Andrena savignyi, SPINOLA, 1838

Kratochwil, Anselm, Mrsn & Mrsn, 2025, Redescription of Andrena savignyi SPINOLA, 1838 (Hymenoptera, Anthophila, Andrenidae), Linzer biologische Beiträge 56 (2), pp. 487-526 : 491-522

publication ID

https://doi.org/10.5281/zenodo.16957194

DOI

https://doi.org/10.5281/zenodo.17484403

persistent identifier

https://treatment.plazi.org/id/010087DE-062A-FFA1-329D-A00FFBC4FA91

treatment provided by

Felipe

scientific name

Andrena savignyi
status

 

Historical aspects of the Spinola collection and the description of Andrena savignyi View in CoL by SPINOLA (1838)

The Genoese entomologist Marquis Massimiliano Spinola di Tassarolo (1780-1857) was a famous, scientifically active entomologist. He was born on 10 July 1780 in Pézenas ( France) and died on 12 November 1857 in Tassarolo ( Italy). He described hundreds of species of Coleoptera, Hymenoptera and Hemiptera ( GESTRO 1915). He personally collected on his estates in Spain and South America and was in intensive dialogue with numerous European entomologists. He invested a lot of money in the purchase of insects and acquired parts or entire collections from other famous entomologists of his time, such as the French hymenopterologist Lepeletier (Amédée Louis Michel Le Peletier, comte de Saint-Fargeau) or the hemipterologist Jean-Guillaume Audinet-Serville (PASSERIN D’ ENTRÉVES 1980, ROSA & XU 2014).

Thanks to the extensive correspondence that Spinola kept, it was possible to reconstruct the collection in the 1980s (CASOLARI & MORENO 1978, CASOLARI & CASOLARI 1980, PASSERIN D’ ENTRÉVES 1983). It was organised according to a classification system of the time: one or more specimens of a species were pinned together, their labels removed from the specimens and a main label attached underneath ( Fig. 2 View Fig ). Depending on the origin of the species, Spinola chose different colours: white (Europe including Northwest Asia, the Near and Middle East), yellow (Asia), blue (Africa), green (North, Central and South America) or pink ( Australia, Oceania) (Fulvio Giachino, MRSN, per litt. 2021).

Each main label shows the genus and species name, the author of the species, the collector and the localities. The original Hymenoptera collection consisted of 69 boxes. The collection was later stored in new boxes, retaining the original arrangement and labelling (CASOLARI & MORENO 1978, CASOLARI & CASOLARI 1980).

SPINOLA (1838) reported that he acquired Hymenoptera from Egypt from the insect tradesman Dr M. Waltl from Passau ( Germany). Specimens were collected in Egypt by the entomologist and traveller M. Fischer. This purchase involved a whole collection of Hymenoptera , the description of which was published in a 110-page publication in the "Annales de la Société Entomologique de France, volume 7" ( SPINOLA 1838).

Among these specimens was a new species, which Spinola named Andrena savignyi after the French zoologist Marie Jules César Lelorgne de Savigny (1779-1851). He deliberately chose this name because J. C. Savigny belonged to a circle of 167 scientists, engineers and artists who accompanied Napoleon Bonaparte’s army to Egypt as part of a "Commission des sciences et des arts" during the "Egyptian campaign" (also known as the "Egyptian expedition") in 1798-1801. Although this "Egyptian expedition" largely failed militarily, it was a great success for natural science and for the research of the ancient Egyptian culture.

The results of the expedition were published under the title "Description de l’Égypte" (1st edition 23 volumes, 2nd edition 37 volumes). J. C. Savigny and the French entomologist Victor Audouin are the authors of this expedition’s entomological results ( JOMARD 1809). The collected Hymenoptera Anthophila were depicted true to scale and very accurately on copper engraving plates, whereby certain details (head, antenna, mouthparts, terminal tergite, tarsi, etc.) were also depicted in detail. Seven printed plates with bees include representatives of the genera Andrena, Anthophora, Ceratina, Colletes, Dasypoda, Eucera, Halictus, Hylaeus, Megachile, Nomada, Nomia, Osmia, Megachile and Sphecodes.

However, a more precise taxonomic characterisation was not carried out prior to this publication. "Mr. Savigny presented a large number of species whose identification is really impossible without notes and coloured drawings. We had to confine ourselves to distinguishing the species on the basis of numbers..." (cited from JOMARD 1809). SPINOLA (1838) refers in his description to J. C. Savigny in JOMARD (1809) and to plate 7, no. 19, which in his opinion represents A. savignyi ( Fig. 3 View Fig ) ( MOUSTAFA 1986).

x The lectotype of Andrena savignyi in the Spinola collection of Turin

GUSENLEITNER & SCHWARZ (2002) assume the Museo Regionale di Scienze Naturali (MRSN) in Turin ( Italy) to be the location of deposition. Indeed, the type of Andrena savignyi is in the MRSN in Turin ( Italy) (Fulvio Giachino, MRSN, per litt. 2021). Fig. 2 View Fig shows box no. 138 with the lectotype of A. savignyi . 1

Figs 4a-c show the three specimens described by Spinola as types and the label (outlined in red in the box). Spinola did not mark any types with a special label. As the specimens originate from Africa, the label should be blue, but it has become greenish over time (information, Fulvio Giachino, MRSN, 2021). The name chosen by Spinola is written on the label: Andrena Savignyi. Behind it is the letter "m", which stands for "mihi" (Spinola as author) and the female sign. Below this is the indication of the publication organ "Ann. soc. ent." (Annales de la Société entomologique de France, Paris). "D. Waltl" describes the origin of the specimens. "D." stands for "Dono" (Italian for "gift") or "Donavit" (Latin), and "Waltl" is the seller. Many specimens in the Spinola collection are labelled with such a "D" (information, Fulvio Giachino, MRSN, 2021). The label also describes the place of origin ("Égypte"). Further details can be found in the original letters of Spinola. The lectotype is dark-coloured on T3 (Fig. 4a); the first two tergites are reddish.

x Description of the females of Andrena savignyi by SPINOLA (1838) and descriptions by other authors

Spinola published in 1838 in the "Annales de la Société Entomologique de France " (volume 7, page 512) the type description of A. savignyi . He did not mention that he had three specimens (females). SPINOLA (1838) characterised the female of A. savignyi as follows (translated from French): "Length 11.3 mm; width 3.4 mm. Antennae, body and legs black. Head, thorax and first abdominal segment covered with long white hairs. Three snow-white hair bands on the second, third and fourth segments. Denser pubescence on the fifth segment, white laterally and black in the centre. Hair of upper anal plate black; abdominal and leg hairs, including tibial brush, white; tarsi with golden yellow hairs. Wings hyaline; wing veins black. The first three segments of the abdomen are more or less reddish, but this colouration is variable. In some, three segments are predominantly red, in others black covers the red more or less, and occupies most of the second and third segments. Hairs on the seventh tergite black; hairs on the hind legs white."

SPINOLA’ S (1838) description is not very differentiated, as was usual for this time. It is based on a mixture of characteristics of A. savignyi and Melitta aegyptiaca. The abdominal hair bands of A. savignyi are not snow-white but rather yellowish-white, and the scopa is not white but golden. Melitta species resemble Andrena species, but among other features, their scopae are restricted to the posterior tibia and basitarsus, and their apical tarsi are broadened ( Figs 4b, c).

