Introduction

Most bark beetles that live in dry, old twigs and branches are of little economic importance. Hiding away in their largely cryptic or forgotten habitats, these beetles are often overlooked in field collections. The diversity of tiny dry-adapted bark beetles is nevertheless high and numerous species have been described from these kinds of habitats, often based just on a single or a few collecting events. In Africa and Madagascar such habitats are typically occupied by species in the genera Hypothenemus Westwood and Afrocosmoderes Johnson and Jordal (Trypophloeini), Glostatus Schedl (Xyloctonini), and several genera in the tribe Micracidini. The taxonomy of tiny and obscure beetles is often in flux, or largely ignored, as one may find it difficult to diagnose species properly. This is mainly the case for Micracidini where recent phylogenetic work has pointed towards many errors in the classification, particularly in the Afrotropical fauna (Jordal and Kaidel 2017).

Members of Micracidini are found in the Neotropics and southern parts of the Nearctic, and in the Afrotropical region, including Madagascar. More than 160 species have been described from the first two regions, whereas Africa and Madagascar are much less explored with fewer species known. The Afrotropics is nevertheless expected to have a similar diversity of species if given more taxonomic attention. Such inventories are accumulating, but the circumscription of meaningful genera has been a challenge. Micracidini included previously 13 genera globally (Wood 1986), with recent additions of two new Neotropical genera (Bright 2010, 2019) and inclusion of two genera from other tribes (Jordal and Kaidel 2017). Five additional new genera are here defined by molecular and morphological characters, and one genus is synonymized, with a new total of 21 genera (Table 1).

Figs. 1–7
figure 1

Various parts of the beetle body important for classification and identification of Micracidini genera. 1 Micracis carinulata head with antennae showing long dorsal horns on the scapus with tuft of setae, and a club with two procurved sutures. 2 Posterior view of declivity and the truncated shape of the last ventrite in Pseudolanurgus harunganae and 3 Dendrochilus sp. 4 Lateral view of elytron and ventrites in Lanurgus podocarpi and 5 Diplotrichus gracilis. 6 Dorsal view of elytra in Lanurgus oleae and 7 Diplotrichus subdepressus

Full size image
Table 1 Classification of Micracidini genera
Full size table

The biology of many species in Micracidini is unusual and deviates from the more typical bark beetles in several aspects. They are often bigamous where two females pair up with a colonizing male. This is a rare mating system in nature; in bark beetles, it seems more common to establish harem polygamy if more than one female should be accepted (Kirkendall 1983; Kirkendall et al. 2015). Females in this tribe are often gracefully ornamented, with long golden setae in the forehead and on the antennal scapus which is often enlarged or extended into a spine (Fig. 1). They use these structures in tactile communication with the male to gain access to a tunnel entrance in the bark. After mating, the two females extend the engravings of two independent tunnels where they deposit eggs in individual egg niches. Secondary sexual characters in the head region are species specific and provide the most useful characters to distinguish closely related species. Males in some genera can be more elaborated on the elytral declivity, with tubercles and mucronate elytral apices. In Cactopinus, and occasional other micracidine species, the male rather than the female has a concave frons, with horn-like projections (Atkinson 2010; Jordal and Kaidel 2017).

Taxonomists have focussed mainly on sexually evolving characters as a mean to define species. As a potential consequence, the classification at higher taxonomic levels became poorly developed. There are now obvious mixtures of species in each genus that belong to different genera in Micracidini (Jordal and Kaidel 2017) or even from other tribes (Johnson et al. 2020). Taxonomic inconsistencies were likely fostered by a general lack of diagnostic characters at the genus level. These problems are perhaps connected to a very ancient origin of Micracidini, which is considered the oldest monophyletic tribe in Scolytinae (Jordal and Cognato 2012; Pistone et al. 2018; Jordal and Kaidel 2017).

Molecular data from five genes are included in this study to provide a robust phylogeny which enable assessment of morphological consistency in groups of similar evolutionary age. New diagnostic features are supported by phylogenetic analyses and incorporated in a new identification key to genera which will improve accurate identification. Molecular data also allow for testing biogeographical scenarios—more specifically the role of Madagascar in the diversification of Afrotropical micracidines. Are species in Madagascar monophyletic with respect to the mainland, with few, or even a single, very ancient origin? If not, are Malagasy taxa paraphyletic, and of more recent origin, with adventive African lineages on the island? Alternatively, multiple recolonizations of the mainland could have happened, which is more in accordance with historical geophysical conditions (Ali and Huber 2010).

Methods

Samples

Type material of the type species for each genus was examined in museum collections in Vienna (NHMW), Tervuren (RMCA), and Paris (MNHN). In the preparation of revisions of the Afrotropical genera, the main types for each species have been studied (holotype, holotype with allotype and/or paratypes, or Egger’s ‘type’, except Miocryphalus ciliatipennis Schedl, 1979). Fresh material useful for DNA analyses (Table 2) was collected in several African countries between 1998 and 2015, after obtaining the necessary collecting and export permits as required at the time of field work.

Figs. 8–17
figure 2

Antennae of female Micracidini: 8 Leiomicracis aurea; 9 Afromicracis setifer; 10 Dendrochilus sp.; 11 Neomicracis squamigera; 12 Phloeocurus africanus; 13 Lanurgus rhusi; 14 Diplotrichus sp.; 15 Pseudolanurgus sp.; 16 Pseudomicracis sp.; 17 Microlanurgus bicolor

Full size image
Table 2 Sequence ID vouchers and their GenBank accession numbers
Full size table

Molecular data

Sequence data were obtained from five genes: Cytochrome Oxidase I (COI, 690 bp), Elongation Factor 1a (EF-1a, 822 bp), Carbamoyl-phosphate synthetase 2, aspartate transcarbamylase, and dihydroorotase (CAD, 458 bp), Poly-A binding protein 1 (PABP1, 435 bp), and the large ribosomal subunit (28S, 720 aligned nucleotide positions). DNA extraction, PCR and sequencing followed Mugu et al. (2018). The ribosomal alignment was made in Muscle (Edgar 2004) given default settings and a portion of ambiguous alignment sites were pruned in GBlock (Castresana 2000) using the most liberal settings possible (allow smaller blocks, allow gaps, allow less strict flanking regions, allow contiguous non-conserved positions). GenBank accession numbers are listed in Table 2.

Figs. 18–27
figure 3

Legs of selected species in Micracidini. Protibiae of 18 Micracis carinulata; 19 Pseudolanurgus sp.; 20 Lanurgus podocarpi; 21 Dendrochilus sp.; 22 Leiomicracis aurea. Metatibiae of 23 Micracis carinulata; 24 Lanurgus sp.; 25 Pseudolanurgus sp.; 26 Leiomicracis aurea; 27 Dendrochilus sp.

Full size image

Morphological analyses

Morphology was studied in a Leica MZ16 stereoscope and photographs made in a Leica 205 C stereoscope with a Leica camera. Multiple photos were stacked and aligned in ZereneStacker. Internal structures were dissected from macerated specimens (Proteinase K, or KOH 8%) and mounted on slides in Euparal. Thirty-six morphological characters were selected for phylogenetic analysis, including flight wings, proventriculus, and male genitalia (Table 3; Figs. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 15, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44).

Figs. 28–33
figure 4

Digestive systems in Micracidini. 28 Maxillae and labium of Lanurgus podocarpi, dorsal view. 29 Proventriculus of Lanurgus podocarpi; 30 Leiomicracis aurea; 31 Neomicracis squamigera; 32 Dendrochilus sp.; 33 Afromicracis sp.

Full size image
Table 3 Morphological variation and information potential in various character states as performed on the Bayesian tree topology
Full size table

Phylogenetic and biogeographical analyses

Nucleotide sequences from all five genes were concatenated and analysed in combination with the 36 morphological characters in MrBayes v 3.2.6 (Ronquist and Huelsenbeck 2003). Best models for each gene partition were selected in MrModelTest (Nylander 2004). Models for morphological characters were given a gamma-distribution based on empirical character variation. Stationarity in likelihoods was visualized in Tracer (Rambaut et al. 2014) and was reached before 3 million generations set as burn-in.

Approximate age for clades were estimated in Beast 1.10.4, with the xml file prepared in Beauti (Drummond and Rambaut 2007). Rates were calibrated with time estimates from a previous analysis of Scolytinae based on 18 genes (Pistone et al. 2018) and therefore given a normal distribution with 5 standard deviations to accommodate for uncertainties in these estimates. Nodes matching the previous published phylogeny included the root which combined taxa from Ipini, Hypoborini and Micracidini at 88 Ma, Micracidini without Leiomicracis at 79 Ma, and the Neotropical clade at 65 Ma.

