Comments:Poppe 28 (Mong.-Tung.), Lee 1958, 113, АПиПЯЯ 19, 33, 292, Rozycki 128. In Mong. one would rather expect *č- in front of -i- as the result of palatalization of *t-; preservation of the voiced consonant may be due to the interaction of expected variants *čiwrin ~ *duirin. The root is no doubt a very archaic one; it is probably attested also in several common Altaic derivatives. Cf. PTM *du-dgu 'couple, spouses' = OJ t(w)otug- 'to marry' (*tub-tu-); PM *dab-ku- 'twice, double, layer' ( > Chag. tapqur, Evk. dapkur etc., see Poppe 1966, 195, TMN 2, 429, Doerfer MT 101), also reflecting a -i̯-less form like *tub-k`V-; PTM *ǯūpti (*ǯub(i)-pti) id.