^{page 763 note 1} Abbreviations: A, B, C, etc. ═ Columns. C ═ Cover-tablet. NI ═ Kharoṣṭhī Inscriptions discovered by Sir Aurel Stein in Chinese Turkestan, o ═ Obverse, r ═ Reverse. U ═ Under-tablet.

^{page 763 note 2} Rapson, NI, p. 301.

^{page 763 note 3} “ On a single place ” Thomas, AO, xii, 45. ekasti 416 o 5 between cakora and uṭi seems to qualify the latter which is called asti also 180 oB 2; again 272 o a horse is called asti; it seems that also corn is asti 272 o 6, 7, as ghrida ═ Skt. ghṛta 159 r 4; on the other hand, asti occurs in conditional sentences with yati and the verb as “ to be ”. Its meaning seems to be that of Skt. āsteya “ extant” ekasti “ single extant ”, therefore vara uṭi 1. nasti is negative, cf. Thomas, AO, xii, 43, 7.

^{page 763 note 4} On eka b(h)iti- see below.

^{page 763 note 5} Cf. Thomas, AO, xiii, 50, 2.

^{page 764 note 1} A similar termination -iṃ occurs in 117 o E: sarvapiṃdaiṃ taṃ caṃ, gavi 20 3, which has been explained as sarvāṇi pingrdgrāni tāni gāvah 23 by Professor Thomas (JRAS, 1927, 545) and as sarvāh pingrdgritā gävah 23 by Professor Lüders (SPAW, 1933, 1010, 2) who prefers to read: piṃdgraiṃtaṃcaṃ. Generally the NI use only sarva piṃdgra, so that a verb Skt. pingrdgray- is less probable, also on account of the expression eka pimda (248 o 8), sarvap. muli or p. muli (590 Uo 4. 428 o 5), or sarvapimda ganaṃnena muli (345 Uo 5) where it must be a noun. On the other hand, 587 Uo 3 offers ciraiṃta bhuma vikrida which may be *cirayitā bhūmir vikritā (land which has not been tilled a long time ? has been sold). Then -iṃta would point to the part, perf. pass, of denominative verbs.

^{page 764 note 2} Instead of ° au. The pronominal termination -e of the plural is found in khula putre 415 Cr. 3. maṃnuṣe 130 Uo 3. vaṣe 534 oB 2. vṛdhe, 326 r 3.

^{page 764 note 3} Cf. surya(surśa t)dade, 572 Cr.

^{page 764 note 4} Cf. Ep. Ind., xx, p. 18, B 2, line 5, etc., p. 27 s.v. ubhayalokahita°. In NI, 107 o 2 is mentioned in the address the alaṃkrgrtaparaloǵamarǵa-.

^{page 765 note 1} In Saka exists tcahaura “ four ”, Sten Konow, Saka Studies, Vocabulary, p. 184.

^{page 765 note 2} aṭa (358 o 6. 637 o 3) is not aṣṭa (Index s.v.), but, according to Thomas, A(cta) O(rientalia), xiii, 67, Hindī āṭā, Skt. aṭṭa “ flour ”. For aṣṭi cf. Noble, BSOS, vi, 453.

^{page 765 note 3} Cf. Burrow, BSOS, vii, 511 on , see also NI, p. 303.

^{page 766 note 1} Cf. Burrow, BSOS, vii, 784.

^{page 766 note 2} 338 r 1 seems to contain tatrisa prasavetu “ (whether)there will be such one in the or not, at any event the must be sent away (let free) ”. vachu is found again 630 o 2, 3 with jaṃna and the verb prasavita; for him fifteen men are to be given, is he perhaps a man from Vaksṃ;u, the Oxus ? For drgrṣṭa appears triṭha, for drgrṣṭvā in 510 tritva.

^{page 766 note 3} BSOS, vi, 453. dhahi is hardly = tathā, as this word occurs o 2 in the form tatha; like duhitrgr by metathesis of aspiration appears as dhitu, dhahi would correspond perhaps to dadhi; because Aphiñanu takes away the camel and the aghita D.V. the milk, Khvarnase prefers to sell the camel. An objection against such an explanation is the masculine sex of the camel; one would expect uti.

^{page 767 note 1} On kilmeci and raja, cf. Thomas, AO, xiii, 63, resp. 45 f.

^{page 767 note 2} masina- may be connected with Iran, masan- (Bartholomae, Altiran. Wörterb., 1154), the adjective derivated from the noun in masan- by haplology instead of ^{page 763 note *} masanan- ? Bartholomae refers to varmin-, nāmín-, and Whitney, § 1230c. masina-could be explained as a-stem or compared with the many proper names with suffix -ina, cf. Thomas, Festgabe Jacobi, 62 f.

