was written alone at the top of the papyrus; and on the analogy of the Florentine papyrus we may restore μετὰ τὴν ὑπατείαν τοῦ δεσπότου

ἡμῶν Φλαουίου) Ἰουστίνου] Φαμεν[ω] τ At the end, [7][s] x[7]][c] would suit the traces, and the next line is clear: ivdixtioros iv výpy. ινδικτίονος ἐν Συήνη. This reading, if accepted, gives the date 6 March, A. D. 573.

The papyri are almost all dated at Syene. In the Munich collection one, P. Mon. 97, was written at Antinoopolis, but since the parties were natives of Syene who happened to be at the moment in Antinoopolis, this document can hardly be regarded as an exception to the statement that all these papyri relate to Syene. A real exception is to be seen in the case of some of the London papyri. The two documents already referred to as the earliest in date of the collection, Inv. 1793 recto and Inv. 1805, come from the neighbourhood of Thebes; and since Inv. 1794, which has lost both beginning and end, resembles 1805 in its formulae and both to some extent in the hand, it seems very likely that it comes from the same district. In 1793 one of the parties lives at Kάoroov Keqapέws tov Θηβών νομοῦ, the other at Κάστρου Μεμνονίων) τοῦ Ἑρμωνθίτου ν[ο]μοῦ; in 1805 two are from Kάoroov [ ], one from [? Κάστρου Μεμνονίζων τοῦ Ἑρμωνθ[ίτ]ου νομού. All these three documents relate to loans, two of them to loans on security; and since the smaller fragments acquired with the collection include several fragments of accounts relating to articles of jewellery pledged as security for loans 1), it seems possible that they too come from the neighbourhood of Thebes, and that the whole of these Theban papyri form part of the papers of a family or families of pawnbrokers or moneylenders. Their presence among the Syene papyri is perhaps to be accounted for by the removal of some member of the family to that place; a supposition somewhat supported by the fact that the two dates preserved are both earlier than the earliest date found in the papyri written at Syene 2).

2. Parties to the Documents.

A number of the London papyri concern the Patermuthius son of Menas and his brother-in-law John son of Jacob with whom Wenger's Vorbericht (p. 13 ff.) has made us acquainted. Several facts as to this family are revealed by the London papyri in addition to those already known. In the first place, a conjecture of Wenger's (Vorb. p. 12) as to the identity of this Jacob father of John with the Jacob son of Dius of P. Mon. 96 is confirmed. The latter is described as x лatoòs síov tov zai Пlaoapari; and in Lond. Inv. 1797 mention is made (the document is

1) Cf. too the account on Inv. 1793 verso, published at the end of this article. 2) For Inv. 1803 (b), another papyrus which probably comes from some other place than Syene, see below, section 3.

addressed to Patermuthius and his wife Kako) of τοῦ ἀκκουβίτου Ἰακώβου Πασαραϊτός σου δὴ πατρὸς Κακὸ καί σου δὴ Πατερμουθίου πενθεροῦ. The genealogy of the two families is then as follows:

Jacobus (dead on 22 Aug.,

Α. D. 585 πρὸ ὀλίγων ἡμερῶν)

Fl. Johannes

Aur. later Fl. Patermuthius Aur. Kako Patermuthius was, as we learn from several documents, a sailor by profession, and a native of Syene, but he became a soldier of the numerus of Elephantine. At what date he entered the army can be determined only roughly and by inference. In Lond. Inv. 1797, of the reign of Tiberius (A. D. 578—582), he appears as Aur. Patermuthius, of Syene. In Inv. 1796, of A. D. 583-4, be is still Aurelius, and is described merely as a sailor. In Mon. 97, of 23 June, 583, he is also Aurelius (Vorb. p. 13). In Inv. 1787, of 12 March, 584, he is again Aurelius and a sailor, and this is still the case in P. Mon. 98, dated 30 May, 585 (Heisenberg). On the other hand, in Inv. 1790, of 22 Aug., 585, he is Flavius Patermuthius, soldier of the numerus of Elephantine, and so he appears henceforth1). Consequently, the date of his entry into the army must fall between 30 May and 22 Aug., 585. There is indeed one papyrus which at first sight suggests an earlier date. This is Inv. 1803 (a)2), a very fragmentary document dated in the 7th year and 3rd of the [post-consulate] of an Emperor whose name is lost, but who must, from the conjunction of these two systems of reckoning, be Tiberius), that is, the year 580-581, or, if the consular reckoning is strict 4), 581. The first line of the body of the document reads, in its present mutilated state, ἐκ πατρὸς Μηνᾶ το[υ] καὶ λεγομένου Βηντέ, and the second dolμov Zvýrys zt2. In a subsequent fragment, occur the words ὁμολ]ογῶ ἐγὼ ὁ προγεγραμμένος Πατερμούθιος. from which it is clear that in the first line 2(covios) Harεquorios must be read before z argós. It might therefore be thought that this is the Patermuthius who occurs so often in these papyri, and it would follow that he was a soldier as early as 581: but since this is contrary to all the other evidence, and the name Bré is not elsewhere recorded for his father, and since moreover, as Prof. Heisenberg points out to me, Patermuthius is elsewhere

