^{page 470 note 2} Abbreviations:

AEW H. Adjarian, Armenisches etymologisches Wurzelwōrterbuch

AIW C. Bartholomae, Altiranisches Wōrterbuch

GEW H. Frisk, Griechisches etymologisches Wōrterbuch

IEW J. Pokorny, Indogermanisches etymologisches Wōrterbuch

KBT H. W. Bailey, Khotanese Buddhisttexts

KT H. W. Bailey, Khotanese texts

LitEW E. Fraenkel, Litauisches etymologisches Wōrterbuch

OssED V. Abaevxs, Istoriko-ëtimologičeskij slovar’ osetinskogo jazyka

TPS Transactions of the Philological Society

^{page 470 note 3} This šav- should go with that in BSOAS, xxm, 1, 1960, 36 ff Google Scholar.

^{page 471 note 1} The problem of rtá-, aša- is again considered byKuiper, F. B. J., Indo-Iranian Journal, v, 1, 1961, 41–2Google Scholar.

^{page 471 note 2} DlcM 756.5, 21; 352.13, bis.

^{page 471 note 3} The older form dlwpwšt in GrBd. 61.4, but more usually without the first -w-, as in GrBd. 61.1, 4; Dd. 36.16, 17, 66; Zātspram 5.1–3; Mēnōk ī xrat 1.31, 13.2; the Pāzand has drūpušt.

^{page 471 note 4} Henning, W. B., BSOS, VIII, 2–3, 1936, 585Google Scholar. The Buddhist Sogdian initial was ambiguous, either n or z. As the rendering of Ind. (Buddhist) Brahmā the Sogdian had ‘zrw’.

^{page 472 note 1} ‘Sutra of causes and effects’ 225.

^{page 472 note 2} SCE 135. Here paθ- ‘to eject’ is connected with OInd. path-, see Benveniste, E., BSL, LII, 1, 1956, 46Google Scholar. Khotanese has phūh- ‘to eject’, see TPS, 1961 (in the hands of the editor).

^{page 472 note 3} P 13.27.

^{page 472 note 4} For starana- note the discussion in Donum natalicium H. S. Nyberg oblatum, p. 15–16.

^{page 472 note 5} Pahlavi texts, p. 112, Draxt asōrīk, as recognized inMarkwart, J., Festgabe J. Szinnyei, 57Google Scholar.

^{page 472 note 6} Morgenstierne, G., Felicitation volume presented to S. K. Belvalkar, 90Google Scholar.

^{page 472 note 7} See the quotations inBSOAS, xxm, 1, 1960, 33Google Scholar.

^{page 472 note 8} As E 25.279 smθtty-upasthāne durna, samyaprahāxṇa/pūrnānu bājo rāddhäpāta kho hälśte; Hedin 21.3 (KT, IV, 34) in a list duna pūmna ḫuśtä besa ‘bows, arrows, spears, shields’.

^{page 472 note 9} EGlossar connected pūrna- with OInd.parvan- ‘knot’ and paruḫa- ‘rough’.

The suffix -un- in the sense of an agent or tool with further suffix -ya- can be seen in Ossetic D älxujnä, I älxui ‘spindle’. The discovery of Chorasmian ‘fa ^*alx ‘top of a spindle’ quoted byHenning, W. B., Z. V. Togan'a armağan, 436Google Scholar, has made it possible to improve upon TPS, 1945, 36. Here Ossetic ^*alxunya- (in which the i-umlaut has prevented the change of -u- to -ŭ- in Iron) and Chorasmian alx disclose an older ^*alxa- or ^*arxa- (possibly from ^*arxra- with dissimilated r-r, in reverse to udda'rt ‘reel’ from ^*uartadra-, see Morgenstierne, G., NTS, XII, 1942, 268)Google Scholar. With different suffix Greek has ἠλεκᾰτη, ἠλακ⋯τη, ⋯λακ⋯τα ‘distaff’ (Homer and later): it is left unattached in H. Frisk, GEW, s.v., andPokorny, , IEW, 676Google Scholar. From these two forms the base would be ^*alek-, in Iranian possibly replaced by ^*alakh- (if no -r- followed the -k-).

The unexplained OInd. RV akrá-, a straight and upright thing, which is compared to a svdru- ‘sacrificial post, yūpa- ‘may be taken for older ^*arkrd- with dissimilated r-r, or as metathetic for ^*arkā-. Earlier attempts to class this akrā- have given ‘elephant’ (with query, Geldner's Übersetzung), ‘pillar’ (Ludwig), ‘banner’ (Grassmann's Wōrterbuch) and ‘acorn’ (Brunnhofer) ; uncertain (Mayrhofer). Sayana and Mādhava's λακ⋯τgarthadīpikā traced akrā- to the base kram-; Yāska offered nakra-.

^{page 473 note 1} For par- seeBenveniste, E., BSL, LI, 1, 1955, 36 ff.Google Scholar, and Festgabefilr H. Lommel, 19.

