Platyceps karelini ( Brandt, 1838 )

Schätti, Beat, Kucharzewski, Christoph, Masroor, Rafaqat, Rastegar, Masroor & Rastegar Pouyani, Eskandar, 2012, Platyceps karelini (Brandt, 1838) from Iran to Pakistan and revalidation of Coluber chesneii Martin, 1838 (Reptilia: Squamata: Colubrinae), Revue suisse de Zoologie 119 (4), pp. 441-483 : 444-460

Platyceps karelini ( Brandt, 1838) – Karelin’s Racer

Coluber (Tyria) Karelini [sic] Brandt, 1838: col. [241] 243, “sur la côte orientale de la mer Caspienne” [ZISP 1695-1700] (origin clarified by Strauch, 1873; type locality possibly in SW Kazakhstan; see Northern Populations).

Choristodonbrachycephalus Severczov, 1873b: 72 [unnumb. Tb.], footnote 2, “okrestnosti Khodzhenta” (vicinity of Khujand, Tadzhikistan) [ZISP 3581] (new replacement name for Ch. sogdianus Severczov, 1873a ; syn. Strauch, 1873; see Northern Populations: first smallprint, Hybrid Racers, footnote 4).

ZamenisKarelinii [sic]. – Strauch, 1873: 110 [272], Pl. III [ZISP 1696] (incl. “ Persien ” [ZISP 1701-02], seeNorthernPopulations: firstsmallprint).

Z.[amenis] ventrimaculatus [sic] ( Gray, 1834) [partim]. – Blanford, 1876: 414 (incl. “Karmán” [Kerman, Iran]; “Kila-i-Fath” [Qala-iFateh, Qaleh-eFath], “Zamrán” [Kamran], see Distribution, Systematics: first smallprints, Hybrid Racers incl. first smallprint, Fig. 8 andTb. 4; Sclater, 1891).

Zamenis ventrimaculatus [sic] [var.] karelini comb. n. – Boettger, 1888: 928 (see Northern Populations, Systematics, footnote 5).

ZamenisKarelini [sic]. – Wood-Mason, 1889: 8 (“ChinRalak” [“AfghanBound. Comm.”], see Boulenger, 1889; Sclater, 1891).

Zamenis rhodorachis Jan, 1863 . – Boulenger, 1889: 102 (“Gulran encampment, Badghis ” [Herat], see Hybrid Racers incl. Fig. 7A, Tb. 4 and footnote 7; Boulenger, 1893).

Zameniskarelinii [sic]. – Boulenger, 1889: 102 (“Chinkilok” [34°31’N 61°52’E, ca. 1’050 m a.s.l., ZSI 13107, Sclater, 1891], “ Helmand ” [River], “Kilki”, “Tirphul” [“Between TirphulandKilki” fideBoulenger, 1893]: incl. BMNH 1886.9.21. 102-103, seeHybrid Racers, Figs 2Aand 6B, Tb. 4). – Boulenger, 1890: 326 (“ Afghanistan ” [incl. Kandahar]: BMNH 1882.3.20. 2, seeMorphologyincl. secondsmallprint, Tb. 3).

Zamenis karelini . – Sclater, 1891: 28 (incl. Quetta [ZSI 11694]).

? Zamenisladaccensis [sic] ( Anderson, 1871) [partim]. – Sclater, 1891: 27 (incl. “Zamran” [ Blanford, 1876: probably ZSI 4616], see Systematics, Hybrid Racers: first smallprints).

Zamenis karelinii [sic]. – Boulenger, 1893: [381, 383] 401 (“New Gulran”, see Hybrid Racers; Boulenger, 1889: asrhodorachis).

Zamenis carelinii [sic]. – Werner, 1893: 92 (head scales).

Zameniskarelini. – Nikolskij, 1897: 335 (“GjarmazinPersiaorientali” [sic] [ca. 34°00’N 59°50’E, ca. 850 m, ZISP 8748]). – Zarudnyj, 1897: 359 (“Germau [...] Khunik” Mountains [Garmab, “Zirkukh”]: NMW 25446.4) 1).

1) Anderson (1999) indicatedthe “ZirkukhRegion”, i.e., “Zirkukh”, “terraZirckuch” or “TerraZirkuch” ( Zarudnyj, 1896, 1897; Nikolskij, 1900, 1916), between 31°00’- 34°01’N [sic] and 60°00’-60°30’EineasternIran. AtleastthevillageShirGug (orShirkuk, 33°01’N 59°30’E) at ca. 2’ 100 m a.s.l. as well as two collecting sites of Platyceps karelini in that region (“Germau” [Garmab], Bamrud) reported by Zarudnyj (1897) and Nikolskij (1900) lie within a much more limited area of Northeast Khorasan-e Jonubi (South Khorasan).

Zamenis karelinii [sic]. – Alcock & Finn, 1897: 563 (Afghan-Baluch border area [ Pakistan, unspecified]). – Nikolskij, 1900: 403 (Bamrud “interraZirckuch” [33°38’N 60°05’E, ca. 930 m], “BenduninSeistano” [“Bendan”, i.e., Bandan, 31°24’N 60°44’E, ca. 730 m], and “Persiaorientalis” [ZISP 9289-92], seefootnote 1). – Annandale, 1904: 209 (incl. “Perso-Baluch frontier”) and 1906: 197 (incl. “Seistan”, see Distribution: second smallprint; McMahon, 1906).

Zameniskarelini. – Wall, 1911: 1034 (Bostan [30°26’N 67°01’E, ca. 1’ 575 m], Gulistan [30°36’N 66°35’E, ca. 1’ 450 m], Mastung, Pishin, Quetta, WaliKhan [29°53'N 66°51'E (vic. MastungRoad), ca. 1’ 650 m]). – Nikolskij, 1916: 10 (Kundar [Kondor, 32°15’N 59°39’E, ca. 1’ 870 m, ZISP 9993]). – Wall, 1923: 618 (summary review, see Morphology: secondsmallprint). – Moricz, 1929: 33 (“Bendan” [Bandan, 34°18’N 57°22’E, ca. 850 m], “Pul-i-Khatum” [Pasgah-ePol-eKhatun, 35°58’N 61°07’E, ca. 400 m], “Turbet-i-Khejdari” [Torbat-eHeydariyeh, 35°16’N 59°13’E, ca. 1’ 350 m]).

Z. [amenis] karelinii [sic]. – Werner, 1929: [65] 71 (review).

ColuberKarelini [ sic]. – Werner, 1936: 198, 201 (Arusan [NEEsfahan, Dasht-eKavir, 34°08’N 55°07’E, ca. 990 m, notinNMW], “TschaSam” [Chah-eSam], “GuluKahak” [“Gulu Cahak”, GuluChahak], vic. Neh [Nehbandan], “Tauran” [Turan, 35°36’N 56°44’E, ca. 1’ 360 m, notinNMW]: NMW 25446.3, 25446.5-6) .

Coluberkarelini. – Smith, 1943: 169 (review, typelocality “S.W. Asia”).

Coluber (Platyceps) karelini . – Guibé, 1957: 139 (“DashBouroun” [DashliBorun], “Sarakhs” [Serakhs]: MHNH 1957.59-60, seeConclusions, footnote 3, Figs 2Band 9A).

Coluberkarelini [partim]. – Leviton, 1959: 454 [462], Tb. IV (“Chah-i-Angir” [Chah-iAnjir]: CAS 84634-36, see Schätti & Stutz, 2005: footnote 1).

Coluber rhodorachis ladacensis ( Anderson, 1871) [partim]. – Leviton & Anderson, 1961: 275 (same material as in Leviton, 1959).

Coluberkarelinii [sic] [partim]. – Raï, 1965: [20, 21] 43 [75], map 9 (“Kerdahan” [“ Iran ”]: MNHN 8722, 1999.8160, seeDistribution, Tbs 2-3).

