Andrena karia, Wood, 2021

Wood, Thomas James, 2021, Fifteen new Andrena species from little-visited arid, Mediterranean, and mountainous parts of the Old World (Hymenoptera: Andrenidae), Zootaxa 4933 (4), pp. 451-492 : 469-474

Andrena karia , spec. nov. (incertae sedis)

urn:lsid:zoobank.org:act:6E7CD3B1-F6D7-4C59-A43A-CE9BB64CC685

HOLOTYPE: TURKEY: SW, Muğla, University campus, 710 m, N37 o 09’39”, E28 o 22’20”, stony steppe with pines, xi.2012 –iii. 2013, 1³ [no collector information] ( OÖLM).

PARATYPES: TURKEY: Same as holotype, 10³ , 12♀ ( OÖLM) , 1³, 1♀, ( TJWC) , 1³, 1♀, ( SMNHTAU) .

Description: Female: Body length 10.5– 12 mm ( Figure 63 ). Head: Black, as wide as long ( Figure 64 ). Clypeus domed, more or less flattened in apical half, apical margin straight, lateral corners forming right angles. Clypeus with clear, large, and shallow punctures, basally separated by 0.5–1 puncture diameter, becoming sparser centrally and apically, separated by 0.5–2 puncture diameters; clear impunctate longitudinal central line present, underlying clypeal surface basally shagreened, dull, becoming weaker centrally and apically, weakly shining. Process of labrum rounded trapezoidal, basally with transverse striations, becoming weaker apically, shiny ( Figure 65 ). Palpi elongate, maxillary palpi with 6 segments, labial papli with 4 segments ( Figure 66 ). Gena broadened, broader than width of compound eye; ocelloccipital distance equals width of lateral ocellus. Fovea poorly defined dorsally, occupying 1/3 rd of area between lateral ocellus and compound eye, long, clearly extending below level of antennal insertions, not separated from inner margin of compound eye. Gena, vertex, face, and scape with moderately long black hairs, longest equalling length of scape, hind face of vertex with few scattered long white hairs. Antenna dark, A5 apically, A6–12 lightened grey below, A3 equals A4+5+6. Mesosoma: Scutum and scutellum strongly shagreened, dull, faintly and irregularly punctate, punctures separated by 1–3 puncture diameters ( Figure 67 ). Pronotum with faint humeral angle, dorsolaterally slightly angulate, almost imperceptible. Episternum and propodeum shagreened, weakly shining, propodeal triangle indicated by slight weakening of shagreenation, basally with weak rugosity. Episternum, and propodeum with long black hairs, scutum and scutellum with shorter intermixed black and white hairs, white hairs restricted to scutellum and anterior margin of scutum. Legs uniformly dark, pubescence mostly black to dark brown. Floccus and femoral scopa white, tibial scopa white ventrally, dark brown above. Wings hyaline, venation black, stigma dark brown, nervulus postfurcal. Metasoma: T1 black basally, red apically, T2–3 entirely red, T4 red basally (can be restricted to small lateral maculations), black apically, T5 entirely black ( Figure 68 ). Terga finely and regularly punctate, punctures separated by 1–1.5 puncture diameters, each bearing a short black hair, underlying tergal surface shagreened, weakly shining. Apical fringe of T5 and hairs flanking pygidial plate dark brown; pygidial plate flattened, entire surface with exception of narrow upturned margins with dense network of punctures, interspaces forming mesh-like network.

Male: Body length 8–9.5 mm ( Figure 69 ). Head: Black, 1.1 times wider than long ( Figure 70 ). Clypeus domed, more or less flattened in apical half, apical margin straight, lateral corners forming right angles. Clypeus with shallow punctures, basally separated by 0.5–1 puncture diameter, becoming very sparse centrally and apically, leaving a large puncture-free area, underlying clypeal surface basally strongly shagreened, dull, becoming weak centrally and apically, shining. Process of labrum rounded rectangular. Mandibles large, robust, long, strongly crossing apically, subapical tooth clearly pronounced on inner margin, basally broadened to 2/3rds width of compound eye, lacking any tooth projecting from ventral surface. Gena strongly broadened, equalling twice width of compound eye ( Figure 71 ), dorsal margin carinate; ocelloccipital distance slightly exceeding width of lateral ocellus ( Figure 72 ). Gena, vertex, face, and scape with long black hairs, some exceeding length of scape, hind face of vertex with several scattered long white hairs. Antenna dark, A5–13 lightened grey below, A3 exceeds A4+5, shorter than A4+5+6. Mesosoma: As in female ( Figure 73 ). Metasoma: As in female ( Figure 74 ). Genitalia simple, gonocoxae with weakly produced and rounded teeth, gonostyli produced to broad scoops apically, internal margin reflexed and raised, dorsal face with brown, plumose hairs ( Figure 75 ). Penis valve broad, occupying majority of space between gonostyli, with broad projecting lateral triangular extensions.