Biogeographical scenarios were tested using model selection in RASP (Yu et al. 2020) using the time tree estimated in the Beast analysis. Areas in Africa were defined using biogeographical regions proposed by Linder et al. (2012): Southern African, Zambesian (eastern Africa), and Congolian (western Africa) regions, in addition to Madagascar and the Neotropical/Nearctic regions. The best model was selected among DEC, DIVAlike and BayArea using BioGeoBEARS, with or without the extra parameter for jump dispersal (+j). Because of the uncertain foundation for testing +j models (Ree and Sanmartín 2018), it was presumed that jump dispersal was more realistically involved in overseas splits, while vicariance motivated models can be more relevant in the inference of widespread taxa on the African mainland. Analyses with or without jump dispersal integrated in the model were furthermore compared to a Bayesian Binary MCMC analysis in RASP (Yu et al. 2015), using default parameters.

Results

Phylogeny

Bayesian inference of 3226 nucleotides and 36 morphological characters resulted in a well-resolved tree topology with high posterior probability for most nodes (Fig. 45). The inclusion of morphological characters did not change the tree topology, but increased node support for three nodes. Micracidini were monophyletic, with the Neotropical genera forming the sister group to all but one of the Afrotropical species. The excepted species formed the sister group to all Micracidini and is here described as a new genus, Leiomicracis.

The remaining Afrotropical taxa were distributed on two main clades. One was composed mainly by species in Afromicracis and its sister group, Dendrochilus. These two genera combined were placed as sister to a single species which is here described in the new genus Neomicracis. The other main clade included the genera Lanurgus and Pseudomicracis and three additional lineages described as new genera. One such lineage consisted of two tiny species with spatulate setae on odd-numbered interstriae only. This new genus, Microlanurgus, is not closely related to any other taxa (Fig. 45). Some species which are currently placed in Lanurgus have bifid setae on the metaventrite (and abdominal ventrites I-II) and all these species formed a lineage described as a new genus, Diplotrichus. This genus was more closely related to Pseudomicracis and a lineage erected as a new genus, Pseudolanurgus, which has a less mucronate elytral apex compared to Pseudomicracis. These three genera together were defined as a group by having a greatly enlarged and prolonged flagellum in the male aedeagus with the loss of tegmen and spiculum gastrale. Saurotocis, here represented by the type species S. dispar (Schedl, 1961), was furthermore nested within Pseudomicracis with maximum support.

A separate parsimony analysis of 36 morphological characters resulted in 68,947 trees. Nearly all genera as supported by molecular data were monophyletic, with two genera forming polytomies not contradicting monophyly (Afromicracis and Diplotrichus). Most characters performed well in terms of high retention indices, with 28 characters obtaining a retention index above 0.6 irrespective of optimization on the Bayes combined analysis topology or one of the parsimony tree topologies. Characters with the lowest performance had RI between 0.4 and 0.6 and these less optimal characters included features of the male frons (character 2), shape of pronotal and interstrial setae (10, 12), shape of the lateral edge of the protibiae (17) and the position of its denticles (18), the shape of the suture and teeth on the apical plate of the proventriculus (32, 33) and its closing teeth (34).

Biogeography

Model selection based on the AIC criterion in RASP using the Beast tree suggested DIVAlike (with or without the additional parameter for jump dispersal) as the best model for biogeographical analysis. The BBM reconstruction (Fig. 46) was similar to the DIVAlike+j analysis. Both of these differed from the DIVAlike analysis primarily by the simultaneously dispersal and vicariance events in accordance with jump dispersal theory, while DIVAlike always estimated dispersal one node before the inferred vicariance event, suggesting expansion of the ancestral area with subsequent vicariant splits in geographical isolation.

Figs. 34–41
figure 5

Male genitalia in selected Micracidini. 34 Micracisella nanula; 35 Cactopinus rhois; 36 Phloeocleptes cristatus; 37 Afromicracis sp.; 38 Neomicracis squamigera (with last two sternites and gut remains); 39 Microlanurgus bicolor; 40 Lanurgus podocarpi; 41 Diplotrichus gracilis

Full size image

BBM estimated an origin of the tribe in Eastern Africa, which occurred about 78.5 Ma (Table 4), and several subsequent range expansions took place from this area. Colonization of the Neotropics occurred in a single event not long after the origin of the tribe, around 69.3 Ma. Madagascar was colonized around 59.6 Ma, from eastern Africa, and thereafter colonized Southern Africa from Madagascar twice (Lanurgus 58 Ma, Diplotrichus subclade 29.5 Ma) and Eastern Africa once (Pseudolanurgus 31.5 Ma). The Afromicracis clade likely originated in Eastern Africa and expanded its ranges towards the west 47.4 Ma and more recently moved several times between various African regions. This clade never expanded its range outside the African mainland.

Figs. 42–44
figure 6

Hind wings of Micracidini species with stippled circles around precostal setae patch and radial cell setae, and arrows pointing at costal setae. 42 Phloeocleptus cristatus (3 precostal setae, 3 costal setae, and 1 radial cell seta); 43 Lanurgus podocarpi (4 costal setae, 3 radial cell setae); 44 Neomicracis squamigera (1 precostal seta, 2 costal setae, 2 radial cell setae)

Full size image
Table 4 Summary of key biogeographical events in Micracidini. Age of dispersal event was estimated in Beast, with median age given, and 95% highest posterior density in brackets
Full size table

Taxonomy

Micracidini LeConte 1876

Type genus – Micracis LeConte 1876.

Diagnosis. Eyes entire, short (longer in Micracis and Micracisella), female scapus usually with a small to large tuft of golden setae, funiculus 5- or 6-segmented, club moderately to strongly flattened, usually with two sutures, occasionally without; labial palps long, basal segment obliquely triangular and thereby separating diverging palps, segment 3 narrow, twice as long as segment 2. Pronotum weakly to strongly dome-shaped, asperate on anterior half. Elytra usually with rows of scale-like setae (hair-like in Cactopinus, some Hylocurus and some Afromicracis). Protibiae either parallel-sided or twisted, with one or more denticles apically on posterior side. Hindwings with either 1 or 3 setae on precosta, 2–9 on costa, and usually 2 (1–3) on distant margin of the radial cell (pterostigma). Scutoscutellar suture parallel to the scutellar grove for most of its length. Pleural suture zigzag shaped. Postnotum separated from metanotum by a membrane. Anterior plate of the proventriculus longitudinally divided and broadly separated (tight suture in some Lanurgus and most Cactopinus), transversely set by sharp teeth, or ridges (Cactopinus), closing teeth plumose or serrated, sometimes compacted. Male genitalia usually strongly elongated by very long apophyses, tegmen y- or ring-shaped, or absent; if present the median struts (manubrium) usually very long (short to absent in Lanurgus, Microlanurgus, Thysanoes).

Leiomicracis Jordal, gen. nov

urn:lsid:zoobank.org:act:4876A2E7-5058-49C8-ADB3-B0DE245EC2A9

Type species. Leiomicracis aureus Jordal, sp. nov., monotypic.

Etymology. From Greek Leios, meaning smooth, referring to the lack of asperities on the anterior part of the pronotum, and the Greek genus name Micracis. Gender feminine as explained for Micracis by Alonso-Zarazaga and Lyal (2009).

Diagnosis. Typical micracidine by the twisted lateral edge of the protibiae with only a single socketed denticle present on the posterior face; inner mucro large, curved posteriorly. Female scapus longer than funiculus, with a triangular tuft of setae on its dorsal side. Antennal funicle 5-segmented; club setose, without sutures. Pronotum asperate in middle, smooth on posterior half, lateral thirds and anterior sixth. Interstriae with a single row of scale-like setae. Setae on ventrites simple, unifid. Hindwing precosta with a single seta, radial cell with two setae. Proventriculus with anterior plate much longer than posterior plate, longitudinally, broadly divided, with bundles of strong, long crop teeth; apical teeth enforced, forming U-shaped structure.

Distribution. Tanzania.

Remarks. Leiomicracis is phylogenetically isolated as the sister group to all other Micracidini. The basal position in Micracidini was also confirmed by adding DNA sequences to the previously published 182 Scolytinae taxa dataset of Pistone et al. (2018). The new genus is distinguished from all other genera in the tribe by the smooth anterior sixth of the pronotum and by the scattered long setae forming a triangular tuft on the scapus.

