Comments:KW 232, АПиПЯЯ 73, Martin 230. Mong. may be < Turk. (cf. Щербак 1997, 127). Cf. also Turk. *Keŕ 'bottom thief' (ЭСТЯ 5, 20), which Tekin (1979, 126) links with Mong. kerčire 'mud, silt': both may actually belong to the same Altaic root.
Comments:KW 232, Poppe 20, 115, Martin 240, ОСНЯ 1, 354. Mong. is hardly borrowed from Turk., despite Щербак 1997, 139. The root tends to contaminate with *gi̯óro q. v. The vowel reflex in Korean is somewhat unclear (*kark- or *kǝrk- would be expected), otherwise all evidence points to *k`irga. The relation of Manchu xergen 'line, scratch, notch' is not quite clear: the word is certainly borrowed (because of -rg-), but the source is unknown; equally unclear is the isolated Manchu qarqa- 'scratch with chopsticks, scrape on the fiyoo instrument' (compared with Kor. by Lee 1958, 113); perhaps in both cases we are dealing with loanwords Manchu < Kor.(?)
Comments:EAS 47, KW 223, ОСНЯ 1, 346, TMN 1, 471-472 ("altes indoeur. Lw." - ?), АПиПЯЯ 57-58, 292. Despite the latter source, Kor. hjǝ́ 'tongue' is to be separated - see a discussion in Miller 1998, who has (following Shiratori), instead compared Kor. kăro- (but, quite unjustly, separated PTM *xilŋü). The Korean reflex remains, however, questionable, since the word can also - following Martin 1996, 75 - be compared with Jpn. *kátár- 'tell', and thus derived from *k`ēro q. v.
Comments:KW 178, Poppe 17, 96 (Turk.-Mong.; but it should be noted that the TM forms (Evk. kalbiŋ etc., see ТМС 1, 365) < Mong. (cf. Doerfer MT 100), while the Kor. kalbi 'rib' does not belong here at all (see under *kéra), АПиПЯЯ 296, Дыбо 6, Мудрак Дисс. 193. In OJ koma ( < *kǝma) would be expected; the diphthong (OJ kwoma) is perhaps due to the influence of the medial cluster.
Comments:АПиПЯЯ 281. A Turk.-Tung. isogloss. The comparison seems quite likely, although the nature of the medial nasal is not quite clear (in TM one has perhaps to suppose a secondary assimilation to two neighbouring velars, or else an original velar suffix *k`i̯ōn(i)-gV).
Comments:In Turkic the root is rather hard to distinguish from *Karɨ < *gàrá 'arm', but the distinction still seems to be necessary. In *K(i)arɨĺ the expected long vowel (reconstructed because of *-ia- in TM after a fricative and high tone in Jpn.) was shortened probably under the influence of *Karɨ.