76 PAUMACARIU 5 + 5 + 3 or 4 + 4 + 5 as the Gaņa scheme of the Apsarovilasita'. None of these schemes applies to 31 la (dihara-kālacakka-haeņa i.e. --UU-U-UUUU), 71 c (jinavara-punna vāya-hayaim i.e. vuuU--u--UUUU), 3 12 1 d (uvasamu jāu sadva-jaņaho j.e. UUUU--11--MUUU), etc. On the other hand the scheme 6 + 4 + 4 given for the Gandhodakadhārā (SC. VI 158 c) is satisfied by all the lines. Of course the theory as usual requires the end syllable to be counted as heavy. The rhyme-scheme seen in the definition stanza of the Gandhodakadhāra in SC. and in the illustrative stanza given in Ch. is: b rhyming with d i.e. the usual rhyme- scheme of the Antarasamă Catușpadi, while in our stanzas a rhymes with b and c with d. But this has little significance, because in their treatment and state- ment Prakrit metricians were far from being precise or exhaustive and they have hardly ever paid due atten- tion to rhyme even though it was a very important feature of the Apabhramśa prosody. The Gandhodaka- dhārā itself is a good instance of this state of affairs. For earlier in Ch. while defining the Jhambataka, Hemacandra has observed" that the Gandhodakadhāră itself that will be defined later on is called Jhambataka when sung. Thus from a metrical point of view according to Hemacandra himself the Jhambațaka and the Gandhodakadhårā" are identical. Now if we glance at the illustrative stanza of the Jhambațaka we find that it has got the rhyme-scheme a/b, c/d (i.e. the same as found in our stanzas) and not that found in the illustra- tive stanza of the Gandhodakadhārā, viz., b/d ! Irregularities: In 3 3 16 the Mss. read cauti- säisayaº (P.), cautīsäisayar (A.) and cautisą aisai (S.). Acceptance of any one of these readings yields one or more moras too many for the line. Hence the text has been emended as cautis'aisayaº. (2) The Kadavaka-commencing stanza of the 13. Sandhi (Dvipadi). Scheme: 6+1-U (ur UUUU) + 4+ 5+ 4+ U-1; (or I'TUT:- sporadic caesura after the 16. mora. This is Duvai or Dvipadi. For a detailed discussion on this metre see HP. 195, Sr. 60. The results obtained by Alsdorf hold good in all particulars for our Dvipadīs also, except in one important point. Against Alsdorf, the 4. Gana is twice Jagaņa: 13 4 1 a is-01:-1:-1: 1 UUU 0,--OUT:--0---0-- and 13 51 bisit:- (-U-UU
These cases clearly show that the form --U(UUUU) was not forbidden for the 4. Gana. U -- U - (1) Hemacandra and Rajasekhara give 6+4+3 also. This would deprive our test of its validity. But in examining the metres of Svayambhu's works, his own metrical work above all else should be given due consideration. (2) vakşyamāņa Gandhoddakadhāraiveyar gāna-vasád Jhambataka-saññar labhate Ch. 380/10-11. 3) Its illustration: pahu tuhe veri aranni gaya, nichhu vi nivasahim jimva sasaya- ghana-kantaya-dūsancarani, tahi jhambadai karira-vanill
