A revision of Haematodes Laporte and Weiserianum Bernhauer (Coleoptera: Staphylinidae: Staphylininae: Xanthopygina)

The species of poorly known but charismatic genera Haematodes Laporte, 1835 and Weiserianum Bernhauer, 1927 are revised. Weiserianum syn. nov. is considered a junior synonym of Haematodes, with Haematodes kuntzeni (Scheerpeltz, 1936) comb. nov. Weiserianum woltersi Bernhauer, 1927 syn. nov. is treated as a synonym of Haematodes tenuipes Kraatz, 1858. Haematodes myteros sp. nov., is described from Paraguay and Brazil. As the type series of Haematodes bicolor Laporte, 1835 is considered lost, a neotype, selected from the original type locality is designated. We also designate a lectotype for H. tenuipes Kraatz, 1858 to stabilize nomenclature for this species, which is similar to H. myteros. As far as known, Haematodes is restricted to the southern Neotropical region and may be nest parasites within Acromyrmex and Atta ant nests as are species of the related genus Scariphaeus, but no direct observations are yet available. We provide a key to the four known species of Haematodes and illustrate their diagnostic features.


INTRODUCTION
The genera Haematodes Laporte and Weiserianum Bernhauer (Staphylininae: Staphylinini: Xanthopygina) contain four obscure and similar species with remarkable morphology, especially their distinctive antennae that are reminiscent of the ant nest beetles (Carabidae: Paussinae: Paussini) (Figs. 1, 2A-2C). These beetles were so notable that Nordmann (1837) even created the family group name Platycnemini (emended by Newton (1995) from the original Platycnemidiformes) for his new genus Platycnemus (maintained here as a synonym of Haematodes). Likely based on the broadly shaped pronotum and wide neck, Haematodes was included in Kraatz's (1857) 'Quediiformes' along with other, distantly related Staphylininae, now known to belong to other subtribes (Brunke et al., 2016). Haematodes and Weiserianum were maintained in a loosely defined Quediina (e.g., Herman (2001)) until Chatzimanolis (2014a) unambiguously treated them as members of the Neotropical rove beetle subtribe Xanthopygina. Within this lineage, they have been hypothesized to be most closely related to the myrmecophile Scariphaeus Erichson (Scheerpeltz, 1936) or the suspected myrmecophile/sphecophile Darwinilus Chatzimanolis (Chatzimanolis, 2014b). Specimens of these genera are rarely collected and are scattered throughout collections worldwide. The species are limited to the southern edge of tropical South America, the 'Chacoan subregion' of Morrone (2014), an area with little recent collection activity and that has been devastated by agricultural development. Morphological features such as thickened legs and antennae, elongate maxillae, and clusters of bristle-like hairs on the pronotum are found in other Staphylinini (e.g., Smilax, Chatzimanolis, 2016), and beetles in general (Parker, 2016), known to be associated with colonies of ants. Scheerpeltz (1937) recognized the close morphological similarity of Haematodes bicolor Laporte to Weiserianum woltersi Bernhauer and his new species, W. kuntzeni, stating that the former differed mainly by the more broadly expanded and flattened hind tarsi, the shorter apical maxillary palpomere and the more compact antennae. These charismatic taxa have never been comprehensively revised and nothing substantial has been published since Scheerpeltz (1937). We here illustrate the diagnostic features of and provide a key to four valid species of this lineage, one of which was new to science.
The electronic version of this article in Portable Document Format (PDF) will represent a published work according to the International Commission on Zoological Nomenclature (ICZN), and hence the new names contained in the electronic version are effectively published under that Code from the electronic edition alone. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix http://zoobank.org/. The LSID for this publication is: E494A584-3784-45A8-8D0A-6C0E024EA5C8. The online version of this work is archived and available from the following digital repositories: PeerJ, PubMed Central and CLOCKSS.

