In [Comrie & Khalilov 2010: 585], 'short' is erroneously glossed as älčah {аьлчагь}, which is in fact alčaʁ {алчагъ} 'low, small in height' [Meylanova 1984: 19], borrowed from Azerbaijani alčag 'low, small in height'.

    Tsakhur-Kum Tsakhur: ǯitʼa-n [Kibrik & Kodzasov 1990: 237].

    The same in the Arsug subdialect: ǯeʁe-d 'short' [Suleymanov 1993: 80]. It must be noted that in [Suleymanov 2003: 81], the Arsug or Khudig form is quoted as ǯaʁˤa-d, which seems erroneous.

    The same in the other subdialects: Tsirkhe, Khpyuk ǯiqːe-f, Kurag ǯeqːe-f, Duldug ǯaqːi-f 'short' [Suleymanov 1993: 80; Magometov 1970: 42; Shaumyan 1941: 171].

    The same in the Khanag subdialect: ǯiqːˈi 'short' [Uslar 1979: 693, 996]. In [Dirr 1905: 170, 231], transcribed as ǯiʁi (and even ǯig-) - this is actually either a form from some Southern Tabasaran subdialect or the beginning of the phonetic process qː > ʁ in Khanag during the 2^nd half of the 19^th century between Uslar's and Dirr' records.

    The same in the Khyuryuk subdialect: ǯiqːˈi {жжикъи} 'short' [Genko 2005: 67].

    The same in the Khiv subdialect: ǯiqːˈi {жжи(к)къи} 'short' [Genko 2005: 67]. The same in Literary Tabasaran: ǯiqːˈi {жикъи} 'short' [Khanmagomedov & Shalbuzov 2001: 162].

    The same in Literary Lezgi: kːürˈü {куьруь} 'short' [Talibov & Gadzhiev 1966: 170; Gadzhiev 1950: 310; Haspelmath 1993: 495, 526].

    Etymologically the same term in the Akhty dialect: Khlyut čːʷerˈi 'short' [Kibrik & Kodzasov 1990: 237].

    In West and East Lezgian a phonetically irregular root *čːiCV- is used for 'short' (reconstructed as *čːi[kʼ]V- in [NCED: 1108]). It is attested as *čːitʼV- in Tsakhur, *čːikV- in Rutul, *čːiqːV- in Tabasaran and Aghul (note that the Aghul forms can be Tabasaran loanwords).

    In Lezgi, 'short' is expressed with the etymologically isolated form, which points to the Proto-Lezgian shape *čːʷerV-.

    In both outliers, inherited forms were superseded with loanwords: Udi < Azerbaijani, Archi < Lak.

    Replacements: {'short' > 'narrow'} (Lezgi)

    Reconstruction shape: The voiced reflex d of Lezgian *tʼ observed in South Lezgian (Kryts gʷädä, Budukh godǝ) is due to influence on the part of Azerbaijani gödäk 'short'.

    Semantics and structure: Primary stative verbal root 'to be short'.

    Tsakhur-Kum Tsakhur: χoče [Kibrik & Kodzasov 1990: 88].

    The same in the Khanag subdialect: bitʼ 'snake' [Uslar 1979: 615, 994; Dirr 1905: 158, 229]. The same in the Khyuryuk subdialect: bitʼ {битI} 'snake' [Genko 2005: 29].

    The same in the Khiv subdialect: bitʼ ~ betʼ {битI, бетI} 'snake' [Genko 2005: 29]. The same in Literary Tabasaran: bitʼ {битI} 'snake' [Khanmagomedov & Shalbuzov 2001: 87].

    The same in Literary Lezgi: ʁülˈäʁ {гъуьлягъ} with polysemy: 'snake / silkworm' [Talibov & Gadzhiev 1966: 95; Gadzhiev 1950: 247; Haspelmath 1993: 490, 526]. A second, less frequent literary word for 'snake' is ilˈan [Talibov & Gadzhiev 1966: 135; Gadzhiev 1950: 247], borrowed from Azerbaijani ilan 'snake'.

    The same in the Akhty dialect: Khlyut ülˈäʁ 'snake' [Kibrik & Kodzasov 1990: 88].

