364 EXCAVATION OF KHAROSHTHI TABLETS [CHAP. XXIII.
the fashion of the modern " treasure-seekers." The ancient records left behind by the last occupiers as so much " waste-paper " (to use an anachronism) were not likely to have been treated by him with more respect than Ibrahim had shown for the collection of tablets he so luckily unearthed in another part of the building.
Below the matting I discovered some more tablets which owed their excellent state of preservation evidently to this safe covering, and then I came upon a small oval-shaped platform of plaster which, judging from the raised rim enclosing it, must have served as an open fireplace. But more interesting and more puzzling too than these structural details were the epigraphical finds. Their variety in respect of shape and size was truly remarkable. The wedge-shaped tablets familiar from the first day's work reappeared again ; but in numbers they were far surpassed by inscribed boards of wood, to which, notwithstanding great variations in detail and proportions, the designation of oblong appears generally applicable. The use . of those which were provided with a handle, usually rounded or pentagonal, and exhibited Kharoshthi writing on both sides, could readily be accounted for by their resemblance to the Indian Takhta (for a specimen, see p. 359) .
Other tablets attaining considerable dimensions in length, up to 30 in., but comparatively narrow, curiously reminded me, by their appearance and the hole regularly found near one end, of the palm leaves which we know to have been used from the earliest times for manuscripts in India. Useful as this hole must have been for handling and storing these tablets of wood, it was nevertheless evident that it could not have been intended, as in the case of palm- leaf manuscripts, for a string to unite a series of them into a sort of
Pothi.' For not only were such tablets of entirely different sizes, but their great majority (close on thirty pieces) showed plainly by the irregular arrangement of their writing, in small columns and often running in different directions and concluding with numerical figures, by the appearance of various handwritings, erasures, bracketings, and similar indications, that they did not contain texts, or even connected communications, but in all probability
IIIF Color HighRes