Taxonomy
Haematodes Laporte, 1835Haematodes Laporte, 1835  Type species. Haematodes bicolor Laporte. Diagnosis. Among genera of Xanthopygina, Haematodes is easily recognized based on the distinct habitus that includes reddish to orange-red coloration and a 'fuzzy' appearance, due to the presence of long, bristle-like setae. It can be distinguished from the (presumably closely related) genera Scariphaeus Erichson and Smilax Laporte based on the combination of the following: antennae forming a long and broad but loose club of transverse segments 3-11 or 5-11 (Figs. 2A-2C); broad neck; abdominal tergites with only anterior transverse carinae, lacking posterior curved lines. Redescription. Habitus as in Figs. 1A-1C; medium large and robust rove beetles; forebody 6.20-10.00 mm long. Body mostly reddish to light orange-red, with darker areas; elytra reddish to entirely dark, abdomen bicolored with basal 2-4 visible segments darker than apical ones. In species with more than a few specimens known, reddish color was observed to vary from dark red to pale orange-red.
Dorsal forebody without microsculpture, except for lateral punctate areas of the pronotum. Head transverse, somewhat hexagonal, with variably dilated temples, disc entirely punctate or with small glabrous area in center of disc, temples with several to many long bristle-like setae. Eyes ranging from 1/3 to 1/2 the length of the head. Ventral head with shallow to strong microsculpture; infraorbital ridge well developed, extending from base of mandible to connect with nuchal ridge (if present) and/or dorsal basal ridge, post-mandibular ridge absent; gular sutures separated but only narrowly at middle; nuchal ridge present in its entirety or in lateral fragments (in specimens of H. tenuipes, H. myteros) or entirely absent (H. bicolor, H. kuntzeni); neck broad in all species. Antennae short (Figs. 2A-2C), antennomere 1 compact, with many long bristle-like setae, at least twice as wide as antennomere 2; segments 3-10 or 5-10 extremely transverse, slightly asymmetrical and glabrous at middle, forming a long club with apical antennomere; antennomeres 5-11 or 6-11 with tomentose pubescence. Labrum well developed, moderately to deeply incised, margins with extensive (H. bicolor) to moderate amounts of long bristle-like setae. Mandibles moderately long and slender, roughly symmetrical, with a single tooth, with dorso-lateral mandibular groove. Maxilla with galea long, longer than maxillary palpus ( Fig. 2D), maxillary palpus 4-segmented; P 2 -P 3 subequal, P 4 slightly narrower and longer than P 3 , P 2 with bristle-like setae. Labium entire, with small apical notch; labial palpi 3-segmented, P 3 distinctly narrower and longer than P 2 , P 2 with bristle-like setae; P 3 apically truncate.
Pronotum (Figs. 3A-3D) wider than head; margin explanate; without post-coxal process; hypomeron not visible in lateral view; inferior line not connected anteriorly to superior line and joining anterior margin of pronotum (inferior line interrupted by coxal cavity in H. bicolor); margin of pronotum, especially anterior angles, with long bristle-like setae; basisternum with multiple macrosetae; prosternum without longitudinal carina. Elytra longer than pronotum, with dense to sparse, uniform punctation. Hind wings fully developed, with MP4 and CuA separate and MP3 present. Scutellum with posterior ridge, densely punctate and setose. Basal elytral ridge present, sinuate, directed anteriad. Humeral row of setae absent. Mesocoxal cavities broadly separated by a depressed intercoxal region, mesocoxal and metacoxal processes short and obtuse to nearly truncate. Legs with 5−5−5 tarsal segmentation; femora, tibia and tarsi flattened; tibia greatly expanded and with several rows of thick spines along outer face; tibia with two long apical spurs, unequal in length; protarsi enlarged and with tenent setae in both sexes; tarsomeres with middle of disc glabrous; empodium with pair of setae.
Abdomen with paired prototergal glands present; abdominal terga with only anterior basal lines (curved and accessory basal lines absent). Abdominal sternite III with transverse basal line forming obtuse angle at middle, not sharply extended posteriad; male abdominal sternite VII with expanded porose structure, situated at midlength of sternite, bearing a dense brush of setae (Figs. 2G-2I); apex of male abdominal sternite VIII with broad, shallow emargination; male abdominal sternite IX robust, with moderately deep apical emargination. Aedeagus short and compact relative to most Xanthopygina; median lobe with one or two teeth more strongly sclerotized than surrounding tissue; paramere short and stout, always distinctly shorter than median lobe, with peg setae. Female with spermatheca not sclerotized. Distributions of Haematodes species, including those that are each others' closest relatives, at least partially overlap. This phenomenon suggests processes other than allopatric speciation acting in this genus, perhaps those related to their potential host ants. This distribution pattern also occurs in the suspected sphecophile/myrmecophile Trigonopselaphus (Chatzimanolis, 2015), which has a similar distribution in the Chacoan subregion.

