^{page 349 note 1} The Piśāca Languages of North-Western India, R.A.S., 1906.

^{page 349 note 2} The Language of the Ashkun Kafirs, in Norsk Tidsskrift for Sprogvidenskap, ii, 1929.Google Scholar

^{page 349 note 3} Published by the R.A.S. in 1929.

^{page 350 note 1} Let me define what I mean by “ Prakrit”. I mean only the various Prakrits described by Sanskrit grammarians, and nothing else. As we shall see, Professor Konow uses the term “ Nbrth-Western Prakrit” in his work on the Kharōshṭhī Inscriptions. Here “ Prakrit ” is used in a different sense, but, as he has so named it, I follow him in using the term for this particular purpose. But this may not be taken as an expression of opinion on my part as to whether this North-Western Prakrit is strictly speaking a Prakrit like, say, Śaurasēnī or Māhārāṣṭtī, or whether we should look upon it as a form of speech allied to them, but differing from them in important points. Here I offer no opinion either way.

^{page 351 note 1} Professor Konow himself (p. xcv) lays stress on its relationship with Dardic and a perusal of the following pages will show how close it is.

^{page 351 note 2} It may be noted that Ṣiṇāa. and Kōhistānī occasionally agree more closely with Indo-Aryan languages than do the other Dardic forms of speech, e.g. in the change of at to tth. This is most evident in Kōhistānī, which is spoken on the Indian frontier, Lahndā being spoken immediately to its south. Kāšmīrī is in a different position. For centuries it has been subject to Indian literary influence, and it is now really a mixed language, Dardic and Indian forms appearing side by side.

^{page 352 note 1} In regard to thia “ spontaneous nasalization ”, see Turner, in JRAS. 1921, 381 ff.Google Scholar; Bloch, J. in Cinquantenaire de I'École Pratique des Hautes ÉtudesGoogle Scholar; and Grierson in JRAS. 1922, 381 ff.Google Scholar Bloch shows that there were traces of a somewhat simil nasalization in Vedic times. In the only Prakrit work written in Kasmīr with which I am acquainted,— the Mahârtha-manjarī of Mahêśvarânanda,-— every word that in Indian Prakrit would contain a double consonant has, in this dialect, a single consonant preceded by anusvāra. Thus the Skr. àtma- appears as aṃpa- (not appa-), Skr. kartā appears as kaṃtārō (not kattārō), Skr. nitya- appears as ṇiṃcamca- (not ṇicca-), and so hundreds of others. Regarding nẹndar, see, contra, Morgenstierne in “ Notes on Torwali” in Acta Orientalia, viii, 296.Google Scholar

^{page 355 note 1} Morgenstierne (Language of the Ashkun Kafirs, 221) connects tanu, etc., with Sanskrit tanu- self . As shown in my Tōrwālī Grammar, § 129, I prefer to connect these words with ātman-, Pr. atta-, appa-, sing. gen. attaṇō, appaṇō. In N.W. Prakrit, also, this word became appa-, atta- (Konow, cv).

^{page 356 note 1} Morgenstierne derives this from Skr. druṇā-, droṇa- (Lang, of the Ashkun Kafirs, 254).

^{page 356 note 2} Notes on the Grammar of Old Western Rājasthānī, § 30.

^{page 358 note 1} There does not appear to be any trace of this treatment in N.W. Prakrit. We may perhaps, however, note the fact that, in it, intervocalic d, ḍh, t, and d are often written ḍr, ḍhr, tr, and dr, respectively. Konow (page c) suggests that this was done to indicate a fricative sound.

^{page 359 note 1} It is perhaps worth noting that, in Chinese, the sound which in Southern Mandarin is pronounced like an English r, is in Pekin pronounced as z (Mateer, xviii).

^{page 360 note 1} These are Bhadrawāhi (Bhad.), Bhalēsē (Bhal.), and Pangī (Pan.). Other Western Pahāṛī dialects which do not fall under this group are Curāhī, Gādī, and Kujuī (Cur., Gā., Kul.). These are quoted for purposes of comparison. It will be noted that in these the letter r is inserted, as in other Dardic languages mentioned above.

^{page 362 note 1} This form would exclude the derivation from puṣya-, as has been suggested above. Cf. N.W. Brakrit puṣa- (Konow, ex).

^{page 364 note 1} Cf. the doubtful sign for this conjunct, a modification of that for ch, in N.W. Prakrit (Konow, ex).

^{page 365 note 1} For the inserted n see p. 352.

^{page 365 note 2} Taken from Lorimer, D. L. R.'s Phonetics of the Gilgit dialect of Shina, § 65Google Scholar; JRAS. 1924, p. 182.Google Scholar