1) Except in Inv. 1791 mentioned below (p. 165).

2) The beginning of Mon. 106 (Heisenberg).

3) Tiberius counted his regnal years from his proclamation as Caesar in 574; cf. P. Lond. 774, III. 280.

4) The post-consular reckoning was usually made to synchronize with the regnal years throughout; ef. P. Cairo Byz. 67151, 3, note, 67159, 1, note.

of the numerus of Elephantine, whereas this person belongs to that of Syene, it may be taken as quite certain that he is not identical with the Patermuthius under discussion.

Inv. 1797, where the dating clause is lost but in which the oath is by Tiberius, shows that Patermuthius and Kako were married at least as early as 582; and the character of the agreement in Inv. 1796 (see below, section 6) suggests that their marriage may have been recent when that document, dated 583-4, was drawn up. The various purchases of house property made by Patermuthius and the loan advanced by him in 586 (P. Mon. 99) to his mother-in-law indicate that he was in fairly flourishing circumstances; but there is some reason to suppose that later in life he may have been less prosperous. In Inv. 1788 (A. D. 611) he and Kako are found borrowing 4 solidi from a sailor of Syene; and in the latest document of the collection, Inv. 1789, dated in 613, Patermuthius, unaccompanied by his wife, borrows 31 solidi from the same person, pledging as security some articles of copper (ópiza2zos). The fact that his wife does not appear in this document may possibly be due to the fact that she had died in the interval.

3

One other point in connexion with Patermuthius must be noted. In Inv. 1791, which is dated in Mesore of the 5th indiction, he is described as Fl. Patermuthius (MS. IIa.), soldier of the numerus of Philae, and a sailor. Since he is a soldier this document must be at least as late as 585. That the 5th indiction is the year 586-587 is suggested by the fact that the document concerns an arbitration by Marcus, who may very likely be the same as the Marcus 670λaorizós who acts as arbitrator in P. Mon. 103, of the year 583 (Vorb. p. 16f.). Now in 585, as we have seen, Patermuthius was already a soldier of the numerus of Elephantine, and so he appears in all later documents, down to 613. We cannot therefore suppose that he was first assigned to the numerus of Philae and later transferred to that of Elephantine, and if the document is not to be dated later than 613 (the earliest possible date would then be 616), which seems unlikely, the most probable explanation is that Por is merely a scribe's blunder1).

As regards John, the brother-in-law of Patermuthius, it may be noted that in Inv. 1792 he is described as oro(ations) teigor keyi[óro]s [ons. The date of the document is 8 March, 584 or 585), and at []..

1) It is, however, possible, since his mother's name is not given, that this is the Patermuthius son of Menas of Inv. 1803 (a) (above, p. 164), who had perhaps been transferred from the numerus of Syene to that of Elephantine. But that Patermuthius is not, in Inv. 1803 (a), described as a sailor.

2) The doubt is due to an inconsistency in the dating clause. The regnal year is the 3rd and the indiction the 2nd. One must be wrong, since March of the 3rd year of Maurice fell in the 3rd indiction.

this time therefore he was a recent recruit. As in P. Mon. 97 of 23 June, 583 he is στρατιώτης ἀριθμοῦ Συήνης (see Vorb. pp. 13, 14), the earlier date is the more probable for Inv. 1792. The name Aurelius assigned to him in the Munich document is presumably a scribe's blunder.

