W KTHS'OtOCV OF TETB TN 00 PACIFIC ISLAMXSk 



• 



is more immpdiotely connpctpd with the th&n with the Tibetic- 

 In thf ChiriPSP, Scythic, Indo-European, Caucasian, Spniitic and AfrioiQ 

 fiimiliet* the prevalent iiaines for the Conr contain ditfi^rciit roots. The 

 Tih'tan name is therefore indppendent. The Fhi loeh, loh is a variety of an 

 archaic form applied to the Jlnrae in I'urro-Tilwtaii, and to the /JtYT in 

 Chinese, and prohably connected with Tibetan forms for the Com only 

 t'hrough the derivation of both from the Pame archaic Ai»iatic ina«ciiline 

 root. 



Ra ffottt ( ra-mtt./'., ra-Ua m.) appears to he the .Miime root. The word 

 ia only found in Bhotian. It is not C'hinese. it seems clttar therefore that 

 Ift was (>ri?rhmlly derived from chan-w, a niasc. form of the root chang-. 



Tho li(|uid root ii? ap[ilied to the //onf. in i^eraito- Miotic lanffUHfres, /w- 

 run Malii-ahj, //-i-on fiani, Aw-rar, rt, re Cialla, ilia Danakil, arre ISaumsiIi, 

 oraha Biilaritlfl, [eru, eri, ere LsoBina kc. k probably a contraction of thCf 

 com. African r-wure, wuli, ( -jtrm &e.]; on-e l^auniuli. But tJjese names do 

 not ajjpear to have any direct coniieetion with the Tihetan. 



Rd Ilm'M' ii* Ugriun, and the pitinral i.H preserved in Ostiak lo^, 

 loch, low, (in other Ugrian lang^uarres lo, In, lyn; wol, wyl, wal; lown. 

 The Uistiak jcutturiil form correspnds not only with the Siifan rd, but with 

 the CMne&e Ink, lu, Id deer. In E, Tibet aiid Sihng it is also u.sed as m 

 gt»iieric vocable for sh'C/fj two kinds of wliich are knowji ha-luk and 

 j)e-Iuk tllodo:i4on J. B. A. 8. %vi, 1008), The root may also lie contiiined 

 m the ?liiinoiede bo-ra, Koria m<tl, mar, Tungruaian mo-ro-/i, rau-ri-H, mu- 

 ri-/, ^lougolian mu-ri-/^j mtj-ri— thid liquid fonn correspondi* with the 

 Horpa rhi. But in tlm prevalent Tartar term the liquid 1;^ either one of 

 the common' finals taken by naonasylJabic roots in theiScythio family or it 

 it* II nGX |H.w:lix, tlie root heijig mo, mti, cognate nnth the Chinese raa. 

 In the U;^man and Turkish ala-sha, Turkisir lo-sha and OuucaMnn uloh 

 ( Mirtjejiari ), it nuiy be a jtubstantive root, m in the I'^aii l» &e. 



From the occnn-ence of the tkiuid as a prelix or postfix in the human sex 

 names and in several namci* of aiumala in Sc^-thic pfloiwaries, it is probably 

 servile and masculine, or wan so orij^dnally. Jt/wtFin al-m&z^ «/-mux, h- 

 grian /o-niini, ?'W-golo:i; Hit^fkatd Fin oZ-raa, M7-naa, rfe-M-man, Mong-oliaa 

 ere, Turkwh ire, eri, er, ir, er-kek, n-.in, ar-ini : Cat tr-my-shak, tr-gek- 

 mvi, Turki.sh (.<o «/^/-p-sliik, ata binn<r Jw/A/ v) ; Dt/ff koi-rw, koi-r^, koi-m 

 Ein, «/-Lschip, fl/-tihip; ?^-t8eLa]) (alfio tschip &c.) Yeniseiiin j Ox or-<^ol 

 TutiiiTi-^ian. 



II wc consider the kbial aa the substantive root in the Seytiu<: 

 mo-ro-rt, mu-ri-w> and ro, ri ai* masc. servilea, winch in the l?^riaii 

 ojr &iv, have iiecome Miibstan lives, the fame linew must be taken ol the 

 Tibetan bo-i"d, rd, which are thuj* ]iluwd in the same class with pa-iff, pha ; 

 chanfr-J'fl, ra ; chu-/«j /*i-chi, rhu; and perhaps r-ta, if the Bhotian }ire- 

 lixuul r-, a-, i/-^ Imj, m is probable, contractions of the origi- 



nally mniit:uline deiinitives la, Iki, si kc. As the form ro id no lonffor cur- 

 rent in the Houthem Scythic !unj5:na{re3, it must belong to the archaic U- 

 prian basis of Tibetinj, like a lurj^-'c proportion of the other vocabloj*. la 

 the Ultmindian vocabularies tho ma-ujuline liuuid root retains the gruttu- 

 ral thud in several diulccta. Indeed all the EJcythic forms arc foujid^— ri, rin, 

 mi, log, lok 



TIT. Tlie sibilant in its anpliciition to the cat appears to he archaic, 

 native* The root i^ founu in ticy thic vocabuJaries for the Muusc {t\ ^ 



