i42 



ETHVOtOGT OF TRl nmO-PAClFIC IStAJfOS. 



Each of these ia n fwculiar Tn odification of ihp prima nr form ; and the devia- 

 tions from it and Inim each other are so l onsiderable. that they mmt be* 

 lonp to an ag-e loiijf Riibaequeiit to that of the exclusive currencr of gTit* 

 turn! form!<and when ilie systf m hsid become prirt!yconfrefed. Whetherthese 

 forms ari«f inated in ojm dialfct or in aevend, their romhiiintion in a sing-la 

 current HVsf-em ciirmnot belonjf to the ear? if r eras of the formation. The 

 form in .1, mm, mm, Ike. must have long" been tlie unit exclusively cur- 

 l-ent in ii Chinese syHf^ra. Tlie form in 4 im a later for in, as it wants the 

 tinal eoiit<onant j the iiivtialj however, having a atron;;er Hf*piirate. The 

 current form wiis probably preeedfd by one closer in the vowel to 3, liko 

 thnt iireserved in Thochu^ IJorfm aitd »ome Southern dia!eet«. The '''tiU 

 later liquid a may have originatetl in a we.stern Cliinei»e dialect which had 

 a similar form, lum, liim» Imitr, lanjtr, Ink, bk, in ih* current unit, and wad 

 associated with tho.-4e ancient 8i fan or East Himnlaic dialectn which pos- 

 Besaed a similar unit and prave it to the South, but it is more coiisiaten* 

 with the other forms of the ayatemtoreferitdireetly tosumjHuk&c. What* 

 ever may he the history of th'e prodnetion of the Chinese numeral Bystem 

 which ultimately had these forms fixed in its 3, 4 and 6, it is clear thHt, after 

 they were so tiled, it l>ecarae the mont influential p^ystem fir^t in China and 

 Tibet, andthen^ throni^h the great Tibetan mipntio'n, in the South also. The 

 prevalent 4 ami anpear to have always been concreted and aubstimtial 

 names in the Himnlaic province, the connection between them and the 

 unit having- been losit before they were received from Chinese. The cun ent 

 form of at, on the other hand, appearn to have retained its unit power, 

 after this late Chinese |ihaiie became that of the Himataic province ulm, 

 aa it undoubtedly remanied current with thi-^ power in 7 and 10, and pro- 

 bably m 1 and 8 also. The slender form of the dual current in fhe Chi - 

 nese 2 in a eontrflctetl form is ao widely prevalent in tlie Himalaie pruvincd 

 in 2 and 7, that it must be tiiwociated, in the form nging. nerik kc, with 

 the later Chino-Himalaieform of 3, 4 and 6. The vocalic but broad 2 of 

 6 — which is almo!it imi verbal — obviously belong to the wime j>hase. 

 The current Chinese 1 and 7 apjiear to be later in form. The labia] 8 

 end 100 were probably not po?<!*es>iedl by the western Chinese dialects, 

 until after the period when they g-ave the secondary forms of the unit to 

 'Tibet. They have no connection with the other Chinese numerals aud 

 no representiitivesin the purer Ilimalaicsystemg, It is prohithle there- 

 fore that this unit was preaerved in a northern Chinese dialect — jierhaps 

 the Kwan-hwa itself— which became that of the predominant Chines* 

 nation. The f I imalaic HYiitems present many examples of a similjir per- 

 eistency of a native or older form. Thus the Kwan-hwa system, in a 

 late fomi and embracing thirs labial 8, has been widely spread over Ul- 

 traindia by 1 he conquerintr I>au tribes, but the 1 and 'J are wantinfr, in 

 the Lau dialect, native names tttkint^- their place. The northern dialectfi 

 have the Chtiieae name in 10t>, but it hna not established itiieli' iu tb« 

 •ouihern. * 



* As the forms of the same root varied from era to era in different 

 numbers and in diflVreiit dialet:ts, and as these forms did not attain the 



rittk of fla»ociat.ng some forms that may never have bten current togetkw 



