A new apterous rove beetle genus (Coleoptera: Staphylinidae) from the Northern Andes with an assessment of its phylogenetic position

A remarkable new apterous genus of Xanthopygina beetles is described here as Ikaros gen. nov. The new genus includes three new species, I. apteros gen. et sp. nov. from Colombia, I. paramo gen. et sp. nov. from Colombia and I. polygonos gen. et sp. nov. from Venezuela. Phylogenetic analyses using molecular and morphological data were performed to assess the phylogenetic position of Ikaros gen. nov. and whether the three new taxa formed a monophyletic group. All analyses, including those with aptery-associated characters removed, strongly supported the monophyly of Ikaros gen. nov. The genus could not be confidently resolved as a member of any of the existing genus-group lineages, likely due to a lack of morphological signal in the backbone of the tree. Further analyses, ideally with molecular data, are needed to determine the position of Ikaros gen. nov.

To our surprise, we recently discovered three Xanthopygina taxa in museum collections that look nothing like a typical xanthopygine: they are dull brown, have short elytra and completely lack hind wings. To our knowledge, these taxa are the only known apterous Xanthopygina rove beetles. The majority of rove beetles are peculiar among Coleoptera Linnaeus, 1758 in having elytra that do not cover their abdomen. Despite the reduced elytra, most rove beetles have fully functional hind wings, asymmetrically folded (Saito et al. 2014) underneath their elytra. While apterous rove beetles are not necessarily rare (e.g., Peng et al. 2013;Puthz 2015;Schomann & Solodovnikov 2016;Clarke 2018;Bordoni 2020), there are few published reports of apterous Staphylinini (sec Żyła & Solodovnikov 2020) rove beetles (Smetana 1995) but flightless species have evolved in all subtribes of Staphylinini (Brunke pers. obs.).
To test whether these three taxa form a monophyletic group and to decipher their phylogenetic position in Xanthopygina, we included these taxa into the data matrix of Chatzimanolis & Brunke (2019) and added three morphological characters. The results of these new phylogenetic analyses are presented in this paper along with the formal descriptions of a new genus and three new species.

Specimens
Specimens were examined using an Olympus ZX10 stereo microscope and photographed using a Canon 40D camera equipped with a MP-E 65 mm macro lens on a Cognisys StackShot 3X macro rail and controller (https://cognisys-inc.com/stackshot3x-rail-pkg.html). Images were automontaged using Helicon Focus Pro ver. 7.5.6 (http://www.heliconsoft.com/heliconsoft-products/helicon-focus/). We took the following measurements: EL = elytra length, measured in lateral view from the anterolateral angle of the elytra to the apex of the elytra FL = forebody length, a surrogate of total body length, measured by adding HL + PL + EL HL = head length, at middle, from the anterior margin of frons to the nuchal ridge HW = head width, the greatest width, including the eyes PL = pronotum length, at middle PW = pronotum width, greatest width We wrote the generic description based on the morphological data matrix presented in this paper. Type labels are included in quotes and separated by a slash '/'. Explanatory text that was not on the original label is included within square brackets [ ]. We produced the map using the online program SimpleMappr (Shorthouse 2010). In this paper, we used the phylogenetic species concept of Wheeler & Platnick (2000)
The taxon sample was that of Chatzimanolis & Brunke (2019), with the addition of the three new taxa described in this paper. No molecular data were available for these new taxa. Phylogenetic analyses were conducted in MrBayes ver. 3.2.6 (Ronquist et al., 2012) (Bayesian) and IQ-TREE ver. 1.6 (Nguyen et al. 2015) (maximum likelihood, ML) with settings exactly as in Chatzimanolis & Brunke (2019). Analyses in MrBayes were run on the CIPRES Science Gateway (phylo.org). Analyses were conducted on the total evidence dataset (4891 characters) of morphological (93 characters) and molecular data (4797 bp), which was initially partitioned into morphology and gene by codon position. Optimal partitioning scheme and model selection were determined using PartitionFinder2 (Lanfear et al. 2012), running on CIPRES with the settings used in Chatzimanolis & Brunke (2019). Clade support for Bayesian and ML analyses was assessed using 1000 replicates of the ultrafast bootstrap (UFB) (Hoang et al. 2018) and an SH-aLRT test with 1000 replicates (Guindon et al., 2010). Nodes with Bayesian posterior probability (PP > 0.90) were considered well supported, nodes with PP = 0.80-0.89 were considered to be weakly supported, and nodes with PP < 0.80 were considered unsupported. In the ML analysis, nodes with support values of both UFB ≥ 0.95 and SH-aLRT ≥ 80 were considered well supported, nodes with one of UFB < 95 or SH-aLRT < 80 were weakly supported, and nodes with both UFB < 95 and SH-aLRT < 80 were unsupported. Convergence of Bayesian analyses was assessed using Tracer ver. 1.6 (Rambaut et al. 2014) and by examining Potential Scale Reduction Factor (PSRF) and Average Standard Deviation of Split Frequency (ASDSF) values in the MrBayes output.
To test whether the three newly added characters were influencing the results of our analyses (perhaps due to convergence of morphological structures), a separate Bayesian and ML analysis (BA-mod, MLmod) was run without these three new characters. The morphological dataset, as well as the concatenated morphological and molecular datasets, are provided as supplemental materials (Supp. file 1, Supp. file 2).

