A taxonomic update to Pachymelus (Pachymelopsis) Cockerell, with a new species from Ethiopia (Hymenoptera, Apidae)

A taxonomic update to Pachymelus (Pachymelopsis) Cockerell, with a new species from Ethiopia (Hymenoptera, Apidae) Thomas J. Wood1 1 Naturalis Biodiversity Center, Darwinweg 2, 2333 CR, Leiden, Netherlands https://zoobank.org/3CDA594E-8D18-4953-9AD8-C2CDFFDDCD3B Corresponding author: Thomas J. Wood ([email protected]) Academic editor: Stephan M. Blank | Received 18 August 2025 | Accepted 5 November 2025 | Published 26 November 2025 Abstract Pachymelus Smith, 1879, is a small bee genus comprising 21 species restricted to sub-Saharan Africa and Madagascar. Despite the small size of the genus, several uncertainties persist within the subgenus Pachymelus (Pachymelopsis) Cockerell, 1905, which contains five species and is restricted to sub-Saharan Africa. Based on a revision of museum material and inspection of contemporary collections, a number of changes are implemented. Pachymelus bettoni (Cockerell, 1910), described in the male sex, was incorrectly considered to be a distinct species; it is actually the male of Pachymelus (Pachymelopsis) conspicuus Smith, 1879, syn. nov., and is returned to synonymy with it. Pachymelus (Pachymelopsis) armatipes (Friese, 1911), stat. rev. is returned to species status, having been incorrectly considered conspecific with P. conspicuus. Pachymelus (Pachymelopsis) haladai sp. nov. is described from Ethiopia. A lectotype is designated for Habropoda capensis Cameron, 1905 (= Pachymelus festivus (Dours, 1869)) from material that was previously considered lost; the species is newly reported from East Africa (Kenya), and the status of Afromelecta lieftincki Eardley, 1991 is discussed. The type locality of Pachymelus (Pachymelopsis) abessinicus (Friese, 1912) is clarified, moving from Uganda to Kenya, which is shown to host five of the now six members of this subgenus. These revisions further illustrate the need for greater survey effort and taxonomic revision of bees in dry parts of Eastern Africa. Key Words Anthophorinae, Eastern Africa, lectotype, sex association, solitary bees, synonymy Introduction The genus Pachymelus Smith, 1879 is endemic to sub-Saharan Africa and Madagascar and contains 21 species, with 15 from Madagascar and six from continental Africa (Eardley 1993; Pauly et al. 2001; Michener 2007; Ascher and Pickering 2025). The phylogenetic position of Pachymelus has recently received attention, as it appears to be basal within the subfamily Anthophorinae (Bossert et al. 2019; Orr et al. 2022), suggesting an African origin for this whole clade of bees. The genus is divided into two subgenera, Pachymelus s. str. (16 species) and Pachymelopsis Cockerell, 1905 (five species), with Pachymelus s. str. found almost entirely on the island of Madagascar and Pachymelopsis restricted to continental Africa. Pachymelopsis was described as a genus to accommodate the species Pachymelus conspicuus Smith, 1879, which was described from “Nyassa”, today probably referable to Malawi, and Cockerell defined it purely in opposition to the Madagascan Pachymelus micrelephas Smith, 1879, the type species of Pachymelus. Since then, a broad consensus has emerged that Pachymelopsis is best treated as a subgenus of Pachymelus (Brooks 1988; Eardley 1993; Pauly et al. 2001; Michener 2007, though see Baker 1993). Eardley (1993) recognised six African species in Pachymelus (Pachymelopsis), and Pauly et al. (2001) recognised 15 Madagascan species in Pachymelus s. str. Contributions to Entomology 75 (2) 2025, 339–353|DOI 10.3897/contrib.entomol.75.e169092 Copyright Thomas J. Wood. