^{page 501 note 1} See Roth, “ Über den Soma ”, ZDMG. 1881, pp. 681–92Google Scholar; and “ Wo wäehst der Soma ? ”, ZDMG. 1884, pp. 134–9.Google Scholar Excellent English translations of both papers were furnished by Mr. C. J. (subsequently Sir Charles) Lyall, I.C.S., with a letter dated 22nd July, 1884, to the Hon. Sir Steuart Bayley, then Member of the Governor- General's Council. Together with notes of Dr. G. Watt they were supplied in print to officers employed on the Afghan Delimitation Commission. I owe a typewritten copy of those translations to the courtesy of the officer in charge, Government of India Records.

^{page 501 note 2} Cf. Macdonell-Keith, , Vedic Index, ii, pp. 475Google Scholar, under the head Soma, for a succinct survey of the numerous widely different opinions recorded by Indologists and others about the identity of the plant figuring in Vedic hymns and later texts.

^{page 502 note 1} See Innermost Asia, i, pp. 263 sqq.Google Scholar

^{page 502 note 2} See ibid., figs. 171, 172.

^{page 502 note 3} Cf. his Appendices, in Serindia, iii, pp. 1351 sq.Google Scholar; Innermost Asia, ii, pp. 996 sq.Google Scholar

^{page 503 note 1} Cf. Innermost Asia, i, pp. 265, 268Google Scholar (L.F. 05, L.F. i, 03).

^{page 503 note 2} Cf. ibid., ii, pp. 736 sq., 740 sq. (for graves L.S. 2, 3, 6); 743, 748 (L.Q. iii).

^{page 503 note 3} Cf. Dictionary of Economic Products of India, iii, pp. 246 sq.Google Scholar

^{page 504 note 1} It fully agrees with Dr. Aitchison's observation about the distribution of the Ephedra that I found a low scrub, known locally by the name of Hūm and closely corresponding in appearance to the above description of the genus, growing plentifully on the gravelly wastes crossed on my journey of 1915 along the Perso-Afghan border between Mujnābād and Durōh. I well remember how bitter the taste was when I tried to chew a little of the green twigs.

^{page 504 note 2} See loc. cit., iii, p. 250.

^{page 505 note 1} I take my information on this point from Wellcome's Excerpta Therapeutica, 1930, pp. 72 sqq.Google Scholar

^{page 505 note 2} Cf. e.g. RV. viii, 48, 1, where the Soma is called the drink “ to which all the gods and men together stream calling it ‘sweetness’ self ’ ”, as quoted by Roth, ZDMG. 1881, p. 683.Google Scholar

^{page 506 note 1} For a lucid analysis of such data and of the references bearing in general on the cult-practices connected with the Soma, cf. Macdonell-Keith, , Vedic Index, ii, pp. 474 sqq.Google Scholar

^{page 506 note 2} Cf. ibid., p. 475, note 14.

^{page 507 note 1} See Yasna x, 11, in Geldner's edition.

^{page 508 note 1} This communication was, I regret to confess, through my fault, not printed in the Proceedings of the Congress. The identifications then proposed were mentioned by me in 1887 to my lamented friend Professor James Darmesteter and readily accepted by him ; see his Zend-Avesta, i, pp. 102 sq.Google Scholar, with notes 30–4. For an independent reference to that communication, cf. Geiger-Kuhn, , Grundriss der Iranischen Philologie, ii, p. 393, note 2.Google Scholar

^{page 508 note 2} The Pahlavi commentary renders the iškata by shikaft “ cave ”. Can this interpretation be connected in any way with the legend of Alexander's Greeks which looked for Prometheus’ cave in the Indian Caucasus, i.e. the Paropanisus ?

^{page 508 note 3} Here, too, as in the case of the phonetic derivation of Tirāh (see below) account may have, perhaps, to be taken of the influence exercised by a local population speaking a Dardic tongue. For the change of initial k > kh > gh cf. Grierson, , Piśaca Languages, p. 93.Google Scholar The change of initial k into kh is regular also in certain East Iranian languages: see ibid.

The Ghōrband valley lies very close to the area where certainly in later times Pashai, a Dard language, was spoken.

It deserves to be noted that the name Ghōrband occurs also as the name of a considerable valley which descends to the right bank of the Indus from the watershed towards Upper Swāt. The valley belongs to a hill-tract where until the Pathān conquest of late mediæval times a Dardic language, akin to those still prevailing in the adjacent Indus Kōhistān, was spoken.

^{page 508 note 4} Cf. Grierson-Stein, , “ Notes on Tirāhī,” JRAS. 07, 1925, pp. 405–16Google Scholar; Grierson, , The Piśāca Languages, p. 133.Google Scholar

^{page 509 note 1} See An Archœological tour in Wazīristān and Northern Balūchistān (Memoirs of the Archæological Survey of India, 1929, No. 37), pp. 2 sq.

^{page 509 note 2} Cf. ibid., p. 2, note 3. For Professor Hillebrandt's identification, see Vedische Mythologie, iii, p. 268.Google Scholar

^{page 509 note 3} Professor Hillebrandt, 's identification, Vedische Mythologie, iii, p. 268Google Scholar, note 3, seems to have been suggested first by Dr. Brunnhofer (Iran und Turan, p. 41). The close phonetic relation between the Vedic and the modern form of this local name is too clear to require specific demonstration.

^{page 510 note 1} For the orographic configuration of this region, the sheets: Afghanistan, Baluchistan, of the Survey of India's Southern Series maps on the 1 : 2,000,000 scale may be conveniently consulted.

^{page 511 note 1} See Dictionary of Economic Products, vi, pt. i, p. 487Google Scholar: “ This species is found on the drier ranges of the Western Himālaya from Kumāon (altitude 14,000 to 16,000 feet) to Western Tibet (altitude 9,000 to 14,000 feet) and is distributed to Afghanistān. … ”

Food.—“ In Afghānistan, the plant is always wild, and appears to grow abundantly in many parts. When green, the leaf stalks are rawash, and when blanched by heaping up stones and gravel around them, they are called chukri; when fresh (in which state they are sometimes brought to Peshawar in spring) they are eaten either raw or cooked. They are also dried for use, to be eaten with other food, and are sometimes made into a preserve.” (Stewart.)

^{page 512 note 1} I quote the relevant passage from Sir Lyall, Charlesx0027;s translation of Roth's paper, ZDMG. 1884, pp. 134 sqq.Google Scholar

^{page 512 note 2} See Macdonell-Keith, , Vedic Index, ii, p. 475.Google Scholar

^{page 513 note 1} Cf. ibid., ii, p. 477.

^{page 513 note 2} See ibid., ii, p. 477, and Hillebrandt, , Vedische Mythologie, i, pp. 219 sqq.Google Scholar, there quoted.

^{page 514 note 1} Cf. Macdonell-Keith, , Vedic Index, ii, p. 478.Google Scholar Their names are Ārjīka, Pastyāvant, Śaryaṇāvant, Suṣomā, the territory of the Pañcajanāḥ.

Apart from Suṣomā the only other locality for which a likely identification might be proposed, is Śaryaṇāvant. Its mention in RV. viii, 7, 29, along with the Suṣomā has suggested to me that its name might perhaps be connected with that of the Harro river, which drains the main portion of the Hazāra District, to the west of Rawalpindi. The phonetic derivation of the modern name would offer no serious difficulty, as the change ś > h is regular in the Indo-Aryan languages of the Indian North-West.

^{page 514 note 2} Cf. my article “ On River Names in the Rigveda,” JRAS. 1917, pp. 91–9.Google Scholar