________________
CHAPTER IV)
BOWER MANUSCRIPT
XXXix
In Part II we have the lotus, reinforced by the comma as well as the double stroke, after verse 10, in fol. la, to mark off the end of the introduction to the treatise. Similarly after verse 24 on fol. 165, we have the lotus by itself to mark the end of a series of short formulæ (verses 18-24), and after verse 39a, on fol. 261, to mark the end of a single short (unnamed) formula (verses 38-39a). And after verse 737, on fol. 2265, we have the wheel to mark the end of the long pippali-vardhamana formula (vv. 716-737). In Part III a disk is frequently used in this way, to mark thc end of a formula; especially in fol. 36, where it occurs not less than seven times, in 11. 1, 2, 3, 5, 7, 8, 9.
Of particular interest is the circumstance that the lotus and wheel appear to be used, in Part II, also to indicate glosses, which the author of the Nâvanitaka himself seems to have added to the extracts from old authoritative works of which his own work is mainly composed. Thus on fol. 2703 (page 67) there is, after verse 879, the obvious gloss prâchinikâ pâlhå enclosed between two wheels (see note 418, on page 162). On fol. 3366-, verse 1109 is enclosed between two wheels, and its purport suggests its being a gloss (see note 490 on page 180a). In the similar case of verse 929, on fol. 2863 (p. 69), which the author had at first omitted to mark as a gloss, he (or rather a subsequent copyist) has afterwards, on revision, inserted the lotus mark between lines 2 and 3. The same practice is observed in Part III, which may be a work by the same author. Here, on fol. 167, the lotus marks what appears to be a gloss, so also on fol. 36. It will be observed that both passages, thus marked, are in prose.
In Parts V-VII, the usago with regard to marks of interpunction is much the same as in Parts I-III. But in addition we meet with three signs which exactly resemble our modern comma, semicolon, and full stop. The comma occurs, e.g., in Part V, fols. 263, 364, 5a", twice even in a reversed position on fols. 3al and 543 (see Table V, Traverse 3), in Part VI, fol. 264, and in Part VII, fol. 123. But it is probable that the comma is really identical with the more usual lengthwise-comma (the numeral one), of which it is an exaggerated cursive form. The semicolon, practically identical with the well-known sign of the visarga, occurs, e.g., in Part V, fols. 6a5 and 663, and in Part VII, fol 4a6. The full-stop, or single dot, is found, e. 8., in Part V, fols. 2a5, 3a, 665, in Part VI, fols. 160, 3al, and in Part VII, fols. 2a4 and 264. As to the ordinary signs, the double stroke does not happen to occur in Parts V and VII, in which the comma, either erect or prone, regularly takes its place. In Part VI the double stroke is found in a slightly modified form, embellished with a hook to the left at the top of the first stroke, as in fol. 4a5, or with a hook to the left and right respectively at the top of the two strokes, as in fol. 46. The lengthwise-comma, or the numeral one, as already observed, is used regularly in Part V. e.g., in fols. 1a3.5. 2a, etc. So also in Part VI, e.g., in fols, 164, 3a, and in Part VII, e.g., in fol. 1a5. Neither the wheel nor the lotus is found in any of Parts V-VII. In their place Part V uses the spiral which is the conventional representation of the sacred saikha, or conch shell, as in fol. 569. Once in fol. 3a®, this spiral is accompanied by the lengthwise-comma, It will be observed that the same spiral appears also in the remark which is appended to Part III (Plate xxxviii, obv.), and which, as has been previously (pp. xxi and xxxv) stated, was written by the scribes of Parts V-VII.
In Part IV the usage with regard to interpunction is as follows. The double stroke is not uncommon. In its plain form it occurs, e.g., in fols. 2a, 3a%; but it is often accompanied with the lengthwise-comma, or numeral one, as in fols. 2a, 3a2, 4al, and occasionally this comma is drawn across the double stroke, as in fols, 3a", 3a, Moreover in the case of