Leiomicracis aurea Jordal, sp. nov

urn:lsid:zoobank.org:act:67223A57-D607-43D7-83DA-3754CB428B1C

(Figs. 47, 50, 53)

Type material. Holotype female: Tanzania, Udzungwa, Megombera forest, NE Mang’ula, GIS: −7.817, 36.980, ex. old twig, 15 Nov. 2009, 15xi-2, B. Jordal leg. Paratypes: same data as holotype (3). Holotype and two paratypes types deposited in ZMBN, one paratype in NHMW.

Etymology. Latin adjective aurea meaning golden or splendid, referring to the golden-brown colour of the most mature specimens, and the elegant golden setae on the female scapus.

Diagnosis. Female scapus with long golden setae on its dorsal side increasing in length towards its distal end, tuft appearing triangular. Pronotum with asperities in middle, reaching only to anterior sixth. Protibiae with a single denticle on its posterio-lateral side. Proventriculus with closing teeth compacted and coarsely serrated.

Description, female. Length 1.1–1.3 mm long, 2.4–2.5 × as long as wide. Colour light golden or bronze brown (possibly young individuals). Frons flat to weakly impressed, surface reticulate, obscurely punctured; scattered fine setae, slightly thicker and denser on epistoma. Antennal scapus longer than funiculus, dilated apically, with a tuft of increasingly long golden setae on its dorsal side; funiculus 5-segmented, pediculus about as long as segment 2 and 3 combined; club flat, 1.6 × as long as wide, pubescent, with a faint trace of one procurved suture. Pronotum distinctly humpbacked, on median half from summit to anterior sixth with sharp asperities, on posterior half, laterally, and anterior sixth smooth, with obscure punctures. Basal and lateral margin rounded. Vestiture consisting of fine, semi-erect setae. Scutellum wider than long, posterior margin U-shaped. Elytra finely rugose, shiny, with irregular rows of very small interstrial and strial punctures. Vestiture consisting of regular interstrial rows of erect, narrowly spatulate setae, and fine, short semi-recumbent strial setae. Ventral vestiture of fine, rather long setae. Legs. Protibiae evenly rounded on lateral and inner sides, ending in an apical central mucro which is rather strong, curved posteriorly; one additional tiny socketed denticle on posterior side close to tarsal insertion. Metatibiae narrow, sides parallel, with distinct, fine, long inner mucro, and three transversely set apical denticles. Proventriculus. Apical plate 2.5 × as long as posterior plate, with regular transverse rows of sharp triangular or obovate teeth. Apical teeth connected in an U-shaped formation with 2–4 longer, sharp teeth. Closing teeth short, rather smooth.

Male. Unknown.

Biology and distribution. Only known from the lowland type locality near the Udzungwa mountains in Tanzania. Six young female specimens were taken from a small twig of an unknown host plant.

Neomicracis Jordal, gen. nov.

urn:lsid:zoobank.org:act:F44DE0EF-DCD7-4285-BFAC-8AA30D30BEF3

Type species. Neomicracis squamigera Jordal, sp. nov., monotypic.

Etymology. The Greek prefix Neo-, new, reflects the designation of a new genus in Micracidini. Gender feminine (Alonso-Zarazaga and Lyal 2009).

Diagnosis. Typical micracidine with twisted protibia which has apical denticles on its posterior side, and a long tuft of setae on a dorsally strongly elongated female scapus. Antennal funicle 6-segmented; club with first suture absent on median 3/4 and the second suture complete, transverse, surface of segment 1 and 2 fused, corneous. Elytral apex slightly inflated. Setae on pronotum in both sexes and on ventrites III-V in the males broad and scale-like. Apical margin of metatibia transverse. Precosta in hind wings bearing a single seta, radial cell with two setae. Proventriculus with fine, dispersed crop spines; apical plate shorter than posterior plate. Male genitalia with tegmen Y-shaped, manubrium long; spiculum gastrale a simple rod.

Distribution. Tanzania.

Remarks. This genus is superficially similar to the new genera Pseudolanurgus and Diplotrichus described below, having an inflated elytral apex, but differ from those genera by the elongated horn on the scapus. Neomicracis is unique in the tribe by having corneous first two segments of the antennal club (first suture interrupted), and by the broad scale-like setae on ventrites III-V in the males. It differs further from the sister group, which consists of Dendrochilus and Afromicracis, by the 6-segmented funiculus.

Neomicracis squamigera Jordal, sp. nov.

urn:lsid:zoobank.org:act:8D5258B2-AC48-4EF2-8C22-607F50C2F3D5

(Figs. 48, 51, 54)

Type material. Female holotype: Tanzania, Udzungwa National Park, GIS -7.868, 36.844, ex small twig, 11xi-7, B. Jordal, leg. Allotype male and paratypes: same data as holotype (5). Holotype, allotype and two paratypes in ZMBN, two paratypes in NHMW.

Etymology. Latin adjective composed by squama, meaning scale, and -ger, meaning bearing, referring to the broad scale-like setae associated particularly with asperities on the pronotum.

Diagnosis, female. Pronotum with very broad scale-like setae. Scapus short, bearing dorsally an elongated spine with long setae pointing inwards; funiculus 6-segmented; frons smooth and slightly impressed on central third; scutellum with four short setae in a transverse row. Male similar to female except scapus elongated, frons flat, reticulated, and ventrites III-V with very broad and long scale-like setae.

Description, female. Length 1.1–1.2 mm long, 2.4–2.6 × as long as wide. Colour brown. Frons flat, weakly impressed and impunctate on central one-third, coarsely reticulated elsewhere; scattered fine setae present around impressed area, slightly thicker and much longer, dense setae on epistoma. Antennal scapus shorter than funiculus, strongly elongated on its dorsal side into a long triangular horn, with a tuft of long forwardly pointing setae; funiculus 6-segmented, pediculus about as long as segment 2–4 combined; club flat, as long as wide, with suture 2 transverse, suture 1 absent in median three quarter, segments 1 and 2 fused, corneous, shiny. Pronotum short, much wider than long, summit slightly elevated, anterior slope short, steep; anterior half with sharp, almost quadrated asperities, each associated on its posterior side with a broad scale-like setae; on posterior half generally smooth, reticulated, in median posterior area slightly granulated; vestiture consisting of fine, semi-erect setae. Basal and lateral margin rounded. Scutellum U-shaped, wider than long, with 3–4 short setae. Elytra finely rugose, finely granulated, on declivity granules sharper; punctures faint, confused. Vestiture consisting of regular interstrial rows of slightly curved, pointed or slightly spatulate setae, and fine, short semi-erect strial setae. Ventral vestiture of fine, rather long setae. Legs. Protibiae with lateral edges evenly rounded towards a large, posterio-laterally curved terminal mucro, appearing slightly twisted; one lateral socketed denticle and one additional denticle on posterior side close to tarsal insertion. Metatibiae narrow, sides parallel, with a distinct, fine, long inner mucro, and two transversely set apical denticles. Proventriculus. Apical plate slightly shorter than posterior plate, with regular transverse rows of sharp triangular. Apical teeth not apparent. Closing teeth smooth, long; femoral teeth sharp, connected to base of closing teeth.

Male. Similar to female except length 1.0–1.1 mm, 2.2–2.5 × as long as wide; frons rugose, lightly impressed on lower half; upper margin of impressed area marked by two small tubercles; antennal scapus short, rounded; setae on abdominal ventrites III–V very broad and long, one transverse row on each ventrite, pointing posteriorly. Genitalia elongated, apophyses as long as penis body, tegmen open dorsally, Y-shaped, with long manubrium; spiculum gastrale a simple rod, as long as penis.

Biology and distribution. Only known from the type locality in the Udzungwa mountains in Tanzania. It was taken from a 1 cm thick twig of an unknown host plant where it was excavating longitudinal but irregular egg tunnels.

The Afromicracis clade

This group consist of two genera defined by a long, narrow scapus with scant setae in both sexes, and a 5-segmented funiculus. The precosta on the hind wing has only a single seta, the inner mucro of the metatibiae is long and thin, and in males the tegmen is minute and Y-shaped, and a very long, thin and often coiled flagellum runs through the aedeagus.

Dendrochilus Schedl, 1957

Type species. Dendrochilus strombosiopsis Schedl, 1957.

(Figs. 49, 52, 55)

Diagnosis. Small, elongated species. Funiculus 5-segmented, scapus at least as long as the funiculus in both sexes, in females with a fine tuft of setae on its dorsal side; antennal club elongated, corneous on both sides, shiny and suture-free. Last ventrite elongated, bottle-shaped and truncated. Proventriculus with closing teeth compact and serrated.

