The millipede genus Glomeris Latreille , 1802 ( Diplopoda , Glomerida , Glomeridae ) in North Africa

In North Africa, the genus Glomeris is shown to encompass 11 species, all of which are keyed. Th ese include: G. troglokabyliana sp. n. from several caves in Algeria, G. monostriata sp. n. from a cave in Libya, G. colorata sp. n., an epigean species from Tunisia, G. anisosticta Brandt, 1841 (still a nomen inquirendum) from Algeria, G. brolemanni Schubart, 1960 from Morocco, G. carthaginiensis Schubart, 1953 (stat. n., elevated from subspecifi c rank) from Tunisia, G. fl avomaculata Lucas, 1846 from Algeria, G. klugii Brandt, 1833 (with G. marmorata Brandt, 1833, G. fuscomarmorata Lucas, 1846, and G. maculosa Verhoeff , 1921 as new junior subjective synonyms) from Algeria and Tunisia, G. mohamedanica Attems, 1900 from TuniZooKeys 12: 47-86 (2009) doi: 10.3897/zookeys.12.179 www.pensoftonline.net/zookeys Copyright Sergei I. Golovatch et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Launched to accelerate biodiversity research A peer-reviewed open-access journal

The millipede genus Glomeris Latreille, 1802 (Diplopoda, Glomerida, Glomeridae) in North Africa Introduction Th e millipede order Glomerida is basically a temperate to warm temperate Holarctic group which contains about 30 genera and some 450 species (Mauriès 2006).Only six genera and about 80 species occur in the Oriental or Neotropical realms, reaching Sulawesi and Guatemala in the South, respectively.Th e genus Glomeris Latreille, 1802, with the type-species G. pustulata (Fabricius, 1781), is the largest, numbering about 100 species and a few hundred subspecies, varieties, forms or aberrations, largely from Europe, but marginally also in the Canaries, North Africa and northwestern Anatolia.Like all Glomerida, in Africa Glomeris species only occur north of the Sahara, along a relatively narrow strip of the Mediterranean coast, in rather humid habitats.Th is genus is characterized by a subquadrate shape of the telopod femur, which is not markedly hypertrophied in relation to the adjacent podomeres and is supplied with a broad distocaudal lobe instead of a distinct process (Mauriès 2006).Th e taxonomy of Glomeris has only recently been improved.In particular, the Central European and Macaronesian faunas were reviewed and keyed (Hoess 2000;Golovatch and Enghoff 2003), with the main species-specifi c characters currently recognized as lying in colour patterns and, to a lesser extent, telopod structure.Th e identities of several Central European congeners were clarifi ed with the use of allozyme electrophoresis (Hoess and Scholl 1999;Hoess 2000), whereas scanning electron microscopy was quite extensively applied to the study of the alluaudi-group of Glomeris endemic to the Canary Islands (Golovatch and Enghoff 2003).
Th e fi rst Glomeris ever to have been recorded in Africa seems to be G. klugii Brandt, 1833, a species originally described from a single female from "Egypt or Syria" (Brandt 1833), but later rectifi ed as coming from "Syria" (Brandt 1840b).Th e holotype, still housed in the Berlin Museum (Moritz and Fischer 1973), was found to have been mislabeled, and G. klugii proved to be a senior synonym of the common European species G. undulata C. L. Koch, 1844(Golovatch 2003).
Glomeris, as well as a few other myriapod genera, was then simply listed as present in Algeria (Brandt 1840a).A few months later, two much more detailed accounts appeared.Th e fi rst of these, Brandt's (1840b) review of Glomeris, mentioned two forms of G. pustulata found in Algeria and Germany, which were described or referred to as "var.microstemma n." and "var.marmorata " Brandt, 1833, respectively.Brandt also stated that both these varieties were very common in Algeria, but occurred much less frequently in Germany.Th e second publication (Brandt 1841a) was an essay specifi cally focusing on the fauna of Algeria.It largely repeated the same information, but the same varieties were instead referred to as G. pustulata "var.anisosticta n." and "var.marmorata"!Just like microstemma, the variety anisosticta was stated to diff er from the European samples of G. pustulata in showing both median spots on the thoracic shield smaller than the lateral ones.Unfortunately, this information was omitted from both the Brandt biblio graphy and the list of Brandt's diplopod taxa in Golovatch and Hoff man (2001), otherwise this confusion would have already been corrected.It is now apparent that these names are strictly synonymous, with microstemma Brandt, 1840 having priority over anisosticta Brandt, 1841 and representing the fi rst truly African glomeridan to have been named.In contrast, marmorata was downgraded from a full species (Brandt 1833) to a variety of G. pustulata, while the samples from Algeria were found to be typical var.marmorata (Brandt 1840b(Brandt , 1841a, b), b), in no way diff ering from their European counterparts.
At the present, the type material of microstemma and anisosticta seems to be lost, since it is missing from the collections of the Berlin (Moritz andFischer 1973, 1978) and St. Petersburg (Golovatch and Hoff man 2001) museums.A neotype designation would therefore be necessary to revive either name as a taxon.Verhoeff (1906) was apparently the last author to use the name microstemma, still as a variety (this time of G. pustulata norica Latzel, 1884), and even incorporated it into a key.In contrast, because anisosticta has since been elevated to the rank of a full species (Brolemann 1921), despite having been published slightly later, we use anisosticta as the valid name.
Although the identity of G. marmorata might appear to lie beyond the scope of the present study, because it was fi rst described from Hercynia (= Harz), Germany (Brandt 1833), it will be resolved below due to its relevance to the North African fauna.Th is taxon has hitherto remained dubious (Golovatch and Hoff man 2001), even though the type material has long been available in the Berlin Museum (Moritz and Fischer 1978).It is highly regrettable that Hoess (2000) did not attempt to revise type material of numerous Central European Glomeris when preparing his otherwise very useful review and key.Otherwise he could not have overlooked the great similarity between G. marmorata, as redescribed and beautifully illustrated by Koch (1863) from samples taken in southern Germany (and probably also based in part on a restudy of type material), and G. undulata which Hoess also very skillfully depicted himself.