    The isolated Caucasian Albanian boˤq 'snake' may regularly originate from a Proto-Lezgian form like *woˤrƛʷ(V), which, in turn, may regularly continue North Caucasian *wHoːrƛʷVɫV (~ *b-) [NCED: 1048] (or rather *wHoːrƛʷV- with the suffix -ɫ- in proto-languages of individual groups). This Proto-North Caucasian stem means 'snake' in the Avaro-Ando-Tsezian branch and 'snail' in Lak; note that, pace [NCED: 1048], specific Nakh forms for 'snake' originate from the word for 'mud', not from the aforementioned North Caucasian stem.

    The second candidate is *ƛːar [NCED: 787], which denotes 'snake' in South Lezgian (Kryts, Budukh) and Rutul, having been lost in the rest of the languages. In terms of distribution, it can be posited at least as the Proto-Nuclear Lezgian expression for 'snake'. Its North Caucasian comparanda are to be reconstructed as prototerms for 'snake' in Nakh, Khinalugh and possibly Proto-West Caucasian.

    Because North Caucasian *ᴌːăɦrV (> Lezgian *ƛːar) has a wider distribution, whereas *wHoːrƛʷV- in the meaning 'snake' seems to be a local Avaro-Ando-Tsezian isogloss, we prefer to postulate *ƛːar as the basic Proto-Lezgian term for 'snake'. The original meaning of Lezgian *woˤrƛʷ(V) (> Caucasian Albanian boˤq 'snake', if the proposed etymology is correct) is unclear, perhaps 'a k. of snake' vel sim.

    In Tsakhur, *ƛːar was superseded with *χʷärčVy [NCED: 1080], whose original meaning should be 'a k. of worm' (cf. the meaning 'worm' for Archi χʷˈarši q.v.).

    In Tabasaran, the form bitʼ 'snake' is attested; it is isolated within Lezgian (Lezgi bütʼrˈük(ʷ) 'small worm' seems to be a Tabasaran loanword, see notes on 'worm'). Its Lezgian protoform is reconstructed as *pːe(m)tʼ (~ b-) in [NCED: 290] with possible external North Caucasian comparanda with the meanings 'snake' and 'worm'.

    In Lezgi, 'snake' is expressed with the root *mulaqʷˤ 'worm' q.v. [NCED: 817].

    In some lects, inherited forms tend to be superseded with borrowings from Azerbaijani ilan 'snake' (Kryts Proper, Aghul). In Archi, a Lak loanword is used.

    Replacements: {'worm' > 'snake'} (Lezgi).

    Reconstruction shape: Correspondences seem regular.

    Semantics and structure: Primary substantive root. The oblique stem is *ƛːara-.

    Distinct from qʼʷˤˈaqʼar-tːu 'narrow, tight (of path, dress)', participle from the stative verb qʼʷˤˈaqʼar 'to be narrow' [Chumakina et al. 2007; Kibrik et al. 1977b: 309] (in [Kibrik & Kodzasov 1990: 238], quoted with a typo: qʼʷˤˈaqʼʷar-).

    Distinct from dar 'narrow' [Kibrik et al. 1999: 872], borrowed from Azerbaijani dar 'narrow'. In [NCED: 522] (Proto-Lezgian *hˤɨ(m)ƛʼːä-), also the Kryts suffixed adjective kɨ-tä- 'narrow' is quoted (not found in other sources).

    Distinct from dar 'narrow' [Authier 2009: 104], borrowed from Azerbaijani dar 'narrow'.

    In [Meylanova 1984: 113, 245], the word nazik {назик} is also quoted in the meaning 'thin'; basic semantics is 'thin 2D' in all found examples. Borrowed from Azerbaijani nazik 'thin 2D/1D' (ultimately from Persian naːzuk 'thin').

    Distinct from more specific taχta-lu {тахталу} 'flat, plane, thin 2D' [Meylanova 1984: 133, 245].

    Distinct from dar 'narrow' [Kibrik et al. 1999: 872], borrowed from Azerbaijani dar 'narrow'.

    Tsakhur-Kum Tsakhur: kʼɨˤwa-n 'thin 2D' [Kibrik & Kodzasov 1990: 239].

    A second term gaˤd-dɨ is observed in [Makhmudova 2001: 95, 183] and [Comrie & Khalilov 2010: 587]. Makhmudova claims that the two terms are opposed as follows: gaˤd-dɨ is applied to animated objects (i.e. 'lean, thin'?), and qʼɨl-dɨ to inanimate ones.