Diagnosis: Placement of A. karia into the existing subgeneric system is highly challenging. Most directly, it can be recognised in the female sex as part of the group of species with elongate palpi, and in the male sex as part of the group with a broadened gena, elongate mandibles that cross apically, and genitalia with a broadened penis valve with projecting lateral triangular extensions. This places it close to a group of species that have hitherto been placed in the subgenera Margandrena and Ptilandrena ( Table 1). Problems with subgeneric classification are discussed below, and diagnosis is made here by direct comparison to morphologically similar species.

Female material can be separated from most of these species by its red-marked terga and by the predominantly black pubescence of the face and mesosoma , placing it closest to A. krausiella Gusenleitner, 1998 and A. menahemella Scheuchl & Pisanty, 2016 . It can be separated from A. krausiella (alternative character state in parentheses) by the more extensively red marked terga, covering all of T2–3 (T2–3 at least partly black on discs), the entirely black pubescence of the face (facial hairs mostly black but with concentration of white hairs centrally around the antennal insertions), strongly postfurcal nervulus (weakly postfurcal, almost interstitial), and the strongly basally striate labrum (with only one or two striations, predominantly shiny). It can be separated from A. menahemella by the strongly shagreened scutum and scutellum (scutellum partially shiny centrally, scutellum extensively shiny), the more extensively red marked terga, covering all of T2–3 (Israeli A. menahemella T2–3 at least partly black on discs, note that in Moroccan A. menahemella T2–3 are entirely red, Wood et al. 2020a), and the strongly basally striate labrum (without striations, predominantly shiny).

Male material can also be placed close to the same two species through the combination of red-marked terga and predominantly dark pubescence. It can be separated from A. krausiella by the pubescence of the face which is entirely black (facial hairs mostly black but with concentration of white hairs centrally around the antennal insertions) and by the strongly postfurcal nervulus (weakly postfurcal to interstitial). Separation from A. menahemella is simple, as the penis valve has clear projecting lateral triangular extensions (penis valve broad but with weak lateral extensions).

Discussion: The molecular phylogeny of Andrena of Pisanty et al. (2020) identified numerous instances of polyphyletic subgenera. One particularly troublesome subgenus has been Ptilandrena (type species, the Nearctic Andrena erigeniae Robertson, 1902 ), which should be the senior name for the large Holarctic group that have traditionally been predominantly referred to as Euandrena (see also Praz et al. 2019), but has been applied to several groups in which the males have a widened gena combined with elongate mandibles. One group includes East Mediterranean species that fly during the autumn through the winter, including A. grossella Grünwaldt, 1976 , A. muscaria Warncke, 1965 , and A. crocusella ( Table 1). However, the phylogeny of Pisanty et al. identifies this group as i) present in the same broad clade (Clade 32) but not falling directly in the true Ptilandrena and ii) mixed together with some of the species placed in the subgenus Margandrena .

The phylogeny inferred by Pisanty et al. (2020) consistently placed A. crocusella together with A. (Margandrena) krausiella , and separate from A. (Margandrena) marginata Fabricius, 1776 (type species for the subgenus) and A. (Margandrena) quinquepalpa Warncke, 1980 . Study of material from the crocusella + krausiella group suggests that the morphology of the male penis valve is potentially a common character that can be used to identify this clade, specifically the presence of a strongly widened penis valve that occupies the majority of the space between the gonostyli, normally but not always with pronounced lateral triangular extensions. This can be seen in six taxa, specifically A. crocusella (with extensions, illustrated in Pisanty et al. 2016), Andrena grossella (with extensions, Figure 76 , see also Grünwaldt 1976), A. karia (with extensions, Figure 75 ), A. krausiella (with extensions, illustrated in Scheuchl 2010), A. menahemella (with weak extensions, illustrated in Pisanty et al. 2016), and A. sibthorpi Mavromoustakis, 1952 (with extensions, illustrated in Scheuchl 2010). In contrast, species around A. marginata appear to have a narrow penis valve with clear space between itself and the gonostyli (see illustrations in Scheuchl 2010). However, the males of A. (M.) elsei Scheuchl & Gusenleitner, 2009 and A. muscaria are unknown making their placement impossible using this character, and COI barcode data from A. (M.) hyacinthina (G. Pisanty in. litt.) suggests placement closer to the crocusella + krausiella group rather than A. marginata , and therefore rendering the genital character of lateral extensions to the penis valve of limited value for group diagnosis as these are not present in A. hyacinthina (see illustrations in Scheuchl 2010).

