42
meaning of its terms. Variation in the pattern itself, instead of variation in content, is illustrated in the last phrase of this line, หนี ญญ่าย พ่าย จแจ. The rhythm is here triple in both members, augmented as they are by the introduction of a short alliterative syllable in each before the main stress. For a general view of this subject of metrical forms in prose, the reader is referred to pp. 19—21 above.
The word ◌ีหน fled, near the end of the line, illustrates one of the difficulties growing out of the fact that when the vowels were given their place in the line, they were not also given their place in the sequence of letters in the syllable. Since the vowel ◌ี must appear at the beginning of the syllable in which it is used, no matter where in the syllable it is actually pronounced, it comes about that 'ni' fled, is spelled exactly as is 'hin' stone, ll. 82, 85. The return of this group of vowels to the places above and below, which they occupied in the Indian scheme, has obviated the difficulty in their cases—indeed, this difficulty may have hastened their return. But a similar uncertainty as to pronunciation still besets the prepositive vowels. Only the context can determine whether แหง spells 'hæng' or 'ngæ'; whether เพลา spells 'phlau' or 'phela'.
As regards the text, S has transformed the second word of l. 6 beyond recognition, and beyond possible pronunciation, by writing an extra ส bodily in place of the เ before the consonant. P leaves off the 'mai ek' both in this word and in ญ่าย further on. But omissions and even alterations of accents by P are far too frequent to call for individual notice. ผ้า is made into ฝ้า, which is probably right as a correction ; but it is not the reading of the stone. ฝ้า does appear once out of five occasions of the use of the word, in l. 21.
7. The vacant space at the beginning of this line was once occupied by some letter, for a trace of a part of its right hand stroke still remains. Whatever it was, it cannot be a part of the following word. S followed the obvious suggestion of making it a part of the previous word, and wrote in a ง—almost at random, it would seem, since it makes a word of no intelligible sense. P follows S both in text and transliter-