    Cf. also qʼɨcʼ-dɨ, quoted in [Dirr 1912: 166] as 'thin' without specification (corresponds to qʼɨcʼ-dɨ 'narrow' in other dialects). Maybe this is the Mukhad word for 'thin (1D)'.

    In [Kibrik & Kodzasov 1990], two Burshag words are quoted as synonyms for 'thin 2D/1D'; the second one is kːu-re-r.

    The same in the Usug subdialect: kʼile-f 'thin 2D', iškːe-f 'thin 1D' [Shaumyan 1941: 143, 184]. The Usug form iškːe- was probably influenced on the part of the Lezgi word for 'thin 1D'.

    Differently in the Khanag subdialect: čʼilːˈi 'thin 2D; sparse (wood, hair)' [Uslar 1979: 974, 1008], opposed to kːu-rːˈi with polysemy: 'thin 1D / narrow' [Uslar 1979: 768, 1008; Dirr 1905: 182, 244].

    The same system in the Khyuryuk subdialect: čʼilːˈi {чIилли} 'thin 2D' [Genko 2005: 186], opposed to kːu-rːˈi {ккурри} 'thin 1D' [Genko 2005: 96].

    The same in the Khiv subdialect: čʼilːˈi {чIилли} 'thin 2D' [Genko 2005: 186], opposed to kːu-rˈu {ккуру} with polysemy: 'thin 1D / narrow' [Genko 2005: 97].

    The same in Literary Tabasaran: čʼilːˈi {чIилли} 'thin 2D' [Khanmagomedov & Shalbuzov 2001: 343], opposed to kːu-rˈu {ккуру} with polysemy: 'thin 1D / narrow' [Khanmagomedov & Shalbuzov 2001: 200].

    The same opposition in Literary Lezgi: qʼelˈečʼ {кьелечI} with interesting polysemy: 'thin 2D / wiry, lean (of human, animal)' [Talibov & Gadzhiev 1966: 201; Gadzhiev 1950: 851; Haspelmath 1993: 503, 528], šükʼˈü {шуькIуь} 'thin 1D' [Talibov & Gadzhiev 1966: 387; Gadzhiev 1950: 851; Haspelmath 1993: 507, 528]. Distinct from two literary words for 'narrow': inherited gütʼˈü {гуьтIуь} 'narrow' [Talibov & Gadzhiev 1966: 91; Haspelmath 1993: 489, 523], and borrowed dar {дар} 'narrow; restricted, tight' [Talibov & Gadzhiev 1966: 107; Haspelmath 1993: 486, 523] (< Azerbaijani dar 'narrow').

    The same opposition in the Akhty dialect: Khlyut qʼälˈačʼ 'thin 2D', škːi 'thin 1D / narrow' [Kibrik & Kodzasov 1990: 239].

    NCED: 639. Distribution: The basic data can be summarized as follows:

'THIN'	Udi	Archi	Kryts	Budukh	Tsakhur	Rutul	Aghul	Tabasaran	Lezgi
*ƛʼɨlä- [NCED: 639]		1D/2D	1D/2D		2D	2D	2D	2D	2D
*hˤɨ(m)ƛʼːä- [NCED: 521]			narrow(?)	1D/2D	1D	1D	1D / narrow	1D / narrow	1D / narrow
*ʔisːal(ː)- [NCED: 752]							narrow	narrow	sal narrow
*qʼːʷˤarV- [NCED: 933]		narrow					lean, emaciated
?	qːačː narrow
?						qʼɨcʼ- narrow
Azerbaijani loanword	1D/2D
Azerbaijani loanword			narrow	narrow	narrow

    For Proto-Nuclear Lezgian, the opposition *ƛʼɨlä- 'thin 2D' [NCED: 639] / *hˤɨ(m)ƛʼːä- 'thin 1D' [NCED: 521] can be reconstructed with safety. It is possible to treat such an opposition as a secondary feature of Nuclear Lezgian (and only reconstruct *ƛʼɨlä- 'thin 2D/1D' for Proto-Lezgian), but, since both roots possess external North Caucasian cognates with the meaning 'thin', we prefer to reconstruct *ƛʼɨlä- 'thin 2D' / *hˤɨ(m)ƛʼːä- 'thin 1D' for Proto-Lezgian.