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Research Article Thomas J. Wood: A taxonomic update to Pachymelus (Pachymelopsis) Cockerell340 However, Michener (2007: 751) considered Pachymelus peringueyi (Friese, 1911) (South Africa and Namibia) to belong to Pachymelus s. str. Following Michener, the two subgenera therefore do not have allopatric distributions, with Pachymelus s. str. not entirely confined to Madagascar. Phylogenetic data from Pachymelus (Pachymelopsis) are also lacking, with all species sequenced to date belonging to Pachymelus s. str. sensu Michener (Bossert et al. 2019; Orr et al. 2022). Additional phylogenetic work is needed, but pending such revisions, the subgeneric concepts of Michener (2007) are followed here. Although the revision of Eardley (1993) was comprehensive, including inspection of all relevant type material, examination of additional museum specimens and recent collections has uncovered novelties within Pachymelus (Pachymelopsis). The opportunity is taken to present new data and expand known distributional ranges, clarify a few outstanding issues including sex associations, illustrate type material photographically, and describe a new distinctive Pachymelus (Pachymelopsis) species from Ethiopia. Methods It was possible to study 133 specimens of Pachymelus (Pachymelopsis) in collections, including relevant type material. Morphological terminology and generic and subgeneric concepts follow Michener (2007), with the exception of the marginal “zones” of the terga, which are referred to as marginal “areas”. Following taxonomic changes, the identification key of Eardley (1993) is updated to cover all Pachymelus (Pachymelopsis) species in sub-Saharan Africa. Photographs were taken using an Olympus E-M1 Mark II with a 60 mm macro lens. Additional close-ups were taken with a Mitutoyo M Plan Apo 10× infinity-corrected objective lens in combination with an Olympus M.Zuiko 2× teleconverter lens, a 10 mm Kenko DG extension tube, and a Meike MK-P-AF3B 10 mm extension tube. Photographs were stacked using Helicon Focus B (HeliconSoft, Ukraine), and plates were prepared in GNU Image Manipulation Program (GIMP) 2.10. Post-processing of some images was done in Photoshop Elements (Adobe Systems, USA) to improve lighting and highlight specific characters. To ensure unambiguous attribution, personal names mentioned in the text are, at first mention, accompanied by their corresponding GND (Integrated Authority File) identifier, when possible, available at https://www.dnb.de/ EN/Professionell/Standardisierung/GND/gnd_node.htm. Abbreviations MNHN Museum nationale d’Histoire naturelle, Paris, France NHMUK Natural History Museum, London, United Kingdom OÖLM Oberösterreiches Landesmuseum, Linz, Austria OUMNH Oxford University Museum of Natural History, Oxford, United Kingdom RMNH Naturalis Biodiversity Center, Leiden, the Netherlands TJWC Personal collection of Thomas J. Wood, Leiden, the Netherlands ZMB Museum für Naturkunde, Berlin, Germany Results Pachymelus (Pachymelopsis) abessinicus (Friese, 1912) Anthophora aethiopica Friese, 1911: 659, ♂ non ♀ nec. Cameron [Kenya, ZMB, not examined] Anthophora abessinica Friese, 1912: 89 nom. nov. pro Anthophora aethiopica Friese Notes. Eardley (1993) summarised the nomenclatural situation surrounding Friese’s names and designated a male lectotype, as the females described by Friese actually belong to an Amegilla species. This lectotype designation fixed the type locality as “Br. O. Afrika, Mulango, 1908, Sauberlich [sic]”. Eardley stated this to refer to Mengo in Uganda. However, consideration of the collector of this specimen must be made to inform this choice. Günther Albin Säuberlich (1864–1946, https://explore. gnd.network/gnd/1068413840) was an evangelical Lutheran missionary who worked in East Africa on and off between 1888 and 1912 (Bauer and Hörnlein 2005). Between 1906 and 1912 he was based at Mulango in southern Kenya, which is located at –1.4272°S, 38.0120°E, approximately 115 km east of Nairobi. The spelling of this place name matches that on the type label, and