Coluberkarelini [partim]. – Minton, 1966: [47] 122 [172] (“QuettaDist.”: nearPishin [incl. SAM 931], 2 miles east of Hanna [Urak Valley]: AMNH 96219-20, see footnote 3).

Coluberkarelini. – Král, 1969: 63, Tb. 1 (vic. AgChah [ca. 36°55’N 66°11’E, ca. 280 m, SNMB (“ SNM ”) 54]) .

Coluber k. karelini . – Mertens, 1969: [3, 10] 56 ( Darzi Chah [ Afghanistan]; Khuzdar, Quetta: SMF 62924, 62940 , 64629 ) .

“ Coluber species ” [ cf. karelini ]. – Clark et al., 1969: 312, Fig. 1 [map] (between Herat and “IslamQala” [IslamQaleh]: CAS 103785).

Coluber karelinii [sic]. – Leviton & Anderson, 1970: [173] 195 (key, distribution).

Coluber karelini . – Tuck, 1971: 61, map 21 (“Khurasan: 1 km. S. Esfideh” [Esfeden]: USNM 148631).

Coluber k. karelini . – Nilson & Andrén, 1981: 139, Abb. 9 (10 km N Mobarakiyeh [Dasht-e Kavir, ca. 35°09’N 51°47’E (vic. Askarabad), ca. 800 m, NHMG (“G.N.M.”) Re. ex. 4422]).

Coluber k. karelini [partim]. – Khan, 1982: [225] 226 (see Distribution: second smallprint).

Coluber (Coluber) karelini [partim]. – Mahendra, 1984: [286] 287 (incl. Platyceps mintonorum [ Mertens, 1969], seefootnote 3).

Coluber karelini . – Khan & Ahmed, 1987: 368, Tb. IV (Mastung, see Morphology: second smallprint).

Coluber karelini [partim]. – Clark, 1990: 33, [unnumb.] Tb. (“10 Km.W Tashkurgan” [Tashqorghan, Khulm] and “45 Km.WHerat”:?CAS 120540, 120714, seeSystematics, HybridRacers, Fig. 7B, Tb. 4).

Coluberkarelini [partim]. – Latifi, 1991: [67] 104, Pl. [Fig. 34, map] (NWEsfahan [“ Central Province ”]: Kashan [33°59’N 51°27’E, ca. 950 m]; Golestan [“ Mazandaran ”]: Aq Qal’eh [37°01’N 54°27’E, ca. 30 m], Gonbad-eQabus [37°15’N 55°10’E, ca. 60 m]; CentralKhorasan [Khorasan-eRazavi]: KalatehNader [37°04’N 56°45’E, ca. 990 m], “Ghoochan” [Quchan, 37°06’N 58°31’E, ca. 1’ 320 m], Mashhad [36°18’N 59°37’E, ca. 1’000 m], Neyshabur [36°13’N 58°48’E, ca. 1’ 200 m], Sabzevar [36°13’N 57°41’E, ca. 920 m]; SouthKhorasan [Khorasan-eJonubi]: Birjand [32°52’N 59°13’E, ca. 1’ 500 m], “Ghaen” [Qa’en (Qayen), 33°44’N 59°11’E, ca. 1’ 450 m], Hoseinabad [Hoseynabad, seeMaterialandMethods], Sarbisheh [32°36’N 59°49’E, ca. 1’ 830 m]; Semnan: “Gharmsar” [“CentralProvince”, Garmsar 35°20’N 52°13’E, ca. 1’ 100 m], Shahrud [Emamshahr, 36°26’N 54°57’E, ca. 1’ 380 m], Torud [35°27'N 55°01'E, ca. 839 m]; Sistan-veBaluchestan: Zabol [31°02’N 61°30’E, ca. 480 m], Zahedan [29°30’N 60°52’E, ca. 1’ 350 m], seeMorphology [dimensions], Distribution). – Khan, 1997: [52] 56 [58], Figs 1, 2A, 6 [map], Tbs 1-3 (incl. Chaman [30°55’N 66°28’E, ca. 1’ 335 m], “PunjPai” [Panjpai, 29°55’N 66°30’E, ca. 1’ 485 m],?”Zob” [Zhob, 31°20’N 69°27’E, ca. 1’ 415 m], seeMorphology: smallprints, Distribution: secondsmallprint, Fig. 5).

Coluberkarelini. – Tunijev etal., 1998: 78 (Khiveabad [ Turkmenistan, 37°11’N 59°33’E, ca. 650 m], see Material and Methods) .

Coluber karelini [partim]. – Khan, 1999: 276, 288 (habitat types, see Distribution: second smallprint). – Latifi, 2000: [136] 261, Pl. [Fig. 34, map] (incl. KermanandYazdProvinces, seeDistribution).

P. [latyceps] karelini comb. n. – Schätti & Utiger, 2001: 935 (see Systematics: second smallprint).

Coluberkarelini [partim]. – Khan, 2002: [23, 30, 45 (“ k. karelini ”, “ k. mintonorum ”), 57] 99, Figs 41-43 [coloured photographs of specimens from “ Turkmenistan ” (ca. 40 km east of AshgabatalongGaragumCanal) and “ Iran ” (26 kmnorthofGonabad, 34°38'N 58°46'E, ca. 900 m, MMTT specimen), resp.], 45a-c [head views, dorsal colour pattern], map 7 (seeDistribution: secondsmallprint, Acknowledgements).

Coluberk. karelini . – Khan, 2004: 196 (“ Pakistan ” [checklist]); Firouz, 2005: 203 (provinces “10, 13, 14, 24, 25”, fideLatifi, 1991 [1985], seeDistribution).

Platyceps karelini . – Nagy et al., 2004: 224, 230, Figs 2-4 (mitochondrial [4] and nuclear [1] genes, see Systematics) and Lawson et al., 2005: 583, Figs 1-3 (DNA sequences [cyt b, c-mos]: CAS 184636 [SW Turkmenistan]). – Schätti et al., 2005: Abb. 8 (molecular data: 2443.03 [ Uzbekistan: Bukhara (Buxoro)], seeSystematics: secondsmallprint).

Platyceps cf. ventromaculatus [partim]. – Schätti, 2006: 677 (Dashli Borun: MNHN 1957.59, seeSystematics, Conclusions, Fig. 2B).

Coluberk. karelini [partim]. – Khan, 2006: [4, 17, 38 (Fig. 28.Biii)] 195, Tb. 10.1 [altitudinal distribution], map (see Morphology: third smallprint, Distribution: second smallprint).

Platyceps k. karelini . – Rastegar Pouyani et al., 2008: 18 (“ Iran ” [checklist, incl. “ P. k. mintonorum ”]).

MORPHOLOGY

Head 2.0-2.47 timeslongerthanbroad. Rostral 1.40-1.80 timesbroaderthan high (see Hybrid Racers: second smallprint). Internasals and prefrontals about equal in size; the latter coalesced in a specimen from Arusan ( Werner, 1936) of unknown whereabouts. Frontal 1.21-1.53 timeslongerthanbroad, 1.18-1.63 timeslongerthan internasals and prefrontals, 0.92-1.22 times as long as parietals. Posterior border of the latter straight, slightly indented at the median suture, or forming an obtuse concave angle.

