62 Dr. llocvn\o — Uemarks on Birch Bark MS. [April, 



varunena must be taken to be equivalent to one long, in order to 

 conform the line to the ordinary rules of a sloka ; so also in the first 

 half line of the eighth sloka in the first extract, where the two shorts of 

 suhliagd must be taken as one long. Many more examples of a similar 

 kind might be quoted. 



" Now as to the age of the MS., I believe it to be very old and 

 written not later than the end of the 5th century A. D. The style of 

 writing is exactly like that which we meet with in the early Gupta 

 inscriptions, between 450 and 550 A. D. These may best be seen in Mr. 

 Fleet's volume III of the Corpus Inscrijptionwn Indicarum. 



" One of the test letters, for the determination of age, of the Nagari 

 alphabet — and the most important one in the present case — is the 

 akshara ^ ya. Its original form was ^ ; this changed to cJ/^ or q^ ; 

 next the left hand loop-line was extended to the point of junction of 

 the perpendicular stroke, k3J . The object of this, of course, was to 

 permit of the letter being written with one continuous movement of the 

 hand. The next step was to dissolve the point of junction, ^ , a 

 natural consequence of quick writing. From this point, the modern 

 form was quickly reached. The whole course of this development 

 is clearly traceable during the period (about 400 to 600 A. D.) of 

 the early Gupta alphabet. The initial forms c^ and ,^^ we find 

 still used throughout in the Allahabad pillar inscription of Samudra 

 Gupta (about 400 A. D., see Fleet, ibid,^ p. 6), "the Udayagiri Cave 

 inscription of Chandra Gupta II (c. 410 A. D., ibid., p. 35), the Bilsad 

 pillar inscription of Kumara Gupta (415 A. D., ibid., p. 45), the 

 Mathura image inscription of Skanda Gupta (454 A. D., ibid., p. 263), 

 the Bhitari pillar inscription of Skanda Gupta (c. 460 A. D., ibid., 

 p. 53), the Kahaum pillar inscription of Skanda Gupta (460 A. D,, 

 ibid., p. 67), etc. The final form <^ is already used throughout the 

 Bodhgaya inscription of Mahanaman (588 A, D., ibid., p. 274), and in 

 the Asphad stone inscription of Adityasena (c. 650 ; ibid., p. 202). 

 The intermediate form ^-^ is only found in the Indor copper plate 

 inscription of Skanda Gupta, of 465 A. D. (ibid., p. 68), the Mandasor 

 stone inscription of Kumara Gupta, of 473 A. D. (ibid., p. 79), the 

 Majhgawan copperplate inscription of Hasfcin, of 510 A. D. {ibid., p. 

 106), and the Jaunpur stone inscription of Tsvaravarman of about 520 

 A. D. (ibid., p. 228). With regard to this intermediate form it is 

 particularly to be noticed, that it occurs side by side with the older 

 forms cV or o^j ^^^ t^^^* i* is exclusively used with the vowels e and 

 0. Thus we have it in the Indore plate in yojyam (line 7) ; again in 

 the Mandasor inscription in yo and 2)riyo (line 14) ; again in the 

 Majhgawan plate in chhreyo (line 14), yo (line 16), ye (line 18); again 