    In both outliers (Udi, Archi), the semantic opposition 'thin 2D' / 'thin 1D' has been eliminated. In the Udi case, this happened under the influence of Azerbaijani polysemy; for Archi, a similar Lak influence is probable.

    Replacements: {'thin 1D' > 'narrow'} (Aghul, Tabasaran, Lezgi), {'thin 2D > 'wiry, lean (of human, animal)'} (Literary Lezgi).

    Reconstruction shape: For *ƛʼɨlä-, correspondences seem regular except for pharyngealization in Tsakhur. In the case of *hˤɨ(m)ƛʼːä-, the situation is more complicated, because this root is frequently modified with adjectival suffixes (-*-tːV, *-rV) that cause reduction of the initial vowel and subsequent simplification of the cluster *mƛ (the Proto-Lezgian nasal phoneme is reconstructed on the basis of external evidence).

    Semantics and structure: Primary stative verbal roots 'to be thin 2D' (*ƛʼɨlä-) and 'to be thin 1D' (*hˤɨ(m)ƛʼːä-).

    Distinct from darɨ-n 'narrow' [Kibrik et al. 1999: 872], borrowed from Azerbaijani dar 'narrow'.

    Tsakhur-Kum Tsakhur: hɨˤkːʸe-n 'thin 1D', [Kibrik & Kodzasov 1990: 239]. Distinct from darɨ-n 'narrow' [Kibrik & Kodzasov 1990: 238], borrowed from Azerbaijani dar 'narrow'.

    Distinct from darɨ-n 'narrow' [Kibrik & Kodzasov 1990: 238; Dirr 1913: 154], borrowed from Azerbaijani dar 'narrow'.

    Distinct from qʼɨcʼ-dɨ 'narrow' [Dzhamalov & Semedov 2006: 167].

    Distinct from qʼɨcʼ-dɨ 'narrow' [Kibrik & Kodzasov 1990: 238].

    Distinct from kːe-tːu-f 'narrow' [Kibrik & Kodzasov 1990: 238], derived from the same root with the adjective suffix -tːu.

    Distinct from isal-f 'narrow' [Kibrik & Kodzasov 1990: 238].

    A second term for 'wind' is kulʸek [Ibragimov 1990: 55, 66; Ibragimov & Nurmamedov 2010: 196], borrowed from Azerbaijani küläk 'wind'.

    A third term is yelʸkan, quoted in [Kibrik et al. 1999: 879, 892] as the only generic term for 'wind', but glossed as 'light wind, breeze' in [Ibragimov & Nurmamedov 2010: 161]. Borrowed from dialectal Azerbaijani *yel-kän vel sim. from yel '(light) wind' (cf. literary Azerbaijani yel-kän 'sail; fan').

    Tsakhur-Kum Tsakhur: mɨc [Kibrik & Kodzasov 1990: 209].

    A second term for 'wind' is kulʸak [Kibrik & Kodzasov 1990: 209; Dirr 1913: 174, 222], borrowed from Azerbaijani küläk 'wind'.

    A synonym is kulʸäk [Kibrik & Kodzasov 1990: 209; Comrie & Khalilov 2010: 59], borrowed from Azerbaijani küläk 'wind'.

    Distinct from inherited xɨbɨl {хьыбыл} 'light wind, breeze' [Dzhamalov & Semedov 2006: 283] (this is incorrectly listed in [Comrie & Khalilov 2010: 59] as the basic Ixrek term for 'wind').

    The same in the Khanag subdialect: mikʼ with polysemy: 'wind / rheumatism' [Uslar 1979: 855, 990; Dirr 1905: 195, 226]. The second Khanag word for 'wind' is kulak [Dirr 1905: 185, 226], borrowed from Azerbaijani küläk 'wind'.

    The same in the Khyuryuk subdialect: inherited mikʼ {микI} 'wind' [Genko 2005: 120] and borrowed kulˈak {кулак} 'wind, whirl' [Genko 2005: 88].

    The same in Literary Tabasaran: mikʼ with polysemy: 'wind / rheumatism' [Khanmagomedov & Shalbuzov 2001: 229]. The second literary word is kulˈak {кулак} 'wind, whirl' [Khanmagomedov & Shalbuzov 2001: 203], borrowed from Azerbaijani küläk 'wind'.