Nostril-eye distance 0.69-0.90 times length of internasals and prefrontals. Loreal usually as long as high, situated on third and posterior part of second supralabials; fused with nasal in AMNH 96220 (both sides) and TMUS 1002 (right); two superposed loreals in USNM 148631 . Preocular entire and in contact with frontal exceptinBMNH 1882.3.20. 2 (rightside), FMNH 141604 , andMNHN 1957.60. Anterior subocular always present; presubocular absent (59% based upon specimens with pertinent data, see Material and Methods) or present (41%, bilateral in BMNH 1882.3.20. 2, 1886.9.21. 102, CAS 103785 , MNHN 1957 .59, 1999.8160, NMW 25446.5, PMNH 761 , RUZM 11 .1, SMF 62924, 62940 , USNM 148631 and 240003 [Fig. 10A] as well as on one side in MHNG 2718 .12, MNHN 8722 , and NMW 25446.4). Mertens’s (1969) remark regarding the presence of an additional scale (presubocular) below the loreal in SMF 64629 (“als ein zweites (unteres) Loreale ausgebildet”) is by mistake; there is no presubocular nor any other supplementary scale between the anterior subocular, loreal and supralabials .

Normally nine supralabials; eight on one side in the unregistered Qom specimen, AMNH 96220 with eight/ten, ten in CAS 103785 and on the ride side of BMNH 1882.3.20. 2, CAS 84634, and PMNH 761 . Fifth supralabial in contact with eye (unilaterally sixth in AMNH 96220 and PMNH 761 ); orbit completely separated from supralabials by a row of three suboculars (i.e., fifth supralabial horizontally divided) in TMUS 1002 . Posterior subocular distinctly larger than anterior. Two postoculars and anterior temporals (see footnote 7); two or three scales in second temporal row; lower anterior scale larger (upper particularly small in, e.g., NMW 25446.6 and SMNS 2381 or on left side of SMF 64629, see also Fig. 4C) .

Khan (1997) notified that “in 97%” (n=27) of Pakistani specimens the fifth supralabial enters the eye and “3% have 5th on one side and 5th and 6th on the other, one specimen has none in contact on one side, one on the other.” However, the latter condition (supralabials unilaterally excluded from orbit) is not tabulated (l.c.: Tb. 1). Wall’s (1911) remark that “the 3rd, 4th and 6th being divided, and the 4th, 5th and 6th touch the eye” is due to a different and incomprehensible terminology for supralabial scales. Leviton (1959) called the posterior subocular “subpostocular”; Minton (1966) and Mertens (1969) as well as many earlier authors (e.g., Strauch, 1873; Boettger, 1888) considered the anterior and posterior subocular scales as the lowest preocular and postocular, respectively.

Ten sublabials (nine in MNHN 1957.60), four in contact with first inframaxillary (three in USNM 148631), sixth largest. Anterior inframaxillaries shorter and broader than posterior pair (about the same length in MNHN 1957.59); posterior chin shields separated by two (occasionally three) rows of elongate or lanceolate scales anteriorly and four to five (three) behind. Gulars in four (three to five) oblique rows of scales between the caudal edge of the posterior inframaxillaries and the first ventral.

Ventrals in examined specimens 196-212 (ƋƋ 196-209, ♀♀ 201-212); preventrals (usually one or two) absent in CAS 84634-35 ; last scale incompletely developed in TMUS 994 and 1002. Anal scute divided, 90-106 (ƋƋ 90-105, ♀♀ 91- 106) paired subcaudals. Sum of ventrals and subcaudals 288-314 (ƋƋ 288-311, ♀♀ 294-314, Tb. 3). Males from the vicinity of Kandahar (no females at hand) differ visà-vis remaining Afghan populations in distinctly fewer ventrals and subcaudals (see also Tb. 2). Kandahar and Pakistani males (n=7) have fewer total body scales than those of more northern and western populations (288-303 versus 303-311) .

Strauch (1873: 116, 272) enumerated two specimens (ZISP 1701-02) of unknown gender from “ Persien ” (probably “Chorassan oder Kirman ”, coll. “Graf Keyserling”) with 199-200 ventrals; the latter has 102 subcaudals (sum 302); “89” subcaudals for ZISP 1701 (sum 288) need confirmation (tail possibly incomplete). More ventrals (“2+216”) and total body scales (320) than in Iranian females examined by us are reported from the Kavir Desert (NHMG 4422, Nilson & Andrén, 1981); these findings require verification. Data for BMNH 1882.3.20. 2 (damaged Ƌ, 198 ventrals, 100 subcaudals) in Boulenger (1893: letter g, 193 and 98, respectively) are without missing ventrals. Wall (1911) notified as few as 192 ventrals (92-99 subcaudals) for “several specimens” (at least six, see chresonyms) deposited in the ‘Quetta Museum’. In view of the proximity of these records to northern Kandahar (Tb. 3), we do not reject a priori this minimum (192); however, ventral counts of close to 190 in populations from Northeast Baluchistan Province are in need of confirmation. The report of up to 111 subcaudals by Wall (1923) is most probably from Central Asian Platyceps karelini as evidenced by the maximum (213) for ventrals (see Northern Populations) 2). The by far highest number of subcaudals (110) for Pakistan ( Khan & Ahmed, 1987) in a specimen of unknown gender from Mastung remains unconsidered in this study. Khan’s (1997) total body scale counts (282-317) are not based upon individual data but are simply the sum of the extremes for ventrals (192-207) and subcaudals (90-110). Brück’s (1968) “ karelinii ” [sic] from the Morghab (Murgab) in Northwest Afghanistan with 147-150 ventrals, 62-90 subcaudals, and including a melanistic individual belong to a colubrid genus different from Platyceps Blyth (see next smallprint).

Dorsal scales with paired apical pits, normally in 19-19-13 rows along trunk (Tb. 1). Males with this formula have the first and second bilateral reductions between ventrals 114.5-128 (56-62%ven) and 117.5-129.5 (57-64%ven), respectively; the third (last) is situated at ventrals 138-165.5 (67-80%ven). The most cranial positions are from TMUS 994; whereas differences versus all remaining males with pertinent data are slight in the case of the anterior reductions, the third (at ventral 138 or 67%ven) is much farther cranial than the second lowest absolute level (146, AMNH 96219) or relative position (72%ven, MHNG 2718.11). All females conform to the standard formula (19-19-13 dsr). With the exception of MNHN 1957.60 (see footnote 3) and USNM 148631 (reduces to 17 dsr at ventral 110.5 or 53%ven and to 13 dsr at 140.5 or 68%), the verified reductions occur between ventrals 116.5 to over 125 in NMW 25446.5 (55-62%ven), 118.5-128.5 (57-63%ven), and 143-166.5 (69-80%ven), respectively.

3)

More often than not, the first reduction is paravertebral followed by a lateral fusion (Tb. 1). Usually, rows 2-4 and 7-9 are involved in the anterior reductions; the sixth row participates unilaterally in MNHN 1957.60 and TMUS 1002 (17-15 dsr); the vertebral involves in AMNH 96220, SMNS 2381, and USNM 148631 (19-17 dsr). The transverse positions of the first and second reductions are mixed in MHNG 2718.12; another male (ZMB 6876) reduces to 17 dsr (‘low-high’) on the left side followed by the reversed sequence to 15 dsr on the right. The third reduction usually involves rows 6-8 (5-7 in AMNH 96219); the vertebral row participates in CAS 84636, 103785, 120714, FMNH 141604, MHNG 2718.12, NMW 25446.4, SMF 62924, SMNS 2381 as well as USNM 148631 and 240003 (SMNS 2381 and USNM 148631 with an additional median reduction farther cranial, see above).

The reductions of SMF 64629 (Ƌ) with 19-19-11 dsr involve rows 8+9, 3+4, 6+7, and 5+6 between 50-89%ven. Two males from near Kandahar (MNHN 8722, 1999.8160) show aberrant overall patterns (Tb. 2). In the former, the reduction to 17 dsr and a vertebral split (immediately followed by a median fusion) occur before midbody (46-48%ven). Without taking account of the addition above ventral 104 (simultaneous with a lateral reduction on the same side), MNHN 1999.8160 exhibits five changes in the number of dsr involving the vertebral row.

