² Hoernle, A. F. R., Manuscript remains, 55Google Scholar; Watanabe, , JRAS, 1907, 263.Google Scholar

³ Thomas, F. W., Tibetan literary texts and documents, I, 105.Google Scholar

⁴ Asia, Major, NS, IV, 1, 1954, 94.Google Scholar

⁵ Fuller details are given in Khotanese texts, IV (now in the hands of the printer).

¹ Nariman, G. K., Jackson, A. V. Williams, and Ogden, C. J., Priyadarśikā, XLIIIGoogle Scholar, with facsimile plate.

² BSOAS, XXI, 3, 1958, 536Google Scholar; ‘Languages of the Saka’, in Handbuch der Orientalistik, IV, 1, p. 137.Google Scholar

³ Hoffmann, Karl, Corolla linguistica, 80ff.Google Scholar

⁴ KT, I, 233Google Scholar, 4 v 2; Sanskrit, ed. J. Nobel, p. 3.

¹ Kha 1.13, 144 r 2, KBT, 5Google Scholar. Sanskrit in Divyāvadāna, 158.Google Scholar

² KBT, 16.154 ff.Google Scholar

³ Kha 1.205 a, 35 b 2.

¹ KT, II, 103Google Scholar, translated in Asia Major, NS, VII, 1959, 24.Google Scholar

² Avestan āsna- in Bailey, H. W., Zoroastrian problems, 10Google Scholar, no. 1; earlier BSOS, VI, 4, 1932, 954Google Scholar; TPS, 1956, 91.Google Scholar

³ Details on the social position of the vazno can be found in Allen, W. E. D., History of the Georgian people, 225ffGoogle Scholar. Janashia, S., Istorija Grusii, I, 90Google Scholar, explained uazno by ‘ne-aznaur-’.

⁴ Dd 36.17, quoted in BSOS, VI, 4, 1932, 954Google Scholar; TPS, 1956, 91.Google Scholar

⁵ The Pahlavī word was misunderstood by West, E. W., SBE, XVIII, 87Google Scholar, as azenâvar ‘unarmed’, a wrong reading; in Hübschmann, H., Armenische Grammatik, 151Google Scholar, the word is changed into azēnāvar with -ē-.

⁶ Svanskie teksty 70 of Nižeradze, Sbornik Materialov … Kavkaza, XXXIGoogle Scholar; Haykakan matenagrout'iun, 1883, 23Google Scholar, in Latin script, part of a personal name.

¹ P 3513, 82 r 4, KBT, 65.Google Scholar

² Boyer, Rapson, Senart, and Noble, Kharoṣṭhī inscriptions, s.vv. No further case in the additional texts BSOS, IX, 1, 1937, 111 ff.Google Scholar

³ For the rank of nobles in Sasanian Persia, see Christensen, A., Iran sous les Sassanides, 2nd ed, 100 and 111 ff.Google Scholar

⁴ Nyberg, H. S., Øst og vest, 1945, 68.Google Scholar

⁵ The Armenian azat-k' are noticed by Markwart, J., Caucasica, VII, 1931, 150.Google Scholar

⁶ Zoroastrian problems, 10Google Scholar, no. 1, and JRAS, 1955, 14.Google Scholar

¹ AIW rightly saw the difference. It is not better explained by Duchesne-Guillemin, J., ‘L'homme dans la religion iranienne’, in Studies in the History of Religions (Supplements to Numen), II, 1955, 98Google Scholar, as from ‘appearance’, with parallel in Latin ‘species’. The proposal to see here the base reduced grade ka- by K. F. Johansson requires more assumptions, see Petersson, H., Arische und armenische Studien, 59 ff.Google Scholar

² References for vispuhr are given in TPS, 1956, 91Google Scholar; they have been widely discussed. The ‘daughter of the house’ is also known. In Sogdian occurs wysδywth (P 6.165).

³ It will be seen below that though it has rendered great services P. Thieme's book Der Fremdling im Ṛgveda (and its sequel Mitra and Aryaman) has not seemed satisfactory. The main thesis which connects in one group arí- ‘foe’, arí- with aryá- ‘īśvara’, aryamán-, and ārya-, is an assumption which seems to lack a foundation. Abaev, V. (Voprosy jazykoznanija, II, 1958, 114)Google Scholar, though favourable, failed to find confirmation outside Indo-Aryan. To alleviate this difficulty he proposed to connect Ossetic äcägälon ‘foreign’. But here a different analysis by the negative a- and čakara- offering a meaning ‘not inmate of the house’ would be more acceptable, on which elsewhere. It hardly seems possible to get good results in a matter as deeply involving Iranian terms as Indian unless the Iranian vocabulary is fully used. Here in this book neither the Avestan nor Ossetic vocabulary has been considered in the interpretation of Iranian arya-which has given Pers. (h)ēr ‘wealth’ and Iranian ar- which has given Ossetic Iron āryn ‘to bear children’.