A. bipartita BRULLÉ (1839) was described based on specimens from the Canary Islands and renamed A. bicolorata SMITH, 1853 (nec Apis bicolorata ROSSI, 1790 ) due to the homonymy with A. bipartita LEPELETIER, 1841 . COCKERELL (1938) also synonymised A. antilope PÉREZ (1895) from Algeria with A. bipartita . It is incomprehensible why COCKERELL (1938) did not use the valid name A. savignyi for the Egyptian populations. COCKERELL (1938) described Andrena bipartita aswanica from Egypt on the basis of a female collected by Capt. K. J. Hayward in Aswan (= Assuan, Egypt) on 29 January 1921 ( Fig. 5b). In contrast to A. bipartita, COCKERELL (1938) characterised A. b. aswanica as follows: "Wings distinctly brownish, but not dark; second cubital cell somewhat contracted above, the hair, which is white in A. bipartita , is distinctly yellowish, first three tergites red." In an identification key, COCKERELL (1938) distinguished the nominate subspecies with the feature "2 tergites red", without mentioning what the nominate subspecies is and where it occurs. Fig. 5a shows a female from the Aswan region, collected by Mohamed A. Moustafa on 5 March 1972, and the type of A. bipartita aswanica ( Fig. 5b). According to WARNCKE (1967), A. b. aswanica COCKERELL is synonymous with A. savignyi .

DYLEWSKA (1983) listed the following characteristic features for A. savignyi in her description: body length 12-13.8 mm; index fovea facialis width to half facial width measured at the centre of the central ocellus = 0.33-0.36; clypeus length to clypeus width = 0.84; scapus length equal to the combined length of Fl1 and Fl2; Fl1 about 2.2 times longer than wide at the end; Fl2 and other flagellomeres about as long as wide, the last Fl about 1.5 times longer than wide; length-width ratio of maxillary palpal segments 1.6: 2.8: 1.8: 1. 7: 1.7: 1.8, of the labial palpal segments 3.0: 1.5: 1.6: 1.7; galea about 0.75 of the length of the clypeus; malar field about 2.6-2.7 times wider than the smallest length; genal field 1.9 times wider than the ocular profile; distance of the lateral ocelli from the posterior margin of the head about 1.6 ocelli size; labral process triangular and shiny; clypeus microsculptured, only at the base strongly shiny and coarsely punctured, distance between the punctures about 1-3 puncture diameters, with unpunctured line in the centre; mesoscutum finely microsculptured, strongly shining, only at the base with strong microsculpture and coarsely punctured; scutellum only microsculptured laterally and coarsely punctured; distance between the punctures about 0.5-4 puncture diameters; postscutellum coarsely microsculptured with greasy shine and coarsely punctured, distance between punctures about 0.5-2 puncture diameters; midfield of propodeum coarsely and deeply veined; terga with fine microsculpture and strong shine, finely punctured, distance between punctures about 0.5-3 puncture diameters. Apical depressions of terga 0.33 of a tergum length; pygidial field broadly terminated, without furrows; body brownish black, more often T1-T3 completely or partly reddish, or only the apical depressions of the terga reddish and translucent; antennae reddish ventrally; face, thorax dorsally, legs, body creamy white and hairy; vertex with golden yellow hairs, T1 whitish and hairy laterally, T2-T4 with broader apical setae; T5-T6 brownish-black and hairy. DYLEWSKA (1983) points out that the width of the abdomen does not always have to be measured, as the Andrena species have similar body proportions, and in her opinion, the measurement of the wing length is also of no help, as this is not a characteristic feature. As the studies on the A. wollastoni group and on species of the subgenus Suandrena show, both parameters are valuable for comparative taxonomic differentiation ( KRATOCHWIL 2020, 2021b).

GUSENLEITNER & SCHWARZ (2002) also described the females of A. savignyi in a broader sense: "The approximately 12 mm long female is characterised by a very shiny, not very densely punctured clypeus, not very wide fovea facialis, fielded centre of the propodeum, orange-red coloured T1-2[3] with continuous bands on T2-4, orange-red coloured tibiae and tarsi of the 3rd pair of legs as well as an apically ‘cut off’ pygidium."

MOUSTAFA & IBRAHIM (1965) published a description of the females of A. savignyi exclusively for Egyptian specimens: "Length 12-14 mm; colour of body black, abdomen basally orange. Facial fovea cuneiform with median depression, covered with moderately dense greyish yellow hairs; Fl1 longer than Fl2 and Fl3, Fl1 and Fl2 brownish black, flagellar segments from Fl3 inclusive to end ventrally orangish; three terminal segments of the antenna dorsally orangish red also; clypeus black; malar space 1/3 as long as wide. Propodeum midfield lightly areolate, color of wings hyaline; prestigma brown, 2/5 of pterostigma in length, pterostigma centrally brown lighter apically, with margins dark; first recurrent nervure well apicad of nervulus. Middle basitarsus covered ventrally with golden hairs, posterior spur of hind tibia strongly curved, expanded basically. Abdominal terga with sparse white hair bands; terga 2 and 3 red, tergum 4 either red or black, terga 5 and 6 (pygidium) black, pygidium bearing long dense brownish black hairs at its posterior end, venter black, basally orange."

The descriptions of different authors differ considerably in many cases and take very different characteristics into account. The main differences relate to the body size, the colouring of the tergites and the hair colouration, as well as the colour of hair bands, among others.

x Redescription of females with special focus on the variation of morphological and morphometric characteristics

4.1 Habitus, body length Figs 5a-b, d-e and Figs 6a-b show dorsal and lateral views of the habitus of females from the Aswan region and Wadi Digla. The head and thorax of A. savignyi are usually black; the antenna is dark dorsally and reddish ventrally. The first pair of legs is reddish-brown; the posterior pairs of legs are reddish. The complete reddish to orange-yellowish colour of T1-T3 is conspicuous in all examined individuals from Egypt ( Fig. 5a). T3 can also be coloured black or reddish-brown ( Fig. 6a View Figs 6 ), usually with a reddish margin in front of the depression. The depressions are reddish-orange and have a greenish metallic shine. The hairs on the head, mesosoma and metasoma are usually yellowish-white. The metasoma has a more or less dense hair band on the first segment and dense yellowish-white hair bands covering the entire depression on T2-T4. The scopae are characterised by golden hairs.

In the following, the different body parts are discussed in detail, and differences to the previous descriptions of the species are shown. A large number of features show variations within the Egyptian specimens. Since numerous individuals from a local population are also present in the entire material studied (e.g. from the Luxor region), intrapopulationspecific variations can also be analysed. The morphometric data refer to the measurement of 25♀♀, indicating the mean value, the standard deviation (SD) and the maximum and minimum values. For morphological traits, the percentages of the frequency of trait characteristics are given.

