Unravelling Amegilla (Glossamegilla) diversity across the Wallace Line: new species, wing morphometrics, and biogeographic boundaries (Hymenoptera, Apidae)

1 Unravelling Amegilla (Glossamegilla) diversity across the Wallace Line: new species, wing morphometrics, and biogeographic boundaries (Hymenoptera, Apidae) Frédéric Carion1, Maxence Gérard1, Guillaume Ghisbain1, Thomas J. Wood2 1 Laboratory of Zoology, Research Institute for Biosciences, University of Mons, Place du Parc 20, B – 7000 Mons, Belgium 2 Naturalis Biodiversity Center, Darwinweg 2, 2333 CR, Leiden, Netherlands Corresponding author: Frédéric Carion (Fr[email protected]) Copyright: © Frédéric Carion et al. This is an open access article distributed under terms of the Creative Commons Attribution License (Attribution 4.0 International – CC BY 4.0). Research Article Abstract The Indo-Australian Archipelago is a large and biologically complex region that straddles the famous transitionary border between the Indo-Malayan and Australasian biogeographic realms called the Wallace Line. This archipelago contains much of the diversity of Amegilla (Glossamegilla) Brooks, 1988 that was previously revised during the 20th century by the Dutch entomologist Maurits Lieftinck, but no modern works on the subject have been made. We present a new revision of the subgenus Glossamegilla in the Indo-Australian Archipelago using wing morphometrics to validate existing species concepts, as well as more complete type revision and examination of undetermined material. These efforts result in the new synonymy of A. bouwmani (Lieftinck, 1944), syn. nov. with A. amymone (Bingham, 1896) and unveiled five novelties, namely A. floresiana Carion & Wood, sp. nov. (Indonesia: Flores), A. suzanneae Carion & Wood, sp. nov. (Malaysia: Sabah), A. celineae Carion, sp. nov. (Indonesia: East Kalimantan), and the unknown males of A. gigas (Friese, 1922) and A. vigilans (Smith, 1860). In total, 20 species are currently known in the Indo-Australian Archipelago. Wing morphometrics were found to strongly support most of the existing species concepts. No individual Glossamegilla species were found to occur on both sides of the Wallace Line, and the subgenus as a whole is not known to the east of the Lydddeker Line. Overall, wing morphometrics can help to determine Glossamegilla species in the Indo-Australian Archipelago but need to be coupled with other characters such as additional morphological features or biogeography. Future research involving new collections of specimens are needed to thoroughly understand the evolutionary history and biogeography of the subgenus in the Indo-Australian Archipelago. Key words: Anthophila, bees, geometric morphometrics, island biogeography, new species Introduction The wild bee subfamily Anthophorinae is a medium-sized subfamily of the family Apidae containing seven genera, with more than 750 recorded species worldwide. This subfamily was previously treated as Anthophorini, a tribe of the subfamily Apinae (Dubitzky 2007; Michener 2007; Bossert et al. 2019; Ascher and Pickering 2020). Within Anthophorinae, the two largest genera are Anthophora Latreille, 1803, with more than 420 described species, and Amegilla Academic editor: Thorleif Dörfel Received: 23 June 2025 Accepted: 5 August 2025 Published: 16 October 2025 ZooBank: https://zoobank. org/48BDF6F0-D09A-47C6-9FFD6B300243BF6E Citation: Carion F, Gérard M, Ghisbain G, Wood TJ (2025) Unravelling Amegilla (Glossamegilla) diversity across the Wallace Line: new species, wing morphometrics, and biogeographic boundaries (Hymenoptera, Apidae). ZooKeys 1256: 1–79. https://doi.org/10.3897/ zookeys.1256.162903 ZooKeys 1256: 1–79 (2025) DOI: 10.3897/zookeys.1256.162903 2 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Friese, 1897, with more than 220 described species, Amegilla being the most closely related genus to Anthophora (Dubitzky 2007; Michener 2007; Ascher and Pickering 2020; Orr et al. 2022). Numerous Amegilla were first described as Anthophora due to the close resemblance between the two genera and these similarities were only finally resolved during the 20th century (Brooks 1988); some examples of these changing taxonomic concepts are found in the works of Lieftinck (1944, 1956). Brooks (1988) was the first to attempt ordering both Anthophora and Amegilla into a system of subgenera, leading to the description of multiple new subgenera, especially for