É. Apor (ed.): Jubilee Volume of the Oriental Collection, 1951–1976. Papers Presented on the Occasion of the 25th Anniversary of the Oriental Collection of the Library of the Hungarian Academy of Sciences.

J. HARMATTA: Sir Aurel Stein and the Date of the Sogdian "Ancient Letters"

87 Chavannes' book he would have spared himself a series of misunderstandings and mistakes. 7. W. B. HENNING: BSOAS 12 (1948) pp. 601-602. 8. Sir Aurel STEIN, Serindia II. p. 735. 9. Sir Aurel STEIN, Serindia II. p. 669. The fragmentary slip is not included among the documents published by E. Chavannes and the inventory number of No. 609 (T.XII. a.II) is obviously incorrect (the serial number of the find is missing). 10. Sir Aurel STEIN, Serindia II. pp. 721-766. 11. The tables were composed on the basis of the books by E. CHAVANNES, Les documents chinois découverts par Aurel Stein, Oxford 1913., by H. MASPERO, Les documents chinois de la troisième expédition de Sir Aurel Stein en Asie Centrale, London 1953., by Sir Aurel STEIN, Serindia II, Oxford 1921., as well as by help of the corrections written by Stein in his copy of Chavannes' book. The table compiled by Chavannes is incomplete: it does not contain every date occurring in the Chinese documents published by him and does not give any reference to their finding places. The latter defectiveness of the table was clearly felt by Stein who wrote in the references to the sites in his copy of the book. In general, studying Chavannes' book, one must constantly consult Sir Aurel STEIN'

s Serindia. 12. The serial numbers represent those of the documents published by Chavannes while the letter M.after them marks the serial numbers of the documents published by Maspero. A question-mark after the serial number means that the date of the document cannot be established with absolute certainty. 13. The inventory numbers of the documents contain at the first place the sign of the watchtowers consisting of the capital letter T, a Roman numeral (-serial number of the watch-tower) and sometimes also a minuscule letter, while the second Roman numeral indicates the different refuse-heaps or other finding places within the same site. Lastly, the Arabic numerals mark the serial numbers of the finds unearthed at the indicated finding place. Thus e.g. the inventory number T.XV.a.III. 27 is to be explained as follows: T.XV.a = sign of the watch-tower, III indication of the finding place within the site T.XV.a, 27 = serial number of the find discovered at the finding place III. The meaning of the inventory numbers became clear even for Chavannes only after Sir Aurel Stein had called his attention to the significance of their different components (cf. his remarks in the Errata of his book, p. 230) 14. Sir Aurel Stein had convincingly shown (Serindia. II.p. 636) that of the two theoretically possible dates proposed by Chavannes, only 94 B.C. can be taken into account from a historical view-point. 15. This document contains two dates: 63 B.C. and 58 B.C. (Cf. also 58 B.C. ) 16. This document contains two dates: 58 B.C. and 63 B.C. (Cf. also 63 B.C. above. ) 17. Cf. MASPERO, op. cit. p. 41. 18. Cf. MASPERO, op.cit. p. 9, n.6. 19. The inventory number T.IV. II. 15 in CHAVANNES, op.cit.p. 77 is obviously a misprint and should read T.XIV.II. 15 (not corrected bv him in the Errata), cf. STEIN, Serindia II.p. 688. 20. Cf. Maspero, op.cit.p. 6, n. 7. 21. This document is not included in the Table compiled by Chavannes, cf. STEIN, Serindia II.p. 686. 22. Cf. MASPERO, op.cit. p. 26. He proposed two alternative dates for this document: 10 B.C. and 115 A. D. Taking into consideration, however, that we have dated documents at the watchtower T.XXII.d only from 47 A.D. on, we must regard 115 A.D. as the only possible date. 23. E. CHAVANNES, op. cit. p. 145 determined the date of this document by the following argumentation. The slip, representing part of a calendar, belongs to a group of documents which includes, among others , a slip, analogous to those containing parts of the 'Chi chiu chang' written between 48-33 B.C. Knowledge of this work on the Tun-huang Limes can probably be assumed in his opinion in the 1st and Ilnd centuries A.D.In this case the date of No.

680 can only be 153 A.D. If we take into consideration that a fragment of the 'Chi