^{page 112 note 1} My journey to Kabul was undertaken in the course of my duties as Research Lecturer in Ancient History at Christ Church, Oxford, and with the assistance of a grant from the College.

^{page 112 note 2} I understand that the inscription of Uruzgan was first mentioned in Kabul by Mr. Benava, a member of the Foreign Press Department of the Afghan Government. I am much obliged to Dr. Kohzad for passing on the news to me.

^{page 112 note 3} Khalilullah Khan, Chief Clerk of the Governor's private secretariat at Kandahar, was deputed to accompany me as guide, a most valuable reinforcement.

^{page 113 note 1} E.g. Cunningham, , Num. Chron. 1894, p. 276 fGoogle Scholar.

^{page 113 note 2} Thomas, F. W., “A Tokhari (?) MS.,” JAOS. 64 (1944), 1Google Scholar. Altheim, F., Aus Spätantike und Christentum, plates 7–10Google Scholar. Hansen, O., in La parola del pasato (Rivista di studi classici), fase. xx (1951), p. 361 (a reference which I owe to Professor H. W. Bailey)Google Scholar. The Uruzgan inscription is no doubt that of which reports reached Mons. Ghirshman at Kabul in 1942. See Ghirshman, R., Les Chionites-Hephthalites, p. 61Google Scholar.

^{page 114 note 1} Tallgren, , “Inner Asiatic and Siberian Rock Pictures,” Eurasia Septentrionalis Antiqua, viii, 183Google Scholar.

^{page 114 note 2} JRAS., 1944, p. 17Google Scholar.

^{page 114 note 3} My readings are principally based onCunningham, , “Later Indo-Scythians,” Num. Chron. 1894, p. 243, and Herzfeld, Kushano-Sassanian Coins, Table III. Not all of the more recent treatments of the subject are on sound linesGoogle Scholar.

^{page 114 note 4} The Hephthalite Greek of the inscription is transliterated in the Latin script, since the normal Greek alphabet has insufficient characters. The following conventions are observed. Greek san (p) is rendered by š. Omicron and alpha are not consistently distinguished at this period, and in transliteration are rendered impartially by o. Omicron can also represent ū, as in the spelling ποβοπο = Shāpūr, for which see Herzfeld, , Kushano-Sassanian Coins, p. 37. Hiatus between vowels clearly implies the presence of a glide, w, not written. The asterisk * represents a letter of uncertain value. Letters damaged or illegible are represented by a dash for each letterGoogle Scholar.

^{page 114 note 5} I cannot accept the view that this complex represents phi.

^{page 114 note 6} Cf. Hansen, L. c., p. 682.

^{page 115 note 1} E.g. upon a gold stater of Huvishka in the collection of Major-General H. L. Haughton, and on a garnet seal in the Fitzwilliam Museum. Both unpublished.

^{page 115 note 2} An acute observation which I owe to Mons. Raoul Curiel.

^{page 115 note 3} Cunningham, , Num. Chron. 1894, pp. 276–8. Kura (Salt Range, Punjab) inscription of Toramana, Epigraphia Indica, i, 238Google Scholar.

^{page 115 note 4} A garnet intaglio in the Fitzwilliam Museum, Cambridge; a Kushan bulla in the Ashmolean Museum, Oxford. Both unpublished.

^{page 116 note 1} Brackets enclose conjectural insertions.

^{page 116 note 2} A welcome confirmation that the correct reading of the first letters of the second line is maho, a point about which I was at one time uncertain.

^{page 116 note 3} A reading which I owe to the quick eye of Khalilullah Khan.

^{page 116 note 4} I am grateful to Mr. Rabin for help with these readings. The second word of the first line is not quite certain.

^{page 117 note 1} “The whole of the great mountainous district of the upper waters of the Helmand and the Kandahar (i.e. Arghandab) rivers was known to the Arabs as Zābulistan, a term of vague application . . .” Strange, Le, The Lands of the Eastern Caliphate, p. 349Google Scholar. The sources have, in fact, little to add to the remark of Hamdullah Mustaufi (Tr. Le Strange, p. 144): “Zabul is a province both broad and long, which was of old a kingdom.” Firdusi makes repeated reference to Zabulistan, and since he wrote at Ghazni, close at hand, must have been well aware of its location. He describes it (Ed. Mohl, , Valash, 1, 27) as grouped with Kabulistan, Bust, and Ghaznin, under the charge of a marzban. Whatever the region's exact boundaries, there can be little doubt that our inscription lies at the heart of ZabulistanGoogle Scholar.

^{page 117 note 2} The petroglyphs represent animals, mostly ibexes, and closely resemble those in the Ghorband Valley, described by Foucher, , La vieille route de L'lnde, ii, 390Google Scholar. For a date, compare Tallgren, “Inner Asiatic and Siberian Rock-Pictures,” in Eurasia Septentrionalis Antiqua, viii, 183Google Scholar.