British East Africa almost invariably refers to modern Kenya, not Uganda, which was most commonly referred to as the Uganda Protectorate or simply Uganda. Kenya must therefore be considered the terra typica, not Uganda. Friese (1911: 660) also mentioned males from Madibura [Madibira] and Murutunguru-Ukerewe [Ukerewe Island] in Deutsch Ost Afrika (= Tanzania), and these are considered valid country records at present. Distribution. Kenya and Tanzania (Friese 1911). Pachymelus (Pachymelopsis) armatipes (Friese, 1911) stat. rev. Anthophora armatipes Friese, 1911: 660, ♀♂ [“Somaliland”, ZMB, not examined] Material examined. Ethiopia • 2♂; Jowaha N. of Debra Sina; 6,000 ft a.s.l.; 7 Aug. 1946; K.M. Guichard leg.; D.B. Baker det. 1979; NHMUK; KEnya • 1♂; Brit. E. Af., Mtito Andei; 2,500 ft a.s.l.; 26–28 Mar. 1911; S.A. Neave leg.; D.B. Baker det.; 1982; NHMUK; • 2♀; E of Contributions to Entomology 75 (2) 2025, 339–353 341 Mwingi; 14 May 2007; M. Halada leg.; T.J. Wood det.; OÖLM/TJWC • 1♀; Taita Hills, Voi river; 14 Apr. 2007; T.J. Wood det.; OÖLM. Notes. The situation surrounding three taxa, P. conspicuus, Pachymelus bettoni (Cockerell, 1910), and P. armatipes, has been confused. Brooks (1988) considered all three to be conspecific, but Eardley (1993) treated P. conspicuus (inclusive of P. armatipes) and P. bettoni as two distinct species. This case requires clarification, as P. conspicuus was described from a single female (Fig. 1), P. bettoni from a single male (Fig. 2), and P. armatipes in both sexes. Friese (1911) provided a diagnosis for P. armatipes against P. festivus (Dours, 1869), citing the larger size, different female colouration, and long hair fringe on the male fore basitarsus. Eardley (1993: 220–221) argued that the female of P. armatipes was conspecific with P. conspicuus, and since the sexes of the type series were “apparently” correctly associated, the male of P. armatipes must therefore be the male of P. conspicuus, as “no evidence was available for the association of the holotype of P. bettoni with that of P. conspicuus”. Eardley’s synonymy of P. armatipes was unjustified, as (1) there are morphological differences in pubescence between the females of P. armatipes and P. conspicuus, and (2) there was no effort to associate sexes across the observed ranges of the possible taxa to provide additional support for or against a particular association. While P. conspicuus females have dense and fine tergal pubescence (Fig. 1D), P. armatipes females have thicker hairs on the terga, giving an almost scaly appearance (Fig. 3D). The scutal pubescence of P. conspicuus is also finely intermixed greyish and dark brown hairs (Fig. 1C), whereas in P. armatipes the pubescence is rich reddish-brown with only a few intermixed black hairs (Figs 3A, 3C). There are also slight differences in body size, with female P. conspicuus measuring around 15–16 mm and P. armatipes around 16–18 mm (see Friese 1911: 661), and in the colouration of the yellow facial markings, which are lemon yellow in P. conspicuus (Fig. 1C) and ivory yellow in P. armatipes (Fig. 3B). Concerning males, examination of specimens has found P. conspicuus females with P. bettoni males flying together at the same site on the same day in four locations (Kenya = 3, Zambia = 1; full material examined listed below under P. conspicuus). In fact, the specimens from Zambia were critical to understanding the sex association, as males displaying the morphology of P. armatipes have only been examined from Ethiopia (Fig. 4) and Kenya, whereas “P. bettoni” males have been examined extending south through Tanzania to Zambia, covering the known range of P. conspicuus based on female specimens. Figure 1. Pachymelus conspicuus Smith, 1879 holotype female (NHMUK). A. Label details; B. Body, dorsal view; C. Head and mesosoma, anterolateral view; D. Metasoma, dorsal view. Thomas