⁴ Inscription of Šāhpuhr I, Parth. 19, Pers. 24. The Greek has Αριαν.

⁵ DW 438.23; West, E. W., SBE, XLVII, 129.Google Scholar

¹ Études asiatiques, 1954, 137.Google Scholar

² A valuable theoretical examination of the case of Ossetic iväzun ‘be outstretched’, ivāzun ‘stretch out’ to represent the expected verbal use not otherwise attested is presented by Benveniste, E., BSL, LII, 1, 1956, 60–71.Google Scholar

³ In kävdäs ‘stall’ we have an older *kafta- with added suffix -sa-, connected with the many words from kap- ‘be contained’ listed in TPS 1954, 144ff.Google Scholar

⁴ Dig. jerun is not yet certainly explained. Possibly older a- was replaced by ä-, whence jä-, je-, with the result that the jer- was associated with other verbs containing -er-, as per- ‘scratch’, c'erun ‘lick’, and serun ‘amble’. In ŭāryn, Munkácsi, A., Blüten der ossetischen Volksdichtung, I, 25Google Scholar, the ŭ- is secondary (further details in TPS, 1959).Google Scholar

⁵ Quoted in Vs. Miller's dictionary s.v. lämāryn, and TPS.

¹ Krause, W., Was man in Runen ritzte, 2nd ed., 1943, 23Google Scholar; Runeninschriften, 539, no. 56.Google Scholar

² Pokorny, J., Celtica, III (Zeuss volume), 308Google Scholar; Études Celtiques, VII, 18.Google Scholar

³ Fuller details in TPS, 1959.Google Scholar

⁴ 56 v 1, KT, I, 147.Google Scholar

¹ Siddhasāra 12 v 5.Google Scholar

² Or. 9609, KT, I, 235.Google Scholar

³ Morgenstierne, G., NTS, XVII, 1954, 281Google Scholar; NTS, II, 1929, 270.Google Scholar

⁴ WP, II, 323.Google Scholar

¹ It has previously been proposed to trace this nīra- to the nī- of nīta-, which, however, is rather to be set with Lettish niju, see BSOAS, XXI, 1, 1958, 43Google Scholar; also to snā- and na- which, however, seem to offer a less satisfactory connexion than this Khot. nai-d-. Rōšānī has nay-: nid ‘to churn butter’.

² The commentary of Bhaṭṭabhāskara-miśra is cited in the Vaidika-padānukrama-koṣa, s.v.

³ Morgenstierne, G., NTS, II, 1929, 269.Google Scholar

⁴ Benveniste, E., JA, 1933, 2, 241.Google Scholar

⁵ Morgenstierne, G., IIFL, II, 232.Google Scholar

¹ Indian Linguistics, XVI, 1955, 119Google Scholar, written before the complexity of bases containing nai- had been realized.

² Indian Linguistics, XVI, 1955, 118Google Scholar. Other words in Schrader-Nehring, Reallexikon, s.v. Käse.

³ Buddhica, Bibliotheca, IV, part iii, 218.6Google Scholar, Mūla-madhyamalca-kārikā, cited by Edgerton, , DictionaryGoogle Scholar, beside Lalita-vistara, ed. Lefmann, 207.17, kurkurā iva gardūra-baddhāḥ, ed. Vaidya, p. 150.

⁴ Quoted in Donum natalicium H. S. Nyberg oblatum, 11.Google Scholar

¹ Weller, F., Zum soghdiscken Vimalakīrtinirdeśasūtra, 62 ff.Google Scholar

² In BSOAS, XIII, 3, 1950, 654Google Scholarpaitrya- was taken to contain dar- ‘hold’, as had been proposed before, but shortly after the meaning of grab- seemed more suitable.

³ TPS, 1955, 76ff.Google Scholar; Morgenstierne, G., IIFL, II, 420Google Scholar. Other forms are cited there.

¹ Wackernagel, J., Altindischa Grammatik, II, 2, p. 463.Google Scholar

² Legends of Brbú occur later, see Vedic index, s.v.

³ Here the Sanskrit text diverges.

