^{page 751 note 1} A Comparative Grammar of the Dravidian or South-Indian Family of Languages, third edition (reprint of second), London, 1913, pp. ix ff.Google Scholar

^{page 751 note 2} See Zeitschrift für Indologie und Iranistik, vol. iii, pp. 81–112. Of the additions I could make to that paper one at least, viz. to the paragraph on numerals, will be found interesting: Tamil pattu, Telugu padi, etc., “ten” is evidently ═ Samoyede bet, büd, bü, etc.; and Tamil, Telugu (etc.) nūŕu “hundred” seems to correspond with Samoyede (Arch.) jur do. (< *ńr´u; cf. Tarn, nāl “day” with Samoy. jālea, jale, etc., do.). (Cf. 0. Donner, Samojedische Wörterverzeichnisse, reprint of 1932, pp. 35, 9, 8, 48, 97.)Google Scholar

^{page 751 note 3} Published in 1935, in the Atti del Congresso di Linguistica tenuto in Roma.

^{page 751 note 4} I use the term Uralian (═ Finno-Ugrian with Samoyede, or ancestor of both)in an indefinite sense, both for the sake of commodity and also to avoid pronouncing an opinion as to the special form or forms of speech with which we are concerned. W. v. Hevesy pleads for the Ugrian, Caldwell was inclined to give Finnish the preference, but we may have to go farther back and perhaps less far for the Mundā than for the Drāvida family. The following abbreviations will be met with: Ka. ═ Kanarese, Ma. ═ Malayālam, Ta. ═ Tamil, Te. ═ Telugu, Tu. ═ Tulu, Sa. ═Santali, Mu. ═ Mundārī, Cher. ═ Cheremisse, Esth. ═ Esthonian, Fi. ═ Finnish, Hu. ═ Hungarian (Magyar), Lp. ═ Lappish, Li. ═ Livish, Mord. ═ Mordvinian, Ost. ═ Ostiak, Perm. ═ Permian, Syr. ═ Syryenian, Veps. ═ Vepsish, Vog. ═ Vogul, Vot. ═ Votiak, Samoy. ═ Samoyede.

^{page 752 note 1} I translate from his excellent paper Zur frage nach der verwandtschaft der finnisch-ugrischen und samojedischen Sprachen (Helsinki, 1915), p. 26.

^{page 752 note 2} See JEAS., 1934, pp. 798 ff. This is a rejection of v. H.'s attempt as is also Sauvageot's review in SSL., 1932, pp. 180 ff., while Figulla (OLZ., 1934, col. 187 ff.) contends merely that the Mundā languages belong to, but not that they are connected with the Finno-Ugrian family.

^{page 753 note 1} In his review in Zeitschr. f. vgl. Sprachforschung, 1928, pp. 145 ff., written after three scholars (P. W. Schmidt, W. Porzig, and H. Güntert) had pronounced, though with a few words only, in favour of my attempt.

^{page 753 note 2} Nobody will accept such equations as L.'s Nos. 2, 8, 11, 14, 19, 23, 36-9, 44, 53, 57. They show his difficulty to reach the number sixty, while my paper has really about a hundred equations. It is not the same thing to draw up lists purposely based on superficial resemblance and such where etymologically impossible equations are sought to be excluded.

^{page 754 note 1} Does Sanskrit pippīlilca ∼ Votiak sippelikas (both ═ “ant”) also belong to this category? A Hindu was delighted to find in Germany the good Indian name Kistenmacher (Krishnamacharya). Of words with but two consonants, but missing in Littmann's list, at least English much, (cf. Gothic milcils) ∼ Spanish mucho (< Latin multus) deserves mentioning.

^{page 755 note 1} Here the rareness, in Dravidian, of the initial palatal nasal is also a point to be noticed (it has been mostly replaced by the dental nasal or y, or has altogether disappeared; cf. the nom. sing, of the first personal pronoun: Ta. yāń (lit.), nāń; Te. nēnu, ēnu; Ka. nānu, ānu, etc., all from ñāń, preserved in Ma. only, which must have been an ancient non-literary form also in Ta., etc.), as is the fact that both words so far as known, do not occur in the sphere of influence of the Mundā languages. As to the well-known vague use of animal names I may note from India that in Tamil nari “jackal” (i.e. “bawler”) is a name of the tiger also, and that in Kūrkū the dog is called chītā which is in reality a kind of leopard.

^{page 755 note 2} Being but quite imperfectly acquainted with Finno-Ugrian linguistics I have had to renounce aiming at a consistent phonetical rendering of the words adduced from that quarter (as I have also simply transliterated in the usual way the Drāvida and Mundā words), but have (as above) endeavoured to give the Finnish and Hungarian words in their modern orthography and have otherwise followed (with simplified spelling) Setälä and Szinnyei so far as they were available to me and, where not, O. Donner and occasionally v. Hevesy. The lists are also certainly not as complete as they could be. Note that j in Fgr. and Samoy. words is identical with the y in Drāv. words (and in Engl. year, etc.) and that the ń of the former is the same as Drāv. ñ.

^{page 756 note 1} For original guttural nasal (lost or replaced by , j, v), see Szinnyei; for Drāv., cf. n < ñ below No. 50 and, above, last footnote but one.

^{page 759 note 1} With loss, on both sides, of the initial sibilant which is preserved in the Votiak, Cheremisse, and Finnish word (sen, šün, suone). For the Dravidian, cf., e.g., the word for eyelash: Ma. cima a. ima, Tu. sim(m)e a. ime, Ta. only imai.

^{page 761 note 1} Unless some such evolution as kaiti > kaicci > kayyi > kāju > kāgu and …kayyi > keyye > khega > could have take place; cf. equation No. 28.

^{page 762 note 1} Suggested by the geographical position of Brāhūī (see on this now G. Morgenstierne in his Report on a Linguistic Mission to North-Western India, Oslo, 1932, pp. 5-6), but necessitated also, I believe, by the pre-Aryan “Mediterranean” component established for pre-historic India by anthropology and archaeology and equally required by certain linguistic considerations on which, however, I am not now prepared to speak.