(C) 2013 Andrey Frolov. This is an open access article distributed under the terms of the Creative Commons Attribution License 3.0 (CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
For reference, use of the paginated PDF or printed version of this article is recommended.
Citation: Frolov A (2013) Stenosternus Karsch, a possible link between Neotropical and Afrotropical Orphninae (Coleoptera, Scarabaeidae). ZooKeys 335: 33–46. doi: 10.3897/zookeys.335.5471
The monotypical orphnine genus Stenosternus Karsch is known from a single specimen of Stenosternus costatus collected on the São Tomé island (Gulf of Guinea). The holotype of Stenosternus costatus Karsch is re-examined and its characters are discussed and illustrated. Although the genus was implicitly placed by Paulian (1984) in the Old World tribe Orphnini Erichson, re-examination of the holotype of Stenosternus costatus shows that it has characters similar to those of the members of the New World tribe Aegidiini Paulian. Placement of Stenosternus in the Aegidiini is supported by the metepisternum widened posteriorly (forming posterior metepisternal lock for closed elytra) and a keel separating basal and anterolateral parts of propleurae. Relationships of Stenosternus with other orphnine taxa and possible ways of origin of São Toméan orphnine fauna are discussed.
Trans-Atlantic dispersal, Africa–America disjunction, biogeography, scarab beetles, orphnines, Orphnini, Aegidiini, São Tomé, Gulf of Guinea, Cameroon line
Orphninae Erichson is a group of predominantly tropical scarab beetles. To date, 15 genera and 195 species of Orphninae are known (
Stenosternus costatus, the only described species of the genus, has a puzzling combination of unique characters with those similar to the members of the Aegidiini. The original description of the genus and another work clarifying its position (
Material used in this work is deposited in or borrowed from the following organizations: Institut royal des Sciences naturelles de Belgique, Bruxelles (IRSNB), Muséum d’Histoire Naturelle, Geneva (MHNG), Museum für Naturkunde, Humboldt-Universität, Berlin (MHUB), Muséum national d’Histoire naturelle, Paris (MNHN), Natural History Museum, London (NHML), Zoological Institute RAS, Saint-Petersburg (ZIN).
Preparation of genitalia follows the common technique used in entomological research. Photographs were taken with a Leica MZ9.5 stereo microscope and a Leica DFC290 digital camera from dry specimens. Partially focused serial images were combined in Helicon Focus software (Helicon Soft Ltd.) to produce completely focused images. Photographs were not altered except for digital enhancing with Adobe Photoshop (Adobe Inc.): levels and tone correction, background elimination. Outline figures were made by tracing features on digital photographs with Adobe Illustrator (Adobe Inc.).
For modeling the distribution of Aegidium Westwood, MaxEnt software (
Type species. Stenosternus costatus Karsch, 1881, by monotypy.
(MHUB). Body length (from anterior margin of clypeus to apices of elytra) 16.1 mm, pronotal width 7.3 mm, elytral width 7.6 mm. Body (Fig. 1) uniformly dark-brown with slight bronze tint. Surface densely punctate, almost rugose. Head and most part of pronotum punctate with oval deep punctures separated by 0.5 to 0.2 their diameters, sometimes almost adjacent. Each puncture has shagreened microsculpture with 1 short (only slightly protruding above surface of pronotum) seta. Intervals between punctures look smooth.
1–3 Habitus of males 4 punctation of elytral disc 5 aedeagus in lateral view and parameres in dorsal view 6 serration of medial margins of parameres 7–10 aedeagus in ventral view 1, 4–7 Stenosternus costatus Karsch, holotype 2 Aegidium colombianum Westwood 3 Orphnus macleayi Laporte de Castelnau 8 Hybalus cornifrons (Brullé) 9 Aegidium parvulum Westwood 10 Orphnus compactus Petrovitz.
Clypeus emarginate anteriorly, with crenate margin, without tubercles. Genal and frontoclypeal sutures absent. Genae not protruding past eyes, indistinct. Frons feebly convex medially. Labrum feebly protruding past clypeus. Eyes relatively small, eye width 1.5 times smaller than distance between eye margin and gula (in ventral view).
Antenna 10-segmented, with 3-segmented club.
