Vedic Metre in its Historical Development c11.68

c11.68

(iii) On the other hand the text regularly gives *matí* as instrumental of *matí*, and there is no reason why this form should not be restored for *matyā* v 58 5d. In the compounds *-matí* is always required by the metre, namely *prámatí* (text *prámatyā*) i 53 5c, *sumatí* (text *sumatyā*) i 31 18d, v 42 4d, x 29 8d: and on this analogy it is quite probable that we should restore (in accordance with the metre) *asání* ii 14 2b, *návyasí* viii 51 3c, and *hiraṇyáyí* viii 1 32b, 78 2c.