J. Wood: A taxonomic update to Pachymelus (Pachymelopsis) Cockerell342 Figure 2. Habropoda bettoni Cockerell, 1910 (= Pachymelus conspicuus Smith, 1879) holotype male (NHMUK). A. Label details; B. Habitus, lateral view; C. Head, frontal view; D. Fore tarsus, dorsal view; E. Hind tarsus, dorsal view; F. Metasoma, dorsal view. Pachymelus bettoni must therefore represent the male of P. conspicuus based on local co-occurrence and overall distributional overlap. The species concepts presented here are consistent with the morphological concepts used by Donald Baker (see listed specimen determinations). Pachymelus bettoni is therefore formally synonymised with P. conspicuus below, and P. armatipes stat. rev. is restored to species status. The males are easily separable, as P. conspicuus has the hind basitarsus apically broadened and thus appearing triangular (Fig. 2E), and the mid trochanters are ventrally rounded, whereas P. armatipes has the hind basitarsus parallel-sided (Fig. 4D), and the mid trochanters are ventrally produced into long, ventrally projecting teeth (see identification key). Distribution. Ethiopia*, Somalia, Kenya* (Friese 1911; Eardley 1993, as Pachymelus conspicuus partim). Pachymelus (Pachymelopsis) conspicuus Smith, 1879 Pachymelus conspicuus Smith, 1879: 117, ♀ [Malawi, NHMUK, examined] (Fig. 1). Contributions to Entomology 75 (2) 2025, 339–353 343 Pachymelus insulanus Stadelmann, 1898: ♀ [Tanzania, ZMB, not examined]. Habropoda bettoni Cockerell, 1910: 215, ♂ [Kenya, NHMUK, examined] (Fig. 2) syn. nov. Material examined. Ethiopia • 1♂; Abyssinia, Busika; 5 Nov. 1911; R.J. Stordy leg.; T.J. Wood det.; NHMUK • 1♂; Abyssinia, Higo Samula; 30 Oct. 1911; R.J. Stordy leg.; T.J. Wood det.; NHMUK; KEnya • 1♂; B.E. Africa; C.S. Betton leg.; NHMUK (holotype of Habropoda bettoni) • 1♂, 1♀; Brit. E. Af., 30 miles from Magadi Junc.; 1–30 Apr; 1912; F.G. Hamilton leg.; T.J. Wood det.; NHMUK • 1♀; Brit. E. Af., Kibwezi; 3,000 ft a.s.l.; 2–4 Apr. 1911; S.A. Neave leg.; T.J. Wood det.; NHMUK • 1♂, 1♀; Brit. E. Af., Mtito Andei; 2, 500 ft a.s.l.; 26–28 Mar. 1911; S.A. Neave leg.; T.J. Wood det.; NHMUK • 1♀; Brit. E. Af., Voi; 8–10 Feb. 1912; S.A. Neave leg.; T.J. Wood det.; NHMUK • 1♂; Brit. E. Africa, E. shore Vic. Nyanza, near Karungu; 28–29 Apr. 1911; S.A. Neave leg.; T.J. Wood det.; NHMUK • 1♂; Brit. E. Africa, Kuja Valley, S. Kavirondo; 4,000 ft a.s.l.; 30 Apr. – 1 May 1911; S.A. Neave leg.; T.J. Wood det.; NHMUK • 1♂, 1♀; Nzoi [probably Nzaui Hill], Ukambani Country; 1 Jan. – 28 Feb. 1889; F.J. Jackson leg.; NHMUK • 8♂, 2♀; Taita Hills, Voi river; 14 Apr. 2007; M. Halada leg.; OÖLM/ TJWC • 1♀; Voi (Tsavo) env.; 22 Nov. – 2 Dec. 1996; Mi. Halada leg.; OÖLM • 1♀; Afrique Orient. Angl., Kibwezi (Wa-Kamba); 1904; Ch. Alluaud leg.; T.J. Wood det.; MNHN; Malawi • 1♀; [label illegible]; NHMUK (holotype of Pachymelus conspicuus); tanzania • 1♂; German E. Africa, Stigi [?Sijiji Hill]; 19 Apr. 1917; Dr. G.D.H. Carpenter leg.; T.J. Wood det.; NHMUK • 1♂; Mkomazi Game Res., Mbula Hill; 25 Dec. 1994; G.N. Stone leg.; T.J. Wood det.; OUMNH • 1♀; Rukwa Valley, Tanganyi a.; 1–31 May 1953; FitzGerald leg.; T.J. Wood det.; NHMUK • 1♀; Tanganyika: Old Shinyanga; 23 Mar. 1952; E. Burtt leg.; G.E.J. Nixon det.; 1958; NHMUK • 1♂; W. Shore of K. Manyara; 1 Feb. – 31 May 1935; B. Cooper leg.; T.J. Wood det.; NHMUK; zaMbia • 1♀; N. Rhodesia, Road to Nangereri; 6 Feb. 1911; Silverlock leg.; T.J. Wood det.; NHMUK • 4♂, 2♀; N.E. Rhodesia, Up. Luangwa Valley; 1800–2000 ft a.s.l.; 8–9 Mar. 1908; S.A. Neave leg.; D.B. Baker det. 1988; OUMNH • 2♂; N.E. Rhodesia, Up. Luangwa Valley; 1800–2000 ft a.s.l.; 22–26 Mar. 1908; S.A. Neave leg.; D.B. Baker det. 1988; OUMNH • 1♀; Siovonga [Siavonga]; 11 Apr. 1981; G.G.M. Schulten leg.; C. Eardley det. 1993; RMNH. Notes. Although P. conspicuus was nominally reported from Ethiopia by Eardley (1993), the