A new Chlerogelloides from northeastern Brazil and French Guiana, with a key to the species (Hymenoptera, Halictidae)

Abstract A third species of the rare augochlorine bee genus Chlerogelloides Engel et al. (Halictinae, Augochlorini) is described and figured. Chlerogelloides nexosa sp. n. is most similar to the type species, Chlerogelloides femoralis Engel et al., in that both have modified midlegs in the males. The former, however, can be distinguished on the basis of its terminalia, which in some respects more closely resembles those of Chlerogelloides simplex Engel and Brooks. Brief comments on the secondary features of males and relationships of the genus are provided. A key to the species of the genus is provided and floral records of Cordia nodosa Lam. (Boraginaceae) and Gonzalagunia dicocca Cham. & Schltdl. (Rubiaceae) are noted.

introduction Numerous lineages of bees within the New World tribe Augochlorini (Halictinae) are known to have characteristically elongate heads, typically involving an elongation of the malar space and clypeus, and often the supraclypeal area, and presumably adaptations for obtaining nectar from flowers with deep corollas, although few floral records are available (e.g., Engel 2000Engel , 2007Engel , 2009Engel , 2010aEngel , 2010bEngel and Brooks 2002;Engel and Gonzalez 2009). Among such bees, one of the more peculiar are the species of the genus Chlerogelloides Engel et al. (Engel et al. 1997;Engel and Brooks 1999), so named for its superficial resemblance to the more diverse and widespread genus Chlerogella Michener (Engel 2009(Engel , 2010a, and putatively related to Chlerogella and Ischnomelissa Engel (Engel 2000). Unlike nearly all of the aforementioned cases of augochlorines with noticeably elongate head, in Chlerogelloides the malar space is unmodified and short, but the clypeus remains distinctively elongate. A complete account of the genus has been provided by Engel and Brooks (1999) and Engel (2000). To hitherto known species of Chlerogelloides have been reported from Brazil (Amazonas), Colombia (Putumayo), Ecuador (Morona Santiago, Napo, Sucumbíos), French Guiana (Roura), and Peru (Loreto) for Chlerogelloides femoralis Engel et al. and Brazil (Amapá) and French Guiana for C. simplex Engel and Brooks (Engel et al. 1997;Engel and Brooks 1999;Moure 2007). As the specific epithet suggests, the type species, C. femoralis, is also noteworthy among the Augochlorini for its peculiarly modified midlegs in males, the function of which remains unknown but are presumably used in mating. Tragically, no biological information of any kind has been reported for any species of the genus.
Herein we report the discovery of a third species in the genus, specimens of which were captured while visiting flowers of Cordia nodosa Lam. (Boraginaceae: sometimes placed in its own order, Boraginales) and Gonzalagunia dicocca Cham. & Schltdl.
(Gentianales: Rubiaceae), and representing the first occurrence of the genus in Pará State, Brazil.

