INTRODUCTION 79 most ticklish. The varieties of the different types of the Ghattās are distinguished from one another by the number of moras their Pādas are required to contain. But addition or removal of even one mora makes a difference in metre. Coupled with this narrow margin of difference in the continguous varieties of the Ghattās is the fact that the nature of the orthography of the Ap. Mss. being in various points confusing, the text preserved by them is far removed from having a mora-perfect correctness. This state of affairs many a time obscures the exact number of moras contain- ed in a particular Ghattā-pāda and as a result it becomes consider- ably difficult to identify with precision the Ghattā-metres. Apart from this, the main cause of obscurity lies in the anceps value of the end syllable of a Pāda. This always causes a difference of one mora and the consequent dubiety. The treatment of this point by ancient metricians does not contribute much to the clarification. Hemacandra offers the follow- ing remarks: Vante G V a krah Pādänte vartamano hrasvo g samjño bhavati. Sa ca pras- tāre takraḥ sthāpyate. 'vā' iti vyavasthita-vibhāṣā. Tena yatra apavādaḥ tatra g sainjño na bhavati. Dhruvāsu vivakşa-vaśād gurutvam laghutan ca. yad äha: Oja-samkhyā yadā'bhiştă, dhru- vasu viratau tadā 'go lata, yugma-samkhye tu, viratau guruta lag- hoặ//Tathā: gurualo) cchiya ekka-lahu-virama-visayammi visama samkhãe'jamala-lahu lahua(o) cchiya, sama-samkhă-samthio hoi// Ch. 1 5 with Com. For the last stanza in the above citation Hemacandra is indebt- ed to Svayambhū, since that very Gatha is found at SC. V 2 with the correct readings guruo and lahuo. The passage means: (1) The short end syllable of a metrical Pada is to be treated as long as a general rule. (2) To this rule there are fixed exceptions. In these excep- tional cases the final short is to have its natural value. (3) In the case of the Dhruvā or Ghattā, the value of the end syllable depends upon the number of moras the Dhruva- pāda is desired to contain. (4) After counting the moras of a Ghattā-päda exculsive of the final syllable, if the Pāda is short by one mora, then the remaining end syllable should be counted as of one mora and accordingly even if in such cases the end syl- lable is actually long, it should be considered short. On the other hand, if the Pāda is short by two moras, then the end syllable should be given the value of two moras and accordingly even if the end syllable is actually short it should be considered long. This rule works well in those cases wherein we know before- hand the metre, but in those cases in which we have to decide the metre by scanning a specimen, we would be faced with as many as, four alternatives regarding the Antarasamā Catuşpadis. A Ghatta with the actual mora count of 13/10, in its odd and even Pădas and wih short end syllables, for instance, can be regarded as contain- ing 13/10, 14/10, 13/11 or 14/11, moras and these are four different metres called Marakatamālā, Abhinavavasantaśrí, Kusumakulama- dhukara and Bhramaravilăsa! The form of the metre has nothing
