^{page 419 note 1} I do not believe in the derivation of the infinitive suffix ṇ from -anīyam. The suffix aṇa, aṇaũ in Apabhraṁśa (Hemacandra, iv, 411) certainly belongs to an ordinary verbal noun in ana, and I am unable to explain the phonetical changes and the development of the sense of the form which must be supposed when adopting the derivation from anīyam. The suffix anīya becomes aṇijja in Māhārāṣṭrī, and Marāṭhī agrees with that dialect in the formation of passive forms. Hindī and especially Braj, on the other hand, show some points of connection with Śaurasenī (compare, for instance, kiyau, done, with Śaurasenī kida, where Māhārāṣṭrī has kaa), and anīya in Śaurasenī becomes aṇīa. I therefore think it probable that an old verbal noun in aṇaa, corresponding to the Apabhraṁśa infinitive in aṇaũ, has in Hindī been confounded with a form derived from the participle of necessity in anīya. Such a supposition would explain the fact that the Hindī form in nā is used both as an infinitive and as a future participle passive. In Marāṭhī, however, the form in ṇ is a pure verbal noun. And its derivation from anīya is, I think, phonetically impossible. A suffix -aṇaaṁ, on the other hand, must become Marāṭhī ṇ , Braj naũ, and so forth. Compare the analogous development of the past participle passive.