^{page 474 note 1} E. Benveniste, Textes sogdiens.

^{page 474 note 2} The same group -nx- occurs in the Sogdian form of nryxšnx, nrxšnx, Parth. nrysf, Pers. nrsyh, Armen. lw. nerseh, Arab. Pers. narsai. There is -n- also in Sogd. nns ‘nose’, early enough to preserve the -⁸-. In an initial syllable we find Oss. D änsurä, I ssyr ‘tusk’, with Sogd. ‘ns'wr ‘tusk’, and Av. tiži.asūra-, quoted BSOAS, xxm, 1, 1960, 28, n. 6Google Scholar. The -š- of ‘rsy may come from Prakrit or be a Sogdian change from -š-. The Sogdian ‘xs'yn-’ ‘blue’ with ^-š- where the other Iranian languages show axšaina-, may be recalled. Khotanese has from Prakrit arja-.

^{page 474 note 3} Jolly, J., Medicin, 1901, 121Google Scholarapasmāra-, unmāda-, 69 graha-; English translation byKashikar, C G., Indian medicine, 1951Google Scholar.

^{page 474 note 4} Edited, KT, III, 135 (Stein 117.1–3). Facsimiles are printed inGoogle ScholarMatsumoto, Eiichi, Tonkōga no kenkyū, 1937Google Scholar, plates cxc, cxci.

^{page 474 note 5} KT, I, 172, Sanskrit text 89 r 3–4.

^{page 474 note 6} To be printed in KT, v. Parts of the Uigur text are inMüller, F. W. K., Uigurica, II, 50 ff.Google Scholar; andMalov, S., Doklady Akademii Nauk SSSR, Ser. B, 1930, No. 5, pp. 88–93Google Scholar.

^{page 474 note 7} Rachmati, G. R., Tiirkische Turfan-Texte, VII kr'x 4.3, 63Google Scholar; k'rx 4.19, 25.

^{page 475 note 1} P 2801.35, KT, in, 66 grahasta-. For -asta- note also P 3510.7.1 KBT, 51 gūttarastä ttīś śstā ‘having family, having brilliance’ from Ind. gotra- and tejas-. Jātahastava 21 v 1 ysirasta- ‘cordial’ is made from a Khotanese ysira- ‘heart’.

^{page 475 note 2} Bartholomae, C., Mittdiranische Mundarten, i, 8 ff.Google Scholar, gathered up passages known to him.

^{page 475 note 3} This comparison with Sogdian has now been made, though with a query, in V. Abaev, OssED, s.v. ärrā.

^{page 475 note 4} BSOAS, XXIII, 1, 1960, 34Google Scholar.

^{page 476 note 1} Pokorny, , IEW, 27Google Scholar; Fraenkel, E., LitEW, 7Google Scholar.

^{page 476 note 2} Kent, R. G., Old Persian, grammar, texts, lexicon, 170Google Scholar, preferred a(h)rilca-. On ari- see TPS, 1959, 109 ff.Google Scholar; M. Mayrhofer, Concise etymological Sanskrit dictionary, s.v. alikd-.

^{page 476 note 3} Hübschmann, H., Persische Studien, 135Google Scholar(rēy in theŠāhnāmah ed. Vullera, , III, 1547, 900)Google Scholar. J. Wackernagel proposed to identify Av. araēka- with Olnd.alīká-, as in the epithet of a snake AV 5.13.5 alīká-, see Wackernagel, J., Ai. Gram., III, p. 427Google Scholar.

^{page 477 note 1} C. C. Uhlenbeck, Kurzgefasstes etynwlogisches Wōrterbuch der altindischen Sprache, compared Greek Š⋯λ⋯ομαι, but in order to translate by ‘eilend’.

^{page 477 note 2} P. Thieme had rightly seen (IF, L, 1, 1932, 70–2)Google Scholar that the meaning of RV 10.71.6 was decisive. Regretfully one must give up the proposal of W. Wiist to see in RV 10.108.7 a place name Ālaka- connected with Ossetic Ālägātä. A name Ālāg, one of the Nārtü, is known (Iron ādämon sfäldystād, V = Pamiatniki narodnogo tvorčestva Osetin, V, 186) in the phrase Nārti Alägän āttäj jeunäg yog ‘Āläg of the Nārtä had one cow’. Ossetic-l- is of ambiguous origin, older Iranian either -r- or -ri-.

^{page 477 note 3} Wackernagel, J., Ai. Gram., III, 427–8Google Scholar.

^{page 477 note 4} Suvarriabhāa-sūtra, ed. Nobel, J., p. 116Google Scholar. Toch. A 324 b 7 āḍakava(ti).

^{page 477 note 5} Hoffmann, H., Āṭāṇātika-sūtra, pp. 55Google Scholar, 65, 69; Khotanese, in Suvarṇabhūsa-sūtra 53Google Scholar v 3 (KT, I, 237).