2) The ventral counts of many specimens mentioned in literature probably include one or two preventrals.

3) The number of dsr immediately prior to the anal scute may be lower (12, FMNH 141604, ♀) or higher (15, BMNH 1886.9.21. 102-103, ♀♀). Guibé (1957) reported 21 dsr on the anterior trunk of MNHN 1957.60 (♀, Fig. 9A); it has 19-19-13 dsr and the reductions behind midbody are situated farther cranial than in most females (Tb. 1). Minton (1966) notified an “increase to 21 on neck” in an unspecified specimen from Pakistan (possibly SAM 931, unexamined, see also Mahendra, 1984). This count was probably taken in front of the 15 th ventral. TABLE 1. Number of ventrals and dorsal scale row (dsr) reduction pattern in Platyceps karelini from Iran to Pakistan with 19-19-13 dsr at the 15 th ventral, midbody (msr), and five ventrals prior to the vent (see footnote 3). The longitudinal positions (ventrals) indicate means for bilateral reductions; no precise data for third fusion of dsr in MNHN 1957.60 (approximate position, transverse level not ascertained) and anterior reductions of NMW 25446.4- 5 (verified level with 19 and 15 dsr in parentheses). Abbreviations: lat (lateral), par (paravertebral), p+v (paravertebral levels and vertebral row).

Positions along trunk Transverse levels

Accession number Ventrals in absolute numbers in %ven or remark (dsr) Gender and origin

AMNH 96219 198 115.5, 119, 146 58, 60, 74 par - lat - par Ƌ, Pakistan

AMNH 96220 205 119, 126,? 58, 61,? p+v - lat -? Ƌ, Pakistan

BMNH 1882.3.20. 2 198 - - 19-19-13 dsr Ƌ, Afghanistan

CAS 84635 204 - -? -? - 13 dsr Ƌ, Afghanistan

CAS 84636 209 - - 19 msr Ƌ, Afghanistan

MHNG 2718.11 205 123, 128.5, 147.5 60, 63, 72 par - lat - par Ƌ, Iran

MHNG 2718.12 208 127.5, 129.5, 165.5 61, 62, 80 mixed - p+v Ƌ, Iran

NMW 25446.6 203 121, 129, 152 60, 64, 75 par - lat - par Ƌ, Iran

PMNH 761 207 - - 19-19-13 dsr Ƌ, Pakistan

PMNH 762 198 - - 19-19-13 dsr Ƌ, Pakistan

SMF 62924 200 117, 119, 148.5 59, 60, 74 par - lat - p+v Ƌ, Pakistan

TMUS 994 205 114.5, 117.5, 138 56, 57, 67 par - lat - par Ƌ, Iran

TMUS 1000 206 120, 128.5, 150.5 58, 62, 73 par - lat - par Ƌ, Iran

TMUS 1002 206 128, 128.5, 163 62, 62, 79 lat - par - par Ƌ, Iran

ZMB 6876 202 121, 124, 150.5 60, 61, 75 (see text) - par Ƌ, Iran

NMW 25446.3 202 - - 19 msr juv., Iran

TMUS 1001 205 - - 19 msr juv., Iran

unregistered (vic. Qom) 204 - - 19-19-13 dsr?, Iran

BMNH 1886.9.21. 102 208 117, 122.5, 153 56, 59, 74 par - lat - par ♀, Afghanistan

BMNH 1886.9.21. 103 208 117.5, 122, 152 56, 59, 73 par - lat - par ♀, Afghanistan

CAS 84634 212 117.5, 124.5, 147.5 55, 59, 70 par - lat - p+v ♀, Afghanistan

CAS 103785 207 116.5, 118.5, 143 56, 57, 69 par - lat - p+v ♀, Afghanistan

CAS 120714 209 121, 125.5, 166 58, 60, 79 lat - par - p+v ♀, Afghanistan

FMNH 141604 202 124.5, 126, 152.5 62, 62, 75 lat - par - p+v ♀, Iran

MNHN 1957 .59 201 121.5, 123, 160.5 60, 61, 80 lat - par - par ♀, Iran

MNHN 1957 .60 203 115.5, 116 (ca. 135) 57, 57 (ca. 67) lat - par -? ♀, Iran

NMW 25446.4 209 (120-128) 161 (57-61) 77? -? - p+v ♀, Iran

NMW 25446.5 208 (125-135) 166.5 (60-65) 80? -? - par ♀, Iran

RUZM 11 .1 204 119, 122, 151 58, 60, 74 par - lat - par ♀, Iran

SMF 62940 204 123, 128.5, 154 60, 63, 75 par - lat - par ♀, Pakistan

SMNS 2381 207 121, 123.5, 159.5 58, 60, 77 p+v - lat - p+v ♀, Iran

USNM 148631 208 110.5, 119, 140.5 53, 57, 68 p+v - lat - p+v ♀, Iran

USNM 240003 204 118, 126, 148.5 58, 62, 73 par - lat - p+v ♀, Iran

TABLE 2. Dorsal scale row (dsr) reduction pattern of Platyceps karelini MNHN 8722 (197 ventrals) and 1999.8160 (196) from the vicinity of Kandahar, Afghanistan. Right (upper line) and left longitudinal positions in absolute numbers of ventrals. Changes in the number of dsr involving median rows in boldface. Abbreviations: v.red . (vertebral reduction, in case of even number of dsr), v.s. (vertebral split).

MNHN 8722

7+8 (91) 2+3 (118) 6+7 (129) 3+4 (169) 4=4+5 (183)

19 17 v.s. (94) 18 17 15 13 11 13

9 (90) 8+9 (95) 2+3 (115) 6+7 (129) 3+4 (167) 4=4+5 (182)

MNHN 1999.8160

8+9 (103) 3+4 and 8 = 8+9 (104) 8 = 8+9 (105) 8+9 (108)

19 18 19 18 16 [17] 16 17

3+4 (97) 3=3+4 (100) 8+9 (101) 3+4 (103) 6+7 (107)

cont. MNHN

1999.8160

3+4 (162)

15 14 v.red . (129) 13 12 13 11

7+8 (127) 3+4 (158) 4=4+5 (160) 4+5 (162)

FIG. 1

Male Platyceps karelini from Hoseynabad-e Mish Mast (34°27’13’’N 51°10’02’’E, Esfahan Province), approximately 35 km southeast of Qom, Iran ( QDE specimen, see Material and Methods) .

Longest specimens approximately 700 mm snout-vent length (Ƌ, NMW 25446.6, tail incomplete) and about 670 + 212 mm (♀, SMF 62940); Khan & Ahmed’s (1987) 680 + 230 mm are from a specimen of unknown gender (see smallprint above). Latifi’s (1991) “ 107 cm; tail, 25 cm ” may rely upon a taxon different from Karelin’s Racer (see Comparison). Smallest individuals 185 mm snout-vent length (CAS 84636, fide Leviton, 1959) and 200 + 65 mm (NMW 25446.3). Tail/body ratio 0.31-0.40 for males and females.