¹ Further forms in WP, I, 468Google Scholar; Pok., IEW, 627–8Google Scholar; Vasmer, M., REW, I, 546.Google Scholar

² Jakovlev, N., Sintaksis čečenskogo literaturnogo jazyka, 77.Google Scholar

³ Genko, A. N., Zapiski kollegii vostokovedov, V, 718Google Scholar; Dumézil-Jabagi, , Textes populaires inguš, 26.Google Scholar

⁴ Pamjatniki narodnogo tvorčestva asetin, II, 1927.Google Scholar

⁵ This problem needs further work. Preliminary notes are given in BSOAS, XXI, 3, 1958, 538–9.Google Scholar

⁶ The practice of adding a herb to drink was noted by Hesukhios, s.v. μελίτιον πόμα τι Σκuθικòν μέλιτος έψομένου σùν νδατι καì πóα τινí.

¹ Minorsky, V., History of Sharvān and Darband, 107ff.Google Scholar

² Herbert, A., Casuals in the Caucasus, 1912, 253.Google Scholar

³ A recent large survey is contained in A. Steiger, in Meier, F. (ed.), Westöstliche Abhandlungen, R. Tschudi zum siebzigsten Geburtstag überreicht, 1954, 87–106.Google Scholar

⁴ Henning, W. B., ‘Mitteliranisch’, in Handbuch der Orientalistik, IV, 1, p. 85Google Scholar, connects with Sogd. xōm, from older hauma-, Avestan haoma-.

⁵ Details are given in Annali, Istituto Universitario Orientale di Napoli, 1, fase. 2, 139.Google Scholar

⁶ Dig. k'insä has the intrusive -n- familiar in Oss. Dig. änsurä, Iron ssyr ‘tusk’, and earlier in Bud. Sogd. ‘ns'wr occurring in the cliché wywšw ‘ns'wr ‘six-tusked’, the usual epithet ṣaddanta- in Buddhist texts (VJ 201, et al.). Avestan tiži.asūra- has the word without -n- (Yašt 14.15). If Avestan mazā- is pre-Sasanian it has a similar -n-.

¹ Further details are given in TPS, 1955, 69–70.Google Scholar

² BSOAS, XII, 2, 1948, 327.Google Scholar

¹ Jacobi, H., Ausgewählte Erzählungen in Mâhârâshtrî, 65.Google Scholar

² For Khot. -ṭh- see JRAS, 1953, 110 ff.Google Scholar, with baṭha- ‘protective mail’ from *v(a)rθra. For kālā, Henning, W. B., Sogdica, 56.Google Scholar

³ Wackernagel, J., Altind,. Gram., II, 2, pp. 480 ff.Google Scholar

⁴ Khotanese in P 3513, 51 v 3, KT, I, 127.Google Scholar

⁵ Gr Bd. 131.11 mōč, Paris MS mōčak. Bud. Sanskrit in JRAS, 1955, 21.Google Scholar

⁶ Morgenstierne, G., IIFL, I, 401Google Scholar; II, 230.

¹ It was proposed in BSOAS, XXI, 3, 1958, 528Google Scholar to connect with gad- ‘to plunder’.

² Gershevitch, I., Grammar of Manichaean Sogdian, p. 162.Google Scholar

³ Details in Annali cited supra, p. 28Google Scholar, n. 5, I, fase. 2, 125.

⁴ WP, I, 537.Google Scholar

⁵ Abaev, V., Ist.-ëtim. slovar' oset. jazyka, 96Google Scholar, citing also zäl and zälyn.

¹ ‘Languages of the Saka’, in Handbuch der Orientalistik, iv, 1, p. 139Google Scholar. The h- also in huṣṣ- ‘grow’, participle huṣṣata-, from vaxš-: uxš-.

² Nārty Kāddžytä, ed. 1946, 86.3.Google Scholar

³ On vákṣas- WP, I, 261Google Scholar, is uncertain, but quotes an older proposal to connect vakṣanā-, vákṣas-, and vankṣaṇa-. On Armen, vak-, Nyberg, H. S., Hilfsbuch des Pehlevi, II, 96Google Scholar. Oss. sinäg, BSOAS, xx, 1957, 59.Google Scholar

⁴ E 14.71; Siddhasara 102 r 5; P 4089 a 2 (KBT, 20).

¹ Vs. Miller, , ‘Die Sprache der Osseten’, GIP, I, Anhang, 65Google Scholar tentatively thought of pat- ‘fall’.

