The Carabus fauna of Israel – updated identification key, faunistics, and habitats (Coleoptera: Carabidae)

Th is key to the Carabus species of Israel is an updated identifi cation key with notes on the distribution and habitats of the species. Substantial additions, corrections and taxonomic changes on the Carabus fauna of the Middle East generated the need of an update of the knowledge of the genus Carabus in Israel. Th e classifi cation and the identifi cation of sibling taxa of the subgenus Lamprostus are still a problem: A zone of sympatry supports the species status of both C. sidonius and C. hemprichi. Th e lack of any evidence of sympatry for the taxa in species rank of the C. syrus group and their variability of the exoskeleton (mentum tooth, tip of aedeagus) requires further systematic and taxonomic studies.


Introduction
Despite the fact that Israel is a small country (about 22,000 km 2 ), it displays an enormous ecological diversity originating from its peculiar biogeographic location in south-western Asia and its great physical variety (Furth, 1975;Por, 1975;Yom-Tov & Tchernov, 1988): It links the desert Sahara-Arabia belt, the Mediterranean region and the high Asian mountains. Th e ground beetle genus Carabus with its preference for humid habitats reaches its southern distribution limit in Israel, and only some species are distributed there (cf. Bousquet et al., 2003).
Th e Carabus fauna of Israel was fi rst described in a fundamental work by Schweiger (1970). Substantial additions, corrections and taxonomic changes were made by Kleinfeld & Rapuzzi (2004) and Deuve (2004;2005) in the last years. Additional records of some species demonstrate the need of an update of our knowledge on the genus Carabus in Israel. Moreover, the increasing interest in the conservation biology, ecology, evolutionary biology and faunistics of ground beetles in Israel (Bar, 1978;Chikatunov et al., 2006;Chikatunov et al., 1999;Finkel et al., 2002;Mienis, 1978a;1978b;1978c;1978d;Pavliček & Nevo, 1996) demands a new identifi cation key and a short synopsis of the present day knowledge of the distribution, faunistics and habitats of Israeli Carabus species. Th e taxonomic confusion in this group prevents us from presenting a fi nal identifi cation key. However, we hope that this short overview stimulates further studies to solve some systematic problems of the Carabus fauna of the Middle East.
Nomenclature of vegetation types for a characterisation of the habitats follows Danin (1988).
Total body length (BL) is measured from the tip of the mandibles to the apex of the elytra as the maximum linear distance.
Line drawings were prepared using a drawing tube attached to a Leica MZ 95 stereobinocular microscope. Dissections were made with standard techniques; genitalia were preserved in euparal or in polyvinylpyrrolidon containing mixture on acetate labels (Lompe, 1989) or without embedding in dried condition.

Identification key with notes on distribution and habitats of the species
Th e members of the genus Carabus are easily recognizable by the lack of a typical antennal cleaner, posterior coxae contiguous in midline of body, mandibles not transversally furrowed, and third antennal segment without keel. Moreover, the species living in Israel are characterized by their body size (BL > 15 mm) and entirely black colour (without any spots or markings or metallic luster). For a general characterisation of ground beetles see Trautner & Geigenmüller (1987) and Ball (2001). 1 (15) Pronotum without marginal setiferous pores (Fig. 1a, b, c)  Labrum divided into three lobes (Fig. 2). Elytra with or without punctuation and granulation, habitus variable. BL: 25-36 mm. An eurytopic species in woodlands (Fig. 15), forests, batha (open and semi-open habitats, Fig. 16), arable land, dunes (Fig. 17), steppe and desert-like, overgrazed, semi-arid habitats of the northern Negev (Fig. 18), up to 1200 m above sea level. In northern and central Israel southwards to the northern Negev south of Be`er Sheva (e.g. Noqdim Plateau). 1 Fig. 1a (7) Elytral sculpture with punctures and striae. Slender species. BL: 25-36 mm.