Dorsal head colour pattern usually absent but often with a fine line along interparietal suture (Figs 1-2, 9A). A dark slanted subocular blotch and a dark oblique streak from the angle of the mouth to the anterior lateral border of the parietals always present. Exterior edge of supraocular diffusely brown or blackish above the eye in BMNH 1886.9.21. 102 (Fig. 2A), MNHN 8722 , 1957 .60, SMNS 2381 , and a QDE Platyceps karelini (Fig. 1, see last paragraph of Comparison section regarding peculiarities of dorsal head pattern in examined specimens from SW Esfahan, Markazi, and Qom). MNHN 1957 .59 (Fig. 2B, see Guibé , 1959) displays conspicuous deep black markings on the pileus made up of ‘eyebrow’ flecks, a broad mid-parietal line with bilateral twirls, a hexagonal protrusion at the fronto-parietal border (with a short median line running to a slightly less distinct dark area of similar size on anterior edge of frontal), a half-moon mark behind the parietals, and these scales encroached upon laterally by the temporal streak (also observed in SMNS 2381 ). The latter extends across the parietal meeting its counterpart mid-dorsally in BMNH 1886.9.21. 102, forming an obtuse angle (Fig. 2A). Some specimens (e.g., MNHN 8722 , 1957 .60) show a roundish, subelliptic or short elongate nape spot. The latter is prominent in MNHN 1957 .59 (see above, Fig. 2B) and connected to the post-parietal fleck by a short bar; a unilateral transverse line runs towards a dark mark behind the posterior exterior border of the parietal. A veritable nuchal streak is present in CAS 120714 and NMW 25446.5 (faint) or a QDE male (Fig. 1) and SMNS 2381 , which have it fused with the first right nape blotch; in BMNH 1886.9.21. 102 the right and left portion of the anteriormost cross-band are bent cranially and converge into a striking mid-dorsal wedgeshaped extension (Fig. 2A) .

Body above creamish, pale grey, tan or buff with 41-60 brownish to black crossbands. They are widest on neck, narrower than interspaces, and distinct along the whole trunk but fading away on tail. Specimens from Afghanistan and Iran as well as AMNH 96220, SMF 62924, and SMF 62940 ( Pakistan) have 41 to 54 complete transverse blotches. More (58-60) are found in three Pakistani Platyceps karelini (AMNH 96219, PMNH 761-62). On the neck, and sometimes farther behind (e.g., NMW 25446.6), the cross-bands may extend to the venter but they normally do not reach beyond the flanks and alternate, at least posteriorly, with a ventrolateral series of dark spots or bars usually encroaching upon the lateral edges of the ventrals. Underside of neck and venter ivory, ochre, or yellowish, sometimes with a pale orange (e.g., CAS 103785), pinkish, or salmon hue (Fig. 1).

The minimum for the number of complete cross-bands (41) is from SMNS 2381 with a vague origin (see Material and Methods) and the maxima for Iran and Afghanistan rely upon CAS 120714 (see below) and SMF 62940 (53). An unexamined QDE specimen has about 52 (Fig. 1, see Comparison). About 42 are present in MNHN 1957.60 (Fig. 9A, see Northern Populations: second smallprint); NHMG 4422 from Tehran [Teheran] Province shows 43 ( Nilson & Andrén, 1981). Afghan specimens with pertinent data available (no counts ascertained for CAS 84634-36) have 45-54 or more complete dorsal bands (see Material and Methods); the maximum based upon CAS 120714 (Herat area) requires detailed definition. A potential Platyceps karelini x P. rhodorachis (CAS 120540) from North Afghanistan (Khulm) with approximately 78 comparatively short transverse blotches fits the number of dorsal markings (68-88) observed in two Tadzhik specimens (MHNG 2442.98, MTKD 16095) collected within less than 100 km airline from Khulm (see Northern Populations, Hybrid Racers). A “ Coluber karelini ” without origin reproduced in Khan (1993: Fig. 13, 1997: Fig. 3, and 2006: Pl. 123) is not this species as evidenced by over 70 cross-bands and the neck pattern. Actually, this picture is manipulated; in reality, two entangled snakes are discernible and the almost completely visible specimen is a Platyceps sp. other than Karelin’s Racer, i.e., the taxon revalidated in this paper (see Systematics, Comparison: first smallprint, Conclusions). The report of a melanistic karelini FIG. 2

Dorsal head pattern in Platyceps karelini from Afghanistan (A) and Iran (B-C): BMNH 1886.9.21. 102 (A, Herat), MNHN 1957 .59 (B, Golestan), and USNM 240003 (C, Esfahan, subadult, see Fig. 10A). Courtesy of Ivan Ineich / MNHN. Not to scale .

from “ Afghanistan ” ( Brück, 1968, see smallprint above) is probably Natrix tessellata ( Laurenti, 1768) .

Maxillary with 13-15 or 16 (USNM 148631) teeth, anterior series subisodont, diastema usually wide, posterior two teeth enlarged, last offset laterad. Palatine with 8 (9), pterygoid 15-16 ( Wall, 1911), and dentary (15) 16-19 teeth. Hemipenis subcylindrical, spinose at base, apical area calyculate; spines subequal in size except for the fringe of denticules along the sulcus spermaticus; border of calyces serrated (Fig. 3).

NORTHERN POPULATIONS

Platyceps karelini from Turkmenistan to Kyrgyzstan and Tadzhikistan (Appendix B) show mostly similar variation of head scales (see footnote 7), the number and reduction pattern of dorsal scale rows, body proportions, dentition (in particular maxillary and dentary teeth), and hemipenis ornamentation as observed in southern populations.

ZISP 19031.1 displays a loreal which is higher than long (Fig. 4C). The preocular (entire in specimens from Iran to Afghanistan and Pakistan except CAS 120540: Fig. 7B, see Hybrid Racers) is divided in MTKD 13602 (right side), MTKD 16095 (both sides), and ZISP 19031.1 (right, with a faint cleft on the left); a short notch is dis- FIG. 3

Sulcate view of everted left hemipenis of Platyceps karelini MHNG 2442 .99. Scale equals 5 mm. Drawing Heidi Laubscher and Andrea Stutz.

cernible in ZISP 17219 (Figs 4C-D). A presubocular is absent in most Platyceps karelini from former Republics of the USSR; only SMF 18219, ZISP 17214.1, 19031.1, and ZMB 38816 show a bilateral scale of variable size in front of the anterior subocular (Figs 4B, 4D). ZISP 14741 (right) and 19031.1 (both sides) possess an additional (third) subocular completely excluding the eye from the supralabials (Figs 4A, 4D); this condition is also reported in two syntypes ( Strauch, 1873: ZISP 1696, 1698). SMF 18220, ZISP 14741 (left), 17219, and 17682 (right) have ten supralabials (fifth or sixth in contact with eye); the same number occurs on both sides of ZISP 1706 ( Strauch, 1873), an unlocated specimen from the lower course of the Murgab ( Boettger, 1888: “10–10”), and on the right of an unspecified individual ( Strauch, 1873: “als Duplicat

4)

4) A small granular scale between the anterior subocular (“Praeoculare inferius”), the loreal, and the third and fourth supralabials ( Strauch, 1873) is observed in the holotype of Choristodon brachycephalus Severczov, 1873 (ZISP 3581) from Khujand (“Khodzhent” or “Chodshent”, 40°17’N 69°38’E, ca. 320 m) in the Fergana Valley, Tadzhikistan. It has 200 ventrals, 85 subcaudals ( Strauch, 1873, see Northern Populations: first smallprint), and the lowest total body scale count for Platyceps karelini reported in the literature consulted by us. This racer with about (“gegen”) 80 short juxtaposed transverse dorsal bars is also noteworthy for the exceedingly developed cuneiform rostral wedged in deeply between the internasals (“Praefrontalia”) and almost completely separating them, as well as the aberration of its head (right side shorter than left, hence the scientific species name) including an elliptic eye ( Strauch, 1873). If not an otherwise aberrant specimen, ZISP 3581 is a malformed P. karelini x P. rhodorachis possibly descending from at least one crossbreed parent. The hybrid hypothesis is supported by comparative data for six P. karelini from Tadzhikistan (Appendix B) with 208-217 ventrals, 105-113 subcaudals (n=4), and a much higher sum thereof (315-323).

