Sec. i] RUIN N. i AND FIRST FINDS OF INSCRIBED TABLETS 325
finish, but were none the less puzzling at the time of their discovery. One series consisted of tablets of rectangular shape varying in length from 6 to i 6 inches, which at the narrower sides of their single inscribed surface invariably showed a raised rim about 2 to i inch broad, exactly resembling a margin, as seen in the reproduction of N. iv. 1J9 (Plate XCVI)13. The writing extending between these rims in neatly written lines, always parallel to the longer side, often showed at the commencement a Kharosthi numerical figure preceded by a word which I very soon made out to be the Sanskrit or Prâkrit sanavatsare ' in the year '. Immediately after came the words mahanuava maharaya ' His Excellency the Maharaja ', as found already in the initial formula of the wedge-shaped tablets, but followed here by what I concluded to be the name of the ruling king, always showing the genitive ending sa. Next there appeared with equal regularity numerical figures preceded by the words mase and divase, ' in the ... month' and ' on the day '. There could be no doubt that these particular tablets contained documents fully dated. But there was nothing in this to explain the peculiar form of the tablets or their special use.
Busily occupied as I was in directing the excavation, and in removing and numbering each of the rapidly succeeding finds, I failed to realize at the time the close connexion between the tablets just described and another class, of which N. iv. furnished even more numerous specimens. They consisted of neatly finished rectangular pieces of wood, varying in length from 34 to 8â in. and in width from 28 to 42 in., with the relatively great thickness of about â to â in. The flat reverse rarely contained any writing, while the obverse in its raised centre invariably showed a square or oblong seal-socket provided with string-grooves, and along its side edges one to three transversely written lines of Kharosthi 14. It was only after the remarkable rubbish-heap of N. xv., described below, had yielded up its treasures that an explanation, as definite as it was simple, revealed itself of these curious seal-bearing tablets and of the rims of the rectangular documents to which they had once been fitted as envelopes 15.
The remains of two small rooms adjoining N. iv. on the south, to which I turned after the latter had been completely cleared, proved to be eroded almost to the floor and furnished no finds. Their condition forcibly demonstrated what would have become of N. iv. if the plaster platforms had not offered some resistance to erosion, and of its relics if my exploration had not been effected in time. The timber-débris covering the eroded slopes north and eastwards plainly indicated that all the extant ruins I had cleared were but a portion of a large dwelling-house ; but apart from the writings found in it there was nothing to throw light on its last occupants. So much was certain, that all these multifarious records on wood could only be the remnants of a collection that had gradually accumulated and been left behind by chance when the place was abandoned. The inference I had from the first been inclined to draw, from the abundance of the documents and peculiarities already noticed in their appearance, as to the ruin having served as the residence of some local official, could in the absence of documentary evidence be little more than a guess. But fortunately the difficult task of decipher-
13 Rectangular ' under-tablets,' to designate them by their subsequently determined character, are N. iv. 13, 16, 18, 3o, 55, 58, 82, 84, 106, 115, 127, 139, 143, 144. The largest in size is N. iv. 58, measuring 16i-x z8 in. The number of lines varies from five (in N. iv. 82) to thirteen (in N. iv. 144)•
" Rectangular covering-tablets or fragments of such are N. iv. 7, 13, 14, 17. a, 17. b, 21, 26, 28, 32, 34, 48, 83, 114, 116, 122, 128, 130, 131, 138, 140, 141, 145. It is thus seen that the number of envelopes' of rectangular documents
found in N. iv. considerably exceeds that of ' under-tablets' of the corresponding type. N. iv. 13 probably retains also a portion of its original under-tablet, but the wood has so completely perished that the character of the two pieces could not be recognized at the time of discovery.
15 See below, pp. 351 sqq. Plates XCIV, XCV, XCVII, CIV illustrate the arrangement of covering- and under-tablets in rectangular documents. For a curiously decayed specimen of the former, N. iv. 2 8, see Plate CV, and above, p. 322.
Character of dwelling N.1.