Included species:

Dendrochilus strombosiopsis Schedl 1957.

Excluded species: Dendrochilus arundinarius Schedl, 1957, Dendrochilus elongatulus Schedl, 1977, Dendrochilus filum Schedl, 1977, Dendrochilus jasminae Schedl, 1957, Dendrochilus mikaniae Schedl, 1957 Dendrochilus robustus Schedl, 1957, all to Afromicracis; Thamnurgus villersi Lepesme, 1942 to Xylocleptes Ferrari, and Dendrochilus intermedius Schedl, 1977, to Acanthotomicus Blandford.

Distribution. Congo, Tanzania.

Remarks. The type species differs from other species previously included in this genus by the elongate, shiny and suture-free antennal club, and by the peculiar bottleneck-shaped apical part of the last ventrite. Two very similar, but undescribed, species grouped as the sister to Afromicracis by maximum support. Both genera share the presence of a long scapus in both sexes, 5-segmented funicle, and a reduced Y-shaped tegmen. Most species assigned to Dendrochilus by Schedl belong to Afromicracis (see below).

Afromicracis Schedl, 1959.

=Miocryphalus Schedl, 1939, synonym by Alonso-Zarazaga and Lyal (2009), name unavailable

Type species. Afromicracis kenyaensis Schedl, 1959.

(Figs. 56, 59, 62)

Diagnosis. Antennal scapus in both sexes straight, longer than funiculus, slightly dilated but not strongly inflated; in females with a fine tuft of scant long setae; funiculus 5-segmented; club on anterior face usually with one procurved suture close to apex. Elytra with scale-like or hair-like setae, always in single rows. Protibiae twisted, with 2 or 3 denticles on latero-posterior side near apex. Ventral setae always simple, never bifid or plumose. Proventriculus with cluster of few crop spines attached to a tubercle base. Male genitalia with a long flagellum, sometimes shorter and broad, sometimes very long and coiled; tegmen reduced to absent, apophyses more than twice as long as aedeagal body.

Included species:

Afromicracis congonus (Eggers, 1940) (original genus Miocryphalus)

Afromicracis convexus Schedl, 1962

Afromicracis dubius (Schedl, 1950) (Miocryphalus)

=Afromicracis angolensis Schedl, 1962 syn. nov.

Afromicacis elongatulus (Schedl, 1977) comb. nov. (Dendrochilus)

Afromicracis jasminiae (Schedl, 1957) comb. nov. (Dendrochilus)

=Dendrochilus mikaniae Schedl, 1957 syn. nov.

Afromicracis kenyaensis Schedl, 1962

Afromicracis longus (Nunberg, 1964) (Miocryphalus)

Afromicracis natalensis (Eggers, 1936) (Stephanoderes)

Afromicracis robustus (Schedl, 1957), comb. nov. (Dendrochilus) (Figs. 57, 60, 63)

=Dendrochilus arundinarius Schedl, 1957 syn. nov.

=Hypothenemus bambusae Browne, 1970 syn. nov.

=Dendrochilus filum Schedl, 1977 syn. nov.

Afromicracis setifer (Schedl, 1957) comb. nov. (Mimips, via Mimiocurus)

Excluded species. The following species have entire eyes and large suture-free antennal clubs and are therefore placed in Eidophelus Eichhoff: Afromicracis agnathus (Schedl, 1942), Afromicracis ciliatipennis (Schedl, 1979), and Afromicracis klainedoxae (Schedl, 1957). Afromicracis nitidus (Schedl, 1965) was originally described in Eidophelus but has emarginated eyes and a general habitus as in Hypothenemus and transferred to that genus (see below). Afromicracis nigrinus (Schedl, 1957) has emarginated eyes and is placed in a new genus Karlsenius near Afrocosmoderes (described below). Six species previously placed in Afromicracis were transferred to Macrocryphalus Nobuchi, Afrocosmoderes, Hypothenemus and Eidophelus by Johnson et al. (2020). One of these, H. attenuatus (Eggers, 1935) is very similar to A. nigrinus and is here placed in the same genus, closely related to Afrocosmoderes and Hypothenemus (see below).

Distribution. Tropical Africa.

Remarks. Afromicracis was removed from synonymy under Miocryphalus by Alonzo-Zarazaga and Lyal (2009), with the latter taxon deemed nomen nudum. Many species were erroneously assigned to Afromicracis (see Wood and Bright, 1992) and are now placed in many different genera in different tribes. Although species are small in size, they do have sufficient characters for reliable classification as noted in the diagnosis.

Microlanurgus Jordal, gen. nov.

urn:lsid:zoobank.org:act:8 AD67315-6B08-43D4-9FF1-296B37B06702

Type species. Microlanurgus bicolor Jordal, sp. nov.

Etymology. Adding the Greek prefix Micro-, meaning very small, to the genus name Lanurgus (masculine), referring to the size of species which are shorter than 0.8 mm.

Diagnosis. This genus consists of very small and stout species (<0.8 mm), with the typical twisted shape of micracidine protibiae which have only one lateral denticle and a single additional denticle on its posterior face. The scapus is much shorter than the 6-segmented funiculus, in females with a little tuft of setae on its dorsal side. The antennal club has two sutures where suture 1 is transverse and suture 2 slightly recurved, appearing ring-like. Unique features include a pronotum which is strongly dome-shaped, with few but large asperities which are not fully reaching the anterior margin; basal and lateral margins rounded. Elytra have scattered, erect, spatulate setae on odd-numbered interstriae only. Scutellum triangular, as wide as long. Male genitalia have tegmen shaped as a complete ring without manubrium, and the apophyses are not longer than the aedeagal body. Proventriculus with apical plate longitudinally divided, much shorter than posterior plate, with few sharp triangular teeth. Lateral, apical and femoral teeth absent; closing teeth lightly serrated, masticatory brush appears spiny.

Distribution. Madagascar (southern parts).

Remarks. This is the only micracidine genus with erect elytral setae only on odd-numbered interstriae. The proventriculus and protibiae are typical micracidine and molecular data firmly placed the two included species in the tribe (Pistone et al. 2018). There are no obvious close relatives based on morphological comparison or by phylogenetic analyses, but one may note that the male genitalia is quite similar to species in Lanurgus.

Included species:

Microlanurgus bicolor Jordal, sp. nov.

urn:lsid:zoobank.org:act:65D3E941-661E-41A5-B287-998CC3052A62

(Figs. 58, 61, 64)

Type material. Holotype, female: Madagascar, Parc National de Tsimanampetsotsa, Forêt de Bemanateza, 20.7 km 81° E Efoetse, 23.0 km 131° SE Beheloka, B. Fischer et al., leg. BLF6254, EH11 sifted litter (leaf mould, rotten wood) [GIS: 23° 59′ 32” S, 043° 52′ 50″ E]. Paratypes: same data as holotype (8); Parc National d’Andohahela, Manantalinjo, 33.6 km ENE Amboasary [24° 49′ 01” S, 046° 36′ 36″ E], ex sifted litter. BLF4810, B. Fischer, leg. (1). Holotype and 4 paratypes in CAS, 2 paratypes each in ZMBN and NHMW.

Etymology. The Latin adjective bicolor, meaning two-coloured, refers to the light yellow and black colours of the elytra.

Diagnosis. Pronotum with about 20 coarse asperities, 2 or 3 of these near the anterior margin. Elytra bicolored, spatulate setae present on odd-numbered interstriae only, each separated by 2–3 times the length of a seta.

Description, female. Length 0.6–0.7 mm long, 2.0–2.2 × as long as wide. Colour yellow and black; primarily yellow on pronotum and posterior part of elytra, venter, and anterior part of elytra black. Frons flat, shiny and impunctate on central half from epistoma to upper level of eyes, reticulate elsewhere; scattered, short, fine setae present around glabrous central area. Antennal scapus much shorter than funiculus, rounded, with a scant tuft of long setae on its dorsal side; funiculus 6-segmented, pediculus about as long as segment 2 and 3 combined; club flat, slightly longer than wide, suture 1 transverse, suture 2 recurved, appears ring-like in dorso-lateral view. Eyes separated by 2.7–2.9 × their width. Pronotum strongly dome-shaped, on anterior two-thirds with large, rounded asperities, only 2–3 of these reaching near anterior margin. Vestiture consisting of mixed spatulate and fine setae, longest setae near pronotal summit. Elytra smooth, shiny, stria not impressed, punctures shallow, large compared to body size, interstrial punctures obscure. Vestiture consisting of rows of erect, spatulate setae on odd-numbered interstriae, and scattered, fine and very short strial setae. Ventral vestiture of fine, rather long setae. Legs. Protibiae with lateral edges subparallel, inner mucro strong, curved posteriorly; two tiny socketed denticle transversely set apically on posterior side close to tarsal insertion. Metatibiae broader apically, with a distinct, fine, long inner mucro, and one apical and two laterally placed thin denticles.

