The inscription, No. X, which is incised on the piece of crystal, shares only two of the peculiarities, just described. Its cha (1. 2) has a tail and its da in °desdnaih (1. 3) and in danam (1. 6) opens towards the left. In other respects its letters, as well as its vowel notation, fully agree with those of Asoka's Edicts. It may he noted that it twice offers
in Samanudesanam (1. 2) with the palatal sibilant exactly in the places where it would stand in Sanskrit. This agreement of its characters with those of the Maurya inscriptions leads to the supposition that it belongs to the same time as the latter.
But the first nine inscriptions are also probably only a few decades later than Asoka's Edicts. They unfortunately contain no historical statements which might be used to absolutely prove the correctness of this estimate. They mention, it is true, the names of a king Kubirako or Khubirako, i.e., Kuberaka, of various families, of gothis or committees and Buddhist saints. But none of them is traceable in any other historical source. Under these circumstances, all that remains is to fall back on arguments deduced from a comparison of other datable inscriptions, which, of course, may be deceptive. If one does this and places on the one side the alphabet of the As'oka Edicts and on the other those of the Nanaghat, Hathigumpha and Bharahut-Torana inscriptions, which belong to about the middle of the 2nd century B.C., one can only come to the conclusion that the Bhattiprolu inscriptions hold an intermediate position between the two sets, but are much more closely allied to the first than to the second. On this evidence, which, I repeat, may mislead, they cannot be placed later than 200 B.C., but may be somewhat earlier.
If this estimate is correct, their characters prove (what, indeed, is also made probable by facts connected with As'oka's Edicts) that during the 3rd century B.C. several well-marked varieties of the Southern Maurya alphabet existed; for they contain a system which cannot have sprung up in a short time, but must have had a longer history. The importance of this result lies therein, that it removes one of the arguments of those scholars who believe the introduction of writing to have happened during the rule of the Maurya dynasty. It has been stated repeatedly that one of the facts proving the Asoka Edicts to belong to the first attempts of the Hindus in the art of writing, is the absence of local varieties among the letters of versions, incised at places which lie at distances from each other of more than a thousand miles. This argument is based, as I have pointed out more than once, on imperfect observation, and it may be met also by the obvious objection, that Asoka's Edicts were all issued from the same office, and that the importance naturally attributed to the writing of the royal clerks at Pataliputra might be expected to influence the copyists in the provinces and to induce them to imitate as much as possible the shape of the letters used at head-quarters. Nevertheless, if the Bhattiprolu inscriptions now show a system of writing, which in some respects is radically different and which may be reasonably supposed to have arisen in Asoka's times or even earlier, they furnish a very great help to those who, like myself, believe the art of writing to have been practised in India for many centuries before the accession of Chandragupta to the throne of Pataliputra.
This is, as far as I can judge at present, the chief value of the new alphabet. I do not think that it teaches us much regarding the history of the Southern Maurya characters and regarding the manner in which they were derived from their Semitic prototypes. There is only one form among them which, I think, may be considered For
