unknown in some Indo-European languages, Mr. Stenkonow considers quite possible, that the Indo-Aryan cerebrals developed quite independently, without there being any special inducing cause. Referring then to the phenomenon in the later Prākṛta speeches, that there is almost a wholesale change of dentals into cerebrals, the learned scholar offers a very reasonable suggestion which I quote in his own words:—
"The cerebral letters, however, form an essential feature of Dravidian phonology, and it therefore seems possible, that Dravidian influence has been at work, and at least given strength to a tendency which can, it is true, have taken its origin among the Aryans themselves."
It has not, however, been noticed by the philologists, that even though cerebral letters prevail very much in Dravidian speeches, these letters are never initials of genuine Dravidian words. No doubt, we observe this very peculiarity in the Vedic as well as in the earliest classical Sanskrit, but we notice that in later Sanskrit as well as in the Prākṛta speeches, there are many words, which though not onomatopoetic in origin, have cerebrals for initials. টঙ্ক (the top of the hill), ঠক্কুর or ঠাকুর (a word of respect), ডমরু (a musical instrument), and ঢুক (to signify entering into) are some examples. As India has been the home of diverse races of men, since remotest antiquity, it is difficult or rather unsafe to particularize definitely the influence of any special non-Aryan race, as the sole cause of any unusual linguistic change.
I must, however, notice in this connection, an important peculiarity of Bengali phonology, which has not to my knowledge, been noticed hitherto by any philologist. I have made it sufficiently clear in a previous lecture, that the people closely allied to the Dravidians, or rather who have to be presumed to be pure Dravidians, form the bulk