Male near identical to female, except fewer and shorter seta on scapus.

Biology and distribution. Many specimens were sifted from leaf mould and rotten wood. It is only known from two localities in in southern Madagascar characterized by dry vegetation types.

Microlanurgus ater Jordal, sp. nov.

urn:lsid:zoobank.org:act:DA34F050-3BDE-4688-BB28-68202E8841F5

(Figs. 65, 68, 71)

Type material. Holotype, female: Madagascar, Mahafaly Plateau, 6.2 km 74° ENE Itampolo, B. Fischer, leg. BLF5758, EH11 sifted litter (leaf mould, rotten wood) [GIS: 24°39′13”S, 043°59′48″E]. Paratypes: same data as holotype (6). Holotype and 1 paratype in CAS, 2 paratypes each in ZMBN and NHMW.

Etymology. The Latin adjective ater means black, referring to the colour of the species.

Diagnosis. Frons impressed below upper level of eyes. Pronotum with about 15–20 coarse asperities, 4–5 of these near the anterior margin. Elytra and pronotum black, legs yellow; spatulate setae present on odd-numbered interstriae only, each separated by 2–3 times the length of a seta.

Description, female. Length 0.7–0.8 mm long, 2.0–2.1 × as long as wide. Colour black, legs yellow. Frons rounded from vertex to just below upper level of eyes, then impressed from a faint transverse rim to epistoma; impressed area glabrous, smooth and impunctate, striate and rugose elsewhere, with fine, short scattered setae. Antennal scapus much shorter than funiculus, rounded, with a scant tuft of long setae on its dorsal side; funiculus 6-segmented, pediculus about as long as segment 2 and 3 combined; club flat, slightly longer than wide, suture 1 transverse, suture 2 recurved, appears ring-like in dorso-lateral view. Eyes separated by 2.4–2.5 × their width. Pronotum strongly dome-shaped, on anterior two-thirds with large, rounded asperities, only 4 or 5 of these reaching near anterior margin; posterior third lightly rugose. Vestiture consisting of mixed spatulate and fine setae, longest setae near pronotal summit. Elytra smooth, shiny, stria not impressed, punctures shallow, separated by 2–3 × their diameter, interstrial punctures absent. Vestiture consisting of rows of erect, lightly curved, spatulate setae on odd-numbered interstriae, and scattered, fine and very short strial setae. Ventral vestiture of fine, rather long setae. Legs. Protibiae with lateral edges subparallel, inner mucro strong, curved posteriorly; two tiny socketed denticle transversely set apically on posterior side close to tarsal insertion. Metatibiae broader apically, with a distinct, fine, long inner mucro, and one apical and two laterally placed thin denticles.

Male near identical to female, except fewer and shorter seta on scapus.

Biology and distribution. Only known from the type locality in dry thorny shrub vegetation in south-western Madagascar where it was sifted from leaf mould and rotten wood. Habitat occupation is similar to the close relative M. bicolor, both were collected in large numbers on two occasions, and localities were less than 300 km apart.

Traglostus Schedl, 1938

Type species. Traglostus exornatus Schedl, 1938.

(Figs. 66, 67, 69, 70, 72, 73)

Diagnosis. Antennal scapus short, funicle 6-segmented, club finely pubescent without sutures. Female mandibles with a pair of sharp spines directed upwards, frons concave in both sexes, female frons with two long tufts of setae curved up and forward. Elytra with confused rows of bristle-like setae of variable length and thickness, male elytra with a pair of spines at base of declivity.

Included species:

Traglostus exornatus Schedl, 1938

Excluded species: Traglostus longipilus Schedl, 1958; Traglostus pubescens Schedl, 1941, to Lanurgus. Traglostus brevisetosus Schedl, 1957 and T. spathulatus Schedl, 1982 were transferred to Lanurgus by Beaver (2011), as synonyms of other Lanurgus species.

Distribution. Africa (Kenya).

Remarks. Pronotum and elytra are very much as in Lanurgus, but female Traglostus differs particularly by the two peculiar long spines on the mandibles directed dorsally. Until DNA data become available it seems prudent to treat this taxon as a valid genus. Four previous members of the genus are now in Lanurgus, two of which are synonymized under other species in that genus (Beaver 2011).

Lanurgus Eggers, 1920

Type species. Lanurgus barbatus Eggers, 1920.

(Figs. 74, 77, 80)

Diagnosis. Antennal club with two procurved sutures; scapus short, with dorsal spine. Interstrial setae scale- or bristle-like, often in multiple confused rows; strial setae broad, confused. Ventral setae simple and hair-like. Protibiae with 3–6 lateral and apical denticles; metatibiae transverse near apex. Male genitalia with apophyses as long as the aedeagal body, tegmen a simple ring. Apical plate of the proventriculus only narrowly separated by a longitudinal suture, lateral teeth reinforced medially.

Included species:

Lanurgus barbatus Eggers, 1920

Lanurgus capensis Schedl, 1965

Lanurgus longipilis (Schedl, 1958) comb. nov. (Traglostus)

Lanurgus oleae Schedl, 1955

Lanurgus podocarpi Schedl, 1955

=Traglostus brevisetosus Schedl, 1957 (syn. Beaver, 2011)

=Lanurgus bicolor Schedl, 1961 (syn. Beaver, 2011)

=Pseudohylocurus caplandicus Nunberg, 1961 (syn. Schedl, 1962)

Lanurgus pubescens (Schedl, 1961) comb. nov. (Traglostus)

Lanurgus rhusi Schedl, 1962

=Traglostus spatulatus Schedl, 1982 (syn. Beaver, 2011)

Lanurgus spathulatus Schedl, 1948

Lanurgus subsulcatus Browne, 1970

Lanurgus xanthophloeae Schedl, 1957

Lanurgus xylographus Schedl, 1963

=Lanurgus oleaeformis Schedl, 1970 (syn. Beaver, 2011)

Excluded species: Lanurgus cribrellus Schedl, 1965 and Lanurgus minutissimus Schedl, 1961 to Pseudolanurgus, Micraciops catenatus Schedl, 1953, Landolphianus elongatus Schedl, 1950, Lanurgus euphorbiae Schedl, 1961, Lanurgus frontalis Schedl, 1953, Lanurgus gracilis Schedl, 1958, Landolphianus minor Schedl, 1950, Micraciops obesus Schedl, 1953, Lanurgus pygmaeus Schedl, 1965, Lanurgus rugosipes Schedl, 1961, Lanurgus subdepressus Sched, 1965, Lanurgus widdringtoniae Schedl, 1962, all to Diplotrichus.

Distribution. South Africa, Namibia, Tanzania, Kenya.

Remarks. Traglostus pubescens share all typical features with Lanurgus and is transferred to that genus. Traglostus longipilus is more atypical by the transverse sutures on the antennal club, very long vestiture, and two hornlike projections on the epistoma. However, it is more similar to other Lanurgus species than the type species of T. exornatus.

The ‘flagellum’ clade

Three genera are defined by an enormously expanded and prolonged flagellum in the male aedeagus, where the tegmen and spiculum gastrale are missing. The last ventrite is slightly to strongly compressed or truncated. Most species are found in Madagascar, the remaining in south to south-eastern parts of Africa.

Diplotrichus Jordal, gen. nov.

urn:lsid:zoobank.org:act:3A8EE90B-38FC-4D0E-93A8-F988F76EB5DE

=Landolphianus Schedl, 1950, syn. – type not designated (Alonso-Zarazaga and Lyal 2009)

=Micraciops Schedl, 1953, syn. – type species not designated (Alonso-Zarazaga and Lyal 2009)

Type species. Lanurgus gracilis Schedl, 1958.

(Figs. 75, 78, 81)

Etymology. From Greek Diplo-, meaning double, and thrix, meaning hair, referring to the split setae on ventral sclerites.