specific male specimens cited were actually P. armatipes. The presence Figure 3. Pachymelus armatipes (Friese, 1911) female (OÖLM). A. Habitus, lateral view; B. Head, frontal view; C. Head and mesosoma, anterolateral view; D. Metasoma, dorsal view. Thomas J. Wood: A taxonomic update to Pachymelus (Pachymelopsis) Cockerell344 of P. conspicuus in Ethiopia is supported by the additional specimens presented above, although the exact localities within “Abyssinia” cannot currently be traced. As discussed above, it is appropriate to synonymise P. bettoni under P. conspicuus as its male. Distribution. Ethiopia, Kenya, Tanzania, Malawi, Zambia, Namibia (Smith 1879; Cockerell 1910; Eardley 1993, as Pachymelus conspicuus partim and Pachymelus bettoni). Pachymelus (Pachymelopsis) festivus (Dours, 1869) Habropoda festiva Dours, 1869: 33, ♀♂ [South Africa, type lost?]. Habropoda capensis Cameron, 1905: 255, ♀ [South Africa, NHMUK, lectotype by present designation] (Fig. 5). Material examined. KEnya • 1♂, 2♀; Brit. E. Af., 30 miles from Magadi Junc.; 1–30 Apr. 1912; F.G. Hamilton leg.; T.J. Wood det.; NHMUK • 1♂; Brit. E. Africa, Magadi Ry.; 1–31 May 1912; F.G. Hamilton leg.; T.J. Wood det.; NHMUK • 1♂; Magadi; 7 Apr. 1912; F.G. Hamilton leg.; T.J. Wood det.; NHMUK • 6♀; Ngong; 1–30 Jun. 1944; V.G.L. van Someren leg.; T.J. Wood det.; NHMUK • 2♀; Nzoi [probably Nzaui Hill], Ukambani Country; 1 Jan. – 28 Feb. 1889; F.J. Jackson leg.; T.J. Wood det.; NHMUK • 1♀; Mission de l’Omo, Machacos [Machakos], Wa-Kamba; 1600 m a.s.l.; 1932–1933; C. Arambourg, P.-A. Chappuis, R. Jeannel leg.; T.J. Wood det.; MNHN; lEsotho • 1♂; Basuto Land, between Motsekuwa [Motsekuoa] and Mafetina [Mafeteng]; 30 Mar. 1902; R. Crawshay leg.; D.B. Baker det. 1979; NHMUK • 1♂, 2♀; Basutoland; R. Crawshay leg.; T.J. Wood det.; NHMUK; south africa • 1♀; Africa: Belmont; 23 Feb. 1934; J. Ogilvie leg.; T.J. Wood det.; NHMUK • 1♀; Cape Colony; ex. P. Cameron collection 1905-192; NHMUK (lectotype of Habropoda capensis) • 1♀; Cape G. H. [Cape of Good Hope]; ex. F. Smith collection 1879.22; T.J. Wood det.; NHMUK • 1♀; Kwazulu, Blood River Mem.; 22 Feb. 2000; G.G.M. Schulten leg.; T.J. Wood det.; RMNH • 1♂, 1♀; O.R. Colony, Bloemfontein; 4,500 ft a.s.l.; 1 Dec. 1894 – 31 May 1895; F.N. Brown leg.; D.B. Baker det. 1988; OUMNH • 1♀; Orange Free State, Donkerspoort; 18 Apr. 1934; J. Ogilvie leg.; T.J. Wood det.; NHMUK • 1♂; Willowmore, Capland; Dr. Brauns leg.; T.J. Wood det.; Accessioned 1912-95; NHMUK • 4♂, 1♀; Colonie du Cap, Steynsburg; 1914; R. Ellenburger leg.; T.J. Wood det.; MNHN • 1♂, 1♀; Colonie du Cap, Steynsburg; 1915; R. Ellenburger leg.; T.J. Wood det.; MNHN. Figure 4. Pachymelus armatipes (Friese, 1911) male (NHMUK). A. Label details; B. Habitus, lateral view; C. Fore tarsus, dorsal view; D. Hind tarsus, dorsal view. Contributions to Entomology 75 (2) 2025, 339–353 345 Figure 5. Habropoda capensis Cameron, 1905 (= Pachymelus festivus (Dours, 1869)) lectotype female (NHMUK). A. Label details; B. Habitus, lateral view; C. Head, frontal view; D. Terga, dorsal view. Notes. Although most type material of Dours is considered lost, P. festivus was described using material from the collections of both Jean Antoine Dours and Frédéric Jules Sichel, the latter of which may still be preserved in the Paris museum, although a recent search could not locate any syntypic specimens. Although the type material is currently treated as lost (Eardley 1993), the concept and use of the name P. festivus are well established in the literature and in collections; only two Pachymelus species occur around the Cap de Bonne-Espérance (Cape of Good Hope, the locus typicus), the description is consistent with its current use, and so it is not considered necessary to designate a neotype. Equally, the identity of Habropoda capensis has been considered straightforward for the same reasons, and