^{page 795 note 1} BSOS. v, p. 137 ff.

^{page 795 note 2} Garbe-Festgabe, p. 24 ff.

^{page 795 note 3} L'Indo-aryen du Veda aux temps modernes, p. 53.

^{page 795 note 4} Die Zigeuner in Europa und Asien, i, p. 459.

^{page 795 note 5} A Comparative Dictionary of the Bihārī Language, p. 93 ff., where appear most of the references to discussions up to 1885.

^{page 795 note 6} Das Saptacatakam des Hāla, p. 556.

^{page 795 note 7} Beitrāge zur Grammatik des Jainaprâkṛit, p. 36.

^{page 795 note 8} Das Aupapātika Sūtra, p. 93.

^{page 795 note 9} Beiträge zur Pali-Grammatik, p. 97.

^{page 795 note 10} Beiträge zur vergleichenden Sprachforschung herausgegeben von A. Kuhn, viii, p. 144.

^{page 795 note 11} Grammatik der Prākrit-Sprachen, § 480, where previous references are recorded.

^{page 795 note 12} Garbe-Festgabe, p. 24 ff.

^{page 795 note 13} iv, 215.

^{page 795 note 14} Saddanīti, ii, Moggāllana-Vyākarana, v, 173, quoted in Critical Pali Dictionary, s.v. acchati.

^{page 795 note 15} A Dictionary of the PǢli Language, s.v.

^{page 795 note 16} Hemacandra's Grammatik der Prâkritsprachen, p. 155.

^{page 795 note 17} Specimen der Nāyādhammakahā, p. 45.

^{page 796 note 1} Comparative Grammar of the Gaudian Languages, p. 366; or to as-.

^{page 796 note 2} Göttingische gelehrte Anzeigen, 1875, p. 627 f., and Beiträge zur vergleichenden Spraehforschung, viii, p. 144.

^{page 796 note 3} Pali Literatur und Sprache, § 135.

^{page 796 note 4} Studj Critici, p. 352, note 49.

^{page 796 note 5} Grit. Pali Diet., s.v. acchati.

^{page 796 note 6} Pali Miscellany, p. 61.

^{page 796 note 7} xii, 19.

^{page 796 note 8} iv, 10.

^{page 796 note 9} Quoted by Grierson, Memeirs of the As. Soc. Bengal, viii, No. 2, p. 88.

^{page 796 note 10} Quoted by Pischel, Gr. PH. Spr., § 480, note 6.

^{page 796 note 11} MSL. xviii, p. 28.

^{page 796 note 12} La formation de la langue marathe, p. 289.

^{page 796 note 13} Crit. Pali Diet., s.v.

^{page 796 note 14} Les Inscriptions de Piyadasi, p. 138.

^{page 796 note 15} IF. iii, 210.

^{page 796 note 16} See especially the evidence collected by Grierson and Hoernle in Comp. Diet. Bihārī, p. 93.

^{page 796 note 17} See especially J. Sampson, Dialect of the Gypsies of Wales, pt. iv, pp. 1–2.

^{page 796 note 18} Śyām Sundar Dās, Hindī-Śabdasāgar, s.w. achnā, āchnā.

^{page 796 note 19} The meanings indriyapralaya and mūrtibhāva given by the Dhātup. for and referred to by Pischel Gr. Pkt. Spr., § 480, are presumably due to the use of with abstract nouns in the accusative, e.g. yuddharaṅgatām (Nalod. 2, 10) ‘ to become the battleground of ’. It is true, as Grierson in Garbe-Festgabe, p. 24, points out, that ‘ to go ’ may thus develop into ‘ to become ’, as in gácchati > Ksh. gaćhun‘ to become ’. But acchati means ‘ abides ’, not ‘ becomes ’.

^{page 798 note 1} Pali Text Society's Pali-English Dictionary, s.v. acchati.

^{page 798 note 2} Pott's ā-sthā- is impossible on phonetic grounds, and was discarded by Grierson and Hoernle.

^{page 798 note 3} E.g. Shina distinguishes Skt. (c)ch and ts as ch and ts(h): .

^{page 798 note 4} BSOS. v, p. 138. For ts(y) we have .