Material and methods
Morphological terminology follows that of Engel (2001, 2009) and Michener (2007, while the format for the description follows those of Engel et al. (1997) and Engel and Brooks (1999). The higher classification of Augochlorini is that of Engel (2000). Abbreviations used for common morphological terms are: S, metasomal sternum; T, metasomal tergum; F, flagellomere; DS, diameter of the antennal scape; and OD, ocellar diameter (based on the median ocellus). Measurements and proportions are adapted from Moure and Sakagami (1962): body length, head length and width, upper and lower interocular distances, and ocellocipital distance. Photomicrographs were prepared using a Nikon D1x digital camera attached to an Infinity K-2 long-distance microscope lens.   Description. ♂: Structure: Total body length 6.60 mm; forewing length 4.50 mm. Head elongate, length 1.80 mm, width 1.47 mm (Figs 2, 3); clypeus longer than maximum width, length 0.60 mm, width 0.47 mm, almost entirety of clypeus (85%) set below lower tangent of compound eyes; frontal line weakly carinate between antennae, becoming a faintly impressed line from there to median ocellus; antennal scape relatively short (excluding basal bulb), length 0.50 mm; pedicel about as long as F1; F1 wider than long, slightly longer than F2; F2-F5 slightly wider than long; F6-11 progressively becoming longer than previous flagellomeres; distance from median ocellus to lateral ocellus 0.05 mm; between lateral ocelli 0.20 mm; ocellocular distance 0.22 mm (1.46× ocellar diameter); compound eye about 3.25x wider than  Engel and Brooks. gena in profile (width of compound eye 0.65 mm, width of gena 0.20 mm), beginning a little below middle of compound eyes. Intertegular distance 0.95 mm; mesoscutellum less than twice as long as metanotum (mesoscutellum length 0.35 mm, metanotum length 0.20 mm); dorsal surface of propodeum faintly concave, elongate (as for the genus: vide Engel et al. 1997;Engel and Brooks 1999;Engel 2000). Mesofemur greatly enlarged, approximately twice as long as wide (measurement made on internal surface: length 1.2 mm, width 0.55 mm), with flattened inner surface (Fig. 5), ventrally ridged along external border from apex basad to ventral tubercle in apical third; mesotibia slightly twisted on internal surface (in posterior view), with two non-contiguous carinae along borders and converging medioapically, one from base of mesotibia and with a small elevation near apex, other more apical on external border, both carinae terminating at a weakly depressed medioapical area from which spur articulates; mesobasitarsus twice as long as wide, flattened on dorsal surface and with carinae bordering weakly depressed inner surface. Forewing pterostigma very long, almost as long as first submarginal cell; hind wing with basal hamuli arranged 2-1, distal hamuli arranged 2-1-2. Male terminalia as in figures 7-10.
Sculpturing: Integument smooth and polished, faintly imbricate in some places, with sparse small to minute punctures, except face above level of antennal toruli with punctures more closely spaced as well as on mesoscutum, mesoscutellum, and metanotum; punctures separated by 2-5× a puncture diameter on mesepisternum; dorsal surface of propodeum smooth, polished, and glabrous.
Coloration: Integument predominantly honey yellow with brownish areas laterally pronotum near pronotal lobe, dorsal surface of propodeum, apex of metafemur, external surface of metatibia, apical third of T1, apical two-thirds of T2-T3, and T4-T7 with yellowish areas slightly darker than elsewhere (darker areas wider and longer in paratype). Head metallic olive green except honey yellow on clypeus (sometimes with some brown areas extending from base in paratypes from French Guiana), labrum, mandible, malar area, and antennal scape; pedicel flagellum dark brown. Mesoscu- tum, mesoscutellum, and metanotum metallic olive green, less shiny than on head; axilla dark brown with metallic reflections; tegula translucent brown; axillary sclerites brown, base of C+Sc honey yellow otherwise wing venation brown to dark brown; pleura honey yellow except mostly brown on hypoepimeral area, with weak metallic green highlights (almost imperceptible in most places); wing membranes hyaline, slightly and faintly infumate apically, with some iridescence depending on lighting.
Pubescence: Pubescence largely consisting of golden simple setae. Head with scattered, largely simple setae, those on supraclypeal area, vertex, gena, and postgena longer; setae dorso-apically on scape longer, between one-third and one-half DS, remainder much shorter; a few short, branched setae in lower paraocular area and surrounding upper border of supraclypeal area; gena with uniformly distributed branched setae, setae with branches arising from one side of rachis in apical two-thirds; setae along borders with compound eyes very short. Mesosomal setae generally simple except more plumose around pronotal lobe and surrounding propodeal spiracle; disc of mesoscutum with relatively short and sparse setae, shorter than those of mesepisternum, although denser than on the latter; disc of mesoscutellum with setae little longer than those of mesoscutum, distinctly longer along posterior margin; metanotum with yellowish, short, plumose setae intermixed with   figure 18)  1.2DS; anterior area formed by carinae along ventral surface of mesobasitarsus glabrous and separate from posterior area covered by minute setae. Wing membranes uniformly pilose. Metasoma generally with sparsely-scattered, simple, long setae, on T1-T3 simple setae mostly distributed across discs, with apical margins glabrous, such setae varying from 1-1.5DS in length, setae becoming progressively longer on more apical segments, on T4-T7 with short, dense, plumose setae intermixed with long setae.
♀: Unknown. Etymology. The specific epithet is taken from the Latin word nexosus, meaning "complicated", and is a reference to the complex morphology of this and other species in the genus.
Comments. A single female, collected in French Guiana with one of the males, is indistinguishable from that of C. simplex (specimen from Patawa, deposited in Brussels). It is possible that it is a female of C. nexosa, but given that definitive C. simplex is known from the same area we cannot be certain that it is a definitive female of the new species. Only further collecting in the region will be able to determine the correct association of females for this complex genus of bees.

