Comments:EAS 79, SKE 155, Poppe 35, 138. The Turkic form presents biggest problems: it must be explained as a result of dissimilation and contraction - *bū(n)ŕ (cf. obvious traces of nasalization in reflexes) < *būĺ(V)ŕ < *mi̯ū́ĺi-ŕV or even *mi̯ūĺi-dVŕV (cf. Mong. *möldür, Evk. melder-); early loss of *-i- would then also account for the back vowel reflex. All other forms are more or less plausibly united under the protoform *mi̯ū́ĺi.
Comments:EAS 79, 147, KW 267, Poppe 35, Lee 1958, 115, Martin 246, ОСНЯ 2, 61, Murayama 1962, 109, Menges 1984, 277-278, АПиПЯЯ 28-29, 69, 86, 278, Rozycki 160. Turkic has preserved the root only within the archaic compound *jag-mur 'rain'. Tone in Jpn. is irregular (probably because of reduction and the position within a compound; suffixless *mi is also attested in OJ, but its accent is unknown); loss of resonant presupposes a suffixed form: *mí < *mi̯ū̀r(i)-gV (cf. Manchu mū-ke).
Comments:Lee 1958, 115 (Kor.-TM), Whitman 1985, 183, 237 (Kor.-Jpn.). Despite SKE 139, hardly a loan in Manchu from Kor. In Jpn. we have to assume *múná < *mura-n (cf. Turk. buluŋ); in Turkic - a secondary narrowing (*buluŋ < *boluŋ).
Comments:EAS 79, 141. Somewhat dubious as a PA etymon: the Turkic reflex is questionable, the Tungus forms may be borrowed from Mong., see TMN 1, 507-508, Doerfer MT 25, Rozycki 159 (although the vocalism is not quite satisfactory for a loan); however, the Kor. word is hard to explain as a Mong. borrowing.
Comments:EAS 98, Poppe 34, 67 (Turk.-Tung.), АПиПЯЯ 53, 280, Дыбо 5, Лексика 234. In Japanese, unfortunately, only Ryukyu forms are attested, which probably underwent an influence of the PJ root *nǝmp- 'to stretch, lengthen'; still, the origin of the Ryukyu stem from this PA root seems probable. The Kor. form is morphologically = PT *bojnak < *mojńo-k`V; cf. also MKor. mǝ̀ŋ-'ǝ́i 'yoke, harness' with preservation of nasal. The TM form also reflects velar suffixation (*mojn-ga-).
Comments:OJ mapa-r- 'turn round' may belong here if it goes back to < *moŋ-bV (cf. Tungus forms), although the vocalism is not quite right; cf. also Mong. (dissimilation?) bömbüger 'round', see EAS 116.