Androctonus bicolor Ehrenberg, 1828

Teruel, Rolando & Kovařík, František, 2014, Redescription of Androctonus bicolor Ehrenberg, 1828, and description of Androctonus turieli sp. n. from Tunisia (Scorpiones: Buthidae), Euscorpius 186, pp. 1-15 : 2-8

Androctonus bicolor Ehrenberg, 1828

( Figures 1–18; Table 1)

Androctonus bicolor Ehrenberg , in Hemprich & Ehrenberg, 1828: pl. II, fig. 4; Hemprich & Ehrenberg, 1829: 358. For further nomenclatural history see Fet & Lowe (2000: 70–71) and Teruel et al. (2013: 5– 10).

TYPE LOCALITY AND TYPE DEPOSITORY. Egypt, ZMB .

TYPE MATERIAL. "Ägypten Ehrberg" [ Egypt, leg. Ehrenberg]; 1♂ lectotype herein designated, 1♀ 5 juvenile paralectotypes, (ZMB-137, Figs. 1–18). See also "Comments" section below for additional paralectotypes labeled ZMB-138 .

OTHER EXAMINED MATERIAL. Egypt, 2006, without further data, 1 juvenile ♀ ( RTOC: Sco.0336), 3♀ ( FKCP); northern part, 2011, leg. A. Tolba, 3♂ 7♀ 1 juvenile ♀ ( RTOC: Sco-0555) . Israel, Hadarom , ' Arad; April 2, 1998, leg. A. Schmidt, 1♂ ( RTOC: Sco-0107) .

DIAGNOSIS (updated). Adult size small to moderately large (males 56–62 mm, females 64–81 mm) for the genus. Coloration basically blackish brown to pitch black; pedipalps and legs apically yellowish, but with carinae infuscate and intercarinal dark reticulations; metasoma with all carinae infuscate. Pedipalps long and slender, with hand conspicuously narrower than patella and very weakly carinate; fixed and movable fingers with 12–13 and 13–14 principal rows of denticles, respectively, basal lobe/notch combination absent. Carapace densely granulose, with anterior and posterior median carinae strong. Sternite V with a large smooth patch in adult males, VI–VII with four carinae. Metasoma very robust and deep (both sexes have segments I– III and frequently also IV wider than long, and all segments wider than deep), with 10-10-10-8-5 complete to essentially complete carinae, most of which are coarsely serrate to crenulate; segments II–IV dorsolateral carinae with 1–2 enlarged terminal denticles; segment V ventrolateral carinae with 5–6 widely flared to lobate denticles and anal arch with 3–5 very poorly defined round lobes; all intercarinal spaces glossy and smooth to sparsely granulose, with a few coarse punctures laterally on V and telson vesicle. Telson slender, with vesicle small but globose, and aculeus longer than vesicle. Leg bristlecombs well developed on I–III, poorly defined on IV. Pectines with 24–28 teeth (mode 25 and 27) in males, and 19–22 (mode 21) in females.

REDESCRIPTION (adult male lectotype, herein designated). Coloration ( Figs. 1–2) essentially uniformly blackish to unaided eye; ocular tubercles, eyes and distal part of aculeus even darker. Chelicerae deeply infuscate to very densely reticulate. Pedipalps and legs progressively lighter apically, becoming yellowish, but with all carinae infuscate and dark reticulations between carinae. Pectines yellowish. Sternite V with the smooth patch pale yellowish. Metasoma with all carinae plus the apical portion of all segments infuscate. Telson uniformly blackish, only with distal half of aculeus darker.

CHELICERAE. With dentition typical for the genus; tegument polished and smooth except on dorsodistal portion of manus, which possesses coarse granules arranged in longitudinal ridges.

CARAPACE. Trapezoidal, slightly wider than long and with all carinae strongly granulose, but not fused into any particular configuration; tegument very densely covered by medium-sized and small granules; anterior margin very shallowly V-shaped, with some pairs of stout macrosetae; median eyes separate by more than one ocular diameter; five pairs of lateral eyes: three large and aligned, plus two small and slightly offset.

