rtpots, retahimg a close rcsemManee to cacli other, aind to ihtme oftTie T9« 



Cabiitaries that have ri-majtii.'d in Jtiid near the priuiiir y ; b * le oi* th^ 

 M Asiatic trifcea. In thi.t woy I woiiM explain the pt- l uhar Ch n%.^« 

 elcmout ofHiraalttiiL', Cauciwian (preserved by the moiintaiiisj, ami Dm* 

 %nro-Ati3tralian, ivnd the secondary lilmaliic element of CfiiicJiJiian, 

 Braviro-Atwtralian tuid other Iftn^mages. The Enst Himalaic tribes pro- 

 bably ooeupied much of what is now ynstero Tibt-t and wiatern China ; wid 

 thdnuh the precise. line of their first wout hem miffrationB cjin hardly be 

 tnu'ed with certaintTT it in DM>rt rotiiiihUnt with Uie general characti-r (rf 

 the Mini- All !itn jjliiHsary, to infer that thi^ first deseetitled mta the lirah' 

 mapiitra bmm by (he routes aftenTurdH IbUowed by the co^iiJate Tilx to- 

 Biirnian tribt'sv and tliencc spread over tite (TaDgefcie valley, mixing with th« 

 prior Dravirisiiis, and, in the eour^pof agcs^elraiinatin;^ tlie Dravirian phy- 

 »ical eteiiic^ti thmi«?h retaitiing Dnivirian pronouns numerals &c. Of course 

 there may have been other more eastern roigratiuns, but the Mon-Anam 

 branch, which predominated and spniad everywhere in Ultraindia prior t9 

 tlie Tib' 'to- Bur man, had ifc* p»-imnry mmtJum home and nursery ui Ben- 

 gal or the Bhriwurtputra-GiMigetic valluy, for it* basis of Dravirian, and of 

 ft *JtH^idarv or corrupt dialect of Dravirian, could have beeu obtained 

 nowhere else/* 



Thfi niitne iei convenient in diatin}j:iii?*hing^ the varioos elements of 

 Aiintif^iAn efhnolc^. . Th*? latent! of the three formsttions of the Indian 

 province haw appropriated it^ only jreneral no me, which is nidically Hi- 

 luuJaic.* Thw ivae rendered it ueceB-sorv to adopt a second name fur that 

 ftirm.' ^on which would otherwise have bad the iirsi claim to the design*- 

 tiun uf Indian^— the Dravirian. A third U required for the titermediatd 

 great formation of northern India and Ultraindia. Tibetan mij^ht lie 

 made to iimlude tlie Indian and Tniu8;jttn;i'etic lanjyfuafrcs of the prt'jjer Ti- 

 betan typt^; butMoU'Anam kn* aaiive characters whieii cannot m con- 

 founded wiih tho«eof the more Scythoid Tibftan, and it ia mmi couTement 

 tou^e n difttiuL-t name for the formaUoD aa a whole. 



Deccttiber lH5(i, 



AdDITIOXB A>D COBOJSCTIOXB. 



ell. V. sec Z^Bronoons and Generic Farticles* 



Paffi' 1 . The cflTciilation of chancef^ here nacribed to Bowp, ia UnnPeTi'd. 

 Alluding to the hyiHrthes^i:? that frnjisHnft of laiiorna»T hud many distinct 

 trieina," he a:iy»^ that the very ronta, full or emf»ty, nnd all tlicir words* 

 IFhether rafmoHvUahic or poly8yllahie, mnat nced^ bo entirely diflerent."' 

 ** There ruiU' besides be some cauuid eoineitlences in real words * but the 

 law of comfufmtioii upfilied t-o the elements of aoinid frives a mathematieal 



}>root', thnt. with all allowances, tliiit cht»ric<j if les^a than one in a million 

 or the same cnrabination of fouuds SK?nifjinfr the aarne precise object." 

 My objeetiou!" to this position hafe been p-ently confirmed by my subse- 

 quent couiparirions of Chinese, Himaluic and Scrthic roots. The number 

 of the elenieufary -Hontidrt that entcreti into primitive lan^ruafre appean 

 to hiive hHen exceediriulv i»m;dl. The same roontwylhiluc roots (phonetic) 

 we repeated Hfiiuu ami litrain, and meet u« ir\ every clofw* »f words. Lilt© 

 all otlier arts, taufrna^re waa, in its eurUer stagies, rnde and narrow. Only 



, Siud (whetice Hiad, ludj i* a Uimaiuic root for river. 