Pronotum trapezoidal, elongated (1.25 times wider than long) while in other Orphninae it is normally wider and shorter (about 1.6 times wider than long, Fig. 2, 3). Pronotum with distinct longitudinal middle depression from base to almost anterior margin. Lateral margin crenulate, base not bordered. Punctation of pronotum is similar to that of head. Propleurae with fine carinae separating anterolateral areas from basal area adjacent to bases of elytra (Fig. 34, arrowed).
Anterior tibiae relatively slender, almost parallel-sided, with 2 short lateral teeth and a smaller medial tooth (Fig. 15). Anterior tarsi absent. Anterior coxa with relatively deep longitudinal fossa on ventral side. Middle and hind legs similar in shape. Middle and hind femora relatively slender, almost parallel-sided, punctate with elongate punctures. Middle and hind tibiae without ridges on outer sides, rugosely punctate. Middle tibiae have 2 apical spurs, outer spur about twice as long as inner one. Hind tibiae with 2 spurs (outer spur about 1.5 times longer than inner one) and with modified spur-like basal tarsomere (Fig. 13). Tarsi of middle tibiae are absent but according to the picture provided by
11, 12 Right mandible in dorsal and apical view 13 apex of metafemur 14 head in ventral view 15–23 protibia of males (arrows in figures 15–17 indicate approximate teeth direction) 24–33 metepisternum 34–37 base of prothorax (arrows in figures 34–36 indicate keel separating basal and anterolateral parts of propleurae) 11, 16, 35 Aegidium colombianum Westwood 12, 37 Orphnus declivis Schmidt 13–15, 24, 34 Stenosternus costatus Karsch, holotype 17, 29 Orphnus macleayi Laporte de Castelnau 18, 36 Aegidinus guianensis (Westwood) 19, 27 Aegidiellus alatus (Laporte de Castelnau) 20, 26 Paraegidium costalimai Vulcano et al. 21, 31 Pseudorphnus hiboni Paulian 22, 30 Triodontus itremoi Paulian 23 Chaetonyx robustus Schaum 25 Aegidium parvulum Westwood 28 Orphnus giganteus Paulian 32 Renorphnus clementi (Petrovitz) 33 Hybalus cornifrons (Brullé).
Elytra somewhat oblong, 1.2 times longer than width. Humeral umbones small but distinct. Elytra without striae but each elytron with low longitudinal ridge from base to about 5/6 its length. Elytra densely punctate with characteristic semicircular punctures each bearing a short setae (Fig. 4). Because of rugose punctation, lateral margin of elytra appear crenulate in dorsal view. Epipleuron with concavity receiving hind margin of metepisternum.
Wings vestigial, about 1/2 length of elytra.
Scutellum 1/20 length of elytra, narrow, rounded apically.
Metepisternum narrow, almost parallel-sided, with rounded distal part which somewhat overlaps epipleuron (Fig. 24). Orifice between mesocoxal cavities absent. Abdominal sternites with irregularly shaped punctures, some punctures V-shaped. Abdominal sternite 8 longer than others, without concavity or tubercle in the middle. Plectrum trapezoidal, with minute seta in the middle near apical margin.
Aedeagus with heavily sclerotized parameres about 2 times shorter than phallobase (Fig. 5, 7). Phallobase symmetrical, with sclerotized ventral plate separated by weakly sclerotized membranous areas (Fig. 7). Apices of parameres with tooth-like rounded processes directed anterolaterally and relatively long and dense pale setae. Lateral sides of parameres with short, sparse, mostly abraded setae. Medial sides of each paramere without membrane in apical 2/3 but with dense, scale-like teeth directed apically (Fig. 6). Basal 1/3 of inner margin of parameres overlapping. Endophallus with about 15 small, tooth-like spinules, without larger sclerites.
Stenosternus can be easily distinguished from other Orphninae by the combination of uniquely modified legs and sculpture of pronotum and elytra. However, this diagnosis is based on a single male specimen and may include characters of sexual dimorphism. Aedeagus with serrated medial sides of parameres, setose apices of parameres, and separate ventral sclerite of phallobase is also highly distinctive.
Mouthparts
In the type specimen of Stenosternus costatus, right mandibular apex is worn and probably broken (Fig. 14) but the mandible shape is more similar to that of Aegidium than of Orphnus. It can be seen that the bristled incisor comb occupies a large part of the inner margin of the mandible.