^{page 798 note 5} In the Bihārī (except Maithilī), East and West Hindī areas we have no evidence for the existence of as the substantive or auxiliary verb. It is not so found to-day. In Old Hindi was a verb of fuller meaning‘ remain, be found, exist'. Śyam Sundar Dās in the Hindī Śabdakoś, s.vv. achnā, āchnā, gives references to Jaisī, Kabīr, and Bihāri: it thus belonged to the vocabulary of literary Awadhī and Braj; and may have entered from the East Rājasthānī dialects, where to-day it provides the substantive and auxiliary verb. That in Old Awadhī was (see, e.g. Grierson and Hoernle's Index to the Rāmāyan of Tulsī Dās, pp. 23, 31).

^{page 798 note 6} Varnaratnâkara of Jyotirīśvara-kaviśekharācārya (reprinted from the Proceedings of the Fourth Oriental Congress, vol. ii), p. 69.

^{page 799 note 1} Grierson, Linguistic Survey of India, v, 2, p. 51, and Seven Grammars of the Bihārī Language, p. 41, gives only the negative na(h)ikh-.

^{page 799 note 2} LSI.v, 2, p. 280.

^{page 799 note 3} LSI. v, 2, p. 305.

^{page 799 note 4} LSI. v, 1, p. 324.

^{page 799 note 5} Ib., p. 327–49.

^{page 799 note 6} Comparative Grammar of the Modern Aryan Languages of India, iii, p. 183 (cf.i, p. 192 f.).

^{page 799 note 7} IF. iii, p. 209.

^{page 799 note 8} Les Inscriptions de Piyadasi, p. 138.

^{page 799 note 9} Aśoka Text and Glossary, p. 54.

^{page 800 note 1} IF. iii, p. 210.

^{page 800 note 2} CII., vol. i (new ed.), Inscriptions of Aśoka, p. 55, note 5.

^{page 800 note 3} JA. 1911, p. 422 f.

^{page 800 note 4} Kharoṣṭhī Inscriptions discovered by Sir Aurel Stein in Chinese Turkestan, p. 302.

^{page 800 note 5} Deutsche Literatur-Zeitung, 1924, p. 1902; CII. vol. ii, pt. i; Kharoṣṭhī Inscriptions, p. ex.

^{page 800 note 6} See Comparative and Etymological Dictionary of the Nepali Language, s.v. bagnu.

^{page 800 note 7} E. J. Rapson and P. S. Noble, Khar. Inscr. Turkestan, pt. iii, Index, s.vv. achati, achati, achatu. Of these the first, in No. 506, read by the editors achati (not actati) is rather, according to a communication from Mr. T. Burrow, to be read as aja vi, which the editors give as an alternative.

^{page 801 note 1} Rapson and Noble, Khar. Inscr. Turkestan, pt. iii, Index, s.w. hachati, hachatu, hachyati. T. Burrow, JRAS. 1935, p. 669, considers ach- to be derived from hach with the loss of initial h- seen occasionally elsewhere in these documents, e.g. astaṁmi ═ ha°, uhati ═ huati. But there seems no reason to doubt that ach- may be the original form.

^{page 801 note 2} See above, p. 799.

^{page 801 note 3} Cf. the invasion of the 2 sg. optative by the imperative ending -su, Pischel, Gr. Pkt. Spr., § 461. But according to Burrow in his thesis A Grammar of the Language of the Kharosthi Inscriptions (deposited in the Cambridge University Library) these, with other forms in -tu, are 2 sg. .

^{page 801 note 4} F. W. Thomas, Acta Orientalia, xiii, pp. 61–2, translates two examples in No. 165 as futures: yo puna tahi karyani ‘ whatever requirements of yours shall come '; ‘ whatever occurrences of good and bad there shall be '.

^{page 801 note 5} R. Pischel, Gr. Pkt. Spr., § 523: gacchaṁ, gacchimi, etc., though Pischel's proposed *gaksyāmi has no foundation.

^{page 801 note 6} BSOS. vi, p. 535; cf. Aśoka Kālsī, etc., ka(c)chati replacing ambiguous *kassati.

^{page 801 note 7} Gr. Pkt. Spr., § 521.

^{page 802 note 1} G. A. Grierson, Seven Gr. Bihārī, pt. ii, p. 51.

^{page 802 note 2} Ib., pt. iii, p. 31.

^{page 802 note 3} Ib., pt. iii, p. 31.