MESOSOMA. Tergites very densely covered by uniformly medium-sized and small granules; I–VI with three carinae (median and submedians), which are strong and granulose, but not projected beyond posterior margin; VII with five carinae (median, submedians and laterals) which are strong and serrate. Sternum standard for the genus: type 1, relatively small, and markedly triangular in shape. Pectines very long (extending around midpoint of leg IV trochanter), narrow and densely setose; tooth count 25/25; basal plate heavily sclerotized and about as long as wide, anterior margin with strong median indentation, posterior margin convex. Sternites III–VI smooth, glossy and sparsely setose, spiracles very elongate and slit-like; III without granulose lateral areas; V with smooth patch very large, widely subtriangular and bulky, translucent yellowish and conspicuously paler than the rest of the plate; VI sparsely granulose, with two pairs of costate to subcostate carinae: submedians weak and coarse, laterals stronger and finer; VII densely granulose, with two pairs of strong and costate to subcrenulate carinae.

METASOMA AND TELSON ( Figs. 16–18). Metasoma very sparsely setose, with all segments very robust and deep (I–III wider than long, IV only slightly longer than wide, and V longer than wide; all segments wider than deep); intercarinal tegument of dorsal surface with granulation becoming progressively reduced to smooth towards distal segments and forming arrowhead-shaped patches on I–III, on lateral and ventral surfaces moderately granulose but not forming a clearly reticulate pattern; dorsal furrow moderately shallow and narrow on all segments; I–III with ten complete to almost complete carinae (lateral inframedian carinae becomes less defined on the basal portion of II–III), IV with eight, and V with five: dorsolateral carinae on I–IV coarsely subserrate to subdentate and progressively raised with 1–2 enlarged terminal denticles, lateral supramedian carinae and lateral inframedian carinae coarsely subcrenate to serratocrenulate and progressively raised distally, ventrolateral carinae coarsely subcostate to subgranulose and even on I–IV, irregularly dentate and progressively raised distally on V with 4–5 triangularly flared denticles, anal arch with 2–3 very poorly defined round lobes. Telson slender and sparsely setose; vesicle small but somewhat globose (1.36 times longer than wide, 1.14 times wider than deep), tegument glossy and essentially smooth, only with some coarse punctures and a coarse but very poorly defined ventromedian carinae; subaculear tubercle insinuated by a coarse granule flanked by the subaculear setal pair; aculeus very long and thick, longer than vesicle, and evenly curved.

LEGS. Legs long, slender, and only sparsely setose; bristlecombs well-developed on I–III, poorly defined on IV (most setae are now missing, but their large and raised areolae warrant recognition); all carinae smooth to subserrate; intercarinal tegument smooth to coriaceous; tibial spurs absent from I–II, well developed on III–IV; prolateral and retrolateral pedal spurs long and slender in all legs; ventral surface of all tarsomere II sharply edged and with two ventrosubmedian rows of spiniform setae, without median row of spinules; claws long and evenly curved.

PEDIPALPS ( Figs. 5–6, 12). Pedipalps long and slender, only very sparsely setose and orthobothriotaxic A-β. Femur slender and straight; all carinae strong and finely granulose; intercarinal tegument finely granulose. Patella long, slender and straight; all carinae moderately strong and finely subgranulose to costate; intercarinal tegument minutely granulose. Chela with hand narrower than patella, oval (1.45 times longer than wide) and with all carinae obsolete to weakly costate and smooth, intercarinal tegument smooth and glossy, with subtle traces of coarse reticulations dorsally and dorsoexternally; fingers long and slender (movable finger 2.48 times longer than underhand), evenly curved and with basal lobe/ notch combination only very subtly insinuated, essentially absent, fixed fingers with 13 principal rows of granules; movable fingers with 14 principal rows of granules and three terminal granules (large terminal denticle not included).

FEMALE (adult paralectotype, Figs. 3–4, 7–11, 13–15, Tab. 1). In general is similar to the male, but there is a slight sexual dimorphism evidenced by: 1) mesosoma relatively wider; 2) pedipalps with hand slightly shorter and less oval in shape; 3) genital papillae absent; 4) pectines much smaller and with non-overlapping, lower tooth counts; 5) sternite V with smooth patch poorly defined and not bulky but concave; 6) ventrolateral carinae of metasomal segment V with the flared denticles large and lobate.