Shortly after Brandt's contributions appeared, Lucas (1846) described three new species of Glomeris from Algeria: G. sublimbata Lucas, 1846, G. fuscomarmorata Lucas, 1846and G. fl avomaculata Lucas, 1846, all listed, redescribed and nicely illustrated after in his atlas (Lucas 1849).Among these species, only G. fl avomaculata was found to be abundant, being represented by fi ve varieties.Since these received no names (they were simply denominated A to E), they have no nomenclatural status.Because type material of all three species is still available in the Paris Museum, they could be revised and, when necessary, lectotypes selected (see below).Brölemann (1913a) provided a brief redescription of G. sublimbata, based on new samples from Algeria.Pocock (1892) referred some fresh specimens from Algeria to G. fuscomarmorata and G. fl avomaculata, but emphasized that probably both were at most only varieties of the European G. conspersa C. L. Koch, 1844and G. connexa C. L. Koch, 1844, respectively. In contrast, Silvestri (1896) identifi ed new material from Tunisia as G. sublimbata and G. fl avomaculata, already formally treating both as just varieties of G. connexa.Th ereafter, Attems (1900), violating all rules of priority, described a new subspecies, Glomeris europaea striata n., within which he distinguished several varieties, including the var.pustulata (Fabricius, 1781), var.transalpina C. L. Koch, 1836 etc., as well as the newly described var.punica n. and var.mohamedanica n., both from Tunisia.Soon after that, Attems (1908) transferred his var.punica to G. connexa and added G. conspersa C. L. Koch, 1844, forma genuina (= conspersa) to the Tunisian list.Both of the originally infrasubspecifi c names by Attems, however, have since become validated as speciesgroup taxa.Th us, in his list of North African millipedes, Brolemann (1921) reported three species of Glomeris from Algeria: G. anisosticta Brandt, 1841, G. fuscomarmorata Lucas, 1846and G. pustulata Latreille, 1804 (sic!); one from Tunisia: G. connexa punica Attems, 1900; and another three from both these countries: G. conspersa C. Koch, 1847 (= marmorata), G. fl avomaculata Lucas, 1846 andG. sublimbata Lucas, 1846.He must have either forgotten to include G. mohamedanica or considered it as a variety not worth mentioning.Schubart (1953), when revising Brolemann's (1921) checklist, treated both G. mohamedanica Attems, 1900 andG. punica Attems, 1900 as full species.Verhoeff (1921) described a further two species from Algeria: G. maculosa Verhoeff , 1921and G. numidia Verhoeff , 1921. Brolemann (1925) added to the confusion by describing from Tunisia the new subspecies G. pustulata trisulcata n., a long preoccupied name (G. intermedia trisulcata Rothenbühler, 1899).To eliminate this homonymy, Schubart (1953) renamed it as G. pustulata carthaginiensis Schubart, 1953.He also described the fi rst congener from Morocco: G. brolemanni Schubart, 1960, and provided some useful comments on the Glomeris fauna of North Africa in relation to a new record of G. fl avomaculata in Algeria (Schubart 1960(Schubart , 1963)).Finally, Abrous-Kherbouche and Mauriès (1996) reported two Glomeris species from a nature reserve in Algeria, and provided an updated checklist of the Diplopoda of that country.
Superfi cially, all Glomeris species, including those from North Africa, can more or less easily be separated into two groups, formerly invalidly treated as subgenera.One group, Eurypleuromeris Verhoeff , 1906, includes the species with a laterally broadened tergum 3 (tergum 4 as counted by Verhoeff (1906Verhoeff ( , 1921)), who considered the thoracic shield to be composed of two terga, 2 nd and 3 rd ).Th e other, Stenopleuromeris Verhoeff , 1909, includes the species with a laterally narrowed tergum 3, in particular its anterior (condylar) part shortened in relation to its posterior part, the two parts being separated by a stria.Th is distinction may still be useful, but it was only after Jeekel's (1971) typifi cation that both these names could be correctly applied.Despite this, all of the numerous nominate subgenera or synonyms of Glomeris (see Mauriès 2006) remain hopelessly heterogeneous.Th us, Glomeris conspersa C. L. Koch, 1847(= G. klugii Brandt, 1833) became the type-species of Eurypleuromeris and, like Glomeris connexa C. L. Koch, 1847, the type-species of Euglomeris Verhoeff , 1906, shows a broadened condylar part of tergum 3. Th e type-species of Stenopleuromeris was designated as Glomeris pulchra C. L. Koch, 1847 which, like G. pustulata, the type-species of Glomeris s. str., or G. dorsosanguine Verhoeff , 1906, the type-species of Xestoglomeris Verhoeff , 1906, has a shortened anterior part of tergum 3.So at the present, follow-ing Hoess (2000), it seems best not to use a formal subgeneric division of Glomeris, referring instead to informal groups in quotation marks.In addition, this character/ distinction appears to be subject to a degree of variation (see below).
Taking into account the two other glomeridans described from North Africa-Eupeyerimhoffi a algerina Brölemann, 1913 from Algeria (Brölemann 1913b) and Glomerellina convolvens africana Ceuca, 1988from Tunisia (Ceuca 1988)-all previous knowledge of the regional fauna of Glomerida can be summarized in the following checklist (Table 1).G. conspersa, reported from Tunisia and Algeria (Attems 1908;Abrous-Kherbouche and Mauriès 1996), is listed under the name G. klugii because it is just a colour morph of the latter (Hoess 2000;Golovatch 2003).
Rich material, including type material of Brandt, Lucas, Attems, Verhoeff and Brolemann, has been amassed from various sources for the present review.Th e following acronyms are adopted here for the relevant repositories: Name: To emphasize the provenance of material from caves in the High Kabylia, Algeria.
Diagnosis: Diff ers from all congeners except G. albida Mauriès & Vicente, 1978 and G. monostriata sp. n. in the clear troglomorphy (in particular, the lack of pigmentation), coupled with two striae crossing the collum and 1-2 striae crossing the thoracic shield, and usually with (6)7+1 ocelli.
Telopods (Figs 1F, G) with a rather high, regularly rounded, bare, central syncoxital lobe fl anked by two setose horns, each crowned with a very small bulb and a long, setiform fi lament.Femur with a large caudomedial outgrowth, subquadrate at base.Tibia with a caudomedial unciform process.Tarsus rather broadly rounded apically.Remarks: Th is species is the only clearly troglomorphic congener to be reported from Algeria.It is noteworthy that some specimens show rudimentary pigmentation of the ocelli.
Based on the shape of tergum 3, this new species can be regarded as somewhat intermediate between the "Stenopleuromeris" and "Eurypleuromeris" types, although closer to the former.