The females have an average body length of 12.17 mm (SD 0.70 mm, max. 13.76 mm, min. 11.12 mm). The body width of the mesosoma is on average 3.75 mm (SD 0.15 mm, max. 4.16 mm, min. 3.45 mm), and that of the metasoma is 3.68 mm (SD 0.18 mm, max. 4.00 mm, min. 3.2 mm). The values of SPINOLA (1838) are in the lower range of our measured values (length 11.3 mm; width 3.4 mm), those of DYLEWSKA (1983) in the range (12.0- 13.8 mm), and those of MOUSTAFA & IBRAHIM (1965) in the upper range ( 12-14 mm).

4.2 Head

Head colour, head shape, compound eyes, ocelli, facial foveae: The head is coloured black, as SPINOLA (1838) also states, but a closer analysis shows that the clypeus has a partly weak but recognisable red colouration. The head is slightly wider than long ( Figs 7a, b, c View Figs 7 ). The head length averages 3.17 mm (SD 0.10 mm, max. 3.38 mm, min. 2.96 mm), the head width 3.47 mm (SD 0.14 mm, max. 3.97 mm, min. 3.32 mm) (head length/head width, index: 0.85). The compound eyes diverge slightly. The upper inner eye distance is larger than the lower one (upper: 2.36 mm, SD 0.07 mm, max. 2.54 mm, min. 2.21 mm; lower: 2.59 mm, SD 0.09 mm, max. 2.83 mm, min. 2.44 mm). The compound eyes have a length of 2.16 mm (SD 0.13 mm, max. 2.32 mm, min.

1.75 mm) and a width of 0.83 mm (SD 0.05 mm, max. 0.91 mm, min. 0.63 mm). DYLEWSKA (1983) gives the distance of the lateral ocelli from the posterior margin of the head as about 1.6 ocelli sizes (diameter). However, the exact measurements of the females examined here show that the ocelli diameter is 0.25 mm (SD 0.01 mm, max. 0.27 mm, min. 0.23 mm) and the ocelloccipital distance is 0.23 mm (SD 0.01 mm, max. 0.26 mm, min. 0.22 mm), and thus the ocelli diameter corresponds approximately to the ocelloccipital distance. As stated by MOUSTAFA & IBRAHIM (1965), the fovea facialis is cuneiform. It has a median depression and is covered with greyish-yellowish hairs.

P u b e s c e n c e: Overall, a yellowish-white pubescence predominates in the head area, although head regions with pure white pubescence (e.g. lower paraocular area, genae) also occur ( Figs 7a, b, c View Figs 7 ). There are also specimens with predominantly white pubescence. A detailed analysis shows the ratios and the colour variation in individual head areas: vertex: yellowish-white (92%), golden-yellow (4%), white (4%), yellowish-reddish (4%); frons: yellowish-white (92%), golden-yellow (4%), white (4%); interantennal area: yellowish-white (76%), white (20%), golden-yellow (4%); paraocular area: yellowish-white above, white below (48%), yellowish-white (28%), white (24%); clypeus (usually only laterally hairier): yellowish-white (84%), white (12%), yellowish (4%); genae: whitish (56%), pale yellow (24%), white (20%).

The greatest colour variability in the hairs occurs in the paraocular and in the genae regions. Colour differences in head pubescence are not site-specific but also occur within a local area, as could be shown by analysing 12♀♀ from Luxor.

The information provided by SPINOLA (1838), who characterised the hair on the head as long and white, is based on confusion with Melitta aegyptiaca. The information provided by DYLEWSKA (1983), who characterised the face as creamy white and the crown as golden yellow, is also inaccurate. The description by MOUSTAFA & IBRAHIM (1965) as greyishyellow in the head area is also incorrect. The authors also did not take into account the intra-species variation and its differences in the various head areas.

A n t e n n a e: WhileSPINOLA (1838) characterised the antennae as black in colour (confusion with Melitta aegyptiaca), DYLEWSKA (1983) did not consider the colouration of the antennae at all. A more detailed analysis shows the following characteristics ( Figs 7d, f View Figs 7 ), although a certain variability can also be found here. The colouration of the scapus is usually reddish-brown, partially metallic greenish and narrowly reddish at the distal margin (72%), but also reddish-brown and distal red (20%), black and distally red (4%) or black and partially metallic greenish (4%). The pedicellus is in most cases black and metallic greenish (72%), or completely reddish-brown (20%), but also black and partially reddish-brown (8%). As with the scapus, the distal margin is often narrowly reddish.

Fl1 is mostly black above with a greenish metallic shine and reddish-brown below (72%) and occasionally also apically reddish (8%). Sometimes only the base is black; the remaining area is metallic and slightly greenish and apically reddish (12%). In rarer cases, Fl1 is completely reddish-brown (4%) or reddish-brown with a reddish apex (4%). Fl2 is mostly reddish-brown, apically slightly yellow-orange (72%); two-thirds reddish-brown, apically yellow-orange (20%); or reddish-brown-orange with an apically black and metallic greenish margin (4%). In many cases, Fl3 and the other flagellomeres are reddish-brown on top and orange (56%) or reddish (36%) on the underside. Occasionally, the upper side is slightly shiny metallic greenish (8%). The flagellomeres become lighter in colour towards the end. According to MOUSTAFA & IBRAHIM (1965), Fl1 and Fl2 are brownish-black, and those from Fl3 to the end are orange-red ventrally; three end segments of the antenna are also coloured orange-red dorsally.

The measurement of the lengths of Fl1-Fl3 shows the following results: Fl1 = 0.54 mm (SD 0.02 mm, max. 0.57 mm, min. 0.45 mm), Fl2 = 0.21 mm (SD 0.01 mm, max. 0.23 mm, min. 0.19 mm), Fl3 = 0.24 mm (SD 0.01 mm, max. 0.26 mm, min. 0.22 mm). Fl1 is about twice as long as Fl2 or Fl3 (length ratio of Fl1-Fl3 = 2.6: 1.0: 1.1). The information from DYLEWSKA (1983) that Fl1 is about 2.2 times longer than it is wide at the end, and that the following flagellomeres are about as long as wide, is largely correct. MOUSTAFA & IBRAHIM’ S (1965) assertion that Fl1 is longer than Fl2 and Fl3 can also be confirmed. Differences in antennal colouration are not specific to a particular locality but also occur within a narrower area.

C l y p e u s: The clypeus has a convex shape and is, with the exception of a puncture-free midline that occurs in some specimens, clearly punctured and – apart from a basal shagreened, microsculptured and metallic greenish-coloured area – smooth and very shiny ( Fig. 8 View Figs 8 ). This was also pointed out by DYLEWSKA (1983). The clypeus has an average length of 1.28 mm (SD 0.07 mm, max. 1.37 mm, min. 1.12 mm) and an average width of 1.98 mm (SD 0.11 mm, max. 2.26 mm, min. 1.73 mm). The length-to-width ratio is 0.64. This value differs from that of DYLEWSKA (1983), who gives a value of 0.84.

In 80%, most parts of the clypeus are slightly reddish to reddish-brown translucent ( Fig. 8 View Figs 8 ); in 8%, only the two sides of the clypeus are reddish to reddish-brown. In 12%, the clypeus is black without any red colouration. MOUSTAFA & IBRAHIM (1965) state only black as the colour of the clypeus. The reddish to reddish-brown colouring varies and is often only visible from a certain angle. The area in front of the clypeus above the labrum shows a metallic, slightly green colour. Differences in clypeus colouration also occur within a local area.