Diagnosis. Antennal scapus inflated, in females rarely with a dorsal spine, funiculus 6-segmented, antennal club with two procurved sutures. All interstriae with a single row of setae (except catenatus group), strial setae very fine, in rows. Protibiae with 3–4 lateral and apical denticles: metatibiae rounded near apex. Ventrites 1 and 2 and part of the metaventrite with bifid setae. Male genitalia with long apophyses, exceedingly long and enlarged flagellum, tegmen and spiculum gastrale missing.

Included species:

Diplotrichus catenatus (Schedl, 1953) comb. nov. (Micraciops)

Diplotrichus elongatus (Schedl, 1950) comb. nov. (Landolphianus)

Diplotrichus euphorbia (Schedl, 1961) comb. nov. (Lanurgus)

Diplotrichus gracilis (Schedl, 1958) comb. nov. (Lanurgus)

Diplotrichus ignotus (Schedl, 1965) comb. nov. (Pseudomicracis)

Diplotrichus minor (Schedl, 1950) comb. nov. (Landolphianus)

=Lanurgus frontalis Schedl, 1953 syn. nov

Diplotrichus obesus (Schedl, 1953) comb. nov. (Lanurgus)

Diplotrichus pygmaeus (Schedl, 1965) comb. nov. (Lanurgus)

Diplotrichus rugosipes (Schedl, 1961) comb. nov. (Lanurgus)

Diplotrichus subdepressus (Schedl, 1965) comb. nov. (Lanurgus)

Diplotrichus widdringtoniae (Schedl, 1962) comb. nov. (Lanurgus)

=Glostatus perplexus Schedl, 1982 (syn. Beaver 2011)

Distribution. South Africa, Madagascar.

Remarks. The bifid condition of ventral setae is unique among Afrotropical micracidines and is the most reliable character matching the molecular data. Some species related to D. ignotus has an extended flange on the elytral apex, but they all have bifid setae and are therefore not Pseudolanurgus. The South African clade of species consist of the type species and one more known species, and several very similar undescribed species, suggesting a recent recolonization of the mainland (see Fig. 46).

Fig. 45
figure 7

Bayesian tree resulting from the combined analysis of four molecular markers and 36 morphological characters. Standard deviation of split frequencies 0.005, potential scale reduction factor 0.002. The tree topology was identical when morphological data were excluded. Posterior probabilities are shown above branches, and below if probability differed in the separate molecular analysis

Full size image
Fig. 46
figure 8

Reconstruction of ancestral areas for Micracidini using Bayesian binary MCMC optimisations (similar to DIVAlike+j). Pie diagrams on nodes indicate the relative probability for of each area. Larger circles are all inferred simultaneous dispersal and vicariance events. In the DIVA-like analysis without the extra jump dispersal parameter, each dispersal event occurred one node earlier than the vicariant event

Full size image
Figs. 47–55
figure 9

Dorsal, lateral and front view of the female holotypes of 47, 50, 53 Leiomicracis aurea; 48, 51, 54 Neomicracis squamigera; 49, 52, 55 female(?) holotype of Dendrochilus strombosiopsis

Full size image
Figs. 56–64
figure 10

Dorsal, lateral and front view of the female holotype of 56, 59, 62 Afromicracis kenyaensis; 57, 60, 63 Afromicracis jasminiae; 58, 61, 64 Microlanurgus bicolor

Full size image
Figs. 65–73
figure 11

Dorsal, lateral and front view of the female holotype of 65, 68, 71 Microlanurgus ater; 66, 69, 72 female paratype of Traglostus exornatus; 67, 70, 73 male holotype of Traglostus exornatus

Full size image
Figs. 74–82
figure 12

Dorsal, lateral and front view of females directly compared to the holotype of 74, 77, 80 Lanurgus barbatus; 75, 78, 81 Diplotrichus gracilis; 76, 79, 82 Pseudolanurgus harunganae

Full size image
Figs. 83–91
figure 13

Dorsal, lateral and front view of 83, 86, 89 female holotype of Pseudomicracis madagascariensis; 84, 87, 90 female specimen (compared to the female holotype) of Pseudomicracis dispar; 85, 88, 91 male specimen (compared to allotype) of Pseudomicracis dispar

Full size image
Figs. 92–95
figure 14

Dorsal, lateral and front view of specimens compared to the holotype of Phloeocurus africanus; 92–94 female; 95 male frons

Full size image
Figs. 96–104
figure 15

Dorsal, lateral and front view of the holotype of 96, 99, 102 Acanthotomicus intermedius; 97, 100, 103 Eidophelus agnathus; 98, 101, 104 Karlsenius attenuatum

Full size image
Figs. 105–113
figure 16

Dorsal, lateral and front view of the holotype of 105, 108, 111 Karlsenius klainedoxae. 106, 109, 112 Karlsenius nigrinum; 107, 110, 113 Karlsenius nitidum

Full size image

Pseudolanurgus Jordal, gen. nov.

urn:lsid:zoobank.org:act:204E9AE5-A4BF-4995-8367-F97EF1CDB9D5.

Type species. Micracis harunganae Schedl, 1961.

(Figs. 76, 79, 82)

Etymology. From Greek, Pseudo, meaning false, referring to the superficial similarity to the genus Lanurgus.

Diagnosis. Antennal scapus short, inflated, in females with a small tuft of setae; club with two moderately procurved and slightly bisinuate sutures, funiculus 6-segmented. Interstrial setae scale-like, in single rows; elytral apex extended, but not sharply mucronate; ventrite 5 truncated, setae on ventrites simple. Male genitalia with long apophyses, an exceedingly long and enlarged flagellum; tegmen and spiculum gastrale absent.

Included species:

Pseudolanurgus bugekeae (Schedl, 1957) comb. nov. (Hylocurus).

Pseudolanurgus harunganae (Schedl, 1961) comb. nov. (Pseudomicracis).

=Lanurgus cribrellus Schedl, 1965 syn. nov.

Pseudolanurgus minutissimus (Schedl, 1961) comb. nov. (Lanurgus)

Distribution. Afrotropical (Tanzania, Madagascar).

Remarks. This genus can be distinguished from the closely related Pseudomicracis Eggers, 1920 by the less procurved and slightly bisinuate sutures on the antennal club and by the slightly extended, but not mucronate, apex of the elytra. It is further distinguished from Diplotrichus by the simple hair-like setae on ventrites, and from Lanurgus by the very different male genitalia. The shared distribution between Madagascar and Tanzania is known for several scolytine genera where more recent colonisations of the mainland have taken place (e.g. Johnson et al. 2020; Jordal 2013).

Pseudomicracis Eggers, 1920

(Figs. 83, 86, 89).

=Saurotocis Wood, 1984 syn. nov.

(Figs. 84, 85, 87, 88, 90, 91)

Type species. Pseudomicracis elsae, Eggers, 1920.

Diagnosis. Antennal scapus short, inflated, in females with a long tuft of golden setae; funiculus 6-segmented, club usually with two strongly procurved sutures, suture along first segment reaching basal and not lateral margin, corneous part usually longer than broad. Apex of elytra strongly mucronate; ventrite 5 truncated or impressed. Setae on ventral sclerites never split.

Included species:

Pseudomicracis camerunus (Hagedorn, 1909) (Araptus)

Pseudomicracis difficilis (Schedl, 1965) (Micracis)

Pseudomicracis dispar (Schedl, 1961) comb. nov. (Saurotocis)

Pseudomicracis elsae Eggers, 1920

Pseudomicracis madagascariensis (Schedl, 1961) (Micracis)

Pseudomicracis pennatus (Schedl, 1965) (Micracis)

Pseudomicracis tomicoides (Schedl, 1961) comb. nov. (Saurotocis)

Excluded species: Pseudomicracis harunganae (Schedl, 1961) and Pseudomicracis bugekeae (Schedl, 1957) (Hylocurus) to Pseudolanurgus; Pseudomicracis ignotus (Schedl, 1965) to Diplotrichus.

Distribution. Afrotropical, mainly Madagascar.

Remarks. This genus is distinguished from all Afrotropical genera, except Phloeocurus, by the strongly mucronate elytral apex, and from the latter genus and the mucronate American genera by the round and twisted lateral margin of the protibiae which has only few apical teeth. Some undescribed species of Diplotrichus have an extended (but not mucronate) elytral apex, and a narrow and elongated corneous first segment on the antennal club, but these species have split setae on the ventral sclerites as in all other Diplotrichus.