Female material in the crocusella + krausiella group cannot currently be consistently morphologically separated from Margandrena . This group has been identified as possessing elongate maxillary (MP) and labral palps (LP), and some species ( A. grossella and an undescribed species from Crete, Pisanty et al. 2016) have additional segments in both the MP and LP. However, all other species have six MP and four LP segments (the typical ratio in Andrena ), so this character is not distinctive. Both groups share a pronotum with a weak lateral carina, fovea that are not noticeably narrowed or widened above or below, the clypeus can be long or short, and the process of the labrum is broad.

The lack of clarity reflects the difficulty of morphological classification in these derived groups (arising 5–10 million years ago, see Pisanty et al. 2020). As such, though the crocusella + krausiella group probably represents a valid distinct lineage, it is not currently possible to unambiguously delineate it morphologically, and so A. karia is assigned to incertae sedis. Given the morphological variation seen within the groups present in Clade 32, the most conservative choice would be to synonymise the subgenera Euandrena , Chrysandrena , and Margandrena with Ptilandrena and transfer all taxa within this clade to this subgenus, but this expanded Ptilandrena would also be difficult to delineate morphologically. Given this incomplete picture, future taxonomic reorganisation will require new sampling programs to discover and describe unknown males to allow for exploration of genital characters, and extensive taxon sampling to facilitate comprehensive molecular analysis. The only exception to these needs is Andrena (Ptilandrena) vetula Lepeletier, 1841 which clearly and unambiguously belongs in the subgenus Simandrena ( Pisanty et al. 2020) .

A final note should be made about the potential floral preferences of A. karia and the wider group. Most members of this putative group have a winter or very early spring activity period, and seem to mostly be associated with geophyte plants, predominantly in the families Asparagaceae , Colchiaceae, and Iridaceae , particularly Crocus ( Table 1). Though much more study is necessary to better understand these poorly-studied species, A. krausiella seems to break from this pattern and collect pollen from a wider range of botanical families ( Table 1, G. Pisanty in. litt.). Because the type series of A. karia was collected using pan traps, no precise flight period information between December and March can be divined, and so it is not possible to say if it may have a foraging niche on Crocus (closer to A. elsei and A. grossella ) or is more widely polylectic (closer to A. krausiella ).

Etymology: The name is derived from the name Caria (Greek: Καρία, Turkish: Karya), the old Greek name for the region which now largely corresponds to Muğla Province.

Other material examined: ( Andrena grossella ): GREECE: Kefalo, 9.xi.1976, 1³, 1♀, det. Grünwaldt, leg. W. u. L. Gross, OÖLM, illustrated Figure 76 ; ( Andrena hyacinthina ): CYPRUS: Limassol, Yermasoyia Dam, 7.iii.2017, 2³, 1♀, TJWC; ( Andrena krausiella ): ISRAEL: Dahlia [Dalia], 6.ii.1948, 1♀, leg. Bytinski-Salz, OÖLM (paratype); Jerusalem, 6.iii.1948, 1♀, leg. Bytinski-Salz, OÖLM (holotype); ( Andrena menahemella ): ISRAEL: Netiv Halamed He, 16.ii.2010, 1♀, leg. G. Pisanty, OÖLM (paratype); ( Andrena muscaria ): GREECE: Delphi, 11.iv.1963, 1♀, leg. K. Warncke, OÖLM (paratype).

Figures

FIGURES 63–68. Andrena karia spec. nov. 63. female profile, 64. female face, 65. female labrum detail, 66. female glossa detail, 67. female scutum, 68. female terga.

FIGURES 69–76. Andrena karia spec. nov. 69. male profile, 70. male face, 71. male gena, 72. male vertex, 73. male scutum, 74. male terga, 75. male genitalia. Andrena grossella Grünwaldt, 1976; 76. male genitalia.