    Differently in the Khiv subdialect, where two words for 'wind' are known: inherited xar {хьар} 'wind' [Genko 2005: 175] and borrowed kulˈak {кулак} 'wind, whirl' [Genko 2005: 88]. The Common Tabasaran term is retained as Khiv mekʼ {мекI} 'rheumatism' [Genko 2005: 119].

    The same in Literary Lezgi: gar {гар} 'wind (in general, but probably not applied to the specifically cold wind)' [Talibov & Gadzhiev 1966: 82; Haspelmath 1993: 488, 529]. Distinct from literary qːa-y {къай} 'cold wind' [Talibov & Gadzhiev 1966: 175; Haspelmath 1993: 501, 529] - a participle from the verb 'to get cold'.

    In the Akhty dialect: no generic term for 'wind' is documented for the Khlyut subdialect, cf. specific gar 'west wind', qːa-y 'east wind' [Kibrik & Kodzasov 1990: 208]. The same in the Khuryug subdialect: gar 'south wind', qːa-y 'cold wind' [Meylanova 1964: 315].

    Based on available data, it is theoretically possible to reconstruct the Proto-Lezgi opposition ɣar 'warm wind' / qːa-y 'cold wind' without a single generic term.

    In Rutul, 'wind' is expressed by the stem *ƛopːol [NCED: 786], which means 'rheumatism' in Lezgi. The original meaning of *ƛopːol seems to have been 'a k. of wind', e.g., 'breeze'.

    The third candidate is *ɬːar [NCED: 758], which means 'wind' in Lezgi (perhaps to be reconstructed as 'warm (south) wind' for Proto-Lezgi) and in Khiv Tabasaran (but not in Proto-Tabasaran), but 'rheumatism' in South Lezgian (Kryts, Budukh). The original meaning of *ɬːar seems to have been 'a k. of wind', e.g., 'warm wind'.

    In Tabasaran, 'wind' is denoted by *meƛʼ 'cold, frost' [NCED: 808], see notes on 'cold' (apparently with the development 'cold' > 'cold wind' > 'wind in general').

    Aghul tireb 'wind' is etymologically unclear.

    In many lects, inherited forms have been completely superseded with Azerbaijani loanwords (Kryts, Budukh, Ixrek Rutul, some Aghul dialects). In Archi, there is a loanword from Lak or Avar.

    Replacements: {'cold, frost' > 'wind / rheumatism'} (Tabasaran), {'wind' > 'rheumatism'} (Udi, Kryts, Budukh, Lezgi), {'wind' > 'hope'} (Aghul, Tabasaran).

    Reconstruction shape: Correspondences seem regular, except for the Tsakhur affricate -c (instead of expected -č).

    Semantics and structure: Primary substantive root. The oblique stem is not reconstructible.

    Distinct from the compound šer-bäħ, quoted in [NCED: 982] with the gloss 'worm, helminth' (the second element -bäħ is not clear).

    There also exists another (less frequent or bound?) term for 'worm': qːurd [Authier 2009: 181], borrowed from Azerbaijani gurd 'worm'.

    Distinct from mulaʡ {мулаъ} 'worm in meat' [Kibrik & Kodzasov 1990: 88; Meylanova 1984: 111].

    Distinct from mɨqˤ {мыIхъ} 'worm' (the exact meaning is unknown) [Ibragimov & Nurmamedov 2010: 270], naʁur ~ laʁur {нагъур, лагъур} 'helminth' [Ibragimov & Nurmamedov 2010: 271] and čʼuˤʔ 'small worms in meat' [Ibragimov & Nurmamedov 2010: 404].

    Tsakhur-Kum Tsakhur: abrawuče [Kibrik & Kodzasov 1990: 88]. Distinct from mɨqˤ 'caterpillar' [Kibrik & Kodzasov 1990: 88], although in [Schulze 1997: 16] mɨqˤ is glossed as 'worm'.

    Suvagil Tsakhur: abrawučʼ 'earthworm' [Ibragimov 1990: 166], distinct from mɨqˤ 'worm' (not specified) [Ibragimov 1990: 166].