^{page 802 note 4} R. A. S. Macalister, The language of the Nawar, p. 36. J. Bloch, who was the first to recognize the existence of acch- in this dialect, Journ. Gypsy Lore Soc., 3rd ser., xi, p. 32, explains this paradigm a little differently as a compound tense containing " le radical (ou l'absolutif ?) du verbe ho-, suivi d'un verbe conjugué signifiant lui-même ‘ etre '."

^{page 802 note 5} ā-ksi- according to Grassmann occurs six times in RV.; BR. give four references to AV. and none to any subsequent text, āksít- once in RV., ánākṣit- in ŚBr. An *ākṣaya- m. ‘ resting-place' possibly survives in Sindhī ākhero m. ‘ bird's nest'. I have found no other surviving verbal form of ksi- in the modern languages, kṣéma- (Pa.Pkt. khema-, Khar. Doc. .‘ welfare', Guj. khem n. (?) ‘ wellbeing ', Mar. khev m. ‘ evil accident', Sgh. semin, hemin ‘ slowly, softly ', kema ‘ magic to avert mischief'; yogakṣemá-, see Nep. Diet. s.v. jokhim) and especially kṣétra- (see Nep. Diet. s.v. khet) have had a considerable fortune.

^{page 802 note 6} Pischel, Gr. Pkt. Spr., § 553.

^{page 803 note 1} Bloch, L'Indo-aryen, p. 243.

^{page 803 note 2} Hultzsch, CII. vol. i, Inscr. Aśoka, p. lxxxi.

^{page 803 note 3} Ib., p. xcv.

^{page 803 note 4} See below, p. 810.

^{page 803 note 5} LSI. viii, 1, p. 59.

^{page 803 note 6} LSI. viii, 1, pp. 440, 441.

^{page 804 note 1} See J. Bloch, Langue marathe, § 104.

^{page 804 note 2} Sten Konow, CII. vol. ii, pt. i, p. cx.

^{page 804 note 3} Pisehel, Gr. Pkt. Spr. §§ 317–320.

^{page 804 note 4} -tr-, preceded by a long vowel, > -t- which subsequently was lost like original Skt. -t- (see W. Geiger, A Dictionary of the Singhalese Language, p. xix); the group .

^{page 804 note 5} W. Geiger, op. cit., p. xxi, and Literatur und Sprache der Singhalesen, p. 42,

^{page 804 note 6} JRAS. 1921, p. 539.

^{page 804 note 7} He instances, op. cit., p. 113. Guj. chũdvũ [but also khũdvũ] ‘ to pound ' ═ Mar. sũdn, but Ass. khundiba ‘ to pound', Hi. khūdna ‘ to trample ' (Skt. kṣuṇatti, Pkt. chuṁdaī, khuṁdaī); Guj. vichalvũ ‘ to rinse' ═ Mar. (cf. Skt. vikṣālita-). Guj. taras ‘ hyena ' ═ Mar. taras (Skt. tarakṣa-, Pk. taraccha-) and ūs ‘ sugarcane ' ═ Mar. us (Skt. ikṣu-, Pkt. ucchu-) are loans from either Mar. or North Guj. where ch > s (JRAS. 1921, p. 540). I find also Guj. cho ‘ plaster, mortar', chovũ ‘ to plaster', but Beng. kho ‘ broken brick', Hi. khoā ‘ broken brick, mortar' (Skt. kṣoda-, kṣódati, kṣodayati; Pkt. khoa- ‘ powder ' ); Guj. lacho ‘ fomenting or burning feet with a hot iron ' ═ Mar. lāds ‘ mark made by cautery ', n. spot, discoloration ‘ (Skt. lakṣá-, Nep. Diet., s.v. lākh).

^{page 805 note 1} Cf. T. Michelson, JAOS. 1910, p. 88, quoted by Bloch, loc. cit.

^{page 805 note 2} CII. vol. i, Inscr. Aśoka, p. 26, note 6.

^{page 805 note 3} See above, p. 799.

^{page 805 note 4} āchi ' is ', āchfti f. ' was ' beside chũ ' am ', LSI. ix, 3, p. 89.

^{page 805 note 5} LSI. ix, 3, p. 89.

^{page 805 note 6} J. Bloch, L'Indo-aryen, p. 81.

^{page 806 note 1} Gr. Pkt. Spr., § 480.

^{page 806 note 2} ch forms have not completely driven out the older forms in Bengali as in the negative substantive verb, see below, p. 810.