FIG. 4

Right lateral head scales of Platyceps karelini from Turkmenistan (B-D) and Uzbekistan (A, Karakalpakstan): ZISP 14741 (A), 17214.1 (B), 17219 (C), and 19031.1 (D). Variation of temporals as well as uncommon or rare conditions, i.e., shape of loreal (D), occurrence of a presubocular (B, D) or divided preocular (D), ten supralabials (C), and fifth supralabial excluded from contact with eye (A, D). Scale equals 10 mm. Drawings Heidi Laubscher and Andrea Stutz.

ausrangirt”, see next smallprint) 5). Eight supralabials are found in ZISP 17386 (right, fourth entering orbit) and 19031.1 (left). The upper portion of the right posterior subocular is split off in NMW 25446.1. The posterior subocular and the lower postocular are coalesced in ZISP 1697 (left) and 1698 ( Strauch, 1873). Boettger (1888) noted a unilateral case of four scales along the posterior border of the eye (“Postocularen […] 3–4”) in a specimen from the Murgab Valley (see Morphology: first smallprint, footnote 5). ZISP 3647 has the left upper postocular fused with the supraocular plate ( Strauch, 1873). We observed eleven (instead of the usual ten) sublabials on one side of MTKD 13436, NMW 25446.2, ZISP 17582 and 19031.2 as well as in ZMB 38816 6).

5) We located and examined four out of six Turkmen Platyceps karelini reported by Boettger (1888), i.e., NMW 25446.1-2 (incl. possible hybrid) and SMF 18219-20. Two specimens of unknown gender obtained along the Murgab (“am Murgab”) with 205-211 ventrals and 102-104 subcaudals, respectively, are possibly deposited in the “k. kaukasischen Museum” in Tbilissi (today incorporated into GNM collections) as notified in the introduction (l.c.: 875).

6) Three out of five northern specimens with eleven sublabials (NMW 25446.2, ZISP 17582, ZMB 38816, see Hybrid Racers) originate from Southeast Turkmenistan. A higher than usual number of sublabials is also found in ZMB 38833 and on one side of BMNH 1873.1.7. 10 (both Platyceps karelini x P. rhodorachis ) as well as a potential hybrid from North Afghanistan (CAS 120540).

Ventrals 201-218 (ƋƋ 202-214, ♀♀ 203-218, 201- 217 in specimens of unknown gender), subcaudals 85-117 (88-117, 92-115, and 85-111, respectively), sum thereof 286-331 (293-331, 296-326, 286-323, resp., Tb. 3). Females of northern groups have slightly more ventrals than males. Northeastern populations (ƋƋ, ♀♀) have higher means of ventrals and subcaudals than Platyceps karelini from any other area. Males from Southwest Kazakhstan to the Kopetdag (see footnote 8) and Bukhara (Buxoro, Uzbekistan) clearly differ from southern and eastern populations in the number of subcaudals (88-103 versus 102-117) and total body scales (293-310 vs. 308-331). Terentjev & Chernov (1940, 1949) and later authors notified 192-220 ventrals and 85-117 subcaudals (identical minima are quoted in, e.g., Bannikov et al., 1977). Their lowest ventral count most probably relies on Wall’s (1911, 1923) data for P. karelini from Pakistan (see Morphology: second smallprint).

ZISP 1705 and 3646-47 ( Strauch, 1873) from “Kenderlinsk” (Kendirli Bay, SW Kazakhstan), “Karatschagly” in the Great Balkhan Range (NW Turkmenistan), and Uzbekistan (“Altes Bett des Oxus”) are classified as females due to their high ventral counts (211-213, 97-100 subcaudals, Tb. 3). The minimum for subcaudals (85) also reported by, for instance, Boulenger (1890, 1893) is from two syntypes ( Strauch, 1873: ZISP 1695-96, “Originalexemplar[e]”; type locality possibly in SW Kazakhstan) and the holotype of Choristodon brachycephalus Severczov from the Fergana Valley (see synonymy, Hybrid Racers, footnote 4). The latter (not allowed for in Tb. 3) accounts for the minimum (285) of ventrals and subcaudals combined in Strauch (1873: 273). Our lowest number (286) relies upon ZISP 1696; the tail of this possibly male syntype portrayed in Strauch (1873: Pl. III) appears to be intact. The morphological description of Platyceps karelini by this author encompasses data of twenty-one specimens including three that had been exchanged with other museums (the “Verzeichniss der […] aufgestellten Exemplare” lists only eighteen deposited in the Imperial Academy of Sciences). This explains discrepancies between subcaudal counts (85-101 according to index versus up to 107 as indicated in the body of the text) and the maximum for ventrals (“bei fünf Stücken [...] mehr als 210”); to conclude from that latter remark, two out of three specimens then no longer in the St. Petersburg collections had more than 210 ventrals ( Strauch, 1873: 113-14, 272-73). Boettger’s (1888) “95” subcaudals for SMF 18220 from Durun is based upon an incomplete tail, and the lowest value (“91”) in the Murgab series is considered incorrect (tip of tail most probably missing, see footnote 5). The maximum for ventrals (220) reported by Terentjev & Chernov (1940, 1949) may originate from one or several specimens collected in the northeastern portion of the distribution range; this count probably comprises one or two preventrals and the actual number may be identical with the maximum for ventrals (218, MHNG 2442.96) observed in this study (Tb. 3).

A female from Uzbekistan (ZISP 13110, see next smallprint) shows 17 midbody scale rows; the first reduction (19-17 dsr) occurs at ventral 71 (35%ven) by fusion of lateral rows. MTKD 16095 (Ƌ) with an irregular reduction formula has 17 dsr on a portion of the anterior trunk, 19 msr, and 11 dsr prior to the vent. A fourth fusion (13-11 dsr) involving paravertebral rows at ventral 166 (81%ven) is present in MHNG 1358.27 (Ƌ).

The verified number of dorsal cross-bands of most specimens from Turkmenistan (e.g., CAS 184636, MHNG 1358.27, MTKD 8281, NMW 25446.1-2, see Hybrid Racers: last paragraph) and MTKD 13602 from Southeast Kazakhstan (42 to ca. 55) is virtually identical to the range of Platyceps karelini from Iran and Afghanistan (41 to at least 54, see Morphology: third smallprint). Strauch (1873: incl. Pl. III) reported 40-48 in specimens from Southwest Kazakhstan, Turkmenistan, and ZISP 3647 from the original course of the Amu Darja (see smallprint above). However,

TABLE 3 . Ventrals , subcaudals, and total body scales in Platyceps karelini from Iran to Central Asia and Baluchistan (Appendices A-B) including literature records. The Kandahar male sample comprisesBMNH 1882.3.20. 2, MNHN 8722 , andMNHN 1999.8160 (seeTb. 2). Parentheses following the sample size (n) in the last column indicate number of specimens with intact tail. Boldface numbers denote literature data from Iran and Pakistan deemed in need of confirmation (see Morphology: second smallprint, Northern Populations incl. first smallprint, footnotes 2 and 4-5) .