Key to species of Chlerogelloides
The present key is modified from that provided by Engel and Brooks (1999). Females are presently unknown for C. nexosa sp. n. Mesotrochanter with strong inner projection (Fig. 4); gena with distinctive tuft of long, plumose setae; face above level of antennae with minute punctures widely scattered over glabrous integument; apical two-thirds of clypeus yellow (Fig. 14);

Discussion
The three known species of Chlerogelloides are all quite similar, immediately noticeable for their modified, elongate clypeus which is basally intruded upon by a deeply acute and projecting epistomal sulcus, such that the narrowed epistomal lobe nearly reaches the clypeal apex. In addition, the species share a short malar space, a pronotum in which the dorsal surface is expanded, and a serrate inner metatibial spur in males and females (Engel et al. 1997;Engel and Brooks 1999;Engel 2000). The bodies are largely yellow but with significantly regions of dark to brilliant metallic coloration, usually green but also blue and coppery (Figs. 2,3,(11)(12)(13)(14). When the genus was first described, the sole species known (C. femoralis) had males with considerably modified midlegs, particularly the mesotrochanter, mesofemur, mesotibia, and mesobasitarsus (Fig. 4). Some of these modifications are now known to be shared with C. nexosa sp. n., such as the greatly enlarged (although somewhat compressed along the anterior-posterior axis) mesofemora with ventral tubercles or teeth and about twice as long as wide, the inner ridges and apically depressed surfaces of the mesotibiae, and the concave inner surfaces of the mesobasitarsi (Fig. 5). The mesotrochanter of C. femoralis is strongly bent and has a strong inner tubercle (Fig. 4), while the same podite in C. nexosa sp. n. lacks such a prominent modification, at most showing a weak swelling (Fig. 5). The mesobasitarsus of C. femoralis is about as long as wide and somewhat triangular, while it is longer than wide and more rectangular in C. nexosa sp. n. In contrast, the midlegs of C. simplex are like those of other Augochlorini (Fig. 6), particularly the normal mesofemora which are about four times as long as wide. In C. nexosa sp. n. and C. femoralis the metallic reflections of the head and mesosoma tend to be green while in C. simplex such areas are more bluish, although green highlights are indeed present in the few known specimens. Males of C. nexosa sp. n. have the gena uniformly covered by plumose setae in which the branches arise from the rachis on one side in the apical two-thirds, while males of C. femoralis have such setae clustered into a distinctive tuft midway along the gena. Other differences between C. femoralis and C. nexosa sp. n. include the smaller ocellocular distance and narrower gena (by comparison to the width of the compound eye) in the latter species. Lastly, the terminalia of C. nexosa sp. n. tend to more closely resemble those of C. simplex rather than C. femoralis.
Detailed descriptions of C. femoralis and C. simplex were given relatively recently by Engel et al. (1997) and Engel and Brooks (1999), and that material is not repeated herein. At present the genus is known only by a meager diversity and relative paucity of material, despite the seemingly wide range of the lineage across South America. Accordingly, it is premature to attempt too contemplative of an investigation into the evolution of this genus, particularly given that the resulting three-taxon statement would have little explanatory power in the absence of more complete life-history or biological data for the constituent species. Nonetheless, if we presume that the modified midlegs of C. femoralis and C. nexosa sp. n. represent a derived feature indicating a close relationship between these two species, then C. simplex would appear to fall basal within the genus. This assumption seems safe for the moment given that such modifications of the midleg podites are not known in related or other genera of Augochlorini (Engel 2000). It would be significant to discover the uses of these elaborate adaptations as used by the male during courtship and mating, and melittologists should be aware of these bees as they undertake observational studies in those regions where individuals are known to occur. Perhaps more interesting and feasible at this stage would be continued work on the relationship of the genus as a whole in relation to other lineages of Augochlorini. Engel and Brooks (1999) and Engel (2000) highlighted a putative relationship between Chlerogelloides and Chlerogella + Ischnomelissa. This hypothesis has not been supported by preliminary molecular work and Chlerogelloides does seem to occupy a rather isolated place among augochlorine genera, at least phenotypically. Intensive sampling is needed in order to increase the material available for the genus (for both molecular and morphological studies), more thoroughly document the distribution of its species, and to bring to light its presumably unique biology.