Comment: Cf. also Lezg. Khl. rič̣ 'bow-string'. 3d class in Bud. In Bud. the word means 'bow-string' and 'a string made of sheep's intestines'. To judge from Tab. rič̣i- (the whole compound means literally "bow-string"+"boards"), the PL oblique base was *reč̣ʷɨ-.
Comment: Obl. base *req̇ʷɨ-, cf. Lezg. raq̇u-, Tab. ruq̇a-, Ag. ruq̇ʷa-. The plural form of this noun is used to denote a lock or a trap: cf. Lezg. raq̇ar 'trap; fetters'; Tab. ruq̇ar, Ag. ruq̇ʷar 'lock, trap'.
Comment: The nominal root *reχ:Iʷa (which is itself an archaic verbal derivate, see below) is present in all languages except Shakhdagh and Udi and belongs to the 4th class in all class-distinguishing languages. Cf. also Lezg. Khl. räχIʷ 'mill', Tab. Kand. raʁI / raʁIin, Düb. raʁIa, Ag. Fit. raχIʷ, Tp. raʕ id., Kryz. räħä-ǯ, Bud. raħa-ǯ 'meal, flour'; in Arch. the root contains -χI:ʷ-, cf. the conjunctional form deχI:u 'and the mill-stone' and the secondary derivate duχI:a-t 'mill'.
The original verbal root is not preserved as such in PL; instead, a secondary denominal stem *reχ:Iʷa- was used. Besides the listed Kryz., Bud. and Ud. forms cf.: Lezg. reʁʷe-z, Tab. ra-ʁI-uz (Düb. ra-ʁI-us), Ag. ruX̀-as (Bursh. urX̀a-s), Rut. -u-ruχIʷa-s, Tsakh. gi-χ:Iʷar-es, Arch. de-χI:a-s (dur. derχI:u-r). The verb belongs to the weak verbal class. Lezghian languages preserve also two other archaic derivates from the same verbal root:
1) PL *χ:Iur / *χ:Iuj 'meal, flour': Lezg. ʁür (Khl. χür), Tab. χIu (Düb. χIuj), Ag. X̀ur (Fit. χIur, Tp. ʕur, Burk. raX̀u < *X̀aruj), Rut. χIur, Tsakh. χIu (χIujɨ-), Ud. χari. The word belongs to the 3d class in Rut. and Tsakh.
2) PL *χ:Iuw(a) 'mill': Kryz. ħuw, Bud. ħuw / ʕuw (4th class in Kryz., but 3d class in Bud.). See Бокаpев 1961, 68; Лексика 1971, 174-175, 205; Гигинейшвили 1977, 121; Магометов 1966, 342.
Comment: Cf. also Lezg. Khl. rɨgɨ-d, Ag. Bursh. jerši-r. Metathesis of *r- into Inlaut in PL was caused by analogy with the numeral "seven" (Tab. urɣu-b, Ag. jeri-d < *uirƛ:ɨ-). Arch. diʎi 'sixty' < *riƛɨ-j.
Comment: Cf. also Tsakh. Gelm. jiq̇, Arch. erg. diq̇i. Obl. base *riq̇ɨ- (Ag. riq̇i-, Rut. rɨq̇ɨ-, Tsakh. jɨq̇ɨ-, Gelm. jiq̇ɨ-). 3d class in Rut., Tsakh., 4th class in Arch.
Comment: For the reflexes of *q̇I:ʷ cf. Lezg. Khl. erg. rüq:ʷedi (lit. with unification: rüqʷedi), Tsakh. loc. jɨqI:e, Arch. erg. diq̇I:i. Ud. (Nidzh.) has a reflex žeq: (with unclear ž-). Obl. base vowel unknown. 3d class in Rut., Tsakh.; 4th class in Kryz., Bud., Arch.
Comment: The Rut. and Tsakh. forms have a *-nV-suffix (which can either reflect the former oblique stem, or - through metathesis - the original root resonant).
Comment: It is not quite clear, whether we should also compare Arch. š:eIle-t:u- 'fat' (with an old reduction of Anlaut and suffixed *-lV- < *-ɨš:V-lV- ?).
Comment: Cf. also Tsakh. Gelm. rɨx́, Mik. rɨIh (having an irregular reflex of *ʎ, probably due to pharyngealization). Obl. base *roʎɨ- (Rut., Tsakh. rixi-); 4th class in both languages.
Comment: 4th class in Kryz. Obl. base unknown (Tab. ruka- is uninformative). A strange form is Tab. Düb. lukar 'bush' ( < *ruk-ar (plur.) with dissimilation?).