Total body

Group and region Gender Ventrals Subcaudals Literature source, remark

scales

ƋƋ 202-207 (4) 88-103 (4) 293-310 (4)

SWKazakhstan 204.5 ± 2.1 96.5 ± 6.6 301.0 ± 7.5

andN unknown 201-209 (14) 85-101 (12) 286-306 (12) Strauch (1873), n=12 (11) Turkmenistanto 205.1 ± 2.7 94.0 ± 5.3 298.9 ± 6.7

CentralUzbekistan

♀♀ 203-214 (13) 92-106 (11) 296-320 (11)

Strauch (1873), n=3 populations SETurkmenistan andSUzbekistan ƋƋ unknownBoettger 205 209 208 209 206... - - 5 8 6 211 213 ± ± ± 2 3 3 ((... 5 4 4 8 2)) 97 102 108 105 102.3.. - - 2 3 116 110 ± ± ± 4. 4 6 ((7.. 5 3 0 2)) 306 315 317 308 307... - - 0 0 8 326 323 ± ± ± 7 8 8 ((... 3 5 7 5 0)) (1888), n=2 Northern SE Kazakhstan ♀♀ ƋƋ 204 207 210 210.. - - 0 2 214 216 ± ± 8 2 ((.. 5 2 9 5)) 112 107 106. - 0 (117 1 ±) 4 (. 4 2) 321 310 314. - 8 (331 1 ±) 7 (. 4 2)

Kyrgyzstan,

Tashkent area (E

unknown 208-217 (3) 105-111 (2) 315-319 (2)

211.7 ± 4.7 108.0 ± 4.2 317.0 ± 2.8

Uzbekistan), and

Tadzhikistan ♀♀ 211 214 -.1 218 ± 3 (. 5 3) 98 106 -115.3 ± (7 4.) 7 320 316. - 5 326 ± 4 (. 4 2)

202-208 (7) 97-105 (5) 303-310 (5)

ƋƋ 204.9 ± 2.0 100.2 ± 3.3 304.8 ± 2.9 unknownStrauch 199-205 (5) 98-102 (3) 300-305 (3) (1873), n=2 (1) Iran 202.0 ± 2.6 100.0 ± 2.0 302.3 ± 2.5

201-209 (9) 92-106 (7) 294-314 (7)

populations Afghanistan ƋƋ

♀♀ 204 205 196 216 206.. - - 3 1 (198 209 1 ± ±) 2 2 ((.. 3 3 9 5)) 98 92 105 104. - 6 100 ((± 1 1)) 4 (. 3 7) 311 303 320 288. - (4 (298 1 1 ±)) 6 (. 3 4) Nilson Kandahar & Andrén area (1981)

Southern ♀♀ 208

197 207.. - 8 0 212 ± ± 1 1 (.. 5 9 0) 96 91 101. - 7 105.3 ± ± 4 (. 3 4 2.) 2 293 300 307.. - 7 8 313 ± ± 5 6 (.. 4 0 1)

ƋƋ 198-207 (5) 90-96 (4) 290-303 (4)

201.6 ± 4.2 94.0 ± 2.7 296.5 ± 6.0

Pakistan (NE

Baluchistan) unknown 202 (1) - - Kahn & Ahmed (1987) 192 -206 92-99 - Wall (1911), n≥6 (?)

♀♀ 204 (1) 97 (1) 301 (1)

ZMB 38816 (♀) collected on the Turkmen-Khorasan border (Sarahs) has as much as about 65 transverse blotches, and even more (66-88) seem to be the rule in Kyrgyzstan andTadzhikistan ( MHNG 2442 .96-98, MTKD 10450 , 11335 , 16095 , NHMB 21058 , ZMB 38591, see also Morphology: third smallprint, footnote 4). Two specimens from the vicinity of Bukhara ( MHNG 2443 .03, MTKD 13944 ) have at least 60 cross-bands .

In addition, ZMB 38816 differs vis-à-vis female Platyceps karelini from Southwest

Kazakhstan, northern Turkmenistan and southeast to Tirmiz (Termez, ZISP 13110 , 204 ventrals, 106 subcaudals) on the Uzbek-Afghan frontier ( Balkh Province ) in, for instance, a higher number of ventrals (216) as typically observed in females from Kyrgyzstan and Tadzhikistan (Tb. 3, see Hybrid Racers, footnote 7). MNHN 1957 .60 (♀, Fig. 9A) collected in the immediate vicinity (Serakhs, Khorasan-e Razavi) of ZMB 38816 across the Tedzhen (Harirud) River has 203 ventrals, 99 subcaudals, and as few as ca. 42 complete transverse dorsal markings (see Material and Methods, Morphology: third smallprint) .

DISTRIBUTION

Platyceps karelini occurs from the northwestern Central Plateau ( Iran) and the northeastern Caspian littoral ( Kazakhstan) to the western Tien Shan region, the foothills of the Pamir, and Southwest Pakistan (Baluchistan). The northern limit of distribution is near 47°N latitude north of the former Aral Sea in Kazakhstan ( Terentjev & Chernov, 1949: map 29; Bannikov et al., 1977: map 114). Brushko (1983) reports a specimen photographed between Mojyunkum (Furmanovka) and Mount Dzhambul, roughly 100 km airline from the southern end of Lake Balqash (Zhambyl Province, Kazakhstan). In the area under consideration, the species is known from east of the Zagros Mountains and the Atrek River (Golestan) in Iran to North Afghanistan (see below) and south through Yazd, Kerman, northern Sistan-ve Baluchestan (e.g., TMUS 1000; Annandale, 1906; Werner, 1936; Latifi, 1991; see next but one smallprint and Material and Methods regarding the origin of SMNS 2381), and western Afghanistan to Baluchistan (Fig. 5).

Mentions from the Arabian Peninsula (e.g., Anderson, 1896; Corkill & Cochrane, 1966; Gasperetti, 1974, 1977; Leviton, 1987; Leviton & Aldrich, 1984) rely on Bedriaga (1879) who erroneously assigned Blanford’s (1876) Zamenis ventrimaculatus [sic] from Ras Musandam (Masandim) to Karelin’s Racer (see Systematics). This specimen (ZMB 10324) from an insular promontory in northernmost Oman (“Cape Massandim, Arabian coast, entrance to Persian Gulf”) belongs to Platyceps cf. rhodorachis (Jan) , the only racer species living in that area.

Raï’s (1965: map 9, p. 46) indication from Kermanshah in Northwest Iran is incorrect (see Comparison: second smallprint) and “Kerdahan” (MNHN 8722, 1999.8160) is Chahar Dahaneh near Kandahar, Afghanistan. Reports from Khuzestan and East Azarbayjan ( Latifi, 2000) are based upon different Platyceps spp. The purported occurrence in “Markazi (Central Prov.)” notified by Firouz (2005: province no. 25) ultimately relies on Latifi’s (1991) P. karelini from Kashan in Esfahan (see chresonyms). However, the westernmost verified collecting site (RUZM 11.1) is indeed from Markazi, i.e., about 20 km roughly southeast of Delijan close to the border with Esfahan. The species is also recorded from southeastern Tehran Province (NW Dasht-e Kavir, Nilson & Andrén, 1981) and adjacent Semnan (Garmsar, Latifi, 1991). Specimens from “Mazandaran” ( Latifi, 1991, 2000) including ZMB 6876 were obtained in Golestan, which formed part of the former province until 1997. The reported presence all over Sistan-ve Baluchestan to as far south as the Gulf of Oman littoral ( Latifi, 1991, 2000: maps) requires confirmation and, in particular, comparison with P. ventromaculatus (see Systematics). Information by local residents towards the senior author regarding the occurrence of P. karelini in Hormozgan is probably due to confusion with Lycodon striatus bicolor ( Nikolskij, 1903) which shows a similar dorsal colour pattern. A recent herpetological investigation of that area by Rajabizadeh et al. (2008) did not provide any evidence for the presence of Karelin’s Racer.