Although the mandible morphology is probably species-specific in some orphnine genera, mandibular characters should be attributed low weight in reconstructing phylogeny of the groups above genus level. The morphology of the mandibles and other mouthparts depends heavily on the type of food utilized by the species. Adaptive radiation might result in parallel modification of the mouthparts in different lineages. Using mandibles in phylogenetic reconstructions is further complicated by their manifold variability not normally found in other structures: interspecific variability, asymmetry, sexual dimorphism, allometric variability in males (i.e. Madecorphnus Paulian), and variability caused by wearing during adult individual life. It is also important that mouthparts be studied as an interdependent complex of structures with strong morpho-functional relationships with each other. Adaptation to a particular food type results in regular and interrelated modifications in all mouthparts including labrum. Undesired redundancy can be introduced in a dataset if these characters are coded separately and, if not weighted, they can overcontribute to the most parsimonious tree topology.
Legs
The fore legs of Stenosternus costatus lack tarsi. The absence of protarsi is often found in Coleoptera including Scarabaeoidea; sometimes only males lack protarsi. However, in the Orphninae well developed protarsi are present in all groups except for Stenosternus. The shape of the Stenosternus protibiae is also unusual for the Orphninae: they are narrow, almost parallel-sided, with three short teeth (Fig. 15). These teeth can be interpreted as the three common scarabeoid lateral teeth with the tooth I being offset medially. However, in all Old World taxa these teeth are directed laterad of the imaginable midline of a protibia (Fig. 17, 21–23; approximate teeth direction are arrowed). In the males of the New World genera (except possibly for Paraegidium, Fig. 20) there is an additional, mostly shorter, medial tooth directed mediad of the protibia midline (Fig. 16, 18, 19). A similar medial tooth can be found in Chaetonyx robustus (Fig. 23), although it is less distinct and apparently subject to a reasonable interspecific variability. Although such medial teeth in male protibiae are known in other scarabeoid groups (i. e. in scarabeine genera Macroderes Westwood, Xinidium Harold and Metacatharsius Paulian), within the Orphninae they may be homologous.
One of the most prominent characters of Stenosternus is the modification of middle and hind tarsi which are reduced to only basal tarsomeres similar to spurs in length and shape (Fig. 13). Tarsal origin of these “spurs” can be seen from their setation and was discussed by
The shape of the legs of Stenosternus costatus suggests that the beetles are poorly adapted to digging in soil. Slender tibiae with short lateral teeth (in protibiae) and without transverse keels (in meso- and metatibiae) are typical for Passalidae and Lucanidae which have apparently ancestral habit as rotten wood dwellers. No information is available about biology of Stenosternus costatus but morphology of the specimen suggests that it too may be a rotten wood dweller. This also agrees with available data about Aegidium cribratum Bates which, at least in Mexico, is found only in rotten logs of a few species of tree (
The adaptive significance of the absence of tarsi is not always clear. It can be speculated that the tarsi are lost when there is no need for them, for example when a beetle does not clamber onto plants. Apparently a large number of Orphninae species, especially flightless ones, do not leave upper soil layer and litter and have no need in clambering. However all taxa, except for Stenosternus, have well developed tarsi in all legs. The absence of tarsi in Stenosternus might be an adaptation for defense from predators like ants or soldier termites.
Pronotum
The pronotum of Stenosternus is distinctive in being longer than in other Orphninae. Its width is almost equal to length (and it looks longitudinal in dorsal view) while in other taxa it is 1.5 times or more wider than long. It lacks ridges or tubercles which are found in males of most Orphninae taxa. A shallow but distinct longitudinal middle depression is found only in Stenosternus. The pleural areas of the pronotum have a feature which is not found in other Old World Orphninae. The propleura have fine but distinct carinae separating anterolateral areas from basal area adjacent to bases of elytra. This area is similar to that in all American genera which have basal area of pronotum adjacent to elytra separated from the rest of propleuron (Fig. 35, 36). Old World taxa have the propleura smoothly convex without any signs of carinae (Fig. 37). The shape of the pronotum of Stenosternus and American genera apparently allows for tighter joining of the pronotum and mesonotum which might be an antipredator adaptation.