Amegilla. These subgenera are still currently in use and are solely based on both morphological characteristics and geographic distributions. This system is challenging to use in practical terms, in large part because it is based primarily on male characters, with female specimens of multiple species remaining difficult to distinguish (Brooks 1988; Michener 2007). Overall, in the system of Brooks (1988), it is sometimes easier to directly place females to species than subgenera. Within Amegilla, Brooks (1988) recognised 11 subgenera and one group of uncertain species (Dubitzky 2007). One of the subgenera established by Brooks is Glossamegilla Brooks, 1988, which the most important characteristic is the presence of a long galea (Brooks 1988). Glossamegilla includes approximately 30 species worldwide, all with an Indomalayan, East Palaearctic, or Australasian distribution. Most of the diversity is found in the Indomalayan region but the subgenus is more broadly distributed from India and China to south-eastern Asia (Brooks 1988; Ascher and Pickering 2020). The present article is geographically centred on Indonesia but also includes the Malaysian part of Borneo, Brunei, and the country of Timor-Leste (thus, the whole island of Timor is considered). This region is also known as the Indo-Australian Archipelago, which consists of approximately 20,000 islands, covering a total of more than 6 million square kilometres (Lohman et al. 2011; Tsang et al. 2019). This archipelago is biologically complex and interesting due to the transition between the Indo-Malayan and the Australasian biogeographic realms as well as the presence of four major tectonic plates converging in the region, namely the Eurasian, Indo-Australian, Philippine Sea and Pacific plates (Bird 2003; Procheş and Ramdhani 2012; Holt et al. 2013; Rueda et al. 2013; Ali et al. 2020; Ali and Heaney 2021). Many biological “lines of separation” have been drawn in this region, mainly based on works concerning vertebrates, attempting to separate the Indo-Malayan and Australasian biogeographic realms, with one of the most well-known called the Wallace Line. The line, originally drawn between the islands of Bali and Lombok as well as between the islands of Borneo and Sulawesi is also one of the most western lines to separate these two realms. However, the precise delineation of this boundary has evolved substantially since the first observations by Salomon Müller (1846), the introduction of the concept of a division between the Indo-Malayan and Australasian faunas by Wallace (1863), and its subsequent reinterpretation by Huxley (1868) – a revision, never accepted by Wallace himself, which used for the first time the term “Wallace’s Line” (Brown et al. 2013; Ali et al. 2020; Ali and Heaney 2021). The current literature does not have a consensus on where the ultimate border between these realms should be and, in fact, the placement of the border seems to depend on the studied group and the authors (Ali and Heaney 3 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line 2021). The lack of consensus for the placement of an ultimate border led to the hypothesis of a transition area, such as “Wallacea”, rather than a strict border, although this is also debated (Ali and Heaney 2021). Concerning the subgenus Glossamegilla of the Indo-Australian Archipelago, only two revisionary works are known, both by the Dutch entomologist Maurits Lieftinck (Lieftinck 1944; 1956). The former work (in 1944) was a first attempt to provide a revision of the anthophorine bees of the Malaysian and Indonesian region; all the Glossamegilla considered in this work were then treated as Anthophora, and only nine species of what came to be considered Glossamegilla were included, although seven of these taxa were newly reported for the region. The latter work (in 1956) reassigned all the species previously treated as Anthophora to the genus Amegilla and established new synonymies inside the genus. It included 17 species, of which three are not found in the Indo-Australian Archipelago, while A. vigilans (Smith, 1860) (described from Sulawesi) and A. gigas (Friese, 1922) (described from Sumatra and Wetar) were not included without any explanation. Morphometric analysis is characterised by the use of statistical methods to study a form, measuring variations in the shape and size of biological structures, as well as their covariations with other variables (Bookstein 