The diameters of the punctures are 20-40 μm; the distances between them are 14-40 μm. There are several smaller, more closely spaced punctures at the base of the clypeus. The clypeus is characterised by a puncture-free midline in 68% ( Figs 8 View Figs 8 a-d); in 32%, it is not puncture-free. Apically, the midline tapers into a triangular puncture-free area in more than half of all cases ( Fig. 8b View Figs 8 ).

L a b r a l p r o c e s s: DYLEWSKA (1983) stated that the labral process is triangular and has a shiny surface. In the specimens studied, the labral process is not triangular, but always trapezoidal ( Figs 9 View Figs 9 a-d). The measurements of the basal and apical width of the labral process confirm the trapezoidal shape: apical 0.17 mm (SD 0.02 mm, max. 0.21 mm, min. 0.14 mm), basal 0.70 mm (SD 0.02 mm, max. 0.74 mm, min. 0.66 mm). The length of the labral process is 0.21 mm (SD 0.01 mm, max. 0.24 mm, min. 0.19 mm). In 58%, the process is slightly apically rimmed, and in 38% clearly rimmed ( Figs 9a, c View Figs 9 ). Only in 4% is it straight. In most cases, a shine is characteristic, and long rugulae (carinae) are visible ( Fig. 9c View Figs 9 ). Differences in the shape of the labrum process also occur within a population.

M a n d i b l e: The mandible is predominantly black in colour, often with a red to reddish-brown tip and a central area that may also be reddish ( Fig. 9d View Figs 9 ).

4.3 Mesosoma

M e s o s o m a w i d t h a n d l e n g t h: The mesosoma, including the tegulae, has a width of 3.75 mm (SD 0.15 mm, max. 4.16 mm, min. 3.45 mm) and a length of 2.17 mm (SD 0.13 mm, max. 2.41 mm, min. 1.82 mm).

C o l o u r, s c u l p t u r i n g a n d p u n c t u r i n g: Thecolourofthemesosomais black, as also indicated by SPINOLA (1838). As already mentioned, DYLEWSKA (1983) considered specimens from outside Egypt in her characterisation, so these data are not comparable. The author characterises the mesoscutum as finely microsculptured, very shiny, with strong microsculpture only at the base, and coarsely punctured; the scutellum as microsculptured only laterally and coarsely punctured with a distance between the punctures of about 0.5-4 puncture diameters; and the postscutellum as coarsely microsculptured with a greasy shine and coarsely punctured, with a distance between the punctures of about 0.5-2 puncture diameters. However, a detailed analysis shows that the anterior mesoscutum is slightly shiny, shagreened, and densely and shallowly punctured ( Fig. 10 View Fig ). The puncture diameter is 20 μm anteriorly, with a puncture distance of 27 μm. The mesoscutum is shiny from the centre to the end, not as densely but more deeply punctured ( Fig. 10 View Fig ). The puncture diameter in the anterior area is 20 μm, with a puncture distance of 27 μm; in the remaining area, the puncture diameters are 14-26 μm, with puncture distances of 27- 41 μm. The scutellum is very shiny; the punctures have diameters of 14-41 μm and puncture distances of 14-27 μm ( Fig. 10 View Fig ).

P r o p o d e u m: The description of the propodeum in DYLEWSKA (1983) (roughly rugose, deeply fielded) and in GUSENLEITNER & SCHWARZ (2002) (fielded) is imprecise. The propodeum has a length of 0.58 mm (SD 0.03 mm, max. 0.65 mm, min. 0.54 mm). The structures are very variable (rugose, ridge-like, honeycomb-like, with longitudinal laminae and fan-shaped structures). There are transitions in the basal area from clear honeycomb-like structures and fielded structures to a dissolution of the structure to ridge-like structures ( Figs 11 View Figs 11 a-d). Apically, ridge-like structures can also dominate, which can transition apically into a clear fan-shaped structure with laminae ( Figs 11a, d View Figs 11 ) but also dissolve in structure.

There are four types, which occur in different frequencies: in 44%, there are some honeycomb-like structures basally, ridge-like structures or laminae towards the centre and a fan-shaped structure apically. Basal honeycomb-like structures are absent in 36% of cases, but there are numerous ridge-like elements that form no or only a fragmentary fan-shaped structure apically. In 12% of cases, there are also no clear basal honeycomb-like structures, but there are ridge-like structures that also have a fan-shaped structure apically. Basal fragmentary honeycomb-like structures, otherwise only ridge-like structures and a very fragmentary fan-shaped structure, are found in 8% of cases.

C o l o u r o f e x t r e m i t i e s: The information given by SPINOLA (1838) on the colouration of the extremities is not correct (confusion with Melitta aegyptiaca). Femur 1 and tibia 1 are red-brown, tibia 1 apically orange-red. Basitarsus 1, mediotarsi and distitarsus are reddish-brown to orange-red; all tarsi are orange-red. Femur 3, tibia 3 and all tarsi of the hind extremities are red-orange ( Figs 5a, b, Figs 6a, b View Figs 6 ). The measurements of the posterior pair of legs yielded the values shown in Table 1.

W i n g s, p t e r o s t i g m a a n d w i n g v e i n s: The wing length is 8.62 mm (SD 0.25 mm, max. 9.12 mm, min. 8.16 mm). SPINOLA (1838) characterises the wings as hyaline and the wing veins as black in colour. The wings of the examined females are light translucent and not dark ( Figs 12 View Figs 12 a-c). Only in very few cases is a very slight tint recognisable. The pterostigma shows a slight colour variation. In most cases, the pterostigma is yellow-orange to orange in the centre (88%) and yellow in a smaller percentage of cases (12%). The margin of the pterostigma is yellow-reddish in 84%, reddish in 12%, and yellowish in 4%. The statement by MOUSTAFA & IBRAHIM (1965) that the pterostigma is coloured brown in the centre is not correct.

The wing veins are in 84% yellow-reddish, in 12% yellow-orange and in 4% reddish in colour and in most cases characterised by dark margins. The information from SPINOLA (1838) that the wing veins are black is based on a confusion with Melitta aegyptiaca.

P u b e s c e n c e: DYLEWSKA (1983) characterises the pubescence very undifferentiated (creamy white hairy dorsal thorax, legs and body). The information of SPINOLA (1838) does not correspond to the characteristics of A. savignyi (thorax covered with long white hairs; white abdominal and leg hairs, including the tibial scopa; tarsi with golden yellow hairs; white hairs on the hind legs) due to the confusion with the two specimens of Melitta aegyptiaca.