the synonymy was made almost immediately (Friese 1905: 233). Eardley (1993: 222) noted that the “holotype” of H. capensis was lost, as it could not be found at the Albany Museum (Grahamstown, now Makhanda, South Africa). However, Cameron (1905: 256) wrote: “Cape Colony. (Brak Kloof); Transvaal.”, clearly indicating that at least two specimens must have been used to describe the species, given the two reported regions. There is also no specific reference to the Albany Museum in the entry for H. capensis, as Cameron stated at the beginning of the paper that specimens dealt with in that work were from various private collectors, the Albany Museum collection, and specimens placed in his own collection. Examination of the NHMUK collection has produced a female labelled in Cameron’s distinctive handwriting as coming from Cape Colony (Fig. 5). The specimen bears the accession number 1905-192, indicating that it was received by the NHMUK in the year 1905, i.e. in the same year as the description of the species. As it was clearly inspected by Cameron and comes from one of the stated localities (Cape Colony), it can be treated as syntypic and is hereby designated as the lectotype. Morphologically, the specimen conforms to the concept of P. festivus employed in the literature (e.g. Friese 1905; Eardley 1993). Finally, specimens of P. festivus were unexpectedly found from southern Kenya in both the NHMUK and MNHN collections (Figs 6, 7). The published distributional range was considered to be South Africa and Lesotho (Eardley 1993), but multiple specimens collected by F. J. Jackson, F. G. Hamilton, V. G. L. van Someren, C. Arambourg, P.-A. Chappuis, and R. Jeannel from a small area of south-western Kenya not far from Nairobi were morphologically consistent with southern African specimens, with very minor yellow markings on the female mandibles (Fig. 6C; see identification key). This emphasises the Thomas J. Wood: A taxonomic update to Pachymelus (Pachymelopsis) Cockerell346 Figure 6. Pachymelus festivus (Dours, 1869) female (NHMUK). A. Label details; B. Habitus, lateral view; C. Lower part of head, anterolateral view; D. Terga, dorsal view. emerging biogeographic link observed in bees between southern Africa and southern Kenya (Ascher and Engel 2017; see Discussion). This has potential implications for a different but potentially ecologically related genus of bees, the parasitic lineage Afromelecta Lieftinck, 1972 (Apidae: Melectini). Originally described based on the species Afromelecta fulvohirta (Cameron, 1905), which is considered restricted to South Africa (Lieftinck 1972), Eardley (1991) described Afromelecta lieftincki Eardley, 1991 from a single specimen from southern Kenya (Fig. 8). Examination of the collecting information reveals that it is from Machakos and was collected by C. Arambourg, P.-A. Chappuis, and R. Jeannel between 1932 and 1933 (Fig. 8A; cited as 1922–1933 by Eardley 1991: 3), which is the same collecting information as for the MNHN specimen of P. festivus. Brauns (1926) cited P. festivus as the host of A. fulvohirta (as Habropoda festiva and Epeolus fulvohirtus, respectively), as mentioned by Lieftinck (1972: 315), although the latter author expressed caution regarding this proposed host association given the diversity of anthophorine bees in South Africa. The concurrent (as far as can be inferred given the lack of precise dates) capture of A. lieftincki and P. festivus in Kenya and A. fulvohirta and P. festivus in South Africa (see also material collected by R. Ellenburger) is suggestive, but given the small number of specimens involved, no action is taken here. The hypothesis that A. lieftincki is simply an isolated population of A. fulvohirta, with both taxa attacking P. festivus, is proposed here but requires additional study and field validation. Distribution. South Africa, Lesotho, Kenya* (Dours 