^{page 304 note 3} On p. 90, he says “ the inquiries as to the meaning of ‘ cerebral’ and the exact place of articulation are so subordinate as to be almost irrelevant”.

^{page 306 note 1} See Wackernagel. Altind. Gr., p. 164.

^{page 306 note 2} Cf. Macdonell, Sanskrit Literature, 38. “They contain a number of minute observations, such as have only been made over again by the phoneticians of the present day in Europe.”

^{page 306 note 3} Cf. Whitney's Sanskrit Grammar, § 45. “The lingual mutes are by all the native authorities defined as uttered with the tip of the tongue turned up and drawn back into the dome of the palate.”

^{page 307 note 1} I could easily produce numerous passages from modern Hindī grammars, written by men whose native language it is, in corroboration of my assertion ; but I should be met by the statement that the descriptions of speech sounds in standard grammars are out of date and cannot be appealed to. This method of disposing of inconvenient evidence has, I admit, the advantage of simplicity. On the other hand, the description of the position of the tongue in uttering a sound of one's native language does not seem to me to be “ an extraordinarily difficult matter ” (Journal, 1925, p. 88) for a serious scholar who has been thoroughly trained in an indigenous system of phonetics of great age. I therefore here give—as an example taken at random from the first Hindī grammar that comes to my hand—the Bhāṣātaltva-bōdhinī, p. 4—“ , ṭa, ṭha, ḍa, ḍha, ṇa, ra, and ṣa are called mūrdhanya because they are sounded by applying the tongue to the mūrdhan, that is to say to a place higher even than the hard palate ” . To anyone who knows who Paṇḍit Rām Jasan, the author of this book, was, and what a master he was of his own language, the suggestion that he was unable to state correctly what he did with his tongue in uttering such a sound as that indicated by ṭ savours of absurdity. The word tālu, though translated in the dictionaries by “ palate ”, means“. the hard palate ”, as opposed to the śuṇḍikā, or soft palate including the uvula.

^{page 309 note 1} It will be observed that Dr. Bailey gives three different, and apparently mutually inconsistent, definitions of a cerebral sound. In one place he says that the point of articulation is anywhere (his italics) on the hard palate behind the teeth-ridge. Again he says that it is never “ further back than the middle of the hard palate ”. Again he says that the tip of the tongue is put against the soft palate. These can hardly all be right, and are the more difficult to reconcile as Dr. Bailey has used cerebrals, all his life and cannot now confuse them with other sounds (Journal, 1925, p. 93).

^{page 310 note 1} Bull. SOS., vol. ii, p. 5.

^{page 310 note 2} This word accordingly is not used here in the meaning given by Dr. Bailey. It is evidently intended as the equivalent of “ cerebral”.

^{page 311 note 1} Besides two statements of Professor Bloch, Dr. Bailey quotes from writings of Professor Daniel Jones, Mr. Lloyd James, Mr. Sutton Page, and Professor S. K. Chatterji; but as he gives no references, I am unable to examine any of the passages cited by him or to test them by their contexts.

^{page 312 note 1} As examples of the opinions of learned Indians who have paid special attention to the point in question, I may refer to two modern books coming from widely distant parts of India, both of which are obtainable in this country. One is Bhāṣā-vijñān, p. 168, by Mr. Śyām-sundar Dās, who is a Fellow of the Benares University, where he occupies the chair of Hindī, and honorary secretary of the Nāgarī Pracāriṇī Sabhā. The other is Gujarātī Language and Literature, pp. 93 and 116, by Mr. N. B. Divatia. This book ia published by the Bombay University as a portion of the Wilson Philological Lectures delivered in 1915–16. Mr. Divatia's remarks on p. 94 on the European pronunciation of r and rh are also interesting. I may also quote the following from Rāma-karṇa's Hindī Vyākaraṇ, p. 10 : and again (p. 13) The author of this lives in Rājputānā, but almost the same words will be found in Śri-lāl's Bhāṣā-candrôday, which is a Hindi grammar published by the Government in Allahabad.

For the classification of r as cerebral, see, for instance, Dāmodara Śāstrī, Bhāṣâdarsa-bālavyākaraṇ, p. 9; Mahādeva-prasāda, Bhāṣā-vyākaraṇ-sār, p. 8; Sudhākara Dvivēdi, Hindī Bhāṣā-kā Vyākaraṇ, p. 5.

For the benefit of those who are not familiar with the work done by Indian phoneticians, I may add that they made a further distinction between cerebral sounds and the sound indicated by ṛ. The latter, they say, is dvisprṣṭa, i.e. having two points of contact (cf. e.g. Sudhākara Dvivēdī loc. cit.). The two points are, of course, one far back on the palate, and the other behind the lower teeth, occurring at the end of the “ flop ” of the tongue.