Biology.
Little is known about species of Haematodes. One specimen of H. bicolor was collected in flight in a 'scrub forest'. One specimen of H. tenuipes was collected at an Agricultural Research Station and may be associated with ants that nest underground in open-habitats, such as the grass-harvesting species of Atta. Nearly all specimens are loaded with soil particles, further supporting a ground-nesting host. Finally, the putatively related genus Scariphaeus is associated with Acromyrmex ants (Scheerpeltz, 1936), has a similar distribution and also often bears soil particles. Scheerpeltz (1937) suggested that the beetles may be associated with Hymenoptera, especially wasps but the morphology of the legs and antennae suggest that ants are far more likely.
Comments. The species of Haematodes and Weiserianum form a morphological grade with regard to features that are likely related to interactions with social insects (thicker appendages, long bristle-like setae, reductions of the nuchal ridge). The type species of Weiserianum, W. woltersi Bernhauer, was discovered to be a synonym of H. tenuipes Kraatz (see below). Although it would be possible to restrict Haematodes to its distinctive type species and move H. tenuipes to Weiserianum, we prefer to treat these taxa as a single, well-defined genus, easily separated from the related, myrmecophile genus Scariphaeus by the distinctive antennae and habitus. This runs in contrast to the usual tendency of taxonomists studying myrmecophiles to erect or maintain genera representing the steps along this grade (e.g., clavigerine Pselaphinae with successively fewer antennomeres). Therefore Weiserianum syn. nov. is considered to be a synonym of Haematodes. We maintain the synonymy of Platycnemus with Haematodes since their type species each correspond to the same taxonomic concept given here for H. bicolor.

Key to species of Haematodes Laporte
1. Antennae extremely broadened into a strong club ( Fig. 2A); pronotum broadly glabrous dorsally (Fig. 3A); hind tarsi markedly robust (Fig. 2E); median lobe with one tooth in lateral view (Fig. 4B) 1B); nuchal ridge entirely absent; pronotum with a pair of broad, loosely organized punctate areas (Fig. 3B); paramere with apex truncate, peg setae in two rows, that do not fuse at middle (Fig. 4E). . . H. kuntzeni (Scheerpeltz) -Elytra and apical two abdominal segments reddish (Fig. 1C); nuchal ridge either complete or with lateral fragments, never absent; pronotum with a pair of punctate areas consisting of single, doubled or tripled rows, or disc of pronotum with only a few punctures (Figs. 3C-3D); paramere with apex emarginate or weakly trilobate, peg setae in two groups that fuse at middle (Figs. 4G, 4I). . . 3 3. Pronotum appearing subquadrate, strongly convex (Fig. 3D); apex of paramere with emargination simple (Fig. 4I); median lobe in lateral view with apex elongate (Fig. 4H). . . H. myteros Brunke and Chatzimanolis -Pronotum appearing transverse, middle portion weakly flattened (Fig. 3C); apex of paramere weakly trilobed (Fig. 4G); median lobe in lateral view with apex short and constricted (Fig. 4F)   Laporte did not indicate the number of specimens studied, nor a type series but indicated that specimens came from Buenos Aires. The type material described by Laporte (1835) is considered lost (K Walker, Museum Victoria, pers. comm., 2018) as these specimens were among those destroyed by a fire at the Smithsonian Institution in 1865 (see Evenhuis, 2012). The second Laporte collection in Museum Victoria, Australia does not contain the type but only a drawing of the type by Laporte (K Walker, Museum Victoria, pers. comm., 2018), which, although crude, matches our species concept for H. bicolor regarding the greatly expanded tibiae. Additional attempts to locate the type series in the Natural History Museum London (M Barclay, pers. comm., 2018) and Oxford University (D Mann, pers. comm., 2018) have failed. Therefore, we found it necessary to designate a neotype to fix the identity of this taxon. A male was selected from the type locality (Buenos Aires, Argentina) and deposited at the CNC.