The type specimens of the type species P. elsae were lost during the world war attack on Hamburg and no other specimens are known from collections. The description states that the apex of the elytra is prolonged and mucronate. The antennal club of the type species has possibly only weakly procurved sutures and may reach the lateral instead of basal margin. The type of P. camerunus is also lost. The description of P. camerunus is in Latin and the diagnostic characters are ambiguous, except body length is 3.5 mm. it will likely remain as a ghost taxon. With the possibility, albeit less likely, that African Pseudomicracis might be Pseudolanurgus, we seek a solution requiring the fewest possible taxonomic changes and thereby nomenclatural stability.

Both molecular and morphological data supported the synonymy of Saurotocis (Fig. 45). The two Saurotocis species transferred here, and Pseudomicracis pennatus, are generally similar in that males have rather deviating shapes of the elytra with excessive tubercles and confused, broad interstrial and strial setae (Fig. 85). The females of these three species (Fig. 84) are more similar to other species of Pseudomicracis in this respect.

Phloeocurus Wood, 1984

Type species. Hylocurus africanus Schedl, 1957. Monotypic.

(Figs. 92–95)

Diagnosis. Antennal scapus short, inflated, with a dense tuft of setae; funiculus 6-segmented, club with two weakly procurved sutures at middle and near apex. Frons convex in both sexes. Body elongated, elytral apex mucronate, surface of elytra asperate with short spines or sharp granules on declivity, interstrial setae scale-like, in rows, intermixed with confused smaller setae. Protibiae broad, squared and flattened, with transverse row of denticles at apex; inner mucro much enlarged, curved posteriorly.

Distribution. Ethiopia to South Africa.

Remarks. This monotypic genus is unique among the Afrotropical members of Micracidini by having a broadly flattened, squared protibiae with apical, transversely set denticles, and a mucronate elytral apex similar to Micracis. Molecular data was not available but based on the strict Neotropical-Afrotropical split between genera, irrespective of morphological similarities, it is hypothesised that the similarity to Neotropical micracidines is not reflecting a close evolutionary relationship.

Key to the Afrotropical genera of Micracidini

Includes species which have protibiae with one or more apical denticles placed on the posterior face of the protibiae, rough asperities on most of the anterior half of the pronotum, entire eyes, and a small to large tuft of setae on the female scapus.

  1. 1.

    Scapus much longer than broad, as long or longer than funiculus; funiculus 5-segmented. .. .. .. .. .. .. .. .. .. .. .. .. . 2

  • Scapus shorter than funiculus, inflated, or extended into dorsal horns, funiculus 6-segmented. .. .. .. .. .. .. .. .. .. .. ..4

  1. 2.

    Anterior sixth and lateral thirds of pronotum smooth, asperate between middle and anterior sixth; female scapus with a long triangular tuft of golden setae (Tanzania). .. .. .. .. .. .. .. .. . .. ..Leiomicracis

  • Anterior half of pronotum with distinct asperities reaching anterior margin and near lateral margins; female scapus with scant, long setae. .. .. .. .. .. .. .. ... .. .. .. .. .. .. ..3

  1. 3.

    Antennal club elongated, glabrous and shiny on both sides, apex of ventrite 5 bottleneck-shaped (Tanzania and Congo). .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .Dendrochilus

  • Antennal club as long as broad, on anterior face with one procurved suture marked by setae near anterior margin, ventrite 5 broadly rounded (Africa). .. .. .. .. .Afromicracis

  1. 4.

    Protibiae very broad, parallel-sided, with a transverse row of apical denticles on its posterior side (Kenya). .. .. .. .. ... .. .. .. .. .. .. .. .. .. .Phloeocurus

  • Protibiae not parallel-sided, twisted, with 1 or 2 lateral denticles and 1–5 denticles on the posterior side near apex. ... .. 5

  1. 5.

    Pronotum in both sexes with broad scale-like setae, in males ventrites IV and V with transverse row of caudally pointed broad setae; antennal club with suture 1 absent on median three quarter, segment 1 and 2 fused, corneous and shiny (Tanzania). .. .. .. .. .. .. .. .. .. .. .. .. .. Neomicracis

  • Pronotum with hair-like or bristle-like setae; antennal club with two complete sutures (or rarely not visible), at least segment 2 and 3 pubescent. .. .. .. .. .. .. .. .. ..6

  1. 6.

    Elytra with erect spatulate setae on odd-numbered interstriae only; pronotum strongly dome-shaped, 10–20 large asperities on anterior half, very few near anterior margin; tiny species, 0.6–0.8 mm; aedeagal tegmen a simple ring (Madagascar). .. .. .... .. .. .. .. .. Microlanurgus

  • Elytra with erect setae on all interstriae, asperities on pronotum more numerous, smaller, evenly distributed; body size more variable; tegmen with median struts. .. .. .. .. .. .. ...7

  1. 7.

    Frons in both sexes concave; each mandible in females with a long hornlike structure pointed dorsally; antennal club pubescent, without sutures. .. .. .. .. .. .. .. ...Traglostus

  • Frons in at least the male, or in both sexes, convex; mandibles without horns; antennal club with two complete sutures (exceptions in some Pseudomicracis). .. .. .. .. .. ..8

  1. 8.

    Elytra with confused rows of bristle-like setae on at least some interstriae; ventral setae simple; male genitalia with complete tegmen and median struts, spiculum gastrale, and apophyses only slightly longer than aedeagal body (mainly South Africa). .. .. ... Lanurgus

  • Elytra usually with single rows of spatulate or bristle-like setae (if confused then ventral setae bifid); ventral setae simple or bifid; male genitalia with enlarged and very long flagellum, apophyses very long, spiculum gastrale and tegmen missing (‘flagellum group’). .. .. ...9

  1. 9.

    Ventrites I, II and metaventrite with bifid setae; elytral declivity gently rounded, rarely slightly mucronate, last ventrite apically rounded (Madagascar and South Africa). .. .. .. ... .. .. .. .. ...Diplotrichus

  • Ventral setae simple; elytral apex slightly extended or inflated to strongly mucronate; last ventrite apically slightly to strongly truncated. .. .. .. .. .. .. ..10

  1. 10.

    Elytral apex strongly mucronate; antennal club with first suture usually strongly procurved, beginning at basal margin (Madagascar and Tanzania). .. .. Pseudomicracis

  • Elytral apex slightly extended and inflated, sutures on antennal club procurved, originating at lateral margins, slightly bisinuate (Madagascar and Tanzania). .. .. .. .. .. .. .. .. .. .. .. ..Pseudolanurgus

Taxa removed from Micracidini

Several taxa are not members of Micracidini and transferred to their appropriate genus.

IPINI

Acanthotomicus intermedius (Schedl, 1977) comb. nov. (Dendrochilus)

(Figs. 96, 99, 102)

The type of this species is partly embedded in glue. The frons is visible and aciculate as in many Afrotropical Acanthotomicus, and the eyes are large and broadly emarginate as in most Ipini. Together with the characteristic humpbacked pronotum typical for Acanthotomicus, this species is transferred to that genus.

DRYOCOETINI

Xylocleptes villiersi (Lepesme, 1942) comb. nov. (Thamnurgus, via Dendrochilus)

For some unjustified reason, this species was transferred to Dendrochilus by Schedl (1957). It is a typical species of Xylocleptes which seems largely mixed with Thamnurgus Eichhoff and Cyrtogenius Strohmeyer as currently classified. A revision of Afrotropical Dryocoetini is therefore much needed.

ERNOPORINI

Eidophelus agnathus (Schedl, 1942) comb. nov. (Miocryphalus)

(Figs. 97, 100, 103)

Eidophelus ciliatipennis (Schedl, 1979) comb. nov. (Miocryphalus)

Species in Eidopheles were recently revised by Johnson et al. (2020). Two species which currently resides in Afromicracis are Eidophelus as recognized by the flattened protibiae with lateral denticles, entire eyes, and antennal clubs without sutures. These species are not Afrotropical (E. agnathus, Australia; E. ciliatipennis, Fiji), which further exclude relationship to the Afrotropical genus Afrocosmoderes.

TRYPOPHLOEINI

Karlsenius Jordal, gen. nov.

urn:lsid:zoobank.org:act:

Type species. Miocryphalus nigrinus Schedl 1957

Etymology. Named after Karl Schedl—a senior Austrian taxonomist—in gratitude for all the mess he made in bark beetle taxonomy which made younger apprentices busy for years in revising dodgy classifications. The Latin suffix -senius is a neuter nominal adjective, meaning senior.