1991; Claude 2008; Adams et al. 2013). More specifically, geometric morphometrics analyses the shape of anatomical structures by using Cartesian coordinates of anatomical landmarks after mathematically removing the effects of non-shape variations (Bookstein 1991; Claude 2008; Mitteroecker and Gunz 2009; Adams and Otárola‐Castillo 2013). This approach was already tested as a tool to discriminate taxa among various taxonomic groups in both Vertebrates and Invertebrates, including bees (Cardini et al. 2009; Ahmad et al. 2022; Casaubon and Riehl 2024; Özkan et al. 2024; Shukri et al. 2024), for which the landmarks are plotted on vein intersections (Francoy et al. 2008; Dehon et al. 2019; Martinet et al. 2019; Gérard et al. 2020; Ghisbain et al. 2021; Soipijit and Sopaladawan 2024). Wing morphometrics offers a valuable method to separate taxa, as the approach is less affected by specimen degradation over time compared to ethological, molecular, or chemical methods, which may require costly reagents, specialised laboratory equipment or rely on traits that can deteriorate or disappear in pinned specimens. In contrast, wings are still well-conserved if the collections are maintained in good condition and are cheaper to study (Francoy et al. 2008; Gérard et al. 2020; Soipijit and Sopaladawan 2024). Moreover, wing morphometrics has the advantage of being a completely non-destructive method (Gérard et al. 2020), making it a highly valuable approach for studying entomological collections. Our current work aims to introduce the use of wing morphometrics in a revision of the subgenus Glossamegilla in the Indo-Australian Archipelago for the first time as well using this approach as a complementary argument within a taxonomic framework. This revision more broadly aims to consider the two species overlooked by Lieftinck (namely A. gigas and A. vigilans), reconsider morphological characters, and provide a revised key with wider consideration of the variation displayed by members of this group. We also take the opportunity to provide distribution maps in order to characterise the distribution of the subgenus and to observe the degree to which these distributions conform to the Wallace Line. 4 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Abbreviations MSNG Museo Civico di Storia Naturale “Giacomo Doria”, Genoa, Italy 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 ZMHB Museum für Naturkunde, Berlin, Germany Materials and methods Morphometric analyses Six species of Glossamegilla, all hosted in the RMNH collection, were studied using geometric morphometrics. These species were chosen because their abundance in the collection was sufficient for ensuring robust statistical analyses (Cardini et al. 2015). The number of specimens studied per sex and per species varies from 13 to 20 (Table 1). The other species considered in this revision were insufficiently abundant to obtain a minimum of 10 male and 10 female specimens, and were therefore excluded from the geometric morphometrics analyses. All the specimens come from the Indo-Malayan region, mainly from Indonesia (all the data underlying the analyses are available on Zenodo following Carion et al. 2025). The right forewing of the specimens was photographed with a standardised millimetre scale placed under the wing. Specimens with wings in poor condition (i.e. broken, missing, or folded) were not used. When the right forewing could not be used, the left forewing was photographed and a mirror symmetry was applied with GIMP 2.10.36 to mimic the shape of a right forewing (~3.6% of wings used were the left forewing). All the pictures of the wings were taken with a Dino-Lite Edge numeric microscope coupled with the software DinoCapture 3.0 v. 1.1.0.0. The “Tps” software© produces a .tps file with the images. First, we created and merged the tps files using tpsUtil32 v. 1.83. We then digitalised eighteen landmarks following a specific pattern (Fig. 1) and set the scale for each wing using tpsDig232 v. 2.31. At this stage, each wing has its own set of Cartesian coordinates, defined as its landmark configuration. The analyses were conducted using Rstudio v. 2024.09.1 and R v. 4.4.2 as well as the packages geomorph, RRPP, ggplot2, MASS, and car. First, a Procrustes superimposition of the data was performed. The Procrustes approach involved standardising size, preserving it as a distinct variable, as Table 1. Number of specimens per