It is obvious from the descriptions of the genus and species (antennae, body form) by Laporte (1835)  Nordmann (1837) did not specify the number of specimens that he studied but referenced specimens collected by D. Sello from southern Brazil. Two syntypes matching this description (except S. rather than D. Sello) were located in ZMHB but we refrain from designating a lectotype as both are females and the identity of this taxon is not problematic. Diagnosis. This species is easily recognized by the glabrous disc of the pronotum (Fig. 3A) and the strongly clubbed antennae (Fig. 2A). Redescription. Forebody length 8.39-10.00 mm. Body reddish orange to yellow orange, appendages, lateral margins of pronotum, scutellum and abdominal segments III-VI dark brown to black.
Male sternite VII with porose structure as in Fig. 2G, broad and thicker than in other species. Male sternite VIII with shallow emargination. Aedeagus as in Figs. 4A-4C. In ventral view, median lobe weakly converging apicad to blunt median projection. In lateral view, median lobe with single tooth and nearly flat ventral face (Fig. 4B). Paramere with apex emarginate and with two groups of peg setae, not distinctly fused together apically (Fig. 4C). Distribution. Widely distributed across the entire range of the genus (Fig. 5A), H. bicolor is the most frequently collected species. Known from southern Brazil, Paraguay, Uruguay and northern and central Argentina, from lowlands to foothills of the Andes (1,325 m) (Fig. 5A). The two historical specimens labeled 'Venesuela' in ZMHB are obviously mislabeled. Bionomics. Laporte (1835) mentioned that his specimen (s) was taken on the wing from the Buenos Aires area in scrub habitat (= ''on la prend volant sur les broussailles'').
Male sternite VII with porose structure as in Fig. 2H, wider than in other species. Male sternite VIII with shallow emargination. Aedeagus as in Figs. 4D-4E. In lateral view, median lobe with two distantly separated subapical teeth and long, relatively broad apical portion (Fig. 4D). Paramere with apex truncate and with two groups of peg setae, distinctly separated (Fig. 4E). Distribution. Known only from Goiás state in Brazil and Guaira department in Paraguay (Fig. 5A).

Bionomics.
Nothing is known about this species' biology. Kraatz, 1858 (Figs. 1C; 2F; 3C; 4F-4G; 5B; (map)) Haematodes tenuipes Kraatz, 1858: 363 Weiserianum woltersi Bernhauer, 1927 The female holotype of W. woltersi is from 'northwest Argentina, Catamarca, Corral Quemado', 2,000 m in the Andes. This is currently a long distance from the other known specimens of H. tenuipes but this species is likely under-collected and H. bicolor has a comparably broad distribution. However, there is a possibility of mislabeling and further collecting will hopefully clarify this issue. Externally, this specimen falls within the range of variability for H. tenuipes. Other material. BRAZIL: 'Brasil', 1940, P. Grossa (1, CNC) Brunke and Chatzimanolis, des. Brunke and Chatzimanolis 2018 [yellow printed label]. In the collection of ZMHB. Diagnosis. This species can be recognized by a combination of: pronotum with at least a few discal punctures (Fig. 3D); elytra reddish (as in Fig. 1C); and pronotum subquadrate (Fig. 3D). Description. Extremely similar in morphology to H. tenuipes and differing only in the following: forebody length 6.20-7.61 mm; head width/length ratio = 1.45-1.57. Pronotum width/head width ratio = 1.11-1.27, width/length ratio = 1.11-1.17, subquadrate (Fig. 3D); disc of pronotum with punctures arranged into dorsal and sublateral longitudinal rows, punctures not doubled (Fig. 3D); aedeagus as in Figs. 4H-4I; in lateral view, median lobe with two approximate subapical teeth and a relatively elongate apical portion that is not constricted near apex (Fig. 4H). Paramere with apex emarginate and with two groups of peg setae, distinctly fusing apically (Fig. 4G). Distribution. Currently known from Cordillera and Cazaapá departments in Paraguay, and Santa Catarina state in Brazil (Fig. 5B). Bionomics. The holotype was collected in a flight intercept trap in San Rafael Reserve. Etymology. The species epithet is the Greek word meaning 'pointed' and refers to the median lobe in this species, which is thinner and more acute in lateral view than that of the other Haematodes.