00 BTHKOLOGT OF THS tN-DO-PACIFlC I8Li.Nl>B. 



The first direction which our search for facta that may help to 

 clear up the obscurities of the South Dravirian system, shoulfl 

 naturally take, ia to the Koi, GaQgetico-UItraindiaii and Aaoucjjiaii 

 Bjatema. There has evidently been «onie displacement and ph<j- 

 uetic modificatioo of roots in the S, Dravirian systeni, and in some 

 points the eorrectness of our analysis cannot be considered as iiilly 

 established by that amount of mutual elueidation which the S. 

 Dravirian dialects themselvea afford. 



The Kol dialects preaen-e a somewhat different numeral system. 

 It appears to have prevaUcd in Ultinindia also prior to the intro- 

 duction of the Tibetan and Sifan modification of the Chinese, for 

 it ia now retained— partially blended with the latter— in those 

 Ultraindian hmguagrea which in pronouns and other words, hare 

 the Btrongeat glossarial affinities to Kol.* A full list of tlie varia- 

 tions which the roots undergo, with Bome remarks on their distri- 

 bution and the probable course of their diiTusion, will be found in 

 the next chapter. 



The Tindyan, like the Soutli Dravirion, numerals po&tfij a pos- 

 sessive definitive, but in place of varying in diflerent terms as it 

 does, to a greater or leas extent, in South Bra>irinn,it is uniformly 

 -ia or -^a (with a few slight phonetic changes and contractions 



• Tlie mafitime posUfcm and haliita of the Man or Peguans. the cTldencfs nf 

 their haviiiK hfcn at one time the chief traifera to the rasfwant on fhe Bay of Bmnal, 

 and ot their haviir^ p^eaily iiiflutmcKil ihu olher UliraindiHH, thi; l*eiiiiiftiilar umi 

 severiil of the Indoriiisiaii nicen, wiih the uiidouhted ispr*"aii af \ in [yft-rJlfraiii(iiRri 

 vocables throu^ih their instriiment iUtY to the cast und south, M inu tu fsuriuif<«; that 

 theworda conimuri lo the Al on- A nam and th« Kol vocahulariiis, had heen earned Ly 

 the Mona from Ultmindia to thei (jantietic basin, rafher tlinn '>v un hihind tnh« 

 It'Xe (he Kold to Ultraindhi, and tins sunuiiM^ ajipGiired to b« sH"eTt;:th> n< d by the 

 peculiarilies of Kol couipared wiilj South Dnivinan. The *2nd jTononn in parti- 

 cular, with tht: Iijvver tirins of the nuHjcrnl system, appusrtrd lo have a Llmiacter 

 comj ielely fort'igo, Amoii'^sl the uii-ccilunt'oud woniti cuiiim' n to Kol and Mon- 

 Anaui ¥ocuhiihiri«;a sniiic were, beyond all doubt, ncn-Dravirian and ol Ultraindian 

 and Tibtao-Uitrsiindiuii origin. Jn the 1 n( riHluctorj' Note to l*art II (an^« vol. 

 vl, p. ti&8) 1 tlicri'fore rt-ioarked that Hic vocables of i he Moii-Anain formation 

 were not only found in Gangetico-UUrain Jiiut Ian,u'iges "but to a rentarkabh! ex- 

 tant £n the Kol dialects, proviniiUiattlioPc;^ formation eiti brum! Lower Jit ngiil and 

 a portion of the Vindyas, alihnntih the i>raviriHn basis was pietK'VVHj in tht; lan- 

 guages uf the letter" ; and In SfC. tHvid. vli, p. 2(>0) it h ioid ' the phonetic basis of 

 the laii^fuiige [Kol] and many pariicle-H and words are Dravirian, but the pronenn-*., 

 WBVernJ ol thti nuuuimla and a larK^ portion oi the words are JUon-Annm." A( the 

 mine time, the mBuciice of the bravlrian pronoroinAl :<y!>tem in Uitraindla v/m in 

 •ttveral jdiu'ea remarked. A more miunti- exaniination nf ihe pronominal and 

 numeral elotuenis of IJravJrian, ot the lordgn eontiiieiital aflinkiii;* tbelornvat ion, 

 and of ita remnants iu Aflonesia, with lljc refeienCiC of tiie KujI iind pronoun like 

 tUe Ut to Dravirian, have eatidil^d me that vvl tie Kul, owin^ to iti^ position, hua 

 buen iiifliiei)ced b.> the libi-to- Chinese lorinations, as the race itaelt haa l.y Th« 

 Tibeto- Ultraindian, the atliiiitiea betwe*^n it tnid the Mon- Anum vocabulariea urr; 

 mainly of prhuary Kol origin. 'Ihe mosit probable con^luMun is that the KoU ar« 