Glomeris monostriata
Name: To emphasize the collum and the thoracic shield each being crossed by only a single stria.
Diagnosis: Diff ers from all congeners except G. albida Mauriès & Vicente, 1978 in being troglomorphic, coupled with only a single stria crossing the collum and thoracic shield; diff ers from G. albida in the evidently bifi d horns and a lower central lobe of the telopod syncoxite.
Description: Length of adults of both sexes (unextended, alcohol material) ranging between 8.0 and 9.0 mm, width between 3.0 and 3.2 mm, up to ca 11 and 3.5 mm, respectively (extended animals); body broadest at thoracic shield.Holotype (unextended) ca 8.2 mm long and 3.2 mm wide.Coloration entirely pallid.
Collum with one (anterior) transverse stria.Th oracic shield with a narrow hyposchism reaching the caudal tergal contour (Fig. 2A); two transverse striae, of which the one starting well in front of the schism crosses the entire dorsum, the other, starting just above the schism, is abbreviated (Fig. 2A).
Remarks: Th is species is the fi rst glomeridan to be recorded in Libya and, given that it is troglobitic, may represent a relictual element.
In general, the condition of having just one stria, rather than two, crossing the collum is extremely rare in Glomeris species.Th e only other congener showing this condition that we are aware of is G. albida Mauriès & Vicente, 1978, a troglobite from Málaga, Spain (Mauriès and Vicente, 1978).Both these species are evidently regressive, apparently in response to cavernicoly.Similarly, G. albida also demonstrates two striae on the thoracic shield, only the anteriormost of which crosses the dorsum.In addition, both these species are pallid, of about the same size, with the same number (5) of ocelli and the same proportions (L/D 2.5) of antennomere 6, and both are devoid of tergal or pygidial sinuosity.Th e main diff erences concern the shape of the telopod  Even more regressive appears to be G. dionysii (Strasser, 1961), a troglobite from Sicily, Italy, which totally lacks striae on the collum (Strasser 1961).