VARIATION. The size differences among adults studied by us seem to indicate the existence of two size-classes in each sex; inside the same class, males are consistently smaller than females ( Tab. 1). Levy & Amitai (1980) stated that adult females may attain a size larger than 90 mm and gave actual measurements of adult male and female 72 and 85 mm long, respectively; these ranges are larger than those we examined, but it is not clear if these authors actually examined more than one species.

Diagnostically relevant morphometric ratios revealed a direct correlation between size and expression of sexual secondary dimorphism: larger specimens invariably showed the highest degree of intersexual differentiation ( Tab. 1). This trend is widespread in the order Scorpiones and has already been observed in other members of this genus studied recently (Kovařík & Ahmed, 2013; Teruel et al., 2013).

Count of teeth per pecten varied as follows: in males 24 (1), 25 (4), 26 (1), 27 (3), 28 (1), and in females 19 (3), 20 (3), 21 (8), 22 (4); as it can be seen, females are more homogeneous for this character. Levy & Amitai (1980) recorded counts of 23–29 (mode 26– 28) in males and 19–24 (mode 21–23) in females, but it is not clear if these authors inadvertently pooled data from more than one species.

The number of principal rows of denticles varied from 12–13 and 13–14 in fixed and movable fingers, respectively; in all cases variation corresponded to poorly defined basalmost rows. Levy & Amitai (1980) gave the same numbers for the movable finger, but provided no data on fixed finger.

AFFINITIES. Amongst the 22 species currently recognized as valid in this genus (Turiel, 2013; Teruel et al., 2013), only A. tenuissimus and A. sergenti Vachon, 1948 , do share with A. bicolor the following combination of three characters in adults of both sexes: 1) coloration uniformly dark reddish brown to blackish; 2) pedipalps very slender, with chelae much narrower than patella; 3) pedipalp fingers without basal lobe/notch combination. All other species of Androctonus which have the same coloration also possess adult pedipalp chelae wider than patella, and with basal lobe/notch combination of fingers present at least in males.

Of them, the easiest to separate is A. sergenti : it differs remarkably from A. bicolor by having a smooth and coarsely punctate metasoma at least in males ( Fig. 35), less attenuate pedipalps, and much more reduced granulation and carination on the body and appendages. carapace it corresponds to posterior width), depth (H), left (L), right (R). Notes: a) telson vesicle length is not given because it narrows definition point; b) incomplete measurements due to aculeus tip broken are marked with an asterisk (*).

Furthermore, A. sergenti is locally endemic in a small area of the high Anti-Atlas Mountains of south-central Morocco (Vachon, 1952).

The closest relative of A. bicolor is A. tenuissimus , but it can be readily distinguished from the former on the basis of the following unambiguous characters: 1) pedipalps conspicuously more attenuate, with femur, patella, and chela much longer: 2) pedipalps with carinae remarkably stronger and intercarinal tegument much more densely granulose, especially on chela; 3) pedipalp patella moderately curved inwards; 4) pedipalp fixed and movable fingers with 13–14 and 14–15 principal rows of denticles, respectively; 5) metasoma less deeper and much more slender (segments II, IV and V longer than wide in both sexes, in females also segments I and III), with carinae much finer and intercarinal tegument densely granulose and never glossy; 6) ventrolateral carinae on metasomal segment V with the flared denticles sharp and much more regular; 7) carapace and tergites with all carinae sharper and intercarinal granulation finer and more regular. Further, the secondary sexual dimorphism in A. tenuissimus is very obvious even to unaided eye, while adults of A. bicolor can be reliably sexed only after microscopic examination of the genital operculum and pectines. Besides, adults of these two species widely overlap in size, color and pectinal tooth counts, but there is a slight but consistent trend in A. tenuissimus to be somewhat larger, lighter (dark reddish brown), and to possess higher counts.

DISTRIBUTION. This scorpion has been widely recorded from North Africa and the Middle East, but its true distribution needs to be clarified because most of those literature records are partially based in misidentifications of closely related taxa (Levy & Amitai, 1980; Teruel et al., 2013). For this reason we do not present a map here pending further studies, but can already disclose that the only occurrences confirmed by us are from northern Egypt through Israel (see the Examined material section above).

