. Accordingly, a civil day of twenty-four hours must have been defined, and, since the definition took place under Roman influence, we may assume that the day began at midnight. We must next consider how the hours of the night were counted. With the single exception already discussed, when an event was recorded as taking place at a definite hour of the night, the hours were counted from sunset: this is the natural interpretation to put on such expressions as the fourth hour of the night' or 'the tenth hour of the night,' and is the habitual practice of Greek and Roman writers. Ptolemy, requiring greater accuracy in the definition of time for astronomical purposes, employed a different method, and gave the time as so many hours before or after noon or midnight : for instance, in Synt. Math. Bk. vII., 8 p.m. is, according to Agrippa, VUKTÒS Wρas Y ἀρχούσης ; according to Ptolemy, πρὸ δ ὡρῶν καιρικῶν τοῦ μεσovukτíov; and again, 4 a.m. is, according to Menelaus, ὥρας ῖ πεπληρωμένης; according to Ptolemy, μετὰ δ ̓ ὡρας καιρικὰς τοῦ μεσονυκτίου. But even to Ptolemy the meaning suggested by ἐπὶ τρίτης τῆς νυκτὸς ὥρας could only have been at the third hour after sunset.' We must therefore conclude that in the British Museum papyrus quoted above, the form of the expression is inaccurate. If then Pawpɩ ā began at midnight, and the hours of the night were counted from sunset, it follows that noon of Phaophi 1 in the Alexandrine calendar corresponded to noon of Phaophi 1 in the old style, and the birth recorded in Ox. Pap. II. ccxxxv. must have taken place on Sept. 29 in 15 A.D., or on Sept. 28 in 16, 17, or 18 A.D. J. GILBART SMYLY. SOME OBSERVATIONS ON THE PEACE OF ARISTOPHANES. HE new Oxford text of Aristophanes, of which the First part, containing the first six plays, has recently been published,' must be judged by the ideal of the Bibliotheca Oxoniensis. A text free from misprints, and as free as possible from emendations; excellent typography; a rigidly exclusive apparatus criticus; abstinence on the part of the editors from making original contributions to the subject. There is much to be said for this ideal, but it is difficult satisfactorily to achieve. Messrs. Hall and Geldart have corrected their proof-sheets with scrupulous and successful care, and (notwithstanding a moment of weakness, which prompted them to make known an unfortunate guess on Peace 612) they have been conscientious in withholding conjectures of their own. They are exclusive and conservative; but they are, or at least seem, arbitrary in their admissions and omissions, in their condemnations and absolutions. They pass lines which are as certainly corrupt as other lines which they obelize; they omit corrections which are as probable and as worthy of consideration as those which they record. First and foremost, I must congratulate them on their general attitude to the metrical canons which have been elaborated in recent years, on the basis of statistical 1 Aristophanis Comoediae. Recognaverunt F. W. Hall, W. M. Geldart. Tomus I. Oxonii, e typographeo Clarendoniano. 2 For corrections adopted in the text, cp. Peace 892, Birds 1356. |