^{page 767 note 3} Cf. Lüders, SPAW, 1933, 1001; Thomas, AO, xiii, 60, 2, and Burrow, BSOS, vii, 510 f., accept the meaning “ wine ”.

^{page 768 note 1} Thomas, NI Index, s.v. śatade, p. 371.

^{page 768 note 2} Thomas, Festgabe Jacobi, 48, suggests it to be Pialma.

^{page 768 note 3} Cf. Thomas, AO, xiii, 56 f.

^{page 769 note 1} trarṃgha, see Burrow, BSOS, vii, 509 f.

^{page 769 note 2} Literally “to-morrow”, Skt. prātah,: pratu (═ prata 361, 499), as punah: punu (puna), or puratah: puradu; -u represents also Skt. -am in likhidu, BS0S, vi, 455; cf. dhitu, pitu, matu.

^{page 769 note 3} āchid means “ take away ” as to be seen clearly from 719 o 2; but here, where in line U o 3 chiṃnaṃti is used, perhaps the meaning “ not care ” is appropriate. acchingrdati in Pāli has not this meaning, but see P. W., s.v. 5.

^{page 769 note 4} tanu, tanuvaǵa, tanuùaǵa being adjectives, see Noble, BSOS, vi, 450 f.

^{page 769 note 5} Though it is uncertain to conclude something from the order of words about the relative rank of the karsenaùa and śatavita (contrast 482 Uo 2, 4 with 590 Uo 8), in the year 17 of king Jiṭugha Aṃguvaka, the mentioned RiciḰǵa is a karsenaùa, and in the year 24 of the same king a śadavita (590 Uo 8. 715 o 4–5), if the two name sakes are identical. Is karsenaùa ═ Skt. karṣangrapa ? But the stem is kriṣi-, though 511 o 5 shows malāprakarsena against malaprakarṣana in line 4. The control of tilling and the kilme-institution reminds the reader of the rules of bestowing land to tax-payers in the first chapter of the second book in Kauṭ. Arthaś (Janapadaniveśa).

^{page 770 note 1} Also a cojhbo is called jeṣṭha 120 r 1.

^{page 770 note 2} One would expect at least the śatavita; the rank and file may be ascertained from 470, where the great king gives his orders to the cojhbo and tomga, then it is passed on to the yapǵu. The daśavita has to do with collection and registration of cattle and corn, etc. (cf. 148).

^{page 771 note 1} That reading according to the Corrections, p. 289.

^{page 771 note 2} vita occurs many times with aśpa, go, paśu; does it mean “ breeding ” in contrast to the vyala uṭa in 341, 420 ?

^{page 772 note 1} Cf. Archiv. Orient., vi, 32 f.

^{page 772 note 2} Cf. Lūders's List, No. 1200.

^{page 772 note 3} Ep. Ind., xviii, p. 156, line 52 f.

^{page 772 note 4} Ep. Ind., xviii, p. 192, line 44; p. 193, line 47.

^{page 772 note 5} Ep. Ind., xviii., p. 216, line 41.

^{page 772 note 6} The reference to Sir Aurel Stein's Serindia, i, 65 (in the Index of the NI s.v. satade, p. 371) for the sense “ measure of land ” is not correct; the passage runs: “ Thus Colonel Trotter, who visited Wakhān with a section of the Yārkand Mission in 1874, distinctly notes that ' Wakhān formerly contained three ” Sads “ or hundreds, i.e. districts, containing a hundred houses each'.” This corresponds well to the śata of the NI.

^{page 772 note 7} Kishori Mohan Gupta, The Land System in South India, Lahore, 1933, 268 f., 275 ff.

^{page 772 note 8} As Mr. Gupta does, p. 271.

^{page 772 note 9} Ep. Ind., iv, p. 250, line 47. Bhandarkar's List, ibid., xx, App., p. 223, No. 1610. Date about A.D. 810, cf. Smith, Early History, 4th ed., p. 413 f.

^{page 773 note 1} AO, xiii, 53.

^{page 773 note 2} JRAS, 1933, p. 550; 1934, p. 97; on taxes and other officials concerned with administration and tax-surveying, cf. ibid., 1934, pp. 99 f., 104 f., 272 ff.

^{page 773 note 3} AO, xiii, 49.

^{page 774 note 1} For the tommi, cf. Thomas, AO, xiii, 53.

^{page 774 note 2} Cf. Thomas, AO, xii, 63 ff.

^{page 774 note 3} Cf. Thomas, AO, xiii, 60.

^{page 774 note 4} Professor Lüders (SPAW, 1933, 999) believes prathanyana (-nā) in 565 r 1 to be a remarkable clerical slip for prathama; should it not correspond rather to Skt. prādhānyānām ?

^{page 774 note 5} Cf. Pischel, Pkt. Gr., § 300.