Diagnosis. Frons sexually dimorphic in at least one species, cuticle on vertex and behind eyes finely strigose. Eyes weakly to distinctly emarginated. Antennal scapus elongated with scattered setae, funiculus 4-segmented, club flat and pilose without sutures. Pronotum declivitous and strongly asperate on anterior half, anterior margin elevated, with 4–6 recurved small teeth; posterior half of pronotum shiny, finely punctured, basal margin with a fine, sharp rim which continues laterally; vestiture consisting of fine hair-like setae. Scutellum triangular, flush with elytra. Elytra with striae not impressed, punctures minute. Interstriae smooth, shiny. Vestiture consisting of single rows of rectangular or spatulate interstrial setae and fine hair-like strial setae. Protibiae variable, narrow to broad, with 4–7 lateral socketed teeth. Mesocoxae moderately separated by little more than the width of scapus.

Included species

Karlsenius attenuatum (Eggers, 1935) comb. nov. (Stephanoderes, via Miocryphalus)

(Figs. 98, 101, 104)

Karlsenius ghanaensis (Schedl, 1977) comb. nov. (Cryphalomorphus, via Eidophelus)

Karlsenius klainedoxae (Schedl, 1957) comb. nov. (Miocryphalus)

(Figs. 105, 108, 111)

Karlsenius nigrinum (Schedl 1957) comb. nov. (Miocryphalus)

(Figs. 106, 109, 112)

Karlsenius nitidum (Schedl, 1965) comb. nov. (Eidophelus, via Miocryphalus)

(Figs. 107, 110, 113)

Distribution

Tropical Africa, Madagascar

New record. Karlsenius nigrinum: Madagascar, Sakalava beach, MA-01-04B-05, Malaise trap, B. Fischer, leg, 2001. New country.

Remarks

Species in this genus have nearly all the salient characters of Afrocosmoderes, except for the fine hair-like setae (vs. scale-like) on a flat, shiny and finely punctate pronotal disc (vs. gradually rounded, granulated), and which becomes abruptly declivitous on anterior third (vs. gradually curved). In these aspects the new genus relates to Hypothenemus but is distinguished by the very different antennal club. One may note, however, that at least one species of Hypothenemus is known to have a rounded club without sutures (Johnson et al. 2016). It seems problematic to place all Karlsenius species in Hypothenemus, particularly because K. nigrinum is dimorphic in the frons, which indicate normal outbreeding. As Afrocosmoderes and Hypothenemus are sister genera (Johnson et al. 2020; Pistone et al. 2018), it seems equally possible that Karlsenius is the closest relative to one or the other.

Karlsenius nitidum has no antennae available on the holotype but otherwise exhibits the general body shape of Hypothenemus and Karlsenius. Because this species is near identical to K. klainedoxae, it is tentatively placed in the same genus. Even though this paper relates to species previously placed in Micracidini, it is worth noting that Eidophelus ghanaensis (Schedl, 1977) (see Johnson et al. 2020) is fairly similar to the type species of Karlsenius and therefore transferred to this genus.

Note added in press: a 28S partial sequence place Karlsenius klainedoxae in Trypophloeini, but clearly separate from both Hypothenemus and Afrocosmoderes.

Discussion

Previous phylogenetic studies have indicated a very early origin of Micracidini which is likely the oldest tribe within the subfamily, dating back to the Cretaceous (Jordal and Cognato 2012; Jordal et al. 2008; Pistone et al. 2018). The time estimates provided here may be slightly exaggerated, as some studies have indicated a younger age for the subfamily, around 90–70 Ma (McKenna et al. 2009; Shin et al. 2017). However, irrespective of the root age, Micracidini will not be as young as tertiary in age in any of the proposed scenarios because the tribe is almost as old as the subfamily. Earlier mistakes in the classification can therefore be explained in part by slow morphological evolution, confounded by an ancient origin and subsequent radiations within the tribe. One may note that genera such as Afromicracis, Lanurgus and Diplotrichus could need further splitting into multiple genera, but no consistent morphological differences have yet been found to justify erection of further genera.

Many new morphological characters were discovered and were helpful in diagnosing genera and groups thereof. It is quite unusual that the shape of setae is diagnostic for a genus, as documented by the split setae on ventrites in the new genus Diplotrichus. This feature is nevertheless the most consistent and is the only reliable external character distinguishing Diplotrichus from Lanurgus and Pseudolanurgus. One may also note that parts of the antennae are very informative. Not necessarily the presence or absence of an elongated horn on the scapus, but rather the length of the scapus compared to the funiculus. Furthermore, the number of funicular segments accurately diagnose clades in Micracidini; we know this feature may vary within some genera (Jordal 1998; Jordal et al. 2004), or even within a species (Johnson et al. 2020). Similarly, the sutures in the antennal club also reflect evolution quite accurately, while past taxonomic work on this tribe never took this information into account.

Despite the recognition of many new diagnostic characters, Afrotropical micracidines are still a challenge to identify correctly. In view of their considerable old age, the limited differentiation of external morphological characters is puzzling, although by no means exceptional among beetles, or other animals. The limited variation in external characters has therefore motivated investigation of internal morphological characters, which has previously proven a rich source of taxonomically informative characters in Scolytinae (Johnson et al. 2020; Jordal 2009; Jordal and Kaidel 2017). Male genitalia differ substantially between externally similar species in the genera Lanurgus and Diplotrichus and reflect more accurately than external characters the ancient split between these two lineages. While aedeagal apophyses are of normal length in Lanurgus and Microlanurgus, and the tegmen and spiculum gastrale are present, the apophyses are extremely elongated and the tegmen and spiculum gastrale absent or not recognizable in the ‘flagellum’ group where Diplotrichus belong. Another internal structure, the proventriculus, is easily characterized for the tribe, but Lanurgus differ from most other genera by the narrow suture and enforced lateral to apical teeth in the anterior plate. The two genera Afromicracis and Dendrochilus have multiple crop spines attached on a tubercle base as another example of easily recognized shared characters. They also have many species with a coiled, thin flagellum in the male genitalia, and together with Neomicracis have only a single seta on the hindwing precosta. All these examples demonstrate the great utility of internal anatomical structures in assessing relationships and placing new taxa in the correct genus, or group of genera. In the molecular era, the application of morphological character is therefore by no means in vain. Rather the contrary, taxonomically informative characters are immediately apparent once a solid phylogeny is established based on broadly sampled molecular and morphological data. This is of great value when molecular data are not possible to obtain.

The Malagasy–African faunal interchange

Madagascar and the Zambesian region of the African mainland stand out as the historically most important drivers of Afrotropical diversity in Micracidini. The Zambesian region is the most likely ancestral area for the tribe, given the endemic distribution of several basal taxa. Subsequent radiations, for instance of the largely Zambesian dominated Afromicracis clade into the Congolian and Southern African regions, further support this view. These are the most widely distributed micracidine genera, but nevertheless restricted to the African continent.

Two particularly important range expansions out of the Zambesian region resulted in markedly increased diversification. The first significant event involved the colonization of the Neotropics (66.2 Ma), which occurred just after the origin of the tribe near the Cretaceous-Tertiary boundary. The diversity of Neotropical (and Nearctic) micracidines is now the largest in the tribe (Wood and Bright 1992). The second significant event was the colonization of Madagascar, which possibly only happened once, in the Palaeocene (59.6 Ma). Both Madagascar and the South American continent had then already been separated from Africa by oceans for a long time and therefore the ancestors of these two lineages must have dispersed over open waters. Broad oceanic barriers have therefore been significant restrictions to further dispersal in Micracidini. The same pattern was found in the wood-boring weevil subfamily Platypodinae which dispersed from Africa into the Neotropics only three times during the Palaeocene to Oligocene, and to Madagascar only once, in the Eocene (Jordal 2015).

Madagascar is particularly diverse in micracidines, with several clades endemic to this island. In addition to the species poor Microlanurgus, the island is likely the ancestral source of Lanurgus, Diplotrichus, Pseudolanurgus and Pseudomicracis. African origin of Malagasy lineages is the most common pattern inferred across plants and animals (Yoder and Nowak 2006) and these data fit nicely with the general biogeographical pattern (see also Jordal 2013). It is also commonly inferred that colonization of Madagascar is a rare event, and that dispersal towards the African mainland occurred more frequently, especially during Miocene and later (Bukontaite et al. 2015; Jordal 2013), consistent with changes in oceanic and wind currents after the Oligocene period (Ali and Huber 2010). In at least three cases micracidines recolonized the African mainland, which has further contributed to increased diversity of the tribe. Only the oldest of these events (Lanurgus) resulted in a rather broad south- to eastern African distribution, with the other two much more restricted. In all its simplicity, it may be taken as yet another evidence for limited dispersal capacity in micracidine beetles.