Introduction] RELATION TO THE MAHABHARATA [xxix with, were composed ¹. About 300 slokas, or approximately one-fourth of the whole work, are devoted to these legends. This is the most important, as well as the most interesting portion of the book, for it comprises the oldest systematic collection of legends which we possess in Sanskrit. Narrated in an epic style, they are historically connected with a number of stories told in the Mahabharata. This historical con- nexion deserves to be carefully examined. Dr. Sieg, in the first part of his meritorious work, 'Die Sagenstoffe des Rg-veda und die indische Itihasatradition 2,' has done this in the case of some legends. I cannot, however, in the present state of our knowledge, agree with him in supposing that the Brhaddevata has borrowed from the Mahabharata. He thinks that the story of Deväpi Arstisena is a clear example of such borrowing. In my opinion, the evidence is all the other way. The single fact that Devāpi's patronymic, Arstisena, has, in the Mahabharata, become an independent name designating another person, but mentioned along with Devāpi, is a clear indication of the posteriority of the Mahabharata form of the story; a differentiation of this kind being a not infrequent phenomenon in mythological development. In the Mahabharata, moreover, a third brother appears, and the name S'amtanu has become S'amtanu. It is, besides, impossible on general grounds that a Vedic work which is undoubtedly earlier than the Sarvanukramani, and not much later than Yaska, should have borrowed from the Mahabharata, which must have assumed the form known to us so many centuries later. The possibility even of interpolation appears to me, as regards the legend in question, to be entirely precluded by the internal evidence of the Brhaddevatā ¹. Our text further contains other matters connected with the deities of the Rg-veda, such as an enumeration of Rishis 5, of female seers 6, of the steeds of the various gods 7, a detailed account of the Apri hymns ", and a full discussion of the character of the Vaiśvadeva hymns". The concluding nine vargas state the gods presiding over various kinds of ritual verses and over the different tones, with some remarks about knowing the deities of the Veda. Important and interesting is the state- ment, occurring in the Anuvākānukramani also, as to a different sequence of the groups of hymns in the first Mandala 10. The statements of the ¹ A list of these is given in Appendix iv, pp. 132, 133. 2 Pp. 150; Stuttgart, 1902. 3 Bṛhaddevata vii. 155-viii. 9. Cp. my review of Sieg's work in the Deutsche Litteraturzeitung, Sept. 19, 1903, columns 2302-4; and my History of Sanskrit Literature (London, 1900), p. 287. 5 ii. 129-131; iii. 55-57. ii. 82-84-Arşinukramani x. 100-102. 7 iv. 140-142. 8 ii. 158-iii. 32. ii. 132-134; iii. 41-52. 10 iii. 125; see note ".