^{page 477 note 6} Lévi, Sylvain, JA, jan.-fév. 1915, 57Google Scholar.

^{page 477 note 7} Kirfel, W., Die Kosmographie der Inder, 224(Jaina)Google Scholar.

^{page 478 note 1} Lüders, H., ‘αντιδωρον Waclcernagel, 302Google Scholar, reprinted Philologica Indica (needing revision); Bailey, H. W., TPS, 1952, 61 ffGoogle Scholar. (add WaxI nrd, Pahl. Psalter n'dy; NPers. nāl from Sogdian, Henning, W. B., BSOS, X, 1, 1939, 98; Sangl.niyōk)Google Scholar.

^{page 478 note 2} On north-western Prakrit, Kroraina, -aǵa-, see Burrow, T., The language of the Kharoṣṭhi documents from Chinese Turkestan, 6, 23Google Scholar; Bailey, H. W., BSOAS, XIII, 1, 1949, 131Google Scholar.

^{page 478 note 3} Ed. Nobel, p. 116.

^{page 478 note 4} Hopkins, E. Washburn, Epic mythology,142, n. 4Google Scholar.

^{page 478 note 5} Pali, Vinaya, i, 91, 15Google Scholar.

^{page 479 note 1} Zeits. f. vergleich. Sprachforschung, LXV, 1–2, 1938, 142Google Scholar. Darmesteter gave ‘méchant’, as equivalent of drvant-.

^{page 479 note 2} Morgenstierne, G., IIFL, II, 224Google Scholar.

^{page 479 note 3} Lommel, loc. cit., p. 40, and K. Barr, Avesta, 137, omit both drva- andranha-. Weller, H., Anahita, 145Google Scholar, followed AIW.

^{page 479 note 4} Digoron k'undäg occurs in Māliti Geuärgi, Iräf, 60: dālä komi i-k'undägi nicäjsāydāj sāu k'äjdor.

^{page 480 note 1} TPS, 1955, 72 ffGoogle Scholar.

^{page 480 note 2} ibid., 72 ff.

^{page 480 note 3} Above, p. 478.

^{page 480 note 4} The verb E 13.75 uskūṣḍe is different and has kauš-, as pyūṣḍe is from ^*patigauṣḍatai. It renders Sansk. auddhatyaṃ dravaṃ prāviṣkaroti ‘he acts frivolously’.

^{page 480 note 5} Siddhasāra 138 r 4, cf. 143 r 3.

^{page 481 note 1} Earlier compared with Zor. Pahl. pan- ‘niggardly’.

^{page 481 note 2} KT, II, 66, no. 30, 6.

^{page 481 note 3} Ch. 0048.30 (KBT, 73) and Kha 1.306b, v 3 (KBT, 8), from tara-tara-.

^{page 481 note 4} KT, II, 75, 49 (and Asia Major, NS, II, 1, 1951, 30); KT, II, 8, 124; KT, III, 135, a3, 5; KT, IV, 92, 8.3.

^{page 481 note 5} BSOAS, XIII, 4, 1951, 926–30Google Scholar ; JRAS, 1942, 14–28Google Scholar; 1949, 2–4.

^{page 481 note 6} Siddhasara, 155 r 3.

^{page 482 note 1} Nor bafgirk’ haykazean lezoui, 1836; H. Adjarian, AEW, s.v. It would suffice to take ŏrr as due to metathesis from -ōr by shortened vowel and lengthened consonant.

^{page 482 note 2} Etymologically uncertain. Adjarian, loc. cit. suggested connexion with diur ‘ease; easy’. If the word should turn out to be Iranian a connexion with a base IE sed-, and nasalized send- will be likely. The Latin sēdāre ‘to quieten’, and Celtic Welsh hawdd ‘easy’ (usually associated with Lat.sedeo ‘sit’) show how such a meaning could develop. Vedic āsand ‘seat’ attests an early sand-.

^{page 482 note 3} Hübschmann, , Armenische Grammatik, 238Google Scholar.

^{page 482 note 4} B. G. Kent, Old Persian, rendered by ‘carnelian’.

^{page 482 note 5} BSOAS, XI, 4, 1946, 781–2Google Scholar.

^{page 482 note 6} This may be cognate with Oss. D kizgä, I yzg ‘girl’.

^{page 482 note 7} Quoted in Hūbschmann, , AG, 238Google Scholar.

^{page 482 note 8} Parthian in Šāhpuhr I inscription 30 θρησκε⋯α τ⋯ν θε⋯ν Persian in KZ 8 kltk'n Z Y yzd'n.

^{page 483 note 1} M. Mayrhofer, loc. cit., discusses possible origin in Dravidian, Austric, and Prakrit.

^{page 483 note 2} Mayrhofer, loc. cit.; Tedesco, P., JAOS, LXXX, 4, 1960, 361–2 Google Scholar.

^{page 483 note 3} Pokorny, , IEW, 988Google Scholar.