KT«N&LOOV OF THE I \ DO-F»AC( FIG ISLANDf. 



of Dravjl'ian forma oocixr in the Milchimang and Limhu irt^fi 

 (idf'iilical witli Kt*! snifl Mun-Anarn furmp), Garo iiing anri Sin^- 

 pho i (bin)] PI.), MikTr nf^ Na^stiiriir Nai^fH iiyi, Kbnri ni and 

 the Na^ forms ncliced in a fjrcvioiia jtatje, * The A rifjamia 

 [Manvflk abo] un<\ tlie Gnnpolif'o-UIfrniniltan ang (Bodo, Gara), 

 anjlka Kirauti are prohahly Edst Tihi Tan, The 2nd pronoun k 

 more 8troii«jly distiiiguishrd in ihn two eys^Mns by its radical vo^vel, 

 which in Draviro-AustraMan k i aa in Cbinene, while irv Tiboto- 

 Ultraindian it U a, as In some of th« archHic Scythic forms. The 

 Uliraindian nicudierd of the Tibeto-UItraindian fitmWy show other 

 Drtvirian affinities in thtnr pronominal sysTenis besides the occa- 

 eiohid adoption or retention of Indiun forms of the roots. Dravi- 

 rian phindii, pofnessiveB and othvr particles oecnr in eeveral lang- 

 »ag( F, Bodo, Dhimal, Naga &c (see chap. IV.) Not only the 

 •oninion Dravii ian plumk in 1 &c are found, but, as we have seen, 

 the* pruTiominid m. 



Amongst the primary ^ffinitiiss of the S* E, Asian languages 

 and DraviHnn raHy be inebided the plural m and the po^sej^sive 

 in i, ni &c. The formeris Chinese -mtin, -men, -me, -niei, -pei and 

 the latter l« Tibetan (yi)» Manyak (i), Bunnan (i), Limbu (in), 

 Bodo and Garo ( ni), as well as Scyihic, Semitico-Lihyan (i ) Zim- 

 bfan (i) kc. The Chinese iraita in the Himalayan and Uttrain* 

 dian languages present great difficulttes. Some are of compara- 

 tively j-ecent East Tibetan origin and in Uttmindia even more 

 modern. Others appear to belong to a counectioa as archaic aiS 

 that between Auj?tralian and Chinese roofs. 



The Draviro-AustraJiim or arcliaic Ludo-Asoaesiau pronijjalom 

 system with ita numerous distinct eieniciita auti combinations, 

 appears to be more ancient or Icsis impaired tban most of tlieays- 

 tema of other hannonic formations of the Old World. From ita 

 general structure it must be considered aa cognate with proto- 

 Scji^liic or Scythico- American. It is richer tJiau Scythic, which 

 has neither seiual forms nor any plurals save the Ordinary 

 generic ones, with the absolute " we" (formed aa in Dravirian^, 

 idthough tlie Scythic power of comhimng such elements as the 

 formation poasesctes ia ainiilar to the Draviro-Australiau, and the 

 position of the subordinate definitives is the same. In some of the 

 *[BrahutL] 