    In [Dirr 1913: 188, 242], the word for 'worm' is muqˤ, but in [Kibrik & Kodzasov 1990: 88] this is quoted as mɨqˤ 'caterpillar'.

    Distinct from laʁur 'helminth' [Dirr 1913: 183, 223].

    In [Comrie & Khalilov 2010: 170], 'worm' is glossed as muq {мухъ} (the modern depharyngealized variant for expected muqˤ).

    According to [Makhmudova 2001: 20], mulüχˤ 'worm' is opposed to the specific term barcʼil 'earthworm', but this seems to be inaccurate, since Ibragimov [Ibragimov 1978: 135] explicitly states that the word barcil (sic!) 'earthworm' is characteristic of the Khnyukh dialect (subdialect of Mukhad) without an etymological counterpart in Mukhad proper. In [Khaydakov 1973: 11] (followed by [NCED: 288]), barcʼil 'earthworm' is labeled simply as "Rutul".

    Distinct from šär-äkʼ 'helminth' [Dirr 1912: 182, 188] and cʼir-uχ 'caterpillar' [Ibragimov 1978: 225].

    Distinct from various terms that are more marginal or specific: šar-akʼ 'helminth' [Dzhamalov & Semedov 2006: 301], seb 'worm; cabbage white butterfly' [Dzhamalov & Semedov 2006: 227] (apparently more correctly glossed as 'caterpillar' in [Dzhamalov & Semedov 2006: 334]), cʼuʁ 'worm' [Dzhamalov & Semedov 2006: 288] (without specifications), čʼiʔ {чIиъ} [Dzhamalov & Semedov 2006: 413] (without specification; missing from the main section of the dictionary).

    Distinct from šːar 'helminth' [Magometov 1970: 23].

    The semantic opposition in the Khanag subdialect is better documented: šːar with polysemy: 'earthworm / helminth' [Uslar 1979: 985, 1009] (missing from [Dirr 1905]). Distinct from Khanag mulˈaqˤ 'worm (in fruit, meat, wound)' [Uslar 1979: 865, 1009; Dirr 1905: 196, 246].

    Similarly in the Khyuryuk subdialect: šːar {шшар} with polysemy: 'earthworm / helminth' [Genko 2005: 192, 221], opposed to mulˈaqˤ {мюляхъ} 'worm' [Genko 2005: 124] (not specified semantically).

    The same in the Kumi subdialect: šːar {шшар}, glossed as 'worm, helminth' [Genko 2005: 192], opposed to mulˈaqˤ ~ mulˈaqʷˤ {мюляхъ(ю)} 'worm' [Genko 2005: 124] (not specified semantically).

    The same in the Khiv subdialect: šar {шар} with polysemy: 'earthworm / helminth' [Genko 2005: 189], opposed to malˈaqʷˤ {мяляхъв} 'worm' [Genko 2005: 124] (not specified semantically).

    In the Truf subdialects: šer {шер} 'worm' [Genko 2005: 189] (not specified semantically).

    Differently in Literary Tabasaran, where šar {шар} is the basic term for 'worm (incl. earthworm and helminth)' [Khanmagomedov & Shalbuzov 2001: 347]. Distinct from literary bitʼruk {битIрук} 'worm' [Khanmagomedov & Shalbuzov 2001: 87] with two examples: "wood worm" and "to become worm-eaten"; perhaps to be analyzed as bitʼ-ru-k - the oblique stem bitʼ-ru- from bitʼ 'snake' q.v. plus the expressive suffix -k (differently in [NCED: 290], where bitʼruk is treated as a compound bitʼ-ruk 'snake' + 'grass-snake'). It seems that literary bitʼruk is a semantic equivalent of malˈaqʷˤ (mulˈaqˤ) from other dialects.

    A very similar situation in Literary Lezgi: šar {шар} 'earthworm / helminth' [Talibov & Gadzhiev 1966: 382; Gadzhiev 1950: 142, 927; Haspelmath 1993: 529]. Distinct from kʷak [abs.] / kuk-rˈa- [obl.] / kukʷ-ˈar [pl.] {квак} 'worm (in fruit, meat, flour, grain, wood' [Gyulmagomedov 2004, 1: 347; Talibov & Gadzhiev 1966: 153; Gadzhiev 1950: 927; Haspelmath 1993: 495, 529] and from the rare word bütʼrˈük [sg.] / bütʼrˈük-ar [pl.] {буьтIруьк} [Gadzhiev 1950: 927], which is explained as a synonym of kʷak in [Gyulmagomedov 2004, 1: 140], but specifically glossed as 'larva of green blowfly' in [Talibov & Gadzhiev 1966: 70] (In [Khaydakov 1973: 11], the assimilated variant pʼütʼrük 'worm' is also quoted). Cf. also literary kʼazun ʁüläʁ {кIазун гъуьлягъ} 'silkworm', literary 'silk's snake (q.v.)' [Talibov & Gadzhiev 1966: 210; Gadzhiev 1950: 927].

