BTSWoiioar or ths indo. pacific rttAwn?* 



of thft Ijower 'Sx^er ns 1, in the compound ptpfix hoo»-war- of 2, 3 and 4- 



B. ib liifilnm. Thir? imiy Lw u IVu-ra ot" the labial unit iu3 in 3 mih, but 

 it is probably only the arclmii: la binl postfix os in 3 and 8, thts root being- 

 elided. 



C. po-niit pu-no, Woratta, Woliiititu in 6 (1^ ,1), 7 (2, 5), 8 (^3, 5)^ 

 and fl(4, 5), fu-n Yiinptro in f). This is tlie labial unit, but it may 

 luive been more )mra<*diatHy derived froui tbe sliP-meii, abe-raon, form of 

 il hy contraction. In thf* h>is.si (i iip:om-pu-n (r>, 1, ); it lias the t^ame fonn. 

 Ho f>u-l, 1, llullufii. It also ocrui:^ iu 7 of Ki-KainW nion-sa {.>, 1 ^jOud 

 Ki-.Nika fiiu^!--ube (Ti, I), and, with n lorn .simihir to tbe latter, in the 

 Tumuli tiiu-ari-an W (10, 1 ) jind ftui^-eii 10. The Darfur tvi-ng of 10 

 (in'20, ^) &.C.) i:* tbe tiaine tern K 



U. lima MalatTdHv, This terra does nut apnetir to )te a 8e«iitct-Afn- 

 can unit. In ^UoJieaia it w iifieJ butb as 'j and itt« a term for "Imrid", 

 tind the ^tter wari jnnbably itn primary njeunin^ in JFahipusy aluo. It is 

 a. E. Asian tertn ; arul in' Zirubsan it ik still current fur ''ting'er" Jiiid "toe*^ 

 ItnUj Uenii, Liam^ leutbu kc. 



Fftrcif^n A0rnitifA. 



Tlie various fom^'^ of tlu" most enBamrm ?4eniitn-A frirfln term, klio-m, 

 *n-m, tu-m, ha-ina, sa-iubo are N. K. j\j*ittUL ko-m, Samoiede so— mb, 

 fio-l>o, Htt-ba. wi-in, aad Monf^uliiui tu-)>uu, ta-bu, in ^\bifb tbe labial ia on 

 firebaic qualitlve ]»mH1x. Fonus witJi final u ti]m mxur, tung-, sun &f* 

 In tbe itrchaie Scytb<>-t "liine^ie or K. Amni sVi^em ft nitmectian is ai-'O 

 olwervrtWe Iw^tween tiiese wames for f> and some of thot^ for ii. Hie 

 Hamoiede and Miai^otitm sa-m, ^*u-bij ta-bun cf 7), uirr*-es« with tbe 3 of 

 Chijiefie sa-ni, and ("4itu.'a.4att sa-nii, .nu-mi ; atid tbt* Tun<rnsittU iuu^-i/u, 

 fon^-iHt, Kamsdiat kail tou-w// 6cv. Cm H, &.e, with tjie ;i (if Cbinei«? sa-n, 

 Yeiii-seian tou^-y/w, douiT'-'w, kiiu ilfjuffoliuu kol-m, kor- fm &c. The 

 Ijifrian and Tnrkish a, vi-.s, vi-t, vii-si, UiM, bi'sb, bi-l-?A- &:c, ia .*J in 

 Turkish wi-we &e,, Jajxtne^ft nn-i--iu, Tbe {-'ancQsiiin i;bu-ba, ebu-tbi, 

 chu-( is similar to the H of Cauciman elii-ba, yu-mi,and of ll{4"rian juhu-d- 

 entf (in f\ ebn-r), kn-m kc. The nilnlajit unit of 5 in Jajwinerte, is^ u like 

 form to the Ejryptian Jkv. {u) and the tioubJe form {J'. ), Tlie labial imit 

 ( C) is Used as '> m t grian aj ul I*]uskariflii ; and tbe iudo-European terna 

 may jMiHftibly be tbe Hfime, 



I'be adjiiJeut Cautasirtn ebn-tbi, chu-t, TTo-ehu-si Georg-., t'chu Circ., 

 chu-ba Awur are analo^ron*! to iSeinitd-.U'rit'an forms. 



Olfs. m the dktrihutioH of the term* 



5 presents a repetition ol'tm-o of tlu? root* for 3, the sibilant^ pruttura! &c. 

 with the labial, and with the hquid, -*-m : s-n. The form of theaibilant 

 &c. with tlie labia! postfix i.-* evidently very archaic. It ib pres*erved in 

 other numerals also, and the coariiate CaucaHiau, Scytbic and N, E, Asian 

 terms are proof* of itfl hiffli antiquity. Tbe Semitic tina! s ( kha-ra-jta &c- ), 

 like the finiil » of (p. Iti), appear.'* to have been addeiT in the Aaiatic 

 bnmoh after the archaic Semitic sy.ntem had been earned to Africa. 



The archaic or prt'-historic u, form of the Semitic unit -appears to 

 have been conveyed by a Jlimyarjtic dialect to Afiica, and to have been 

 spread from the Nilotic provirsce to the Atlautic, over the Northern and 

 Middle rtit£it>ti^. It probably [trecfded thr later Semitic a, i, forma in Oal- 

 la and Zimbiau aho. See the remarks on the ligyjitian '^i, ante p. 4. Since 

 ihit page was printed the fuU^Babyloiiiaii seiiea' ot numerals, as given by 



