but one at least survived intact as sufficient proof that the double tablet had never been opened. Plate XXiI showing the double rectangular tablet N. xxiv. viii. 8o before and after its opening at the British Museum will serve as an illustration. I was at first inclined to treat the broken state of one or more of the string folds by the side of others still intact as merely accidental, until its repeated occurrence in documents not showing the least trace of wear and tear suggested the possibility that it had some significance in this ancient official routine. We shall see presently that there is, perhaps, some textual support for this surmise.
Among the remaining documents five, N. xxiv. viii. 73, 75, 86, 87, 90, showed the string folds at the back of the under-tablet no longer in their original position but obviously cut and tied up again in varying fashions. This operation was certainly facilitated by the fact that the string folds passing through each of the three seal socket grooves were always double, and that by cutting these folds on opposite sides in each pair the requisite lengths of string could readily be secured. Considering how ingeniously all technical details of fastening, etc., in this ancient wooden stationery were thought out and arranged,9 it seems possible that the doubling of the string folds may have been partly designed for this very purpose. In addition there remained the end of the string which, as the reproductions in Plates XXIII (N. xxiv. viii. 71) and XXI (N. xxiv. viii. 85) show, was allowed to remain loose and of considerable length after the third double fold had been fixed. This was always available to help in the operation of re-tying a double rectangular tablet after it had been once opened for inspection.
In order to secure again such a document against any subsequent tampering or unauthorized inspection, all that was necessary would be to fix a fresh clay seal on the knot in which the strings were tied at the back. Such a supplementary clay seal is actually found, though in a broken condition, on the reverse of the under-tablet of N. xxiv. viii. 73. This document is of interest also as it has been wrapped up subsequently in yellow silk for the better protection of the seals and then tied round again with coarse string as the reproduction of the covering-tablet in Plate XXIII shows. It is clear, however, that in the absence of a protecting socket such secondary sealing in clay had far less chance of escaping injury and decay. It is noteworthy that in N. xxiv. viii. 90, one of these re-tied documents, the original sealing in the cavity on the covering-tablet was found completely destroyed. Evidently this had been done when the document was first opened. The same was true in the case of N. xxiv. viii. 89 where, however, the fastening was not cut. In this instance it is possible that the seal was first broken to permit of the string being systematically unfolded, and that after the text within had been read the string was fastened round the two tablets again in the regular fashion. It is certainly curious to observe that the five documents, N. xxiv. viii. 86, 87, 88, 89, 90, which were found either re-fastened, or open or with the seal cavity empty, form a regular series which, as the numbering shows, were discovered in close proximity to each other. It almost looks as if the depositor of the little archive had intentionally put down here a separate packet of documents disposed of in one way or another.
Turning to the seals which, as seen, were essential requisites for the authentication of these documents on wood, it is of importance to note that all records found in the deposit, with the exception of the two pieces, N. xxiv. viii. 89, 90, already referred to, still retain their seal impressions in clay. Only very few of the latter (N. xxiv. viii. 75, 85) have suffered any damage, while in a few cases the seals do not appear to have fully taken (N. xxiv. viii. 78, 87, 92). On fifteen documents the seals used are single ; among the rest five, N. xxiv. viii. 73, 79, Si,