________________
270
THE INDIAN ANTIQUARY.
(SEPTEMBER, 1882.
occurs a wonderful confusion of the sibilants, and in ancient times sa and sa are in Praksit equivalents. The one occurs constantly for the other. If we therefore find and my in the nu. merals for 100, the conclusion is that the people were in that case as careless as in others, and pronounced indiscriminately su and bu, because they were accustomed to do this in common life. The same was probably the case for the numeral syllable for five, where side by side with forms which clearly are nu, others occur which must be read no (UT)-Bhagwanlal's tables.
(4) Fouthly, there are other cases where cer. tain sects, or the Pandits of certain countries, have misread the ancient signs, and have substituted wrong syllables for them. The best cases are: (a) that of the numeral syllable for 10, which in the oldest forms is la Lc in the south, and dut I in the north. This has been rendered by and (Iri and la).
(6) That of the numeral syllable for 100. The Nepalese have misread (as has sometimes been done by modern epigraphists also) the 34=su, as =a.
In these cases the important point, which shows that the people pronounced syllables, and not the mumerals when reading the signs, is that they always made a new syllable of the old sign, not a mere unintelligible symbol. Had they pronounced sata for they would have left it, and not have written a clear for it.
(5) The fifth argument is that down to the present day the numeral syllables are called aksharapalli, vis. letter table,' by the Jainas, and are known to represent syllables. Mallinatha (circa 1150 A.D.) speaks distinctly of such a syllable as a sabda "a word.'
II. The system of numeral syllables as we find it in the oldest inscriptions was settled by the Brahmanical schools. The proof of this assertion lies in the use of the signs c (la) gi, ep) ph, (a) kh, which occur in Brahmanical books, and speak alone. Nobody but a Brahman could bave dreamt of making the Anundsika l, the Jihramaliya , and Upadhmaniya B, serve for numerals. The circumstance that the three strokes , are intended foru, 14, 6-, (hrasva, dirgha, pluta) points to the grammatical schools being the originators of the system. The proof for the assertion that the strokes too have a vowel value lies in the manner in which they are used with the hundreds and thousands-100 being
expressed by my (bu), 200 "T and 300 "F; 1000 by
(dha). 2000 by T. and 3000 by $. If the strokes had a mere numerical value, the marking would be wrong and unintelligible. We should then require for 200 F, and for 300 F. If we pronounce bu, sú, sú-u, the difficulty disappears. Hence, I conclude that wherever we find the strokes , =, E, these, too, are intended as symbols for a vowel, and for the vowel u, because in all grammars the vowel u is used to illustrate the three stages, short, long and treble (pluta). Panini's sutra is , &, 6-1-hrasvad trghaplutah,' i.e.' vowels having the duration of u, , -u are called long, short and pluta.' The origin and meaning of these marks seems to have been forgotten very early, and in the Bauddha and Jaina books, eka, dvi, tri, or sva, sti, brk, om, na, mo (the latter being the usual initial three syllables of books), are sub. stituted.
III. Though I claim for the Brahmans the oldest form of the Aksharapalli," I do not claim for them its invention. We constantly find in India that something foreign imported into the country is made to assume native Indian forms, and disguised so cleverly that one would swear it was a native invention. As I believe that the Indian alphabets are foreign inventions introduced into India long before the historical times, I think it probable that the numerical system too came from a foreign country. I believe the Southern Indian alphabet came to India from Arabia or from the Persian Gulf, vid Suppårå (Sopheir) or Bharoch, and that the Southern Indian numerals came by the same road. But I think that, in spite of the resemblances pointed out by Deecke between the Himyaritic and Maurya letters, we have not yet found the alphabet from which the Southern Indian characters are derived. I think that there is much less chance of making out anything about the numerals, and of saying from what other Bystem they are immediately descended. But it is not at all clear that originally they may not have come from Egypt, but probably through some Arabian traders either from the Arabian Coast or from the Gulf.
NOTES AND QUERIES. 8. DIAMONDS AND PRECIOUS STONESSince my Economic Geology of India was pub. lished I have found in several old travels references to the localities where diamonds and other precious stones were found. Some of these I am unable to identify, but I think it possible
da is the northern equivalent for la in the language difficult to believe that its signs are borrowed from four also.
or five different sources.-E. C. B. Dr. Biller is convinced that the Aksharapalli is of * Zeitschrift der Deutschen Morgenländischen gesellan origin extrageous to India, though he still finds itschaft for 1877, vol. XXXI, p. 598.
