Andrena (Micrandrena) omnilaevis Wood, 2020

Wood, Thomas James, Cross, Ian & Baldock, David W., 2020, Updates to the bee fauna of Portugal with the description of three new Iberian Andrena species (Hymenoptera: Apoidea: Anthophila), Zootaxa 4790 (2), pp. 201-228 : 205-208

Andrena (Micrandrena) omnilaevis Wood spec. nov.

Holotype: PORTUGAL: Confurco , Várzea Cova, 14.v.2019, 1♀, leg. Wood. Deposited in the OÖLM.

Paratypes: PORTUGAL: Confurco, Várzea Cova , 14.v.2019, 2♂ ; Castro Laboreiro , 3 km south, 1♂, leg. Wood ; Lindoso , 1.5 km E, 41.8720, -8.1792, 13.v.2019, 1♀, leg. Wood ; Guarda, Vale do Rossim , 24.vi.1987, 1♀, leg. F. Torres (reported as Andrena semilaevis by Dardon et al. 2014) , University of Salamanca ; SPAIN: Leon, Villablino, Puerto Leitariegos , 1400 m, 12.vii.1987, 1♂, 5♀, leg. M. Schwarz ; Leon , La Magdalena, 1100 m, 11.vii.1987, 1♀, leg. M. Schwarz. Paratypes are deposited at the OÖLM, and the personal collections of T. J. Wood and M. Schwarz (Ansfelden, Austria) and in the Collection of Zoology, University of Salamanca .

Other material examined ( Andrena omnilaevis ): PORTUGAL: Trás-os-montes, Viade [de Baixo], 30.vi.1977, 1♀, leg. Ph. Pronk, Naturalis ; SPAIN: Ávila , Sierra de Gredos, La Plataforma, 1800 m, 19.v.1995, 4♂, leg. H. & J.E. Wiering, Naturalis ; Ávila , 6 km S of Santa Cruz del Valle, 800 m, 26.v.1995, 1♂, leg. H. & J.E. Wiering, Naturalis. ( Andrena semilaevis ) : SPAIN: Gerona , Vilallonga de Ter, 1100 m, 16.vii.1970, 1♀ , leg. V.S. v. d. Goot & J.A.W. Lucas, Naturalis, Leiden .

Diagnosis: This species belongs to the subgenus Micrandrena Ashmead that contains small black bees with ungrooved clypei and (usually) with a strongly and entirely reticulate propodeal triangle. Andrena omnilaevis can be recognised in the female sex by the unique combination of shiny, unshagreened terga, fovea that are not narrowed below, and tergal margins that are uniformly and regularly depressed, shiny, and unpunctured ( Figure 9 ). Only two other Iberian species have shiny terga (though this is common in central and eastern European species e.g. Andrena enslinella Stoeckhert, 1924 Schmid-Egger & Scheuchl 1997 ), Andrena nana (Kirby, 1802) , which has fovea that narrow to half their maximum width at the level of the antennal insertions, and Andrena floricola Eversmann, 1852 , which does not have shiny depressed margins on T2–4 (in common with A. nana ). Andrena omnilaevis is structurally closest to Andrena semilaevis which has the same regularly depressed, shiny, and unpunctured margins on T2–4. However, these strongly contrast with the surface of the tergal discs which are shagreened and dull ( Figure 10 ), the feature behind its scientific name semilaevis (partially shiny). The male can be recognised with A. semilaevis by the combination of shiny depressed tergal margins ( Figure 11 ) and an antennal segment 4 that is as long as wide, and then separated from it by the shiny, not dull, tergal discs ( Figures 11–12 ). The genitalia are very similar, but the gonocoxal teeth are slightly less pronounced in A. omnilaevis in direct comparison ( Figures 13–14 ).

Description: Female: Body length 6 mm ( Figure 7 ). Head: Head 1.3 times wider than long. Clypeus densely but somewhat shallowly punctate, space between punctures less than 1 puncture diameter, underlying clypeal surface shagreened, weakly shining. Impunctate central midline generally but weakly present. Process of labrum trapezoidal, slightly wider than long, glossy. Fovea normal, at their zenith occupying half the space between compound eye and lateral ocellus, slightly narrowed below becoming 80% of maximal width at level of antennal insertions. Pubescence of face white, particularly around clypeus and antennal insertions to whitish brown on vertex. Antennae uniformly dark. Mesosoma: Scutum moderately punctured, punctures separated by 1–2 puncture diameters. Underlying surface slightly shagreened therefore moderately shining, particularly on scutellum. Scutum, scutellum, and immediately adjacent areas with short, light brownish hairs, episternum and sides of the propodeum with longer white hairs. Entire dorsal surface of propodeum reticulate in typical Micrandrena fashion, with propodeal triangle outlined by inconspicuous carina. Propodeal corbicula with stout, simple hairs. Legs dark, with white to light brownish pubescence. Scopal hairs white. Metasoma: Terga uniformly dark, without lightened margins, lacking shagreenation or reticulation, shiny ( Figure 9 ). Tergal discs uniformly punctate, those on T2–4 separated by 1 puncture diameter, those on T1 by 2 puncture diameters. Apical margins of T1–4 impunctate and glossy, those on T2–4 strongly and uniformly depressed, most pronouncedly on T4. T2–4 with loose hair fringes of white hairs, those on T2–3 widely separated, that on T4 complete, though sparse. General pubescence of terga and sterna white, except for central apical fringe of T5–6 around pygidial plate which is dark brown. Pygidial plate small, pointed, with central longitudinal raised area.

