Notes: A Darg.-WC isogloss (see Mészarós 1934, 271, Abdokov 1983, 117). Phonetically and semantically quite plausible, thus the PNC antiquity is possible. But a similar root exists in Altaic (Turkic *takɨku / *tagɨku 'hen, cock', Mong. takija etc.), and an early loan is not excluded (note an obvious late loan: Darg. Tsud. t:awqa which stands quite apart from all other Darg. forms < Turkish tavuq).
Notes: Reconstructed for the PEC level. The vocalism is hard to reconstruct in a trisyllabic root with some irregular phonetic changes (due to the root's expressive nature).
It is interesting to note Azer. dälä 'squirrel' (possibly < *däläg), which has no Turk. etymology and thus may be borrowed from an EC source (the Azer. word, in its turn, is borrowed in Kryz. dälä, Khin. delä 'squirrel').
Notes: This root is probably reflected also in HU: Hurr. taɣǝ 'man, male' (see Diakonoff-Starostin 1986, 26). There is not enough information to reconstruct the vocalism. *-r- is obligatory in the reconstruction to account for the preservation of uvular *-q̇ʷ- ( > Av. -ʕ-) in Inlaut; without a previous liquid *-q̇ʷ- regularly yields -ḳ-.
Notes: Reconstructed for the PEC level. The comparison seems probable enough (although we would rather expect *t:arz in Lak.; lack of spirantization is probably due to an early assimilation: *dwälc̣_V > *dwälc̣_wV).
Notes: An expressive (but obviously archaic) root. Irregular is -b- in PTs (and -b- in Tab.): -m- should be expected. It is not quite clear, whether the protoform should be *dwānʔV (with transfer of labialisation to -n- > -m- in some languages) or *dāmʔV (with a reverse process).
Notes: Reconstructed for the PEC level. The form *dwirχE, reflected in PD and PL, probably goes back to an earlier *u_V-dirχE (with the 1st class prefix *uV-); cf. PHU *wutqi,*witēqi 'son' reflected in Hurr. futqi, fitēqi (see Diakonoff-Starostin 1986, 20).
Notes: Except for metathesis in PTs (*χ:ɨda < *dɨχ:a) correspondences are regular. PN *tχi is a reduction < *tŭχi; in PWC there occurred an assimilation (*tʷǝχ́V > *tʷǝχ́ʷV).
Notes: The comparison between EC and WC forms seems quite plausible both phonetically and semantically. Labialisation of *-q̇w- is suggested by PD *-ḳ- (but it may be secondary, due to progressive labialisation). There is also a possibility of a reflex in Lak.: cf. Lak. ut:u-s:a '(wooden) beam', which may go back to *q̇wɨ̆d(w)V with metathesis.
Notes: Reconstructed for the PEC level. One of several expressive roots with the structure TVNKV / KVNTV meaning 'hole, pit'. Cf. also Osset. tugur 'attic, loft, ceiling' (of Caucasian origin, see Abayev 1979, 313).
Notes: In most EC languages the root means actually 'shady slope of a mountain' (i.e. part of the mountain heavily covered with trees). The root is also attested (with reduplication) in HU languages: Hurr. fāvanǝ, Ur. vāvānǝ 'mountain' (see Diakonoff-Starostin 1986, 19).
Notes: Reconstructed for the PEC level. We can also reconstruct the obl. base *fiw-dV- (*fib-dV-), reflected in PA *xʷidV- = Lak. hat:a- / ħat:a-. = PL *χu-t:.
Notes: The WC form has a prefix of unknown origin; despite this, and the lack of the expected palatalisation (*-x́ʷǝ would be rather expected), there can be no doubt in the common origin of the EC and WC forms. The rules of palatalisation are violated in WC numerals (cf. also EC *ƛHĕ 'three' : WC *ʎ:V), possibly because of original Ablaut.
See Trubetzkoy 1922, 239, 243; 1930, 274-275; Абдоков 1983, 155.