FIG. 5

Distribution records of Platyceps karelini (solid circles) in Afghanistan, Iran, and Pakistan including five entries (symbol X) based upon two P. karelini x P. rhodorachis from Afghanistan (Herat, Nimruz) and three supposed or potential hybrid racers from Iran (Kerman) and Afghanistan (Arghandab River, Nimruz and vic. Khulm, Balkh). The arrow in Southwest Afghanistan points to Kamran (see Systematics: first smallprint). The question mark in Northeast Baluchistan Province ( Pakistan) accentuates reports from near Zhob ( Khan, 1997). Triangles show collecting sites of examined Iranian Coluber chesneii Martin except MNHG 1359.12 (extralimital) and published records from Bahrain (various localities), Iraq (Faw Peninsula), Kuwait, Saudi Arabia, and Sir Bani Yas Island, UAE (see Schätti, 2006: 677-78, footnote 2); open symbols denote two intergrades from Fars, Iran (see Conclusions, footnote 10). The encircled area in the Baluchistan Region delimits the assumed distribution range of P. mintonorum ( Latifi, 1991; Schätti & Stutz, 2005). The stippled line along the Indus Valley and the Makran coast indicates the approximate western distribution limit of P. ventromaculatus ( Schätti & Schmitz, 2006: Fig. 3); two specified collecting sites (stars) in the Baluch littoral are Gwadar (25°07’N 62°20’E, BMNH 80.11.10.201) and Rumra (25°23’N 63°44’E, ZSM 222.1989, hoc loco). See text including Material and Methods, Hybrid Racers, and Appendices (A, C) for further explanations. Drawing Andrea Stutz.

In Afghanistan, Platyceps karelini is documented from Balkh (relies upon potential hybrid), Helmand, Herat, Jowzjan (Ag Chah), Kandahar, and Nimruz (see Systematics , Hybrid Racers : first smallprints). The species certainly occurs in Farah and two northwestern frontier provinces (Badghis, Faryab), may be found in parts of Ghor and Sar-e Pol ( NW Afghanistan), and probably extends as far east as Kunduz. Boulenger’s (1889) “ Badghis ” record based upon BMNH 1886.9.21. 104 is from Herat (see Hybrid Racers) .

Reliable Pakistani records are largely confined to Kalat, Khuzdar, Mastung (incl. Wali Khan), Pishin (incl. Bostan), Qila Abdullah (Chaman, Gulistan), and Quetta Districts in Northeast Baluchistan. Khan’s (1997) mentions from “Zob” (“Loi Banda” and “Muslim Bazar”), i.e., Zhob (Fort Sandeman, 31°20’N 69°27’E), are in need of confirmation (Fig. 5). Mertens (1969) indicated Platyceps karelini from Northwest and Central Pakistan (“aus dem nordwestlichen bzw. mittleren Teile W-Pakistans”) but his specimens only corroborate the presence at Darzi Chah in Afghanistan (SE Kandahar, “ 40 mi WNW Nushki”, SMF 64629) close to the border with Pakistan as well as around Quetta and Khuzdar, the southeasternmost record of Karelin’s Racer. The species certainly lives in the Nushki area as evidenced by SMF 64629. Sturdy or trustworthy reports, and in particular precise collecting sites, are lacking for Northwest Baluchistan Province, i.e., Chagai and northern Kharan where Karelin’s Racer most probably occurs.

Parts of Chagai District were explored by, for example, the Afghan Delimitation Commission ( Aitchison, 1889: map 1) and Alcock & Finn (1897). Annandale’s (1904, 1906) determinations of ZSI specimens from unspecified localities in the border triangle of Afghanistan, Iran and Pakistan (Northwest Baluchistan) rely on Alcock & Finn (1897) and Arthur Henry McMahon. The inclusion of “ Coluber karelini ” among reptile species typical of “The Northwest Upland” (i.e., “from the high plains around Kalat and Quetta northeastward through Waziristan into the lower valleys of Swat, Dir, and Chitral”) by Minton (1966: 40, map 5) is misleading; Platyceps karelini is not recorded from beyond Northeast Baluchistan Province. Khan’s (1977) “ karalini ” [sic] from Darapathar near Rabwah (31°45’N 72°55’E, ca. 170 m a.s.l.), referred to as “an aberrent [sic] race” of “[t]he nominated [sic] taxon” from “arid Punjab ” ( Khan, 1982), is P. rhodorachis (see Comparison). Seven localities (“Boostan”, Chaman, “Punj Pai”, “Peshin”, Quetta, and in the “Zob” area, see paragraph above) listed by Khan (1997: Appendix I; see also Wall, 1911; Minton, 1966) result in two or possibly three map entries on Pakistani territory along the frontier with Afghanistan; collecting sites as, for instance, Gulistan or Mastung ( Wall, 1911; Khan & Ahmed, 1987) are not plotted. Khan (1999: 276, 288) regarded karelini to be a “widely distributed [mountain] species” also enumerated among the “large number of endemics” of the “Chagai-Kharan desert” (“Herpetologically riches [sic] part of Pakistan ”). Khan (2002) report- ed P. karelini (as Coluber auct.) only “from Quetta and Pishin area” but three out of four entries on his map lie west of ca. 66°30’E longitude including the vicinity of Nushki (ca. 29°33’N 66°01’E, roughly 1’000 m) and Dalbandin (28°54’N 64°25’E, ca. 850 m), and thus are likely to refer to P. mintonorum (see Khan, 2002: 45). Not a single record from east or roughly north of Nushki is shown in Khan (2006: map) where merely three unspecified places are pinpointed; the southernmost and inexplicit locality mapped in Khan (2002) is missing, and the two western collecting sites most probably refer to P. mintonorum from Dalbandin and Nok Kundi (28°50’N 62°45’E, ca. 680 m) in Chagai (Schätti & Stutz, 2005: Fig. 1). According to Khan (2002: 99), P. karelini (type locality “Southwest Asia”, possibly fide Smith, 1943), called the ‘Banded desert racer’ or ‘spotted racer’, “frequents plain deserts between 1500-3000 m of elevation in northwestern [sic] Balochistan ”; earlier, the same species was denoted as the ‘transversely striped desert racer’ ( Khan & Ahmed, 1987). It is further declared that “[s]pecimens have also been collected from northwestern [sic] Punjab, from Sulaiman Range” ( Khan, 2006); a virtually identical remark (“I collected two specimens from the Sulaiman Range […] in a rocky area”) is hawked under “ Coluber karelini mintonorum ” and the putative origin left no mark on the accompanying map (see also Schätti, 2006: 683, 685; Schätti & Schmitz, 2006: smallprint p. 761). The occurrence of Karelin’s Racer above 3’000 m as indicated by Khan (2006: Tb. 10.1) is unsubstantiated.

The confirmed altitudinal distribution on the Iranian Central Plateau ranges from approximately 700 m above sea level east of the Tahi Plain (TMUS 1001) in extreme eastern Yazd (Khorasan-e Jonubi border area) to about 1’ 870 m near Kondor in adjacent South Khorasan ( Nikolskij, 1916; see Material and Methods as to USNM 240003). Similar altitudes (ca. 800 to at least 1’ 700 m) are attained on the northwestern Plateau. Lower elevations are inhabited around Zabol (ca. 480 m), along the Harirud (Tedzhen) River (Pasgah-e Pol-e Khatun, ca. 400 m; Serakhs, 280 m), and in Golestan where Platyceps karelini lives near sea level, for instance at Aq Qal’eh ( Latifi, 1991). Afghan records are from ca. 280 m in Jowzjan (Ag Chah) to at least 1’000 m in the west and south (e.g., “Chinkilok” [Herat] or the vicinity of Kandahar including Chahar Dahaneh). To conclude from collecting sites in the immediate border area (Qila Abdullah District, Pakistan), the species most probably occurs at elevations of at least 1’ 200 m in Kandahar Province and elsewhere in Afghanistan. Three Pakistani specimens (AMNH 96219-20, SAM 931) were obtained close to 2’000 m “in a cultivated section of Urak Valley” (vic. Hanna, Quetta) and at ca. 1’ 530 m “in similar terrain near Pishin” ( Minton, 1966). Altitudes nearing 2’000 m are inhabited around Kalat (PMNH 761). P. karelini from Quetta, Mastung, and Khuzdar (see Appendix A) were collected between ca. 1’200-1’ 700 m.