ECOLOGICAL NOTES. In accordance with the poor taxonomic knowledge prevailing in A. bicolor to date, very scarce information has been published on its ecology. The most detailed data are available from Levy & Amitai (1980), who stated that in the Middle East this scorpion lives under large rocks in a wide variety of soils from sand dunes to rocky desert, it is more common inland than along Mediterranean coast, it sometimes enters houses, and females give birth to 30–40 offspring mostly in August.

COMMENTS. The type locality of A. bicolor and the exact composition of its type series require some clarification. As timely elucidated by Braunwalder & Fet (1998), the original description is a plate with a Latin binomen included, but in the text subsequently issued as the intended description, Ehrenberg (in Hemprich & Ehrenberg, 1829: 358) declared to have examined 13 specimens and gave the distribution as "Libyan coast, Syria and Sinai ", but also explicitly stated in the same page that the plate which actually represents the original description was based on specimens from Egypt. In the only two major revisions published to date on this species (Levy & Amitai, 1980; Fet & Lowe, 2000), both statements have been confused or overlooked: the type series has been declared to contain only nine specimens and/or the type locality has been given as the coastal plain of Libya, Syria, and Sinai.

We requested on loan the type series of A. bicolor and received two vials. The one coded ZMB-137 contains one adult male, one adult female and five juveniles, and its label reads Egypt as the collecting place ( Figs. 1– 4). As we designated this male as the lectotype of A. bicolor , Egypt becomes its type locality according to Recommendation 76.2 of the Code (ICZN, 1999: 71); by the way, this concurs with the original statement of Ehrenberg (in Hemprich & Ehrenberg, 1829: 358). The second bottle coded ZMB-138 is labeled "Syria- Ehrenberg" and contains only two very badly preserved, small juvenile paralectotypes (total length around 26 mm), which are impossible to identify accurately; thus, we decided not to cite them inside the type material of A. bicolor , especially because they could belong to another species of Androctonus .

The whereabouts of the remaining four specimens originally mentioned by Ehrenberg (in Hemprich & Ehrenberg, 1829) are currently unknown, but based on external evidence we suspect that they correspond at least to two parcels that could be lost now. For example, Levy & Amitai (1980: 23) mentioned that " A young specimen of A. bicolor was collected by Hemprich & Ehrenberg (1829) in Beirut, Lebanon ", which seems to indicate that it was one of Ehrenberg's missing syntypes. And as the syntypes examined by us did not bear any label indicating Sinai (the other toponym mentioned in the original description), we think this most likely corresponds to a second parcel with the remaining three missing syntypes. Already, Braunwalder & Fet (1998) mentioned only the same syntypes studied by us (seven coded ZMB-137, two coded ZMB-138); thus, it seems that the remaining syntypes have long been missing or misplaced.

In the original description of A. tenuissimus, Teruel et al. (2013: 6–10) discussed the identity of all taxa that have been described or regarded at least once as subspecies of A. bicolor . All arguments exposed therein remain intact after our present examination of the types of the latter, thus, it is unnecessary to repeat them here and the interested reader is referred to that paper for further information. Nevertheless, we will discuss the status of A. aeneas , whose taxonomic history can be very briefly summarized as follows: it was described from "North Africa" at species level by C. L. Koch (1839), but its status has been highly controversial, being regarded either as a valid species (e.g., Vachon, 1952) or as a junior synonym of A. bicolor (e.g., Kraepelin, 1899; Lourenço, 2005), sometimes with subspecies-level (e.g., Vachon, 1958; Fet & Lowe, 2000); last, Teruel et al. (2013) stated that it should be regarded best as a nomen dubium.

Most of this problematic situation is due to the loss of the types (Fet & Lowe, 2000) and that the original description of C. L. Koch (1839) lacks any useful characters that allow the distinction of A. aeneas from its close relatives. The only other choice would have been to examine topotypes, but the type-locality is simply "North Africa", a very large and ecologically complex area that actually hosts several closely related species of this complex. Further, Vachon (1952) attempted a detailed taxonomic interpretation of A. aeneas , but unfortunately it was not based on the (already lost) types and it clearly included more than one taxon. Thus, as we have demonstrated that the original binomen Androctonus aeneas C. L. Koch, 1839 cannot be applied to any actual taxon unambiguously (or at least not arbitrarily), that its types are lost, and that the type locality is not precise, we regard it formally as a nomen dubium.