The mesoscutum ( Figs 5, 6 View Figs 6 , 10 View Fig ) is in most cases (68%) only marginally hairy. There are often only a few hairs in the centre. Seventy-six per cent have yellowish-white hairs, and 24% have whitish hairs. If the centre of the mesoscutum is hairy (32% of the specimens examined), 75% of the hairs are yellowish-white and 25% are white. The scutellum and postscutellum are hairier than the mesoscutum. In 84%, the pubescence is yellowish-white, in only 12% whitish, and in 4% yellowish-golden. Concerning the mesepisternum, 44% of the specimens had whitish and 36% yellowish-white pubescence. Twelve per cent showed pure white hairs; 8% had yellowish-white hairs in the upper part of the mesepisternum and white hairs in the lower part. The trochanteral flocculus and the femoral flocculus are identical in the percentage of hair colouration (whitish-yellow 88%; whitish 8%; white 4%). The tibial scopa is golden with only a few exceptions (92%; whitish-yellow 8%). MOUSTAFA & IBRAHIM (1965) also refer to the golden hairs of the basitarsus.

4.4 Metasoma

M e t a s o m a w i d t h a n d l e n g t h: The metasoma has a width of 3.68 mm (SD 0.18 mm, max. 4.00 mm, min. 3.20 mm) and a length of 6.29 mm (SD 0.43 mm, max. 7.04 mm, min. 5.36 mm).

C o l o u r o f t e r g i t e s: The information on the tergite colouration of A. savignyi varies widely in the literature. According to SPINOLA (1838), T1-T3 are more or less reddish. The lectotype, however, has only two reddish tergites ( Fig. 4a). DYLEWSKA (1983) emphasised that T1-T3 is completely or partially reddish in colour, which is confirmed by the results obtained here. GUSENLEITNER & SCHWARZ (2002) pointed out the orange-redcoloured T1 and T2 as well as occasionally equally coloured T3. COCKERELL (1938) described A. bipartia aswanica from Egypt with three red tergites. MOUSTAFA & IBRAHIM (1965) characterised the abdomen only with basal orange colour. Tergite-specific information on colouration is not given. The statement by DYLEWSKA (1983) for the apical tergite depressions that they are translucent reddish is not quite correct, as they are clearly metallic greenish and orange.

According to the results presented here, the tergites’ colouration and pattern show considerable variation ( Figs 13 View Opposite , 14). The sequence of the examples follows an increasing red colouration:

T1: Dorsally in 60% of the specimens, the colour is orange-reddish but not spotted ( Figs 13e View Opposite , f-h; Figs 14a, c, d, h View Figs 14 ), in 16% orange-reddish and slightly spotted ( Figs 14b, c, e View Figs 14 ), in 16% orange-reddish but clearly spotted ( Figs 13 View Opposite a-d, g; Figs 14b, e, g View Figs 14 ), and in 8% yellow-orange ( Fig. 14f View Figs 14 ). Some specimens have a narrow brown-red margin apically on T 1 in front of the depression ( Figs 13b, e, f View Opposite ). On the front, 60% have a small area of reddish-brown colouring ( Figs 13b View Opposite , e-h; Figs 14 View Figs 14 b-d), 16% have a larger area of reddish-brown laterally, 16% are orange-reddish, similar to the centre ( Fig. 13a View Opposite , Fig. 14a View Figs 14 ), and 8% are yellow-orange ( Fig. 15c View Figs 15 ). The punctures on dorsal T1 have a diameter of 20 μm and a puncture distance of 20 μm. The tergite surface is very shiny.

T2: In 32%, T2 is orange-reddish and slightly or more strongly spotted ( Figs 13a, c, e View Opposite ; Figs 14b, e View Figs 14 ); in 28%, it is only orange-reddish ( Fig. 13f View Opposite ; Figs 14c, h View Figs 14 ); in 24%, it is orange-reddish with a central more or less distinct spot ( Figs 13d, g, h View Opposite ; Figs 14a, d, g View Figs 14 ); in 8%, orange-reddish with two central spots ( Fig. 3b View Fig ; Fig. 14e View Figs 14 ); and in 8%, uniformly yellow-orange ( Fig. 14f View Figs 14 ). Punctation and shine are as in T1. The depression is reddish-greenish iridescent. In 96%, there is a small black spot laterally ( Figs 15 View Figs 15 a-c), and only 4% are without a spot.

T3: Of all tergites, T3 shows the greatest variability in colouration and patterns. In 28%, black colour dominates, which is characterised by a bluish shine ( Figs 13a, b, d View Opposite ; Fig. 14a View Figs 14 ) or, in some cases, reddish parts ( Figs 13b, c View Opposite ), which are often restricted laterally. Twenty-eight per cent of T3s are orange-reddish ( Fig. 14e View Figs 14 ), either completely or with a dark area in the centre ( Figs 14a, d, g View Figs 14 ). Dark reddish-brown colouring dominates up to 20% ( Figs 13g, h View Opposite ), wherein the centre may also have a dark area. Red to brown-reddish colouring dominates in 16% ( Figs 13e, f View Opposite ; Fig. 14b View Figs 14 ), and yellow-orange in only 8% ( Figs 14f, h View Figs 14 ). The base is coloured black with a bluish shine in 64% of cases ( Figs 13b, e, g View Opposite ; Figs 14b, g View Figs 14 ), dark reddish-brown in 20% ( Figs 13f, e View Opposite ), and orange-reddish to yellow-orange in 16% ( Figs 14a, c, d View Figs 14 ). The apical margin is in 50% orange-reddish ( Figs 13a View Opposite , c-e, h; Figs 14 View Figs 14 a-e, g), in 29% reddish ( Figs 13b, f, g View Opposite ), in 13% orange-reddish and metallic slightly greenish, and in 8% yellow-orange. The depression is s orange-greenish iridescent. The punctation corresponds to T2.

T4: In 56%, T4 is black with a bluish shine ( Figs 13a View Opposite , c-h; Fig. 14b View Figs 14 ); in 20%, black with a bluish shine but a red apical margin ( Figs 14a, c, d View Figs 14 ); in 12%, also black with a bluish shine but laterally reddish brown and with a red apical margin ( Figs 13b, g View Opposite ); in 12%, yellow-orange to orange-reddish and apically somewhat reddish-brown ( Figs 14e, f, h View Figs 14 ). The punctation corresponds to T2. The depression is uniformly metallic and greenish.

T5: In 48%, T5 is black and reddish-brown ( Figs 13a, b View Opposite , d-h; Fig. 14a View Figs 14 ); in 24%, reddish-brown ( Figs 14c, f, g View Figs 14 ); in 12%, bluish-black, and laterally reddish-brown ( Fig. 13c View Opposite ; Figs 14b, d View Figs 14 ); in 12%, the base is black and metallic green with reddish depression ( Fig. 14e View Figs 14 ); in 4%, yellowish-orange mixed with some reddish-brown areas ( Fig. 14h View Figs 14 ).

A total of 7 types of tergite colouration can be distinguished:

- T1, T2: orange-reddish; T1, T2 partly spotted; T3, T4 black with bluish shine; T5 black to reddish-brown ( Figs 13a, b View Opposite ).

- T2: orange-reddish margin in front of the depression ( Figs 13c, d View Opposite ).

- T2, T3: reddish to brown-reddish, partly with a reddish margin ( Figs 13 View Opposite e-h).

- T2: orange-reddish, partly spotted ( Figs 14a, b View Figs 14 ).

- T3: orange-reddish, partly spotted ( Figs 14c, d View Figs 14 ).

- T4: orange-reddish ( Figs 14 View Figs 14 e-g).