Glomeris colorata
Coloration (Figs 3, 4) variegated, mostly rather vivid, background usually dark brown, but sometimes mostly yellow due to expanded light spots.Head mostly light brown, evidently marbled near ocelli, with 1+1 and 2+2 small pallid spots against a light brown background just above antennal sockets; labrum pale yellowish; antennae largely dark brown, only tip pallid.Collum with a large, mostly marbled, yellowish central spot (Figs 3,4).Th oracic shield with 2+2 large light spots, sometimes interconnected with a transversely oval marbled area in-between, but separated by a more or less evident, often incomplete, light axial line.Subsequent terga, except pygidium, with 1+1 more or less wide, light paramedian spots, usually arranged into clear stripes, a mostly interrupted and sometimes vague, light, axial line, and a pair of marbled, transversely oval, lateral areas.Pygidium with 1+1 light, paramedian, mostly coalesced spots (Fig. 4), less frequently nearly entirely light grey-brown.Venter and legs contrastingly light yellow.
Collum with two transverse striae.
Th oracic shield with a narrow hyposchism almost reaching caudal tergal contour (Fig. 5A); four transverse striae, of which 1-2 anteriormost starting well in front of

B A
schism and crossing entire dorsum, the other two always abbreviated (Fig. 5A); anteriormost stria only rarely slightly interrupted dorsally.
Telopods (Figs 5D) with a high, rather regularly rounded, bare, central syncoxital lobe fl anked by two setose horns, latter each crowned with a minute bulb and a short setoid fi lament.Tarsus quite narrowly rounded apically.
Remarks: Based on the narrow tergum 3, G. colorata sp.n. can be attributed to the "Stenopleuromeris" type.Remarks: Unfortunately, no new material of this species could be obtained for study.It is stated to be close to G. pustulata, but diff ers in the paramedian spots on the thoracic shield being much smaller than each lateral spot (Brandt 1840b(Brandt , 1841a, b), b).For the time being, we treat this taxon as a full species, but its separation from G. pustulata remains to be confi rmed.
Remarks: Unfortunately, no new material of this species could be obtained for study.It is stated to diff er from congeners in having a black-brown ground coloration, with the thoracic shield showing only a single, large, pale spot in the anterior part, each of the subsequent terga showing a pair of light, lateral, transverse-oval spots, and the pygidium completely dark (except for the usual pale margin, which is slightly wider laterally) (Schubart 1960).It belongs to the "Stenopleuromeris" type.Remarks: Nearly the entire type series of G. fl avomaculata has faded completely, probably due to the long preservation in alcohol.Fortunately, however, the paralectotypes from near Oran and the ?paralectotype ♂ from Lucas' collection (1850) have retained their coloration (Figs 6, 7), which matches quite closely the pattern well depicted by Lucas (1849), based on a then rather freshly collected syntype (Fig. 8).We are certain, however, that all diplopods printed in colour by Lucas (1849) in Plate 1 are too dark and red compared to their natural coloration, not only because pertinent alcohol material shows this, but also in view of the specifi c name itself, fl avomaculata, clearly indicating the presence of light spots on the body.Th e most likely explanation is that Lucas used dry material of G. fl avomaculata which had become somewhat darkened (see also below).

