tions, the versification is good, and that the Tristubh and Jagatï stanzas are of a more ancient type than is found in mediaeval Sanskrit works. Regularly there is a secondary pause after the fourth or fifth foot, just as in Vaidic prosody; before the pause a short syllable has the value of a long one, in which case we find in the MSS. the quantity often marked by writing e. g. c o or c a for c a, but not always. This rule is unknown to Pingala 1, a sufficiënt proof, in my opinion, that the manual under his name is not old.
The Saddharmapundarïka is a composition bearing the traces of having been compiled from more than one source. This circumstance, combined with the work of revisers qualified to modify the text agreeably to the particular tenets, or may be crotchets of their school or sect, prevents us to go back to what may be called the very first text of the book. The editors of the present book could do no more but try to give a readable, not decidedly faultive text, leaving it to every reader to consult the copious various readings and choose for himself what he would prefer.
Before finishing this note, it behoves me to pay my hearty thanks to Professor S. d'Oldenburg, for his unwearied efforts to further in every- respect my work and for the many marks of kindness I have received from him in the course of years.
1 See A. Weber, Indische Studiën, VIII (1863) p. 371, ff.