________________
CHAPTER VI.
ON THE SOURCES AND THE DATE OF THE NAVANITAKA.87
The name and identity of the author of the Návanitaka are not known. The final colophon which perhaps would have supplied that information is, together with the last chapter of the work, unfortunately missing in the manuscript. But from the sources which the author utilized in making his compilation, it is possible to estimate approximately the time wh n his work was written. So much is certain that the date of writing the work cannot be the same as that of writing the manuscript in which it has come down to us. The latter is not an autograph. This is proved by a number of clear indications. For example, on page 28, in verse 45, we find, in one of the medical formulæ, three dots marking the omission of three syllables (ante, Chapter IV, p. xlii). At the time or editing the text the emendation triņi cha[vya-palani] was suggested by me, but in the meantinie the true reading paicha cha[vya-palani] has been discovered by Dr. P. Cordier in an ancient medical compendium, called Bhêda Savihita, from which the Nâvanitaka has quoted the formula in question, Obviously the substitution of the dots shows that the writer of the Bower ManuScript had a defective original from which he copied. Again, on page 58, in verse 723 of the pippali-vardhamâna formula there is the curiously blundered phrase yavad-dasa-varshâs, instead of yâvad-avakarshas. Such a blunder is unthinkable in an original writer: it could proceed only from one wlio copied from a defective original. Again, on page 67, to verse 879 we find appended the gloss prâchiniká pâthå, for the purpose of explaining an unusual name of the drug commonly known as på ha. Such a gloss is not likely to have proceeded from the author himself. As usual, it must have stood originally on the margin of the manuscript, or perhaps between the lines. By a subsequent copyist it was transferred, in the body of the manuscript, to the position where we now find it in the Bower Manuscript. The writer of the latter may, or may not, have been the first to make that transfer; but, in any ense, the present position of the gloss shows that the existing Bower Manuscript was not copied from the author's autagtaplı, but from some intermediate copy of that autograph. The conclusion whichi, indeed, is already suggested by the three dots and the blundered phrase, is that there must have been some interval, perhaps of not inconsiderable duration, between the writing of the autograph and the copying of the existing manuscript. The date of the latter, as explained in Chapter V, p. lii, must be referred to the third quarter of the fourth century, somewhere between 350 and 375 A.D. This supplies us with the lower limit for the date of the Nâvanitaka, which, in view of the above-mentioned necessary interval between the autograph and the existing manuscript, may be placed provisionally in the beginning of the fourth century, or about 300 A.D.
The upper limit is determined for us by the circumstance that the Charaka Samhita and the Susruta Sashita are two of the sources from which the author of the Návanitaka quotes
# For a somewhat fuller treatment of the subject, sce the Journal of the Royal Asiatic Society for 1909, pp. 857 fr.
* See his Recentes Decouvertes, p. 21. The three missing syllables are paricha cha.