^{page 806 note 3} M. Shahidullah, Les Chants mystiques de Kānha et de Saraha, p. 116 (No. 11). From the other Caryās S. K. Chatterji, Origin and Development of the Begali Language, p. 931, quotes 1 sg. acchahu, acchami, 2 sg. acchasi.

^{page 806 note 4} S. K. Chatterji, Bengali Language, p. 118.

^{page 806 note 5} Shahidullah, op. cit., pp. 99, 203.

^{page 806 note 6} op. dt., p. 28.

^{page 806 note 7} op. cit., p. 119.

^{page 806 note 8} Indian Linguistics, v, p. 352.

^{page 807 note 1} S. K. Chatterji, Varṇaratnakara, p. 1.

^{page 807 note 2} G. A. Grierson, Introdtiction to the Maithilī Language, pt. ii, p. 34.

^{page 807 note 3} See references in A. F. R. Hoernle and G. A. Grierson, Comp. Diet. Bihārī, s.v. achh-.

^{page 807 note 4} LSI., v. 2, p. 27.

^{page 807 note 5} G. A. Grierson, Seven Gr. Bihārī, pt. ii, p. 44.

^{page 807 note 6} See below, p. 810.

^{page 807 note 7} See below, p. 809.

^{page 807 note 8} LSI. ix, 3, pp. 1–-201.

^{page 807 note 9} LSI. ix, 3, p. 35. In the specimen on p. 36 occur 1 sg. 3 sg. tvdee hai, Tcarayo he, kamyo che.

^{page 807 note 10} LSI. ix, 3, p. 85.

^{page 808 note 1} LSI. ix, 2, pp. 315–16.

^{page 808 note 2} JRAS. 1927, p. 232 ff.

^{page 809 note 1} In view of the phonetic weakness of the verb ' to say ' in several languages this derivation is far more probable than that it is .

^{page 809 note 2} Beames, Comp. Gr., ii, p. 257; and S. K. Chatterji, Bengali Lang., p. 761.

^{page 809 note 3} T. N. Dave, A study of the Gujarātī Language in the Sixteenth Century V.S., p. 161.

^{page 809 note 4} Altindische Grammatik, § 108.

^{page 809 note 5} Langue marathe, p. 289.

^{page 809 note 6} A Dictionary, Maráṭhí and English, B.V.

^{page 809 note 7} The root is more properly ah-. The nasalization, when it appears, depends upon the nasalization of a terminational syllable: thus āhe: .

^{page 809 note 8} Principles of Murathee Grammar, pp. 113, 114.

^{page 809 note 9} J. Bloch, op. cit., p. 35.

^{page 810 note 1} G. A. Grierson, Seven Gr. Bihari, pt. iii, p. 32.

^{page 810 note 2} LSI. v, 2, p. 148.

^{page 810 note 3} LSI. v, 2, p. 280.

^{page 810 note 4} J. D. Anderson, A Manual of the Bengali Language, p. 15.

^{page 810 note 5} LSI. vi, p. 22.

^{page 810 note 6} LSI. ix, 1, p. 93.

^{page 810 note 7} LSI. ix, 1, p. 402.

^{page 810 note 8} LSI. ix, 1, p. 483.

^{page 810 note 9} I was clearly wrong in Nep. Diet. s.v. hunu in connecting these forms with Hi. hona, Nep. hunu, etc., which are < Skt. bh´vati, though the two verbs have exercised mutual influence on each other. There is much chance of confusion when the two words differ only in their vowels: .

^{page 811 note 1} Mandeālī, LSI. ix, 4, p. 724; Mandeālī Pahārī, p. 746; Cameālī, p. 780; Gādī p. 799; Curāhī, p. 825; Paṅgwālī, p. 851; Bhadrawāhī and Bhalesī, p. 893; Padarf, p. 906.

^{page 811 note 2} Bloch (Ind. Ling., ii, p. 32) showing that the s of Kashmiri 1 sg. , supports Grierson's assumption (Garbe-Festgabe, p. 30) that this tense with its variation for gender is derived from a past participle, Pkt. acchia- . But with the exception of this and perhaps the Hunza Dom verb referred to below (note 4), the forms of the present tense everywhere seem to be derived from the present tense of Sanskrit (BSOS. v, p. 138). Where, as in Maithilī, there is differentiation of gender, it has been introduced secondarily through the influence of the participial tenses. In the Nepalī paradigm

^{page 812 note 1} BSL. xxxiii, pp. 55 ff.