    In the Akhty dialect: two terms for 'worm' are documented for the Khlyut subdialect, both without semantic specifications: šar and kʷak [Kibrik & Kodzasov 1990: 88].

    In the Usukhchay subdialect of the Doquzpara dialect (Samur group) the generic term for 'worm' is pʼarkʼʷ [Meylanova 1964: 223].

    The available data are too scant for detailed Proto-Lezgi reconstruction, but the generic opposition šar 'earthworm, helminth (i.e. large worm)' / X 'worm in (rotten) organic (i.e. small worm)' is the same as in the neighboring Tabasaran language. It must be noted that the rare Gyune and literary word bütʼrˈük(ʷ) 'small worm' seems actually a Tabasaran loan, cf. Literary Tabasaran bitʼ-ru-k 'small worm' q.v.

    The term *šːar [NCED: 982] can be reconstructed as 'helminth' at least for the Proto-Nuclear Lezgian level: 'helminth' in Kryts, Rutul, East Lezgian (Aghul, Tabasaran, Lezgi), additionally extended to mean 'worm in general' in Tabasaran, Lezgi.

    The term *kamk [NCED: 206] could mean 'worm in (rotten) organic' at least at the Proto-Nuclear Lezgian level: Kryts 'worm (in general?)', Lezgi 'worm (in wood, fruit, wound)', Tabasaran 'wormhole, ulcer', Aghul 'itch, mange', Tsakhur 'grub, larva'.

    The original meaning of *čirč- [NCED: 348] is not clear. This is the basic term for 'worm' in South Lezgian (Kryts, Budukh), whereas the suffixed stem *čirč-Vl means 'lizard' in Tabasaran and Lezgi.

    For Proto-Tsakhur, *mulaqʷˤ can be posited as 'worm (incl. earthworm)', although in modern Tsakhur dialects, this stem shifted to the meaning 'caterpillar', having been superseded with the descriptive term 'blind' or the etymologically obscure abrawučʼ(e) ~ abrawuč.

    In Khnyukh Rutul, *mulaqʷˤ was superseded with the etymological isolate barcʼ-il, which can originate from the protoform *pːalcʼ- (thus [NCED: 288] with a dubious Tsezian comparandum). Actually, Khnyukh Rutul barcʼ-il can hardly be separated from Usukhchay Lezgi pʼarkʼʷ 'worm (in general?)' which implies a protoform like pʼalcʼʷ (~ pː- ~ -r-), although the correspondence Rutul b- / Lezgi pʼ is not regular (assimilation or dissimilation between two root consonants). The original meaning of this root is unclear.

    Replacements: {'blind' > 'worm'} (Gelmets Tsakhur), {'helminth' > 'worm'} (Tabasaran, Lezgi), {'worm' > 'caterpillar'} (Tsakhur), {'worm' > 'snake'} (Lezgi).

    Reconstruction shape: Correspondences seem regular.

    Semantics and structure: Primary substantive root.

    A second term for 'year' is ilʸ [Kibrik et al. 1999: 878], borrowed from Azerbaijani il 'year'.

    Distinct from yeš 'age, years' [Kibrik et al. 1999: 879; Ibragimov & Nurmamedov 2010: 163], borrowed from Azerbaijani yaš 'age, years'.

    Tsakhur-Kum Tsakhur: sʸenʸ [Kibrik & Kodzasov 1990: 209].

    The same in the Arsug subdialect: is 'year' [Shaumyan 1941: 143].

    The same in the Khanag subdialect: yisː 'year' [Uslar 1979: 753, 992; Dirr 1905: 180, 226]. The same in the Khyuryuk subdialect: yisː {йисс-} 'year' [Genko 2005: 80].