Integument and punctation
Stenosternus costatus has a characteristic sculpture of the dorsal body surface. Its head and pronotum are densely and rugosely punctate with relatively large, mostly adjacent punctures. In other Orphninae, the punctation of the head and especially pronotum is normally finer and sparser. More distinctive is the punctation of the elytra which are densely covered by U-shaped punctures directed caudally (Fig. 4). Orphninae show reasonable variability of elytral punctation but in most taxa punctures are more or less round. Somewhat similar U-shaped punctures are known in Paraegidium but they are directed apically and their homology is unlikely. The elytra of Stenosternus costatus also have characteristic longitudinal ridges. Similar ridges are found in Aegidium but in this genus each elytron has 2 ridges (Fig. 2). Other taxa do not feature such elytral ridges.
Brachyptery
The type specimen of Stenosternus costatus has vestigial wings about 1/2 the length of elytron. Flightlessness is quite common in the Orphninae and the total number of flightless species may be up to 20 per cent (Frolov, unpubl.). Hybalus Brullé and Chaetonyx Schaum comprise completely apterous species with no visible wing rudiments whereas a number of Orphnus MacLeay species have vestigial wings similar to those of Stenosternus. Flightless species are also known among Aegidium (i.e. Aegidium parvulus Westwood from Guadeloupe).
Metepisternum
The metepisternum shows reasonable variability among orphnine genera (Fig. 24–33). In most taxa it is more or less triangular and tapering caudally (in this paper only the part of metepisternum not covered by elytron is discussed and illustrated) (Fig. 28–33). In most of these groups the metepisternum has a slightly widened anterodorsal angle slightly overlapping epypleuron which is somewhat concave in this place (Fig. 29–32). This structure, anterior metepisternal lock, apparently serves to hold the closed elytra more securely. It is absent in Hybalus (Fig. 33) which might be a secondary loss due to aptery, but it is also indistinct in Orphnus giganteus Paulian (Fig. 28) which has fully developed wings. In the New World taxa, the anterior metepisternal lock is less developed and sometimes indistinct but there is a similar structure in posterior angle of metepisternum which is rounded to triangular and is situated in the distinct concavity of epypleuron (Fig. 25–27). Such a posterior metepisternal lock is not found in the Old World genera except for Stenosternus. The metepisternum of Stenosternus is dissimilar to that in other orphnines in being narrow, almost parallel-sided in most part, but its posterior angle is rounded and overlaps slightly the epypleuron (Fig. 24). A shallow but distinct concavity can be seen on the epypleuron. It is possible that the posterior metepisternal lock is reduced in Stenosternus costatus due to brachyptery and it was more developed in its flying ancestors. In the preliminary phylogenetic analysis of the Orphninae (
Aedeagus
The aedeagus of Stenosternus costatus has a few distinctive features. Scale-like serration on medial sides of each paramere in apical 2/3 is unique at least among Orphninae. The phallobase of Stenosternus costatus can be classed as one of the four types of phallobase found in the Orphninae. It has a ventral sclerite about 2 times shorter than dorsal sclerite. The two sclerites are separated ventrolaterally by thinner, feebly sclerotized membranes (Fig. 7). This type of phallobase is similar to that of Hybalus where ventral sclerite is also distinct, but in Hybalus it has a complex shape (Fig. 8). In the four American genera, the phallobase is tube-shaped with strongly sclerotized ventral side but without differentiation of ventral and dorsal sclerites (Fig. 9). The majority of other Orphninae genera have a phallobase of the forth type which is strongly sclerotized dorsally and with a thin membrane ventrally (Fig. 10). Apices of parameres of Stenosternus costatus are densely setose. Similar setation can be found also in the monotypical American genus Aegidiellus, and Mediterranean genus Hybalus. Apparently in the Orphninae, setation of parameres is genus-specific, but its homology and phylogenetic value is not clear due to high variability of paramere shape which is, in general, species-specific.
The only known specimen of Stenosternus costatus was collected by Richard Greeff on the São Tomé island in the Gulf of Guinea. São Tomé is the largest island of the oceanic sector of the Cameroon Line of volcano-capped swells. It lies 240 km offshore and has never been connected to mainland Africa. The age of São Tomé is estimated as 13 MY (
Precise collecting locality of the holotype of Stenosternus costatus is unknown.