^{page 775 note 1} Burrow, BS0S, vii, 790, who suggests “one another ”.

^{page 775 note 2} Though the “ fourth ” entry really comes in the next line B 3, the first and second stand in A 4 and 5. The same occurs in 105 oE 1: amna triti vara Kaṃcakasa muli 4, because the same man is mentioned D 2 and 4.

^{page 775 note 3} For this reading, see Corrections, p. 290, and for the meaning of saste, cf. Sten Konow, CII, ii, 1, p. 152, and Burrow, BSOS, vii, 515, 783.

^{page 776 note 1} Comparing (M. A. Stein, Ancient Khotan, ii, plate civ) the form of sa in line 1(isa) and of ya in line 2 (khaniyaṃmi) there is no doubt that the reading adopted in the text is correct.

^{page 776 note 2} Cf. CII, ii, 1, p. 22; sodaśe; pamcadaśe, (pp. 65, 79). But it is doubtful whether panicami, p. 119, is a pronominal termination, or influenced by such one, as Sten Konow supposes (pp. cxiii and 119); pamcame is regular, ibid., p. 28, line 1. No doubt, local peculiarities of the dialect must be taken into consideration, as in NI, 661, for the change of see Noble, BSOS, vi, 453, and for the use of cases, p. 451 f.

^{page 777 note 1} Thomas, AO, xii, 44 f., n. 7; Burrow, BSOS, vii, 512. anada may be an equivalent for ājnāpitam, formally it is ājnāpitam, cf. anatena; but it could be explained also as ājnā-tah, cf. ana 39 Uo 4. 492 Uo 3.

^{page 777 note 2} See Burrow, BSOS, vii, 787. ko pi varaga is, according to Burrow, l.c., ko pivaraǵa “ fat ”, cf. Thomas, AO, xiii, 61. In 667 r 5 varaǵa is not clear, and seems to have the same sense as varayaṃ in 206 Cr 7, where it may correspond to Skt. varāya, as the next sentence ma mahi toṣa kariṣyatu makes likely.

^{page 777 note 3} See Burrow, l.c.; Thomas, AO, xiii, 60.

^{page 777 note 4} Should it be guma(ṃ)ca ═ kumbhāh ?

^{page 777 note 5} It would correspond to Skt. ^* puksama- which seems to be unknown. In modern East Turkestan exists the word bodžuma “ knot, bundle ”, according to Radloff,Versuch eines Wörterbuchs der Türk-Dialekte, qazag . “ 1, Kleiner Teppich, 2, Gepäckstück aus Wolle,” čagat. “ Packtuch ”, s. Menges, SPAW, 1933, 1272, s.v. Perhaps connected with that is (pochu)“ long nightgown-like garment worn by Kāshmīrīs, when made of cotton cloth is called ”, Sir ǵ A. Grierson, Dictionary of the Kāshmīrī Language, p. 808.

^{page 778 note 1} krataǵa may be Skt. kṛtayah, a weapon like knife or dagger.

^{page 778 note 2} kojava ═ Skt. kauśeya, Thomas, AO, xii, 54.

^{page 778 note 3} tavastaǵa, Burrow, BSOS, vii, 512.

^{page 778 note 4} M. A. Stein, Ancient Khotan, 367; Burrow, l.c., 510.

^{page 778 note 5} raju.

^{page 778 note 6} kansiya, in note 4, -ṅki-.

^{page 778 note 7} Thomas, AO, xiii, 78; for potrī, cf. J. J. Meyer, Arthaś. trs., p. 656, note 1 (Arthaś., xiv, 3, 178).

^{page 778 note 8} pota according to Halayudha Abhidh., ii, 393 (not pota, as PW, s.v. 3, says, only Medini has that form and Aufrecht's Glossary, p. 276, in his ed. of Halāy.) is “ cloth ”; varaka again is explained (PW, s.v. 1) as potācchādana “ cloak, stuff, tent ”.

^{page 778 note 9} Cf. Thomas, AO, xii, 61, 5, but here it looks like a place-name as in 272 o 6, 7, the locative of which is regular Kuùvaute;baniyaṃmi against the locative of the term khuùaneṃci (489 r 2).

^{page 778 note 10} For the use of the cardinal with vara instead the ordinal see above; that the enumeration of items in a (royal ?) document was used can be seen from the fragmentary text 376 where is found biti, triti, cathurtha harya (o 1, 2, 4).

^{page 779 note 1} So already Thomas, AO, xiii, 78. In 83 Cr 6 biji pata it is “ part ” (bījī pādah) as probably in 152 Cr 4; 666 o 2–3 and 669 o 2 has been restored in the Index, p. 368, s.v. vaṃtati (-de) to: śirsā pada (= Skt. pādayoḥ) v.