________________
No. 9.]
DHAR PRASASTI OF ARJUNAVARMAN.
97
times (in rishabha, 1. 22, and riju, 11. 34 and 81). The initial e resembles the letter pa ; compare e.g. mae, 1.4, with prapa in the same line. The initial o ends with the same flourish as į and the first form of « (see e.g. loo, 1. 6); it bears a little hook at the right top, which is omitted, perhaps accidentally, in two cases (chdo, 1. 44, and osaht, l. 63). The initial au occurs once in autsukyam, 1. 57).
Among the consonants, jha closely resembles the initial ri. It occurs seven times singly and four times in combination with ; (e.g. in majjhanna-samjha, 1. 30). The consonant tha either has its usual shape, or its vertical portion is crossed by a thin carved line (e.g. in kanthirava, 1. 7). The primary form of na consists of a horizontal line to which three vertical lines are attached, the middle one being slightly shorter ; but the secondary form of na resembles that of la. The doubling of na is marked by a horizontal cross-line in four cases (nishanna, 1. 26, sunnan, 1. 52, kanndharana, 11. 52 and 75); the primary and the secondary forms are combined in one instance (ayannehi, 1. 21); and, if my reading is correct, the secondary form is improperly used twice in another (thenna, 1. 75). The last mode is generally adopted if the doubling takes place after r (e.g. in 'vatirnnasya, 1. 1); but the cross-line is then used in five cases (6.g. nirvaranayami, 1. 62), and the primary combined with the secondary form once (in akarnnanta, 1. 1). The usual form of tha is not very different from nua (see e.g. kathamchid, 1. 1). The same form is employed four times after t (e.g. in hatthe, 1. 17); but in the majority of cases the secondary form of tha is identical with that of chha ; compare e.g. ittha and sthané, 1. 6, with uchchhavammi, 1. 16. If my readings are correct, there are three instances (thakka-thakkida, 1. 5, and thenna, 1. 75) where even for the primary tha the same symbol is used as for the primary chha (seo chhachcharana, 11. 5 and 75, and lanchhanan, 1. 16). The group kkha is written in two Ways: In viyakkhana, 1. 45, the kha is regularly attached to the k; but in pariklchalanta, l. 42, the kha seems to be ingerted between the two loops of the k. The group ddha looks like dva in most cases (e.g. yuddha, 1. 17); but in two places (=Ddhard-, 1. 32, and muddhdo, l. 47) its dha is distinguished from va by an additional horizontal line. Similarly, ndh looks like nu in tandhun, 1. 14, but not in samvandh-ochita, 1. 32.
The viráma is employed below l (11. 10 and 61), t (e.g. 1. 9) and (e.g. 1. 20). The avagraha occurs five times- twice after 8 and three times after a (11. 1, 2, 3). If at the end of a line there was no space left for the next akshara, the engraver filled up the line by a symbol which looks either somewhat like a reversed Någari ta, or like & narrow U;' in the transcript these signs are denoted by vertical line in round brackets. Finally it has to be noted that the apper portions of a few letters of the first line are ornamented with scrolls or flowers.
The engraver has done his work with considerable care and has committed comparatively few real mistakes, part of which he has corrected himself on second thoughts. As regards orthography, it must be stated that b is nowhere distinguished from v. The dental is used instead of the palatal sibilant in Sarada, 1. 3, Hiranyakasipu, 1.7, kdemira, 1. 20, aṁsa, 1. 21, and spásena, 1. 33 f.; the visarga instead of the lingual sibilant in chatuhpatha, 1.3, nihkrámati, 1. 58, and nihkrárta, 11. 15, 28, 31, 39, 58, 81 and 82; and t instead of d in atbhuta, 1. 49. The spelling of ujvala, 1. 13, datva, ll. 13 and 21, and patra, 11. 25 and 61, is not correct, but frequent in inscriptions and manuscripts.
The languages of the inscription are Sanskrit and Pråkpit. There are 76 Terses ; the remainder is in prose.
* In four cases (og. martharan, 1.4) the top-line of this form of tha is indistinct.
The kkof pachchakkho, 1. 64, reems to be a modification of the same type. • The second variety occurs only in 11. 22 and 42. • I have added a Sanskrit translatiou of the Prakrit panager at the end of the toxt on p. 117 ff.
