________________
THE INDIAN ANTIQUARY
[CHAPTER V.
the test, and that Indian writings must be referred before or after 600 A, D., according as they show the use of the old or of the new form of that letter.
(10) Seventh century in Nepal.--All the known Nepalese inscriptions are from the neighbourhood of Kâtmândû, Long. 85° 60', which is within the eastern area. The north Western new form of y, comes into these records first in the second half of the seventh century, in an inscription of 677 A.D.,86 (Bendall's Journey in Nepal, No. III, p. 77), and in another undated, but slightly earlier, of about 655 A.D. (Indraji's No. II, in the Indian Antiquary, Vol. IX, p. 174). It is always in its modern variety, and from the beginning it appears independent of the original rule, being used with any vowel as well as with either kind of stroke. Thus we have modern ya in No. III, 1. 18, yathâ; in No. 11, 1, 5, yah; modern ya in No. III, I. 21, dêya, and in No. 11, 1. 13, prandlikâyás, modern yi, in No. 11, 1, 2, kshộbhayitva; modern yu, in No, III, 1. 29, yuvarâja, and in No. 11, 1. 1, yukta. Again we have modern yê, with the superior stroke in No. III, yê, 11. 25-26; modern yai with the superior stroke, in No. 11, 1, 23, kayaité; modern yo, with the superior stroke, in No. III., 1. 12, yo, and in No. 11, 1, 22, bhûyo, but with the lateral stroke in No. 11, 1. 4.9.
The statistics, given in the foregoing paragraphs, may be summarised as follows. The distributive rule referred to in them is based on the two facts, (1) that the new form is used only with the syllables yé, y ai, yo, yau, while with other syllables the old forta is used; and (2) that the new form is used with those syllables when they are made with the lateral stroke, but when they are made with the superior stroke, the old form is used. About 372 A.D., this rule is "in the making"; about 400 A.D. it is in full force; from about 425 AD. it gradually obsolesces, about 550 it has become inoperative. This information enables us to sketch, with considerable precision, the progress of the fashion of applying the new form of y, which was already in use in ligatures, to that letter when it occurred as a non-conjunct.
(1) This fashion arose in the western portion of the northern area of the Gupta script, about the middle of the fourth century A.D. Thence, in the latter, half of the sixth century (in India, but of the seventh century in Nepal), it spread into the eastern portion of that area.
(2) The fashion was at first limited to the syllables ye, yai, yô, yau, when their vowel was written with the lateral stroke. This is shown by the way in which the new form is used in the Bower Manuscript; and the period of this stage of the fashion is fixed by the epigraphic records of Northern India (ante, Nos. 1 and 2) as the second half of the fourth century A.D.
(3) Thie limitation was soon abandoned. From early in the fifth century (ante, No. 3), the fashion of using the new form began to extend to any vowel combination, and to either the lateral or the superior stroke.
(4) By the end of the sixth century the new form had become so fully established in all conditions of the latter y, as to extrude altogether, in all ordinary writing, the old form (ante, Nos. 8-10).
The preceding sketch of the chronology of the origin and spread of the new form of the letter y determines the time of the writing of the Bower Mannscript as having been in the second half of the fourth century A.D. And it is probable that it should be sought rather nearer the beginning than the end of that period. The Table, given on p. xlviii, shows that in the earlier portion (Parts I-III) of the Bower Manuscript the letter y occurs 583 times (col. III) in the vowel combinations ye, yai, yo, and yau, Outside these combinations, it occurs no less than 1,028 times (col. I). If at the time of the Bower Manuscript the fashion of extending the use of the new form of y to cases outside those combinations had already begun to develop, it is hardly conceivable that not a single example of such an extension should occur among those 1,028 cases. The probability, therefore, seems to be that the writing of the earlier portion of the Bower Manuscript should be placed about 350-375 A.D. And seeing that the three later portions of the Bower Manuscript (Part IV, Parts V and VII, and Part VI) must be, as shown in Chapter III, p. XXXV ff., practically contemporaneous with the earlier portion (p. xlvi), it follows that the production of the whole of the Bower Manuscript must be referred to the third quarter of the fourth century A.D.
86 According to the local era, discovered by Professor Sylvain Lévi; see Bp. Ind., Vol. V, Appendix, P. 73, note. By the Harsha era it would be 688 A.D.