Glomeris fl avomaculata
Th is species belongs to the "Stenopleuromeris" type.
Th e syntype of G. fuscomarmorata examined here has faded completely, apparently due to the long preservation in alcohol.Fortunately, most of the other samples from North Africa, especially fresh ones, have retained their coloration (Figs 10,12), which matches quite closely the pattern depicted by Lucas ( 1849), based on a then recently collected syntype (Fig. 13).We are certain, however, that this species must also have been printed somewhat too dark and red compared to its natural coloration in Lucas' (1849) Plate 1, likely because Lucas used dry material (cf Figs 11 and 13).On the other hand, G. klugii in Europe is known to exist in two colour morphs, the dark "undulata" and the light "conspersa" (Hoess 2000), of which "conspersa" is much more widespread   and occupies peripheral parts of the species' distribution area, including North Africa.Could material of "undulata" have served, at least in part, for Lucas' (1846Lucas' ( , 1849) ) descriptions and illustrations? Hoess (2000) marked as questionable populations of "conspersa" from a few small, outlying areas in the Balkans, near Algiers and Tunis, but we can confi rm the presence of G. klugii in North Africa.We suggest that it could well have arrived there, particularly at the largest sea ports, through commercial activities, which have been going on since prehistoric times throughout the Mediterranean.Th e synonymy of G. klugii and G. fuscomarmorata proposed here therefore appears fully justifi ed.
Furthermore, we must give due tribute to Brandt (1840bBrandt ( , 1841a, b) , b) who, already at the very beginning of diplopodological explorations in North Africa, wrote that his G. marmorata from Germany and Algeria were identical.Although he failed to recognize that his own G. klugii and G. marmorata actually represented the same speciesapparently because the holotype of G. klugii (surprisingly) did not show any striae on the thoracic shield, retained (in alcohol) its generally light coloration and was thought to have come from Egypt or Syria-he was essentially correct in thinking that the same species could exist on both continents.Likewise correct have been the very few subsequent records of "conspersa" in Algeria and Tunisia (Attems 1908;Abrous-Kherbouche and Mauriès 1996), whereas most other authors believed that the North African fauna, including that of Glomeris, is fully endemic.
Th e syntype of G. maculosa is a "conspersa" specimen of G. klugii, with the colour pattern still well traceable.Hence the synonymy of these names is also proposed here.
Th e fact that G. klugii belongs to the "Eurypleuromeris" type provides an additional indication of its probable introduction to North Africa from Europe.Moreover, even though G. conspersa is the type-species of Eurypleuromeris, Verhoeff (1921) mistakenly attributed his G. maculosa to the "Stenopleuromeris" type.Indeed, the anterior part of tergum 3 in the syntype of G. maculosa that we have examined is probably a little narrower than is usual for European or other North African "conspersa" specimens of G. klugii, but this variation seems too modest to be considered a reliable distinction, perhaps even refl ecting individual rather than geographical variation.Th is is another good reason to abandon the subgeneric division of Glomeris.Type material: Algeria, Philippeville (now Skikda), date ?, leg.et det.Lucas ?, ♂ lectotype (here designated) (MNHN CC 094), paralectotypes: 1 ♂, 4 ♀ (MNHN CC 094).Th e designation of a lectotype seems advisable in order to fi x the type locality and to ensure that the name-bearing type shows the diagnostically important characteristics of the male sex, particularly as Lucas (1846) spoke about this species being recorded from several localities in Algeria.
Th is species is certainly among the largest and darkest in North Africa.It resembles the widespread and similarly uniformly very dark Western European G. marginata (Villers, 1789), yet G. sublimbata belongs to the "Stenopleuromeris" type.On the other hand, the degree of development of the anterior part of tergum 3 in relation to its posterior part may again prove to vary even intraspecifi cally, as is apparently the case in G. klugii (see above).
Remarks: Th is taxon is here considered to be a full species.Th e type series, said to have originally consisted of 13 specimens of both sexes (Brolemann 1925), is now   incomplete (only 5 specimens left), and strongly faded, but still with a detectable colour pattern.Variation in coloration and pattern modest, the most characteristic feature being the presence of a paramedian pair of especially large, light spots on tergum 6 and a complete absence of lighter markings thereafter until the pygidium .Th is species belongs to the "Stenopleuromeris" type.