Male: Body length 5.5– 6 mm ( Figure 8 ). Head: Head 1.3 times wider than long, compound eyes slightly convergent below. Clypeus densely punctate, spaces between punctures 1 puncture diameter, underlying clypeal surface weakly shagreened, shining more strongly than in female. Impunctate central line absent. Process of labrum cuboidal, as long as wide, weakly emarginate. Pubescence of head entirely white. Antennae uniformly dark, antennal segment 4 as long as broad, as long as segment 3. Mesosoma: Scutum moderately punctured, punctures separated by 2 puncture diameters. Underlying surface weakly shagreened therefore moderately shining, scutellum with shagreenation much weaker to absent and therefore clearly shiny. Pubescence entirely white. Dorsal surface of propodeum entirely reticulate as in female, propodeal triangle also marked with shallow carina. Legs dark, pubescence white. Metasoma: Terga uniformly dark, without lightened margins, shiny, particularly T1, with very weak shagreenation on discs of T2–5 ( Figure 11 ). Tergal discs moderately punctate, punctures separated by 1–2 puncture diameters. Tergal margins impunctate and glossy, those on T2–4 strongly and uniformly depressed, most pronouncedly on T4. T2–4 with very loose white hair fringes, those on T2 widely separated, those on T3–4 complete but sparse. General pubescence of terga and sterna white, with only apical fringe of T6 darker brown. Genitalia simple ( Figure 13 ), apical half of gonostyli with golden hairs, penis valve slightly flared out before base but not markedly widened.

Distribution: From the Sistema Central in western Spain (Sierra de Gredos) to central Portugal (Serra da Estrela), and north to northern Portugal (Serra do Gerês) and north-western Spain (Muniellos mountains).

Floral preferences: Six pollen loads are available, all of which contained pure Sedum pollen ( Crassulaceae ). Pollen loads were collected on the 13 th and 14 th of May (north Portugal), 24 th of June (central Portugal, Serra da Estrela), 30 th June (north Portugal), and the 12 th of July (northern Spain, Galicia), so they span the known flight period of the species. This contrasts strongly with the foraging ecology of A. semilaevis , where analysis of 97 pollen loads from southern England showed that this species is polylectic with a strong preference for Apiaceae , collecting 73.5% of its pollen from this family and with Apiaceae pollen in 97.9% of analysed loads ( Wood et al. 2016). The habitats where A. omnilaevis has been found are rocky, generally acidic uplands with little flowering Apiaceae , at least during May (TJW, pers. obs.). Instead, there is a great profusion of blossoming Sedum at this time. In all three Portuguese locations where the bee was captured in 2019, A. omnilaevis was flying with Flavipanurgus kastiliensis (Warncke, 1985) , one of the only two currently known European bees that are specialists of Sedum species (Wood and Cross 2018) along with the recently described Hoplitis galichiae Müller, 2016 ( Müller 2016) . It is too early to say if A. omnilaevis is a specialist of Sedum , and indeed this may not be the case since the majority of Micrandrena species are polylectic even if some show preferences for certain botanical families at certain time of the year ( Westrich 1989; Schwenninger 2009; Wood et al. 2016, but see Westrich 2010 for an example of oligolecty). What is clear is that based on the habitats in which it was found and the pollen analysis results so far, A. omnilaevis does not appear to share the foraging niche of its sister species.

Discussion: Andrena omnilaevis appears to be a univoltine bee with males and females emerging in May and flying into the beginning of July. It is known from upland areas of north-western Iberia, often in open acidic areas supporting flowering Sedum species, so far the only confirmed pollen host. We are confident that there is no previous type material referring to this taxon as north-western Iberia is historically under-recorded, and the only published synonym of A. semilaevis has its type locality in England ( Andrena saundersella Perkins, 1914 ; Gusenleitner & Schwarz 2002), and that this species was therefore previously undescribed. Interestingly, in a survey of the Serra da Estrella, Kuhlmann (1996) reports an Andrena (Micrandrena) nov. spec. recognised by Warncke, but after taking possession of this material he died and it cannot be found in his collection (M. Kuhlmann in litt.). Given the location (Penhas da Saude, approximately 15 km from the Vale do Rossim) and the flight period (3–7.vi.1989) it is highly likely that this was the same taxon as A. omnilaevis .

It is beyond the scope of this paper to reproduce a modified key to Iberian Micrandrena , but following the key of Dardon et al. (2014) female material would key to couplet 3 where it can be easily separated on the basis of the depressed and shiny tergal margins. Male A. omnilaevis material would key to couplet 19 at which point its characters conflict as the base of the penis value is not widened, but the tergal margins are impunctate. Andrena omnilaevis males can be separated at couplet 19 by the fourth antennal segment which is as long as broad, whereas for the other two taxa here it is shorter than broad.

Etymology: Andrena semilaevis means the one who is partially shiny, a reference to the shiny depressed rear margins of the otherwise shagreened terga. Because of the entirely shiny terga of this new taxon, the name omni (all) + laevis (shiny) was chosen both to reflect its different appearance and to refer back to its sister species.

Figures

FIGURES 7–14. Andrena omnilaevis spec. nov. 7. female profile, 8. male profile, 9. female terga, 11. male terga, 13. male genitalia; Andrena semilaevis Pérez, 1903, 10. female terga, 12. male terga, 14. male genitalia.