Phylogenetic analyses
For the molecular data, the partitioning scheme and corresponding models selected by PartitionFinder were exactly the same as Chatzimanolis & Brunke ( . The overall tree topologies were largely congruent with that of Chatzimanolis & Brunke (2019) and will not be discussed here.
While the monophyly and composition of Ikaros gen. nov. was consistent across analyses, the sister group of Ikaros gen. nov. differed between the Bayesian and ML analyses and could not be resolved with confidence (Fig. 1). The exclusion of morphological characters 92-94 did not impact these

Diagnosis
Ikaros gen. nov. belongs in Xanthopygina based on the following two synapomorphies present: inferior marginal line of the pronotal hypomeron continues as a separate entity beyond anterior pronotal angles and curves around them, and hypostomal cavity moderately delimited (i.e., cavity surface without microsculpture or punctation different from the rest of nearby head surface). The genus can easily be distinguished from all other Xanthopygina based on the reduced elytra (exposing the 2 nd tergum), absence of hind wings and the shape of the abdomen: constricted anteriorly and expanded posteriorly.

Etymology
The word Ikaros is an alternate spelling of the word Icarus, the son of Daedalus who (in the Greek mythology) constructed the Labyrinth. Icarus and Daedalus escaped the Labyrinth by flying with wings made of feathers and glued by wax. However, Icarus flew too close to the sun and the wax melted. The name is rather appropriate for this genus considering these are species found in high altitudes that have lost their wings. The name is masculine.

Etymology
The specific epithet is derived from the Greek word 'άπτερος' ('without wings') and refers to the lack of hind wings. It is treated here as a noun in apposition.

Type material
Holotype (

Description
Forebody length 7.1 mm long. Coloration reddish brown with head and mouthparts slightly darker brown. Head transverse, HW/HL ratio = 1.18. Epicranium mostly impunctate, with few large punctures posteriorly, medially and around margin of head and eye; with faint polygon-shaped microsculpture. Labial palpus with palpomere 3 widest apically, subparallel-sided. Antennomeres with crown-like macrosetae shorter than length of antennomere. Pronotum longer than wide, PW/PL ratio = 0.91; surface of pronotum impunctate except two punctures on each side of median area and few punctures around margin; with faint polygon-shaped microsculpture. Elytra shorter than pronotum, EL/PL ratio = 0.75. Elytra with large, deep contiguous punctures and dense polygon-shaped microsculpture. Abdominal terga 3-5 without arch-like carina. Male secondary sexual structures with shallow but broad emargination on sternum 7; with deep, broad V-shaped emargination on sternum 8; borders of emargination on sternum 7 and 8 appearing 'shaved' (with no setae). Aedeagus as in Fig. 3; in dorsal view paramere longer than median lobe; converging to elongate, narrow tip; apex of paramere with small emargination; in lateral view paramere concave, becoming narrower. Median lobe in dorsal view narrowing to rounded apex; in lateral view median lobe becoming narrower near flattened apex, with no subapical tooth.

Distribution
Known from the National Park Chingaza in the department of Cundinamarca, Colombia (Fig. 6). The holotype was collected in the Republic of New Granada (1831-1858), which refers to a region that included primarily Colombia and Panama, and smaller areas from the countries of Brazil, Costa Rica, Ecuador, Peru and Venezuela.

Etymology
The specific epithet is derived from the name of the ecosystem (páramo) where the specimens were collected. It is treated here as a noun in apposition.

Description
Forebody length 5.3-5.9 mm long. Coloration reddish brown with head and mouthparts slightly darker brown. Head transverse, HW/HL ratio = 1.21. Epicranium with numerous large punctures, except impunctate centre; punctures not contiguous, distance between punctures varies; with faint polygonshaped microsculpture. Labial palpus with palpomere 3 widest apically, subparallel-sided. Antennomeres with crown-like macrosetae shorter than length of antennomere. Pronotum longer than wide, PW/PL ratio = 0.88; surface of pronotum with a median impunctate area as wide as 4-5 punctures; with 2-3 rows of disorganized punctures in addition to rows flanking impunctate centre; with faint polygonshaped microsculpture. Elytra shorter than pronotum, EL/PL ratio = 0.79. Elytra with large, deep punctures, distance between punctures equals to width of 0.5-1 punctures; elytra with faint polygonshaped microsculpture. Abdominal terga 3-5 without arch-like carina. Male secondary sexual structures with shallow emargination on sternum 7; with shallow, narrow U-shaped emargination on sternum 8; borders of emargination on sternum 7 and 8 appearing 'shaved' (with no setae), but less so than in I. apteros gen. et sp. nov. Aedeagus as in Fig. 4; in dorsal view paramere subequal to median lobe; paramere broad, converging to rounded tip; in lateral view paramere more or less straight. Median lobe in dorsal view broad, narrowing to rounded apex; in lateral view median lobe becoming narrower near apex, with small subapical tooth.

