their validity might be established in court. Since, as shown in that discussion, the clay seals were the only means used to authenticate this type of ancient Kharosthi document, it was obvious that the seal impressed in the centre of the covering-tablet, as well as the string passing under it and holding the under- and covering-tablets tightly together, had to be preserved intact if the text written on the inner surfaces was to be kept safe from any tampering and to retain its value as legal proof of the transaction recorded.
For the assumption thus arrived at years before, and not yet tested by the results of slowly progressing decipherment, I could scarcely have wished for better archaeological proof than that which the little archive discovered in N. xxiv. viii now furnished. Here was a large series of documents carefully hidden away as deeds, bonds, and similar legal instruments of value would be, when left behind in case of necessity, all of them on double rectangular tablets and the great majority of them still secured under their original sealings. What rapid examination of the writing I could spare time for then or later, seemed to show that the docket-like entries found here on the obverse of almost all the covering-tablets, • above and below the seals, were not of the usual address style I knew from the envelopes of certain rectangular tablets, found on my former visit, which manifestly contained letters.6 Thus the only question still open seemed to be : were these documents, hidden away with their seals and fastenings intact, deeds, or agreements which the official residing in the house had in safe keeping, or did they refer to land and property of his own ?
The answer, I was well aware, could only be supplied by the complete decipherment and elucidation of these strangely recovered records. However, in the meantime it was encouraging to me to note that the very exceptions seemed to support my conjecture in its main outline. When late in the evening after the discovery I examined the two documents, N. xxiv. viii. 77, 88, which alone in the whole series had turned up open with the fastening completely severed, I found that both were letters addressed in due form to the ` great Cojhbo Somjaka', ` whose sight is dear to gods and men '. The name of the same officer had appeared in the addresses of many of the wedge covering-tablets, presumably belonging to letters with brief office orders, which had come to light in such numbers from the floor of the same room.6 I wondered at the time what the contents of these two letters might have been to induce the last owner of the house, whether Cojhbo Somjaka himself or his heir, subordinate, or successor, to keep them along with what I assumed to be deeds, as ` papers ' of value.
Before I proceed to show how strikingly the results of actual decipherment have since confirmed the general conclusion reached on the spot and indicated in my Personal Narrative,' it will be convenient to record here the instructive data which the examination of the outer appearance alone of these documents can furnish. In the first place it is noteworthy that out of a total of twenty-six double rectangular tablets no less than eighteen were found still under their original sealed fastening and certainly unopened. Out of these thirteen, N. xxiv. viii. 71, 72, 74, 76, 78, 79, 81, 82, 83, 85, 91, 94, 95, still retained intact all of the six folds in which the string was passed through the grooves of the seal socket of the covering-tablet, and across the reverse of the under-tablet, in the regular fashion previously explained.$ The reproductions of N. xxiv. viii. 85 in Plate XXI and of N. xxiv. viii. 71 in Plate XxIII fully illustrate this condition. In the other five documents, N. xxiv. viii. 80, 84, 89, 92, 93, one or more of the string folds were found broken,
in lgot at N. xv ; cf. Ancient Kholan, i. p. 346, note 5.
7 See Desert Cathay, i. pp. 282 sqq. ; also Geogr.Journal, July, 1909, pp. 23 sq.
8 Cf. Ancient Kholan, i. p. 352.