Two other families may be briefly mentioned. The genealogy of the first is:

Another member of the family was perhaps a certain loárns o καὶ Καυμᾶς”) λεγόμενος αυλιζόμενος ἐν τῷ μοναστηρίῳ Πανιπάνης τανῦν de er Ty Zvýry expedic, who in Inv. 1797 (578-582) appears as representative of Tsia, then a¶§. That he was a relative is somewhat suggested by the evidence of Inv. 1787 (A. D. 584). That document is addressed to Patermuthius by a certain Ἰωάννης υἱος Πατεχνουμίου ἐλάχιστος μονάζων ἀπὸ τῆς Συήνης ὁρμώμενος. He refers in it to some claim. apparently touching property sold by him. likely to be made by the sons of Constantius". From the genealogy given above it appears that Tsone and Tsia were great grand-daughters of Patechnumius. and that the former was married to a son of Constantius. Since the John guardian of Tsia was also a monk, it is at least possible that he is to be identified with the John of Inv. 1787. In that case he was the great-uncle of his ward. The other family referred to has the following genealogy:

1) He is described merely as τοῦ προπάππου ἡμῶν, and may therefore be the ancestor of Apa Dius, not of Rachel. It is to be noticed that Tsone and Tsia describe themselves both in Inv. 1797 and in Mon. 101 as daughters of Rachel ἐκ πατρὸς Απα Δίου. This placing of the mother first may perhaps indicate that they were illegitimate children.

2) Cf. P. Lond. IV. Index. Kortrés (cf. P. Lond. IV, 1449, 6) is also possible but less likely.

3. Soldiers and Sailors.

In the London collection, as in that at Munich. the majority of the persons met with are soldiers. As Wenger remarks (Vorb. p. 22), the fact is significant for the importance of the military element in this frontier station.

The soldiers who occur, as parties, subscribers, witnesses, or scribes. are for the most part simple soldiers, whether αριθμού οι λεγιῶνος Συήνης, but there are also many of other ranks or numeri. Patermuthius, as we have seen, was 2 στρατιώτης ἀριθμοῦ Ἐλεφαντίνης, and is once described, apparently by error, as of the numerus of Philae. John, his brother-in-law, is in one document described as στρατιώτης τείρων λεγι[ώνο]ς [Στήλης. The other titles which occur are κεντυρίων, Αύγου στάλιος, ἀπὸ βικαρίων), ἀδιούτως), ακτουάριος (also ἀπὸ ἀκτουαρίων), and τvụлavάo(10g) (drummer). It is possible that in one case a otationrágios is also mentioned, but the reading is here very doubtful1).

One papyrus (Inv. 1803b) raises a problem which I am at present unable to settle with any confidence. The subscriber and witnesses (the first half of the document is lost) are described as soldiers of the numerus of Ba or Ba-. M. Jean Maspero has suggested in a letter that this is to be read as Bardizor), and he referred to Proc. Bell. Vand. II. 14, τοὺς γὰρ Βανδίλους, οὓς Βελισάριος ἐς Βυζάντιον ἤνεγκε, κατεστήσατο βασιλεὺς ἐς καταλόγους ἱππικοὺς πέντε, ὅπως ἐν πόλεσι ταῖς ἑφαις τὸν ἅπαντα ἱδρύσωνται χρόνον. He added however, „Je n'ai pas grande confiance en cette conjecture", suggesting that the names might throw light on it. They are mostly illegible; those which can be read are Βίκτωρ Ἰωάννου and Ἰακώβ Ἰωάννου Φαρ. The conjecture is in any case greatly to be doubted. In the first place, doduós is in the other papyri followed by the name of a place; and moreover, if there were Vandals at Syene they would hardly occur exclusively in this single document. The strongest argument I have, however, discovered since writing to M. Maspero. The scribe calls himself rouz(òc) Ba. Ka/ could also be read, but Ba is the more probable in view of the other evidence, and must be a place-name. The only likely extension I can suggest is Ba Brioros). If that is correct, we must suppose either that the parties to the document were natives of Syene temporarily at Babylon), or that

1) In the quotation in Vorb. p. 22, лgaiñóóitog (this is the right reading) στρατιώτης ἀριθμοῦ Συήνης, the first word is probably not to be taken, with Wenger, as a title but as a name; cf. Lond. Inv. 1791, P2(covios) Пgairósitos υἱὸς Θεοδόρου στρατιώτης) ἀριθμοῦ Συήνης.

2) Cf. Oxy. IX 1190, wo die Rekruten nach Babylon überstellt werden (Heisenberg).