in course of time, an increasing number of homonyms which required to be treated in separate word -books of a different kind, Even the early nika--u works contain, as already indicated, separate sections on homonyms. These sections are solely lexicographical in function; and is is here that we find a long course of development, ending at last in the adoption of the alphabetical order.

The arrangement of homonyms in the 11 ^th section of TivÆ’karam is,as in TolkÆ’ppiyam,quite arbitrary. There are 348 homonyms in this work; and the printed editions give them in an alphabetical order, dividing the section into two parts, onewith the homonyms at the beginning, and the other at the end, of the sutras. But this order must have been introduced by later redactors, homonyms in its 10th section and it may be noted that the number of homonyms has increased to 1091. Theprinted editions of this work also adopt the alphabetical order, through none of the manuscripts PiÅ’kalantai treats of these homonyms in its 10th section and it may be noted that the number of the homonyms has increased to 1091. The printed editions of these work also adopt the alphabetical order, though one of the manuscripts support it. The original order of the homonyms in this section seems to have been based on the respective order in which they occur in the preceding nine sections.

It may also be pointed out that TolkÆ’ppiyam and TivÆ’karam met only casually the requirements of etukai in the arrangement of homonyms. In both the works, the homonyms were intended to be memorised. Etukai was adopted not as a constructive principle to facilitate reference but as a decorativeprinciple to help the sutra style. In the 11 ^th section of C‡-Æ’ma-i-nika--u, we find, for the first time, the principle of etukai expressly adopted ( vide the introductory stanza ), and the section divided into as many sub-sectionsas there etukai. Thus some help was provided for easy reference also. Being the first attempt of its kind, this section exhibited a few defects in its arrangement. Sometimes, homonyms of different etukai were jumbled together in the same stanza ¹ A good number of options adjuncts had also been used for the sake of rhyme or metre. The number of works treated in this section is about 1575.

Even in nika--u works written for popular use, no development in the plan, leading to a change of emphasis form 'memorizing' to 'reference', is found. In one or two works ( e.g.,kayataram ), autÆ’ti-t-to-ai is adopted to help the memory.2In the number of homonyms also, is no increase. On the contrary, Uaiccol-nika--u contains only about 150, kayataram about 850 andÆ’potikai-nikantu (10th section of the 1st part ) about 1350 homonyms.

ARUMPORUI-VILAKKA-NIKANTU

Alongside of these, appeared works which dealt exclusively with homonyms and improved the Œ treatment in many directions. The earliest work of this kind, so far know to us, is Arumporu˜- vilkka-nikantu. Its author was Arumarundiaya Desikar. He was a Vira Saivite and his guru was Pacchai-k-kanda Desikar of Childambaram. The work is dated

---

¹ Later redactors collected ll such stanzas into a separate sub-section and named it kalappetukai ( mixed etukai ) ² Mr.A. Shanmukkam Pillai of so‰vandƒ‹ is mentioned as having written an antÆ’ti-t-to-ai-nika--u< ( vide Ta‹i-c-ceyyu--cintÆ’ma-i ).