Discussion
Interestingly, the proportion of Glomeris species and populations with a ventrolaterally narrowed tergum 3 increases towards the South.In Central and Eastern Europe, most (if not all) species show a broadly rounded tergum 3, i.e. belonging to the "Eurypleuromeris" type, whereas in the Mediterranean area, including North Africa, the majority of species have a shorter anterior part of tergum 3, i.e. they belong to the "Stenopleuromeris" type.A ventrolaterally narrower tergum 3 can be seen as an adaptation for tighter body enrolment, possibly providing better protection from desiccation in the South.In North Africa, only G. mohamedanica and most (if not all) of the G. klugii populations have a somewhat broadened tergum 3, whereas it is considerably narrower in the other species, including even the somewhat intermediate G. troglokabyliana sp.n.Most of the North African species of Glomeris demonstrate the same or a very similar structure of the telopod syncoxite, in which the central lobe is rather high to very high, bare and rounded, whereas the coxal horns are crowned with a small bulb/ lobule and a setoid fi lament, the latter either rounded or pointed at its apex.Only two species, G. klugii and G. monostriata sp.n., show evidently bifi d tips of the coxal horns, a character state which may be evidence of closer ties to European rather than North African counterparts.In the case of G. klugii, we are rather inclined to admit its early introduction from Europe, naturally from "conspersa"-type populations, through human agency to the major sea ports of Algeria and Tunisia.Active trade throughout the Mediterranean has begun at latest with the Phoenicians, ca 3,500 years ago.In the case of G. monostriata sp.n., the closest relatives are probably G. albida (Spain) and G. dyonisii (Sicily), although this might just refl ect convergent adaptations to the cave environment.
As elsewhere, there are two striae on the collum in most of the North African Glomeris species, reduced to one in G. monostriata sp.n.Th e number of striae on the thoracic shield is more variable, typically ranging from two to four, of which the anterior one or two (rarely none) cross the dorsum.
Th e distribution of Glomeris in North Africa  shows that all of the species are, as would be expected, confi ned to a narrow strip along the Mediterranean coast.Th e distribution of G. anisosticta remains unmapped because we only know it occurs, and is common, in Algeria (Brandt 1840b(Brandt , 1841a)).Th e proportion of cavernicoles (most likely troglobites) is increased (two of 11 species), which is hardly surprising given the predominantly harsh environments these normally meso-to hygrophilous Diplopoda face in North Africa.Among the ca 100 Glomeris species known to date, very few occur obligatorily in caves, i.e.only two from Europe (one in Spain, the other in Sicily) and only another two from North Africa (one in Algeria, the other in Libya).
Much more material is required to properly assess the North African glomeridan fauna.Despite the long history of exploration in Algeria, several taxonomic problems remain, such as the status of G. anisosticta.On the other hand, a country like Morocco, which includes most of the Atlas Mountains, will certainly be found to contain more species of Glomeris   than just G. brolemanni.In addition, the discovery of G. monostriata sp.n. in Libya hints that this country might well harbour a richer fauna of Glomeris, particularly in caves.Only the relatively small country of Tunisia can claim to be fairly adequately prospected.
Genetic investigations, e.g.allozyme analyses which have been very successfully applied to the study of European Glomeris by Hoess and Scholl (1999), might provide further insights into the taxonomy and relationships of the North African species.Scanning electron microscopy can also prove very useful in search for new characters (Golovatch and Enghoff 2003).
An updated checklist of the Glomerida in North Africa is given in Table 2, followed by a key to the species of Glomeris occurring in the region.
strongly marked in the bifi d tip of the horns and the relatively low and regularly rounded central lobe in G. monostriata sp.n.

Figure 30 .
Figure 30.A map showing the distribution of Glomeris sublimbata (cross) and G. fl avomaculata (open triangle).

Figure 31 .
Figure 31.A map showing the distribution of Glomeris punica (open circle) and G. carthaginiensis (fi lled triangle).

Figure 32 .
Figure 32.A map showing the distribution of Glomeris mohamedanica (open square) and G. klugii (fi lled star).

Table 2 .
An updated checklist of the Glomerida in North Africa (M: Morocco, A: Algeria, T: Tunisia, L: Libya).