Distribution
This species is only known from Iguaque Fauna and Flora Sanctuary (SFF Iguaque) in the Boyacá Department of Colombia (Fig. 6).

Remarks
The holotype is currently in the collection of SEMC but due to the collecting permit requirements (Z. Falin pers. com.), it will be transferred to the IAVH collection in the near future.

Etymology
The specific epithet is derived from the Greek word 'πολύγωνος' ('polygon') and refers to the stark polygon-shaped microsculpture of the head, thorax and elytra. It is treated here as a noun in apposition.

Description
Forebody length 5.2 mm long. Coloration of body reddish brown, with head and pronotum having undertones of metallic green-brown. Head transverse, HW/HL ratio = 1.16. Epicranium with numerous large punctures, except impunctate centre; punctures not contiguous, distance between punctures typically width of 1-2 punctures; with stark polygon-shaped microsculpture. Labial palpus with palpomere 3 widest before apex. Antennomeres with crown-like macrosetae at least twice as long as antennomeres. Pronotum longer than wide, PW/PL ratio = 0.9; surface of pronotum with a median impunctate area as wide as 2-3 punctures; with 4-5 rows of punctures in addition to rows flanking impunctate centre; with stark polygon-shaped microsculpture. Elytra shorter than pronotum, EL/PL ratio = 0.8. Elytra with large, deep contiguous punctures and dense polygon-shaped microsculpture. Abdominal terga 3-5 with archlike carina. Male secondary sexual structures with shallow emargination on sternum 7; with shadow, small U-shaped emargination on sternum 8; borders of emargination on sternum 7 and 8 appearing 'shaved' (with no setae), but less so than in I. apteros gen. et sp. nov. Aedeagus as in Fig. 5; in dorsal view paramere slightly longer than median lobe; paramere broad, converging to broad, rounded tip; in lateral view paramere narrower apically. Median lobe in dorsal view becoming narrow to small apex; in lateral view median lobe concave, becoming narrower near curved apex, without subapical tooth.

Distribution
Known only from the type locality in Sierra Nevada National Park in Venezuela (Fig. 6).

Remarks
The holotype is currently in the collection of SEMC but due to the collecting permit requirements (R. Anderson pers. com.), it will be transferred to the MIZA collection in the near future.
Key to the species of Ikaros gen. nov.

Discussion
Until recently, there was no comprehensive phylogeny of the subtribe, but Chatzimanolis & Brunke (2019) using molecular and morphological data placed all existing genera of the subtribe into a phylogenetic context and established eight groups of genera: the Plociopterus, Isanopus, Xanthopygus, Smilax, Elmas, Gastrisus, Ocyolinus and Trigonopselaphus groups. The phylogeny presented in this paper is largely congruent with that of Chatzimanolis & Brunke (2019). The major difference is in the composition of the Gastrisus group of genera. In the Bayesian analyses, the addition of the genus Ikaros gen. nov. to the dataset splits the Gastrisus group into two subgroups (one composed of taxa in the genus Gastrisus Sharp, 1876 and the other composed of species in the genera Nausicotus Sharp, 1884 and Torobus) and renders them in a polytomy along with the Elmas, Ocyolinus and Trigonopselaphus groups and Ikaros gen. nov. It is possible that additional molecular or morphological data may unite the taxa of the Gastrisus group with the species of Ikaros gen. nov. as a clade, although the exact placement of Ikaros gen. nov. among the various groups cannot be conclusively determined given our data.
The three new species described here formed a clade regardless of the analyses method (Bayesian or maximum likelihood) or whether the three new morphological characters were included or excluded. Initially, we had hypothesized that the three species presented here may not be closely related, and their similarities may be due to convergence associated with aptery. One morphological feature that caused suspicion of monophyly was the presence of the arch-like carina on terga 3-5 in I. polygonos gen. et sp. nov., which is absent in the other two species. However, our analyses do not support this hypothesis and perhaps further molecular data can clarify this situation.
The most likely reason for the loss of hind wings observed in Ikaros gen. nov. is their high-altitude habitat in the Northern Andes. Loss of hind wings is common in rove beetles found in high elevations (e.g., in Atheta Thomson, 1858: Assing 2004Dasycerus Brongniart, 1800: Wheeler & McHugh 1994Lathrobium Gravenhorst, 1802: Peng et al. 2013). The switch from bright metallic coloration to dull brown is harder to explain but it might also be due to the high-altitude habitat. Although some other Xanthopygina lack bright coloration (e.g., species in Gastrisus, Oligotergus Bierig, 1937 andXanthopygus Kraatz, 1857), most Xanthopygina are presumed to be predators (Chatzimanolis 2003), and deciphering the role of coloration here will probably require direct observations of the natural history of Ikaros gen. nov.