    The same in the Khiv subdialect: yis {йис} 'year' [Genko 2005: 80]; cf. also the Khiv expression yis-qʼˈub {йискьуб} 'the whole year' [Genko 2005: 80]. The same in Literary Tabasaran: yis {йис} 'year' [Khanmagomedov & Shalbuzov 2001: 185]; an additional synonym is literary yis-qʼˈub {йискьуб}, glossed as 'year, the whole year' in [Khanmagomedov & Shalbuzov 2001: 185].

    The same in Literary Lezgi: yis [abs.] / s-a- ~ yis-ˈa- [obl.] / s-ar ~ yis-ˈar [pl.] {йис} 'year' [Talibov & Gadzhiev 1966: 146; Gadzhiev 1950: 144; Haspelmath 1993: 494, 529].

    The same in other dialectal groups: Khlyut (subdialect of Akhty, Samur group) yɨs [abs.] / yɨs-ˈa- [obl.] / yɨs-ˈar [pl.] 'year' [Kibrik & Kodzasov 1990: 209]; Akhty proper (Samur group) yus [abs.] 'year' [Meylanova 1964: 381]; Yargun (Quba group) yis [abs.] / s-a- ~ is-ˈa- [obl.] / s-ar ~ yis-ˈar [pl.] 'year' [Babaliyeva 2007: 18, 53].

    The first one is *sːän [dir.] / *sːänɨ- [obl.] [NCED: 975], meaning 'year' in Archi, South Lezgian (Kryts, Budukh) and West Lezgian (Tsakhur, Rutul), lost in the rest of the languages.

    The second is *yisː [dir.] / *yisːä- [obl.] [NCED: 968] which denotes 'year' in East Lezgian (Aghul, Tabasaran, Lezgi); in these languages, the derived adjectival stem *yisːV-rV- 'old' is also present. In Udi and West Lezgian (Tsakhur, Rutul), the paronymous adjective *yisːV 'old' is attested (see forms in [NCED: 968]).

    Finally, the Caucasian Albanian-Udi form usen 'year' should originate from something like *(y)Vsän.

    Two internal Lezgian scenarios can be proposed, neither of them completely satisfactory.

    1) The non-East Lezgian forms for 'year', reconstructed as *sːän in [NCED: 975], actually look like the nominal derivative *yisːä-n with reduction of the first syllable in all languages except for Caucasian Albanian-Udi. Caucasian Albanian-Udi usen 'year' can only originate from *yisːä-n, but not from *sːän (the development *yi- > Udi u- is attested in other roots as well, e.g., 'bone' q.v., 'heart' q.v.). The nominal word-formative suffix -n is well attested in Lezgian, see [Alekseev 1985: 44 f.]. This suffix, however, was originally the genitive ending. Thus we should suppose that two synonymous stems coexisted in Proto-Lezgian: *yisː [dir.] / *yisːä- [obl.] 'year' (retained only in East Lezgian) and *yisːä-n 'year' < *'of year' (retained in all the other languages, including both outliers). The Proto-Lezgian adjective *yisːV 'old' was derived from the same root ('year' > 'old'); in East Lezgian, it was additionally modified with the common r-suffix: *yisːV-rV- 'old'.

    2) Two unrelated stems are to be reconstructed for Proto-Lezgian: *sːän 'year' and *yisːV 'old'. East Lezgian *yisː [dir.] / *yisːä- [obl.] 'year' is a back formation due to contamination of two original roots (the derivation 'year' > 'old' is typologically not so rare, but the opposite direction seems more problematic). After such a contamination at the Proto-East Lezgian level, the new Proto-East Lezgian adjective *yisːV-rV- 'old' was introduced. The situation with Caucasian Albanian-Udi usen 'year' (which implies a protoform like *yisːän) is similar, although less obvious: usen 'year' could represent a contaminated hybrid of *sːän 'year' and *yisː-V 'old'.

    External comparison suggests that the second scenario is preferable: cf. Lak šin 'year'.

    Replacements: {'year' > 'old'} (Aghul, Tabasaran, Lezgi).

    Reconstruction shape: Correspondences seem regular, except for the Caucasian Albanian-Udi hybrid form usen.

    Semantics and structure: Primary substantive root. The oblique stem is *sːänɨ-.