1869; Cameron 1905; Friese 1905; Eardley 1993). Other material examined. (Afromelecta fulvohirta) south africa • 1♀; Colonie du Cap, Steynsburg; 1914; R. Ellenburger leg,; T.J. Wood det.; MNHN • 1♂; Colonie du Cap, Steynsburg; 1915; R. Ellenburger leg,; T.J. Wood det.; MNHN; (Afromelecta lieftincki) KEnya • 1♂; Mission de l’Omo, Machacos [Machakos], Wa-Kamba; 1600 m a.s.l.; 1932–1933; C. Arambourg, P-A. Chappuis, R. Jeannel leg.; C. Eardley det.; MNHN (holotype). Pachymelus (Pachymelopsis) haladai sp. nov. https://zoobank.org/BEDC341D-BA3E-43E7-A167-6EDB2B24568A Holotype. Ethiopia • 1♂; 60 km SSW Negele [probably actually 5.5333°N, 39.4333°E], 1480 m a.s.l.; 24 May 2015; J. Halada leg.; OÖLM. Diagnosis. Pachymelus haladai can swiftly be placed in the subgenus Pachymelopsis, as the tarsi lack arolia, the Contributions to Entomology 75 (2) 2025, 339–353 347 Figure 7. Pachymelus festivus (Dours, 1869) male (NHMUK). A. Label details; B. Habitus, lateral view; C. Head, frontal view; D. Hind basitarsus, dorsal view. first flagellar segment is shorter than the scape, the labrum presents a weak transverse preapical ridge, and T2 has a moderately strong gradulus which does not reach the apical margin of the segment. At a specific level, it can be swiftly recognised due to the combination of a long dense fringe of hairs on the anterior basitarsus (Fig. 9C; as in P. armatipes, P. conspicuus, and P. festivus) and the mid trochanters being strongly produced into long tooth-like projections (Fig. 9E). This places it closest to P. armatipes, as P. conspicuus and P. festivus have the mid trochanters rounded. Separation of P. haladai can be made by the shape of the hind basitarsus (Fig. 9F) which in P. haladai has a strongly produced acute projection at the base, with the inner surface of the basitarsus distinctly concave (in P. armatipes without such a spine, and with the inner margin of the hind basitarsus straight, more or less parallel with the outer surface, Fig. 4D). The shape of the hind basitarsus also allows for separation from P. conspicuus (Fig. 2E, hind basitarsus strongly broadened apically, thus triangular) and P. festivus (Fig. 7D, hind basitarsus with a small tooth apically, but not basally). Description. Female. Unknown. Male. Body length: 18 mm (Fig. 9A). Head: 1.5 times wider than long, integument dark with yellow maculations on scape, majority of clypeus with exception of black mark basomedially, labrum, and mandibles basally, apexes of mandibles olive-yellow (Fig. 9B). Labrum with weak preapical transverse ridge, surface between preapical ridge and apical margin covered with short buff hairs. Eyes large, inner margins curved, dorsally and ventrally with margins diverging. Galea moderately long, shorter than length of head. Head covered with whitish to buff hairs, hairs not equalling length of scape. Gena approximately 60% diameter of compound eye; ocelloccipital distance slightly exceeding diameter of lateral ocellus. Antennae with dorsal surface dark, ventrally surface from apex of A3 onwards lightened orange, ventral surface of A4–13 additionally with presence of greyish scales. A3 exceeding length of A4, shorter than A4+5; A3 also shorter than antennal scape. Mesosoma: Scutum densely and shallowly punctate, punctures confluent over majority of surface, becoming sparser posteromedially, separated by 1–2 puncture diameters. Mesepisternum with relatively larger punctures, punctures predominantly dense, separated by <0.5–1 puncture diameter. Mesepisternum with whitish hairs, becoming buff to golden brown dorsally. Fore leg with femur entirely and tibia anteriorly lightened orange. Fore basitarsis slightly broadened, ivory white with anterior surface polished, posterior margin with conspicuous fringe of long golden hairs, fringe greatly exceeding diameter of basitarsus (Fig. 9C); remaining tarsal segments brownish.