The six recent regional faunas of the Orphninae (
The core fauna of the islands of the Gulf of Guinea is of African origin. Bioko (Fernando Po) is the largest island with the most diverse biota though it is of continental origin and was connected to mainland Africa during the last glaciation maximum. Consequently its fauna differs slightly from the fauna of the mainland Cross-Sanaga-Bioko coastal forests. Of the true oceanic islands, São Tomé is the largest (850 km2), separated from mainland Africa by 250 km, has a few climatic regions, complex terrain profile, and supports rich biota. Príncipe, although much older than São Tomé (originated in early Oligocene) has six times smaller area and two times smaller altitude range. Annobon is a small island (17 km2) lying 340 km off African shore and having relatively poor biota. Because the islands were apparently a target for colonization for a long time (since the Miocene in case of São Tomé), adaptive radiation resulted in high endemicity at specific and generic levels. The islands contain some of the highest percentage of endemicity in the world (
Atlantic currents might facilitate dispersal of African biota to the oceanic islands of Gulf of Guinea (
Of the all Gulf of Guinea islands, Orphninae have so far been recorded only from São Tomé. It is possible that some Orphnus species (e.g. Orphnus gilleti Benderitter) known from the neighboring mainland region will be found in Bioko. Annobon is unlikely to have any Orphninae due to its small size and apparently unsuitable habitats. However few studies are available on the insects of the Gulf of Guinea and scarab beetles of the islands were not surveyed.
Available data do not allow to soundly hypothesize about the way of separation of the New World and Old world Orphninae faunas. Of the all barriers between regional orphnine faunas, the Atlantic barrier was apparently the least permeable for the members of the group. South America–Africa disjunctions are commonly explained by either vicariance or long distance dispersal. Vicariance of the previously Gondwanan group due to the continent breakup seems the least probable hypothesis in the case of the orphnines. Distribution of the extant taxa shows no pattern of Gondwanan groups: they are absent from regions where Gondwanan relicts are commonly found including Notogea (Australasia) and Patagonian Province of Neotropical Realm. Afrotropical region is the contemporary center of diversity and possible center of origin and diversification of the group. In Africa, Orphninae are mainly distributed in the central part of the continent in savannah and forest–savannah mosaic regions. Few species were recorded south of 25°S.
It is possible that orphnines dispersed from Western Africa to north-eastern South America via the Atlantic ocean. Zoogeographical relationships of Gulf of Guinea islands with South America have been discussed in a few works and modern disjunctive ranges of a number of plant (
To analyze distribution pattern of the Orphninae, MaxEnt distribution models were created for a number of genera including Aegidium (Fig. 38). It is beyond the scope of the present work to analyze Aegidium model in detail but one aspect of it is relevant here: the model predicts high probability of environmentally suitable biotopes in the islands of Cameroon line including São Tomé. The model is based on current climatic conditions but despite reasonable fluctuation of climate in the Northern Hemisphere, especially in the Pleistocene, and fluctuation of the size of equatorial rain forests in Africa and South America, climatic conditions are thought to be relatively stable in the Gulf of Guinea and the islands are considered a rain forest refuge since the Miocene. The distribution model suggests that Aegidium and Stenosternus share similar type of habitat. This can be an additional argument for considering the two genera phylogenetically related rather than superficially similar.
Representation of the MaxEnt distribution model for Aegidium Westwood. Warmer colors show areas with better predicted conditions. Black dots indicate known collecting localities of Aegidium used for the model. Insert shows the Gulf of Guinea and the São Tomé island. Other explanations are in text.
The data available to date do not allow Stenosternus to be definitely classifiedas a member of either Orphnini or Aegidiini.
Phylogenetic relationships of Stenosternus require further research because the material currently available is limited to a single specimen. The study of a female would reveal sexual dimorphism of this taxon. Especially important would be the study of premature stages because the larvae of the New World taxa, at least the genus Aegidium, differ significantly from those of the Old World taxa (
I am thankful to Dirk Ahrens for a discussion about scarab morphology and phylogeny and for help in obtaining the type specimen of Stenosternus costatus, Alaine Drumont (IRSNB) for help with literature and loan of material, Giulio Cuccodoro (Muséum d’Histoire naturelle, Geneva), Johannes Frisch (Museum für Naturkunde, Humboldt-Universität, Berlin), Olivier Montreuil (Muséum national d’Histoire naturelle, Paris), Maxwell Barclay (Natural History Museum, London) for loan of material. Robert Angus (Natural History Museum, London) and an anonymous reviewer provided valuable comments and corrections to the draft manuscript. This work was supported by Russian Foundation for Basic Research (grant 13-04-01002-a) and Scientific & Technological Cooperation Programme Switzerland-Russia.