Fig. 17 is peculiar for its multitude of symbols, most of them known to us by their occurrence on other coins. This specimen is also of the Ventura collection. [See, further, Ariana Antiqua,' xv., 28 to 31.]

Fig. 18, a coin in Mr. Tregear's possession. Several of the same kind have been before introduced into my plates, but hitherto the figure under the elephant has been supposed to be a prostrate elephant vanquished by the upper animal. The multiplication of specimens has at length shown us the true character of the doubtful part, and that it merely consists of two of the common symbols of the series.

But we must now turn to plate xx., in which, thanks to Col. Stacy, I have been able to attempt a more methodical classification from his abundant supply of this Buddhist series of coins.

STACY'S EARLIEST HINDÚ COINS.

(plate xx.)

It is an indisputable axiom, that unstamped fragments of silver and gold of a fixed weight must have preceded the use of regular coin in those countries where civilization and commerce had induced the necessity of some convenient representative of value. The antiquarian, therefore, will have little hesitation in ascribing the highest grade of antiquity in Indian numismatology to those small flattened bits of silver or other metal which are occasionally discovered all over the country, much of change of alphabetical symbols might be incident upon the use of a more perfect language, as compared with the requirements of the local Páli. Another point of important bearing on the main question, is the probable modification the written or cursive literal signs were subjected to as opposed to the stiff and formal outlines of the characters of the rock inscriptions. Up to this time, it has been usual to consider the old Páli writing as the basis of all Sanskrit alphabets: we need not contest this inference, but we may fairly inquire if we have reached the date of the first use of that character in the epoch assigned to the early Buddhist inscriptions? The reply would reasonably be in the negative?' [Prinsep himself, it will be seen, accepted a margin of about two centuries of anterior antiquity, and the estimate might be almost indefinitely extended]. This system of writing, in its sufficiency for all purposes of its own linguistic expression, may well have continued for a lengthened period unchanged, as far as inscriptions were concerned, at the same time that there may have been a progressive advance in the cursive hand, of which we have no immediate record.”—Jour. As. Soc. Beng.', vii. of 1854.]

either quite smooth, or bearing only a few punch-marks on one or both sides; and generally having a corner cut off, as may be conjectured, for the adjustment of their weight. Many instances of this type have been given in Col. Mackenzie's collection (figs. 101 to 108 of Wilson's plates), who describes them as 'of an irregular form, bearing no inscription, occasionally quite plain, and in any case having only a few indistinct and unintelligible symbols: that of the Sun, or a star is most common; and those of the lingam (?), the crescent, and figures of animals may be traced.' The Colonel's specimens were chiefly procured in South India: others have been dug up in the Sunderbans: and many were found at Behat (fig. 14).

But the few selected specimens in Col. Stacy's collection, figs. 25-29) yield more food for speculation than the nearly smooth pieces above alluded to. On all these we perceive the symbol of the Sun to be the faintest of those present: in two instances (figs. 28, 29) it is superposed by symbols which may be hence concluded to be more recent. These are, severally, the Chaitya, the tree, the Swastika, and the human figure; besides which, in fig. 26, we have the elephant, the bull, and the peculiar symbols of figs. 34-37. They are all stamped on at random with punches, and may naturally be interpreted as the insignia of successive dynasties authenticating their currency.

In one only, fig. 30, does there appcar any approach

1 Their average weight is fifty grains, or the same as the ¿ tank (= 3 máshas) of the ancient Hindú metrology. Indeed the word, tank-sala, mint,' goes far to prove that these are the very pieces fabricated for circulation under that name.