INTRODUCTION 25 plausible that the lost portion of SC. contained only the wanting Varnavfttas from the Ukta to Atijagati division along with some Introductory matter and nothing else? Let us examine it. As unfortunately the stanzas are not numbered in the Ms., we have to take resort to inference. Each side of the folios of the Ms. of SC. contains about 8 lines with 42 letters to each line on an average. This gives us a round figure of 330 to 340 letters per side. Now the first few folios contain on an average four metres of the length of the Atijagati or Śakvari class with their definitions and illustrations. On the basis of these data we can form a rough esti- mate as to how many metres on an average were contained by each side of the folios. Some calculation done with the help of the above data shows that about six metres having from 8 to 12 syllables per line along with their definitions and illustrations can be contained by each side of the folio. To form an idea as to how many folios could possibly have been occupied by the treatment of these metres, we should know their number according to Svayambhū. The extant portion of the Varnavrttas treated by Svayambhu can favourably compare in its extent with the corresponding por- tion of the Kavidarpana' or the Vșttaratnākara. The latter of these two works treats a slightly smaller number of metres. On the other hand, Hemacandra's Chandonuśásana is elaborate. Thus Svayambhūcchandas in this matter stands midway between the Kavidarpana and the Chandonusāsana, somewhat nearer to the former than the latter. Now among the missing Varnavșttas those having from 1 to 7 letters per line could not have occupied more than one side altogether. Corresponding to the remaining Classes upto to the Rucirā variety of the Atijagati class, the Kavidarpaņa has 47 metres, the Vștta- ratnākaru has 59. Accordingly SC. could not have treated more than 90 or 100 metres at the most. Counting six metres per side. the space occupied by these metres would come to about 16 sides or 8. folios. Taking into consideration the space devoted to the introductory matter and to the treatment of the metres containing from 1 to 7 letters per line, we can quite reasonably conclude that the treatment of the missing Varnavrttas from the Uktā to the Praharșiņi or the Mattamayūra variety of the Atijagati class, could not have occupied a space of more than nine or ten folios of SC. This conclusion is supported by another consideration too. In any treatment of the Varnavșttas the space occu- pied by the first 13 classes of metres cannot be possibly more than about half of the space occupied by the rest of the classes down to the Utkyti, along with the Dandakas, the Ardhasama Vșttas and the Vişama Vșttas. Now this latter portion in SC. occupies 22 folios (from 23 to 44). Hence the missing Varna- vșttas could not have occupied 10 or 11 folios at the most. If the missing Varņavſttas were contained in ten of the missing folios only, what were the remaining twelve folios devoted to? Let us consider the matter. The Aryă, Galitaka and Khañjaka Praka- raņas of the fourth chapter of the Chandonusāsana deal with some 85 different metres. Making allowance for the greater elaboration and the compilatory character of the Chandonušāsana, we can assume that the corresponding number of the metres of these classes, if treated by Svayambhů, would be in the vicinity or . Counting even as three metres per side on an average, these petres can (1) Velankar, 1935-1936, 51-60.
