________________
EPIGRAPHIA INDICA.
(VOL. XX.
for a crack at the middle, is fairly well-preserved. The execution is neat and calligraphic. The Iecord is a prasasti drawn in florid Sanskrit and, but for the initial symbol and the words api cha coming between the sixth and the seventh stanzas, is entirely in verse. The characters, in which it is written, belong to the northern class of alphabets and present a very marked development in contrast with those of the contemporary and even somewhat later inscriptions which have been found in Northern and Eastern India. They largely resemble the characters of the Aphsad stone inscription of Adityasēnal and would, thereby, indicate that the development, we notice in them, must have taken place not later than the first half of the sixth century of the Christian era, i.e., the time to which this inscription belongs. Till now, the Gayā inscription of A.D. 588-89 was considered to be the earliest inscription to illustrate such forms. But the epigraph under publication is decidedly anterior to that record and, therefore, becomes the earliest known inscription to represent that development. The alphabet to which these characters belong may well be called acute-angled (Bühler) or Siddhamätsika (Berüni) in preference to the term nailheaded' or kutila'. Devanagari is an outgrowth of this lipi. The present record uses the bipartite form of ya throughout, as does the Bödh-Gayā inscription of Mahānaman. Nalanda is not far off from Bödh-Gayā. This epigraph, therefore, will not countenance the supposition that the use of the bipartite ya in the Mahânāman inscription was premature." The tripartite form of ya is found even in the Udaypur inscription of Aparājita, which belongs to the Samvat year 718 (=A.D. 661). It is to be met with not only in the inscriptional records named above, but also in the Horiuzi palm-leaf manuscripts of Japan which are believed to have existed in the second half of the 6th century (A.D.). To regard an inscription as late or early merely on the ground that it uses the bipartite or the tripartite form of ya would not, consequently, be quite accurate. The alphabet used in this inscription is, to a large extent, identical with the modern Dēvanagari or Nägari. The chief points of difference which it presents are these : The a-mātrā is generally indicated by a sort of small wedge attached to the right of the top line, whereas in Năgari it is expressed by a full perpendicular stroke. In the case of nă, however, & somewhat different mode is adopted by slightly bending the right end and extending it upwards beyond the top towards the right side. The u-mātrā is slightly different in shape, as, e.g., in bhūrio, 1. 2. The e-mātrā is marked in two different ways, sometimes in an elongated form of the Nagari symbol with a bend at about the middle, ag in dehinām, l. 1, and sometimes by a short curve appended to the left of the top line of the akshara to which it belongs, as in "ārthine, I. 1, or in karena, 1. 21. Similarly, the ai-mātrā is expressed in two ways. In os=ch-aisha, occurring in the penultimate line, it is practically similar to the Nägari symbol, whereas in the majority of cases it is expressed by the hook-like mark attached to the left of the top line and the slanting e-mätra, as, e.g., in ouchchaih, 1.21. The 6-matra is shown by the symbols of the à and the e-mätras combined, see, for instance, mokshaya and yo in l. 1. The aumätrā is also different, e.g., see gaura in 1. 6, or Sauddhodanér- in 1. 9.
As to the initial vowels, only the à and the i retain their comparatively earlier forms. The former keeps the hook attached to its leg as in ādēsāt°, 1. 12, and the latter, the two small circles placed above the reversed crescent, as in itys, 1. 20.
In the case of consonants, the omission of the top line of the sa, the ma and the ya symbols is noteworthy. So, also, are the forms of the letters bha, ha, gha and ja. The symbol for dha is also dissimilar and still preserves the old form, as does the letter tha. The form of r in conjunction with a following consonant is also antique and noteworthy; e.g., see &Endrair-yah in l. 1 or padārthao in 1.2.
"C. . I., Vol. III, No. 42, plate XXVIII. * Indian Palmography (Ind. Anh, Vol. XXXIII), p.49.. . Ibid. 0.1.1., Vol. III, plato XLI A. "Ep. Ind., Vol. IV, plato facing p. 30.