- T5: yellow-orange with partial red-brown colouring ( Fig. 14h View Figs 14 ).

S c u l p t u r i n g a n d p u n c t u r i n g o f t h e t e r g i t e s: DYLEWSKA (1983) characterised the tergites as finely microsculptured, very shiny and with fine puncturing. The distance between the punctures is about 0.5-3.0 times the puncture diameter. According to our own measurements, the puncture diameter is usually 20 μm, with a puncture distance of 20 μm.

P u b e s c e n c e o f t e r g i t e s: SPINOLA (1838) pointed out that there are three snow-white hair bands on T2-T4, with denser pubescence on T5 with white hairs laterally and black hairs in the centre. The hairs of T6 are described as black. Here, too, the description is based on a confusion with Melitta aegyptiaca. According to DYLEWSKA (1983), T1 has whitish hairs laterally, T2-4 broader whitish apical bands, and T5 and T6 brown-black hairs. A detailed analysis shows a large variation in the tergites’ hair colouration.

- T1: yellowish-white patchy hairy (68%), whitish (16%), white (12%), patchy hairy (4%).

- T2: basal yellowish-white patchy hairy (72%), whitish (16%), white (12%).

- T3, T4: basal very short yellowish hairs (52%), basal very short whitish hairs (20%), whitish (16%), white (12%).

- T5: central brown hairs, lateral yellow-white hairs (56%); central reddish-brown, lateral yellow-white hairs (20%); central reddish, lateral yellow-white hairs (16%); central brown, lateral white hairs (4%); central reddish, lateral white hairs (4%).

- T6: reddish-brown hairs (64%), dark brown (24%), brown (12%).

- T2-T4: white closed bands (64%), yellowish-white closed bands (36%).

P y g i d i u m: AccordingtoGUSENLEITNER & SCHWARZ (2002), the pygidium is apically truncated, which is correct. The pygidium also varies in colour ( Figs 15d, e View Figs 15 ). It is in 84% black, in 8% reddish-brown, in 4% black and reddish-brown, and in 4% black with a reddish tip.

C o l o u r a n d p u b e s c e n c e o f s t e r n i t e s: The sternites also show a variation in their colouring. The sequence of examples in Fig. 16 View Figs 16 follows an increasing red and orange colouration:

- S1: reddish-brown to reddish-orange, often with a dark central area.

- S2: reddish-brown, orange-reddish or orange, often with a dark central area, basal or lateral orange to orange-reddish, often with an orange-greenish shine, depression orange-greenish iridescent.

- S3: centrally or laterally black with bluish shine or centrally reddish-brown or orange-reddish, often with small dark areas with orange-greenish shine, partly apically with narrow reddish margin, occasionally basally or laterally reddish; S3 also completely orange-reddish or orange, depression orange-greenish iridescent.

- S4: centrally black with bluish shine, partly basally and laterally reddish brown, or completely reddish brown with orange-greenish shine, partly basally orange and centrally dark coloured over a large area; S4 also completely orange or with a small basal dark spot, depression orange-greenish iridescent.

- S5: red-brown to orange-reddish, occasionally basally with a small dark spot, depression orange-greenish iridescent.

Previous descriptions of males of Andrena savignyi

A description of the males of A. savignyi is only available from WALKER (1871) under the synonym A. munda collected in Cairo. The male was characterised as follows: "black. Head, thorax, legs and base of the abdomen clothed with white hairs. Abdomen with four bands of white pubescence on the hind borders of the segments, these bands narrower on the underside; hairs brown above the tip. Wings cinereous; veins black, stigma piceous. Length of the body 5 lines" ( WALKER 1871). Five lines correspond to 10.6 mm. The description of the males by SAUNDERS (1904) refers to specimens from the Canary Islands, and the description by DYLEWSKA (1983) to a specimen from Tunisia. A detailed description of the males of the Egyptian specimens is presented here for the first time.

GUSENLEITNER & SCHWARZ (2002) characterised the males as follows: "In males the red colouration of the tergites is strongly reduced, sometimes only the depressions are lightened. In addition to the clear banding, the fielding of the midfield of the propodeum, the formation of the ‘truncated’ pygidium, which is also present here, and the at least red colouring of the tarsi of the 3rd pair of legs, it is above all the structure of the antennae that helps to characterise the species. From the 3rd flagellar segment onwards, the antennae are mostly orange-red (slightly blackened at the top) and saw-like bulges on the underside. The most dominant element of the genital capsule is the penis valve, while the gonostyli are narrow and somewhat twisted and are covered with longer hairs in the centre. The dorsal gonocoxite teeth are also distinct but short and narrow."

x Redescription of males with special focus on the variation of morphological and morphometric characteristics

6.1 Habitus, body length Figs 17a-c shows the habitus of a male from the Fayed region in dorsal and lateral view. Head and thorax are black as in the females; the antennae are dark dorsally, reddish ventrally and, in contrast to the females, saw-like on the underside (see also GUSENLEITNER & SCHWARZ 2002). Leg pairs 1 and 2 are predominantly reddish-brown, and femur 3, tibia 3, basitarsus 3, mediotarsi 3, and distitarsus 3 are reddish and sometimes even orange-red (see also GUSENLEITNER & SCHWARZ 2002). The colouration of the tergites can vary greatly in all the examined specimens from Egypt. T1 can be black to reddish-brown, with a red terminal margin and reddish-brown and metallic greenish depressions. In many cases, the black colour has a bluish shine. T2 resembles T 1 in most specimens. T3 is usually black or black to reddish-brown, while T4 and T5 are black with a greenish-bluish shine. All depressions are reddish and metallic greenish. The pubescence of the head, thorax and abdomen is usually yellowish-white, sometimes also white, as described by WALKER (1871). The abdomen has long hairs on T1 with a more or less dense hair band; on T2-T4, the entire depression is covered with dense yellowish-white hair bands.

The males have an average body length of 11.19 mm (SD 0.62 mm, max. 11.84 mm, min. 10.56 mm). This is within the range given by WALKER (1871) ( 10.6 mm). The body width of the mesosoma averages 3.29 mm (SD 0.14 mm, max. 3.51 mm, min. 3.06 mm), and that of the metasoma, 3.29 mm (SD 0.20 mm, max. 3.36 mm, min. 2.96 mm).

6.2 Head

H e a d c o l o u r, h e a d s h a p e, c o m p o u n d e y e s, o c e l l i: The head is black, but some males have a clypeus with a faint but recognisable red colouration ( Fig. 18d View Figs 18 ). As with the females, the head is slightly wider than long ( Figs 18 View Figs 18 a-d). The average head length is 2.96 mm (SD 0.12 mm, max. 3.02 mm, min. 2.73 mm); the head width is 3.44 mm (SD 0.14 mm, max. 3.58 mm, min. 3.19 mm) (head length/head width index: 0.86). As in females, the compound eyes diverge slightly. The upper inner eye distance is larger than the lower one (upper: 2.38 mm, SD 0.10 mm, max. 2.47 mm, min. 2.21 mm; lower: 2.40 mm, SD 0.13 mm, max. 2.50 mm, min. 2.11 mm). The compound eyes are smaller than those of the females and have a length of 1.88 mm (SD 0.10 mm, max. 1.98 mm, min. 1.69 mm) and a width of 0.76 mm (SD 0.04 mm, max. 0.76 mm, min. 0.72 mm). The ocelli diameter is 0.25 mm (SD 0.01 mm, max. 0.26 mm, min. 0.23 mm), similar to the females, whereas the ocelloccipital distance is 1.69 mm (SD 0.19 mm, max. 1.90 mm, min. 1.40 mm).