² Hājjiābād inscription, edited Nyberg, H. S., Øst og vest, 66.Google Scholar

³ BSOAS, XVIII, 1, 1956, 32.Google Scholar

⁴ BSOAS, XI, 2, 1944, 291Google Scholar, the text now in KT, III, 81.Google Scholar

⁵ Vasmer, M., REW, in, 150.Google Scholar

⁶ Morgenstierne, G., NTS, II, 1929, 244Google Scholar; xv, 1949, 257.

⁷ JRAS, 1953, 100.Google Scholar

¹ The explanation attempted in Asia Major, NS, II, 1, 1951, 37Google Scholar, can thus be completed. Kuci taur ‘dust’, if named as ‘covering’, could be native or Iranian.

² Rugged flanks of Caucasus, 11, 147.Google Scholar

³ Mayrhofer, M., Sanskrit etymological dictionaryGoogle Scholar, leaves toraṇa- unconnected.

⁴ Vogel, J. Ph., ‘The Sanskrit pratoli and its new-Indian derivatives’, JRAS, 1906Google Scholar, Here the gloss rathyā viśithā is rejected.

⁵ Kādäg Igory stäryl, 1956, 15.Google Scholar

¹ Abaev, V., Osetinskij jazyk i fol'klor, I, 469Google Scholar: zurun ‘kružit'sja’.

² Iron Literaturä, 10-äm k'lāsän, 1957, 87.Google Scholar

³ Abaev, V., Iz osetinskogo ëposa, 61, 304.Google Scholar

⁴ Abaev, V., Osetinskij jazyk i fol'klor, I, 469, 449.Google Scholar

⁵ TPS, 1953, 36ff.Google Scholar; TPS, 1954, 152Google Scholar; TPS, 1955, 56ff.Google Scholar; BSOAS, XXI, 3, 1958, 525Google Scholar. Ācārya Limaye has called my attention to the variant between Atharvaveda 10.8.6 gúhā járan and the quotation in Muṇḍaka upaniṣad 2.1 guhā caraṃ. A parallelism of śray- ‘mount’ and jar- can be seen in RV 7.72.4 ūrdhváṃ bhānúṃ savitā devo aśred bṛhád agnáyaḥ, samídhā jarante. Note as to zar- that Burrow, T.has called attention to anujīrṇa- translated ‘followed’ in Annals of Oriental Research, University of Madras, XIII, 1957Google Scholar, ‘Nirvacanāni’, § 1. The word uäl-āzār, equivalent to uäl-āxez ‘prevosxod’, has been hardly suitably placed with āzār ‘terror’, which I had carefully avoided, in Abaev, V., Ist.-ëtim. slovar' oset. jazyka, 96.Google Scholar

⁶ Vs. Miller, , ‘Die Sprache der Osseten’, GIP, I, Anhang, 4–5.Google Scholar

⁷ Abaev, V., Osetinskij jazyk i fol'klor, I, 316.Google Scholar

⁸ On kā-, Khot. khā- ‘to open’ (in Siddhasāra 150 v 4) details elsewhere.

⁹ JRAS, 1953, 113.Google Scholar

¹ BSOAS, XXI, 3, 1958, 527.Google Scholar

² E. Leumann in E gave the word doubtfully the meaning of Sansk. tudati.

³ Vs. Miller, , Osetinskie ëtjudy, in, 150Google Scholar, gave tan-, unsuitable in meaning.

¹ bū- with ū as pū- in pūheitä, pūhaṃjākä, vū- in vūmūva. In a basic syllable būṣṣ- from older baxš-.

² The word hamaysā is not explained. By etymology it could mean ‘crushed small’ from *fra-mazaka- (assuming that here śikarä is plural) from maz- (BSOAS, XXI, 3, 1958, 523)Google Scholar like Lit. māžas ‘small’.

³ The word pvaicā-, pvecā- has not yet been explained. A word of similar form is haṃbecā- ‘summary’, older participle haṃbirsta-, haṃbista- ‘combined’. The word yaḍama is Turkish yolma. For the meaning note Avestan vark- ‘pull; wear’, and the similar Baltic velk-.

⁴ This would improve upon the explanation in Bailey, H. W., Zoroastrian problems, 220, 224.Google Scholar

⁵ The attempt to find in änsonun the word san-, sān- ‘rise, raise’ is hardly satisfactory for the meaning, see Abaev, V., Ist.-ëtim, slovar' oset. jazykaGoogle Scholar, s.v.

¹ This word kṣoṇi- needs full treatment elsewhere.

² BSOAS, XXI, 3, 1958, 527.Google Scholar

³ Yule, H., Marco Polo, II, 488Google Scholar; Vs. Miller, , Osetinsko-Russko-Nemeckij slovar', s.v. som.Google Scholar