Notes: A Lezg.-WC isogloss - but seems reliable both phonetically and semantically (same meanings and semantic development in both subgroups). It is interesting to compare this root to PK *č̣iṗ- 'navel' (note that its descendant, Megr. č̣aṗan "navel", was borrowed in Abkh. as a-č̣́aṗán). The comparison of PK *č̣iṗ- with PEC *ʒ_ŏnʔŭ 'navel', proposed by Klimov (1965 255), is obviously untenable.
Notes: The root is not very widely spread (cf. also Gunz. malu 'hoof' - probably from old Avar.), but seems to be reliable. *-r- in PN is a normal reflex of *ɫ (and *l) before a velar (either as part of root or as a suffix). A possible trace of this root in Lezg. is the Agul (Fit.) form kuš-mul 'hoof (of artiodactyla)' (for kuš- see *kwăśV).
Notes: The semantic correlation 'pipe' : 'horn' is quite usual, and the comparison of the PEC and PWC forms seems plausible (*pǝ̃ša with assimilative unvoicing < *bǝ̃ša; labialisation is lost in PWC, as often, after the initial labial). The PN reflex *marš raises some doubts: we should regularly expect *barš. This, and the isolated position of the Chech. word, suggests a loan, possibly from some Av.-And. source (see comments on the PA form, suggesting the existence of some Av. dialectal form like *marš:u).
Notes: Correspondences are regular. Abdokov (1983, 82) compares PAK *c̣ʷa- with reflexes of PEC *ƛ̣ōrV 'bone' (q.v.) - which is quite impossible phonetically.
Notes: Correspondences are regular. The Av., Lak. and Darg. forms probably reflect a former plural *bŏlcĒ-rV̆- (cf. also Tab. marc-ar); hence the Av. paradigm C.
Notes: The correspondences are regular (note, however, a usual delabialisation in PWC after the initial labial: *bŏlćwĭ > *bŏnčĭ > *mVč́V). The root originally denoted a species of wild millet (used for making brooms), cf. the meanings in PTs and PD. Note that Darg. Ak., Ur. muči 'millet' (and other dialectal forms) must be regarded as an old Avarism.
Notes: There occurred a metathesis either in PEC or in PWC (rather usual for roots of this type); otherwise the correspondences are all right (in PWC there's a usual delabialisation before the following labial) and the comparison seems quite reliable.
Notes: Not quite clear is the source of PWC pharyngealization; in fact, the PWC root may be a contamination of two PNC roots (cf. also PEC *mħĕlq̇wV 'root, stake' q.v.).
Notes: Cf. Urart. burg-ana- ( > Arm. burgn) 'tower, fortress' (see Diakonoff-Starostin 1986, 18). An interesting common NC term - which has parallels in some European languages (an old Wanderwort). Abdokov (1983, 125) compares the Kab. form with heterogeneous EC material (Darg. dirq, Tab. aqI) which is hardly justified.
Notes: Reconstructed for the PEC level. The stem contains a suffixed *-lV (absent in some Av.-And. languages). The Av.-And. forms are compared by Abdokov (1983, 129) with PAK *ʎájǝ / *ʎawájǝ 'stairs'; this seems dubious, because the Adygh forms are obviously derived from *ʎa 'foot'.
Notes: Reconstructed for the PEC level. Not very reliable, because of the strange behaviour of the stem in Lezghian languages; besides, labialised -ʒw- should not have yielded -ṭṭ- in a cluster in PN. Contaminations of originally different roots may be the reason (see above).
Notes: Reconstructed for the PEC level. The root is onomatopoeic (like most birdnames), but the correspondences are regular and the reconstruction seems to be quite secure.
Notes: PL (Arch.) *noƛ̣: probably reflects an earlier *moƛ̣: (with variation *m-/*n-) which would be a normal reflex of *bŭlƛ̣_V.
Reconstructed for the PEC level; cf. also HU: Hurr. purli, Urart. pur(u)li 'house' (see Diakonoff-Starostin 1986, 17). The root is sometimes confused with PEC *bēŁ_V̆ 'shed' q.v., but the two roots should be strictly distinguished.
Notes: The root is obviously borrowed from an Iranian source (cf. Shugn. bǝruǯ, Osset. bärz / bärzä 'birch-tree' etc.); but since the reflexes correspond to each other well phonetically, are present in distant EC subgroups and have substantially modified the original semantics, the borrowing should be probably dated by a very early (common EC) period.
Notes: The word is not very widely spread, and - despite regular correspondences between languages - may be borrowed (cf. Pers. bōr 'bay, reddish', Osset. būr 'yellow' or Mong. bora 'grey'. Cf. Hurr. pawro 'brown' (see Diakonoff-Starostin 1986, 18) - which strengthens the EC etymology - but on the other hand, may be itself a loan from Indo-Iranian *babhru-.
Notes: Quite regular phonetically (although the vocalic reconstruction is somewhat dubious: the *-ü- vowel is not surely reconstructed for PEC and PNC). The EC forms go back to an intermediate protoform *büŁ_a. PWC *bǝɣA can not (despite Shagirov 1977, Abdokov 1983, 78) be compared with Av. muʁ, Chech. buq̇ 'back' (note that these forms have nothing to do with each other, too). See also *=ĕƛ̣_Ĕ.