P u b e s c e n c e: As in females, a yellowish-white pubescence predominates in the head area. This applies to the vertex, frons and area between the antennae ( Figs 18 View Figs 18 a-d). However, there are also areas of the head with pure white hairs. Thus 36% of the males have yellowish-white hairs in the upper paraocular region and white hairs in the lower region, and 73% have white to whitish hair colouration in the genae region. According to WALKER (1871), the head hairs are white, which, in his description, applies to the entire hair colouration of the male. In most cases, the clypeus is hairier ( Figs 18 View Figs 18 a-d) than in females ( Figs 7 View Figs 7 a-c).

A n t e n n a e: Only GUSENLEITNER & SCHWARZ (2002) described the colouration of the antennae and pointed out that from Fl3, the antennae are mostly orange-red and somewhat blackened at the top. Their underside is characterised as saw-like and bulging. This is largely compatible with our results. A more detailed analysis shows the following characteristics ( Figs 19 View Figs 19 a-c), although there is a certain variability between specimens. The upper side of the antenna is black throughout, sometimes also reddish-brown ( Fig. 19b View Figs 19 ), whereas the underside is yellowish to yellowish-orange from Fl2 onwards. The colouration of the scapus is black throughout on top with a metallic, slightly greenish shine ( Fig. 19b View Figs 19 ). The pedicellus, on the other hand, is partially black or apically reddish-brown, sometimes with a metallic greenish shine ( Fig. 19c View Figs 19 ). Fl1 is black with a metallic greenish shine, often with a reddish-brown colour apically ( Fig. 19c View Figs 19 ). Fl2 is dorsally black with a metallic greenish shine, ventrally initially reddish-brown, then orange-yellowish. The reddish-brown area is more pronounced basally and tapers downwards ( Fig. 19c View Figs 19 ). The remaining flagellar segments are dorsally black to reddish-brown with a metallic, slightly greenish colouring, and ventrally clearly orange-yellowish ( Figs 19a, c View Figs 19 ).

The measurement of Fl1 to Fl3 shows the following results: Fl1 = 0.34 mm (SD 0.02 mm, max. 0.34 mm, min. 0.30 mm) (shorter than Fl1 of females), Fl2 = 0.32 mm (SD 0.02 mm, max. 0.34 mm, min. 0. 30 mm) (longer than Fl2 of females) and Fl3 = 0.34 mm (SD 0.02 mm, max. 0.37 mm, min. 0.30 mm) (longer than Fl3 of females).

C l y p e u s: As in the females, the clypeus has a convex shape, is clearly punctured with the exception of a puncture-free midline and, apart from a basal shagreened, microsculptured and metallic greenish-coloured area, is smooth and very shiny ( Fig. 18d View Figs 18 ). The clypeus has an average length of 1.17 mm (SD 0.05 mm, max. 1.20 mm, min. 1.02 mm) and an average width of 1.88 mm (SD 0.08 mm, max. 1.90 mm, min. 1.77 mm). The length-to-width ratio is 0.62.

The clypeus is black and often reddish in the apical part ( Fig. 18d View Figs 18 ). The area in front above the labrum shows a metallic, slightly green colouration. As in females, the diameters of the punctures are 20-40 μm, and the distances between the punctures are 14-40 μm. There are several smaller, more closely spaced punctures at the base of the clypeus. The clypeus is characterised by a puncture-free midline in 55% of cases ( Figs 18 View Figs 18 a-c); in 45%, it is not puncture-free.

L a b r u m: The labrum is flat and rounded, with yellowish-white (golden) hairs on the left and right sides. In the centre, there is a depression with yellowish-white (golden) hairs, which is surrounded on the left and right sides by a hair-free area.

M a n d i b l e: The mandible is black distally and red apically ( Fig. 18b View Figs 18 ).

6.3 Mesosoma

W i d t h a n d l e n g t h: The mesosoma, including the tegulae, has a width of 3.29 mm (SD 0.14 mm, max. 3.51 mm, min. 3.06 mm) and a length of 2.01 mm (SD 0.15 mm, max. 2.11 mm, min. 1.59 mm).

C o l o u r, s c u l p t u r i n g a n d p u n c t u r i n g: Thecolourofthemesosomais black, slightly shiny in the anterior shagreened area, and densely and shallowly punctured. The mesoscutum is shiny from the centre to the end, not as densely but more deeply punctured. The puncture diameter is 27 μm in the anterior third, with a puncture distance of 13 μm; in the remaining area, the puncture diameter is also 27 μm, and the puncture distances are 27-60 μm. The scutellum is very shiny, deeply punctured in the centre with a distance of up to 67 μm, and more densely spaced towards the sides.

P r o p o d e u m: The propodeum has a length of 1.35 mm (SD 0.07 mm, max. 1.46 mm, min. 1.24 mm). The structures are also very variable in males (ridge-like, with rugulae or laminae, honeycomb-like, and fan-shaped structures). There are transitions in the basal area from clear honeycomb structures to a dissolution of the honeycomb-like structure to ridge-like structures ( Figs 20 View Figs 20 a-f). Apically, ridge-like structures can also dominate, which can transition into a clear fan-shaped structure ( Figs 20a, b View Figs 20 ). There may also be a dissolution of the clear structures.

C o l o u r o f e x t r e m i t i e s: GUSENLEITNER & SCHWARZ (2002) only refer to the red colouration of the tarsi of the 3rd pair of legs. Femur 1 and tibia 1 are reddish-brown, as is basitarsus 1. The mediotarsi and the distitarsus are reddish-brown to reddish; the tarsi are reddish. The femur of the third pair of legs is reddish-brown; the tibia is reddish-brown to reddish. The tarsi are in 36% reddish-brown and in 64% orange-red ( Fig. 17). The measurement of the posterior pair of legs yielded the values shown in Table 2.

W i n g s, p t e r o s t i g m a a n d w i n g v e i n s: The wing length is 7.55 mm (SD 0.24 mm, max. 7.70 mm, min. 7.00 mm). The wings are light, translucent and not tinted ( Fig. 17). The pterostigma shows a slight colour variation. In most cases, it is yellowish in the centre (64%), and orange in a smaller percentage of cases (36%). The margin of the pterostigma is always reddish. The wing veins are in 64% yellowish-orange and in 36% orange.

P u b e s c e n c e: According to WALKER (1871), the thorax is white-haired. The specimens studied here show a variation in hair colouration: the mesoscutum, scutellum and postscutellum are in 64% yellowish-white and in 36% yellowish-golden, while the mesepisterum has whitish hairs in 64% and yellowish-white hairs in 36%.

6.4 Metasoma

W i d t h a n d l e n g t h: The metasoma has a width of 3.29 mm (SD 0.20 mm, max. 3.36 mm, min. 2.96 mm) and a length of 5.75 mm (SD 0.44 mm, max. 6.16 mm, min. 5.20 mm).

