________________
THE INDIAN 'ANTIQUARY.
[MARCH, 1889.
(cf. Sp. Hardy, loc. cit), one would be inclined to the first solution ; but, as the idea of a triple uposatha in each half lanation is expressly borne witness to by the Mahavagga (II. 4, 2), I have no hesitation in considering that such is also the intention of the king in this passage. It is true that great uncertainty appears to have prevailed in the tradition about the uposatha. The same work, a little further on (II. 14, 1) only admits 'two-upôsuthas, those of the 14th and of the 15th,' but, on the other hand, another passage (II. 34, 3-4) speaks expressly of the patipada uposatha, that is to say, that which corresponds to the first day of the month (the amawaka of Sinhalese terminology). I do not doubt, however, that Piyadasi considered this day as hallowed by a religious consecration. It is on this one day that the difference between the generic expression, dhuvayé anuposathark, of series A., and our series B., depends ; if this more concise expression is not repeated here, it must necessarily be so in order to exclude some element which it contains, and that element can only be the pratipad. With regard to the rest of the list, I would refer to what has been said about series A. and the plural tísu chátunmásisu; here again, all the full-moons being comprised in the dates chávudasáyé and pasinudasiyé, the terms tisaye and chátonmásísu have no use except as determinatives of the last word, sudivasáyé. I regret that this last term is obscure to me, for I know of no parallel examples of the technical use of the word. We evidently want here something different from a vague astrological expression corresponding, I suppose, to the Vedic sudinatré ahndun (cf. Weber, Die Véd. Nachrichten von den Naksh. II. 315). A comparison with the other lists ought to guide us. We shall subsequently see that the acts successively forbidden by the king necessarily constitute a series of decreasing gravity. It is therefore à priori more than probable that the lists of reserved days, admitting the fact that there is a distinction) should be reduced in parallel lines: the second should contain less than the first; and the third less again than the second; but all the days excepted in the two last should be included in the first. In a general fashion, this conjecture is at first sight justified. Between list B. and list C. it is verified in detail, provided that chatuimdsipakhd can be included under the last head of B.-tisu chátumásisu sudivasá, for the full-moons of Tishya, Panarvasú and the chaturmasyas are included under the two first terms chdoudará and pannadasa. On the other hand, to establish an analogy between A. and B. the last portion of B., tisayé to sudivasá must be included in A., either in the last term, dhuvayê chá anuposathaw, or in the last but one, tisu-palipaddy. In the first case, the three first terms of B. include all the days of uposatha except the patipada uposatha, and sudivasá onght to designate the first of the month, the first of the light half of the month of which the full-moon is in conjunction with Tishya, or Punarvasů, or one of the three months of cháturndsya). In the second case, it would designate the first of the dark half which follows (the full-moons in question). To sum up, therefore, C. appears to require that sudivasd should designate the 16th of the months above referred, and A. permits this interpretation. The conclusion follows that we are driven to admit that B. practically had in view the days which come after the fall moons in conjunction with Tishya and with Punarvasg, and after the fall moons of the months of cháturmásya.' It may seem, perhaps, somewhat surprising that the name sudivasa, lucky day' should be applied to the first of the dark half for in general it is the light half, which is considered as particularly auspicious; but the scruple must necessarily vanish before the positive fact, witnessed by the perfectly clear testimony of our first list, that the day in question, at least in the lunations specified, was considered as having a religious consecration.
This newussary agreement between our three lists upon which I have just insisted, leads us to one last remark. The expression tisáyain puinamásiyas of A, should surprise the reader: tisáyash alone would be sufficient, as all the following lists attest. We should rather expect to find pusinamasi added to tisu chatun másísu, the first full-moons indicated, if it were added any. where. On the other hand, the full-moon in conjunction with Puparvasd plays so important part in the subsequent lists that it is out of the question that it should not be here also. How could it be permissible to slaughter animals on a day on which it was not permissible even to mark thom P I have therefore no hesitation in maintaining that, instead of purnamasiyan, punávasuyam should stand here. I do not deny that such a correction may appear bold, in the