Discussion
Ten species of the genus Carabus are known from Israel. Th e presence of C. phoenix in Israel -fi rst records known from the surrounding of Sasa in Upper Galilee (Kleinfeld & Rapuzzi, 2004) -can be confi rmed by several records from the Meron area (Upper Galilee, cf. Timm et al., 2008) 2 . At several locations in Galilee (including a site close to the Sea of Galilee, about 200 m below sea level) we detected Carabus syrus populations. Th e previously known distribution area in Israel covers the Golan Heights, parts of the Mount Hermon and the Upper Jordan Valley close to Qiryat Shemona (Schweiger & Rapuzzi, 1970). We believe that larger parts of Galilee (including Lower Galilee), Golan Heights and Judean Foothills are still under-represented in faunistical studies. Th erefore it seems most likely that additional populations and perhaps species can be detected. From Jordan and Lebanon new species of the subgenus Lamprostus were already described in the last years (C. pseudopinguis Heinz, 2000;C. lecordieri Deuve, 1992;C. rostandianus Deuve, 2005;cf. Deuve, 2005;Heinz & Staven, 2000). Despite the still incomplete faunistic inventory of Israel, the records of C. syriacus seem to decline, especially in the last decades. Coleopterists, also those collecting mainly in the northern parts of Israel, have not found this largest Carabus species in the Middle East for many years (e.g. Rittner, personal communication). Urbanization, habitat fragmentation and large-scale changes of land use (especially the transformation of natural and semi-natural habitats, e.g. sclerophyllous woodlands and batha, to pine stands and arable fi elds) might be a reason for this decline. Species of the subgenus Procerus show a remarkable decline not only at the southern limit of their distribution area but also in Europe: C. gigas Creutzer, 1799 was once distributed in Styria and Carinthia. At present the species is extinct in Austria (Paill, personal communication), in Slovenia the species is still occurring, but clearly declining (Drovenik, personal communication; Turin et al., 2003). A similar decline seems to occur in some places in Italy (Brandmayr and Casale, personal communication). -If one or several populations are rediscovered, an action plan to conserve the relict populations at the most southern limit of this species (and subgenus) will have to be developed.
A clear problem for identifi cation are the sibling taxa of two Lamprostus groups: Th e characters given in the literature to separate C. hemprichi from C. sidonius and C. syrus from C. lacordieri show a remarkable variability within and between populations; this is true for both the mentum tooth and the apex of the median lobe of aedeagus. Sometimes it is impossible to classify some specimens exclusively from the exoskeleton. While a zone of sympatry is known for C. hemprichi und C. sidonius in Lebanon, any evidence of sympatry is still lacking for the members of the C. syrus group (including C. lacordieri, C. pseudopinguis and C. rostandianus).
Th e results of Pavliček & Nevo (1996) on C. sidonius demonstrated a small-scaled genetic diff erentiation, similar to some other Carabus species (Assmann, 2003;Assmann 2 Records from Sasa are not considered in the distribution map of C. phoenix given by Kleinfeld & Rapuzzi (2004). & Weber, 1997). Th e morphological diff erentiation (from eye inspection) refl ects this strong geographic diff erentiation on another level and should encourage us to study the species complexes morphometrically in order to solve the taxonomic problems (but for this approach still more material is necessary than is available at the moment).
In general one has to keep in mind that diff erences in the aedeagus, especially those of the apex (and not of the internal sac) of this organ, do not seem to be useful to classify taxa at the species level (see for a detailed discussion: Assmann et al., 2008). Th e taxa C. violaceus violaceus Linné, 1758 and C. v. purpurascens Fabricius, 1787 of the subgenus Megodontus can be easily distinguished by diff erent forms of the aedeagus tip (and by lack or presence of striae on the elytra). But both taxa form several broad hybrid zones in north-western Central Europe (Assmann & Schnauder, 1998). An excessive gene fl ow is documented also by molecular techniques (allozymes and mtDNA haplotypes; Eisenacher et al., in prep.). In the light of these results the species rank of some taxa of the C. syrus group should be critically reconsidered.