C o l o u r o f t e r g i t e s: GUSENLEITNER & SCHWARZ (2002) mentioned that in males, the red colouration of the tergites is strongly reduced and only the depressions are lightened. According to the results presented here, the colouration and patterns of the tergites also show considerable variation in the males ( Fig. 21 View Figs 21 ). This results in the following characterisation:

- T1: In 45%, T1 is black to red-brown and has a red terminal margin. In 27%, the red colouring of the terminal margin is absent. In 18%, T1 is bluish-black to reddish-brown, sometimes with a red end margin. In 9%, T1 is reddish-brown with a red end margin. In all cases, the depressions of T1 are reddish-brown and metallic greenish.

- T2: In 46%, T2 is black to reddish-brown and has reddish, usually metallic greenish areas laterally. Thirty-six per cent of specimens have a reddish end margin in addition to the black to reddish-brown colouring. In only 9% is T2 purely black to reddish brown, and in another 9%, it is bluish-black. In all cases, the depressions of T2 are reddish-brown and metallic greenish.

- T3: In 64% of the specimens, T3 is black to reddish-brown, one-third of them with reddish areas laterally, one-third with a red end margin, and one-third only black to reddish-brown. Thirty-six per cent of specimens show a black colour with a bluish shine. In all cases, the depressions of T3 are reddish and metallic greenish.

- T4: In most cases (46%), T4 is black with a bluish shine, but in 27% with a greenish shine. In 18%, T4 is black to reddish-brown, sometimes with a reddish lateral colour or with a red end marginally. Only in 9% is T4 completely bluish-black. The depressions of T4 are reddish and metallic greenish.

- T5: Like T4, T5 is in 55% black with a bluish shine, but only in 18% with a greenish-bluish shine. T5 is in 18% black to reddish-brown, either with reddish spots laterally or with a red end margin. In 9%, T5 is red to reddish-brown. The depressions of T5 are reddish and metallic greenish.

S c u l p t u r i n g a n d p u n c t u r i n g o f t h e t e r g i t e s: Thepuncturediameter is usually 20 μm, with a puncture distance of 20 μm.

P u b e s c e n c e o f t e r g i t e s: A detailed analysis shows the following variation of hair colouration of the tergites:

- T1: yellowish-white with patchy hairs (68%), whitish (16%), white (12%), patchy hairs (4%).

- T2: basally patchy hairs that are yellowish-white (72%), whitish (16%), white (12%).

- T3, T4: basally very short yellowish hairs (52%), very short whitish hairs (20%), not short whitish hairs (16%), not short white hairs (12%).

- T5: centrally brown, laterally yellow-white hairs (56%); centrally reddish-brown, laterally yellow-white hairs (20%); centrally reddish, laterally yellow-white hairs (16%); centrally brown, laterally white hairs (4%); centrally reddish, laterally white hairs (4%).

- T6: reddish-brown (64%), dark brown (24%), brown hairs (12%).

- T2-T4: white hair bands (64%), yellowish-white hair bands (36%).

P y g i d i u m: In contrast to the females, the pygidium does not vary in colouration. It is reddish ( Figs 15d, e View Figs 15 ).

G e n i t a l i a: This is the first description of the male genitalia of A. savignyi from Egypt. Genitalia can also have a slight variation in colouration and shape ( Figs 22 View Figs 22 a-f). The genitalia are relatively broad. The gonocoxes can be black, reddish-brown or reddish in colour. The relatively long apically arranged spines on the gonocoxes, which diverge slightly, are conspicuous. GUSENLEITNER & SCHWARZ (2002) characterised them as distinct but short and narrow.

The gonostyli are indented, as is typical for the genus Suandrena , and have yellowish-white to golden hairs on the apical margin (see also GUSENLEITNER & SCHWARZ 2002). The penis valve is extraordinarily broad and flat with lateral thickenings.

In KRATOCHWIL (2021b) the genitalia of Andrena cyanomicans PÉREZ, 1895 , A. fratella WARNCKE, 1968, A. maderensis COCKERELL, 1922 , A. mirna WARNCKE, 1969, A. notata WARNCKE, 1968, and A. portosanctana COCKERELL, 1922 (all Suandrena species) are illustrated in Fig. 24. WOOD (2021) published photos of the genitalia of A. planiventris DOURS, 1872 , and the newly described A. inaquosa from Palmyra ( Iran). The morphological comparison shows that the genitalia of A. inaquosa are identical to those of A. savignyi . Thus, the name A. inaquosa is a synonym, which was already assumed in KRATOCHWIL (2021b). A more detailed morphological and morphometric analysis of the type at the OLML is planned. A male from the OLML, collected in Syria in 1899 ( A. savignyi , det. Warncke), was analysed by A. Kratochwil.

x Synopsis

There is only one specimen of Andrena savignyi which can be regarded as a lectotype. The previous descriptions of the female of A. savignyi are extremely incomplete and are not sufficient to describe the species with morphological precision. A description of the male of A. savignyi was not available at all yet. An analysis of a multitude of morphological and morphometric parameters and a sufficient number of females and males of different populations from the locus classicus ( Egypt) was therefore necessary.

The results of this study show that an extraordinarily large variation of individual characteristics can be found within Egypt in different populations, but also within one population. This applies, for example, to the extent of the red colouration of the tergites, different structures of the propodeum and the pubescence colour. Nevertheless, in addition to features that clearly characterise the subgenus Suandrena ( WARNCKE, 1968) , there are also many features that clearly distinguish A. savignyi from the other Suandrena species despite variation (see KRATOCHWIL 2021b).

On the basis of these results, it is now possible to make comparisons with other populations of A. savignyi outside Egypt, within the overall distribution area from the Canary Islands to the western border of China. In particular, multivariate morphometric analyses will have to be used. These broad comparisons will be presented soon (KRATOCHWIL, in prep.).

Kingdom

Animalia

Phylum

Arthropoda

Class

Insecta

Order

Hymenoptera

Family

Andrenidae

Genus

Andrena

Loc

Andrena savignyi

Kratochwil, Anselm, Mrsn & Mrsn 2025
2025
Loc

A. bipartita

, COCKERELL 1938
1938
Loc

A. bipartita

, COCKERELL 1938
1938
Loc

A. bipartita

, COCKERELL 1938
1938
Loc

A. wollastoni

Cockerell 1922
1922
Loc

A. antilope PÉREZ (1895)

PEREZ 1895
1895
Loc

A. bicolorata

SMITH 1853
1853
Loc

A. bipartita

LEPELETIER 1841
1841
Loc

A. bipartita

LEPELETIER 1841
1841
Loc

Andrena bipartita

LEPELETIER 1841
1841
Loc

A. bipartita BRULLÉ (1839)

BRULLE 1839
1839
Loc

A. savignyi

SPINOLA 1838
1838
Loc

A. savignyi

SPINOLA 1838
1838
Loc

A. savignyi

SPINOLA 1838
1838
Loc

bicolorata

ROSSI 1790
1790
GBIF Dataset (for parent article) Darwin Core Archive (for parent article) View in SIBiLS Plain XML RDF