species and per sex used for the morphometric analyses. Species Females Males Amegilla cinnyris (Lieftinck, 1944) 14 20 Amegilla cyrtandrae (Lieftinck, 1944) 20 20 Amegilla feronia (Lieftinck, 1944) 13 17 Amegilla insularis (Smith, 1857) 20 19 Amegilla pendleburyi (Cockerell, 1929) 20 20 Amegilla sumatrana Lieftinck, 1956 16 20 5 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line well as minimising positional discrepancies. In practice, the landmark configuration of each wing is first scaled to unit centroid size and then translated to a common centroid. The landmark configurations are finally optimally rotated around their common centroid based on the sum of squared distances between related landmarks across specimens. Wing size was estimated using centroid size, which is defined as the square root of the sum of squared distances between each landmark and the centroid of the landmark configuration (Gérard et al. 2018). Wing sizes were then compared using a one-way analysis of variance (ANOVA), followed by a Tukey’s Honest Significant Difference (post-hoc) test to identify pairwise differences between species. To assess wing shape variation a principal component analysis (PCA) was conducted, allowing the visualisation of species clustering within the subgenus. Finally, a linear discriminant analysis (LDA) was performed to assess the discriminating power of the wing among the studied species. This was followed by a Procrustes ANOVA to determine if significant differences in wing morphometry existed among species. Pairwise comparisons were then conducted to identify which species pairs were significantly different from each other. To evaluate the classification accuracy of the linear discriminant analysis (LDA), we performed a leave-one-out cross-validation (LOO-CV) procedure. In this method, each observation was iteratively excluded from the dataset and used as a test case, while the remaining data were used to build the model. The percentage of correct classification for each species was then calculated by comparing the predicted species identity to the true identity of each observation. This approach provides an estimate of the model’s ability to correctly assign individuals to their respective species based on wing morphometry. Taxonomic changes Morphological terminology follows Michener (2007). The following abbreviations are used in the species descriptions: A = antennal segments, S = metasomal sterna, and T = metasomal terga. Specimens were measured using a Zeiss SteREO Discovery.V8 binocular coupled with a Zeiss objective Plan Apo S 1.0 × FWD 60 mm. The scale of the binocular was calibrated using the same calibrated scale as for geometrics morphometrics at the Figure 1. Right forewing of Amegilla vigilans (Smith, 1860) with the eighteen landmarks digitised in order to study the wing morphometrics. 6 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line lowest zoom. All the measurements were then taken at the same zoom for all the specimens. Ratio measurements of the labrum are given using the following format “width:length” where the width is measured from the right side (near the right compound eye, when viewed frontally) to the left side of the labrum (near the left compound eye, when viewed frontally) while the length is measured from the base (at the junction with the clypeus) to the apex of the labrum. Both width and length were measured on the central axis of the labrum. Length of the specimens was measured with the same equipment in profile view from the apex of the clypeus to the apex of the last tergum. Interalar width was measured with the same equipment by measuring the shortest distance between the inner margins of the tegulae. Geometric morphometric analyses were conducted on the wings of the males of A. amymone and A. bouwmani, using the same method as for the other species (vide supra) within the framework of the synonymy between these two species. Nine males of A. amymone and 20 males of A. bouwmani were used for these analyses, all the specimens come from the island of Sumatra, in Indonesia and are hosted in the RMNH collection. Checklist of the Indo-Australian species of Glossamegilla and specimen photography The update to the checklist of the Indo-Australian Glossamegilla was compiled by examining the RMNH collection, leading to the discovery of undescribed species, and integrating this information with published works (Smith 1857; 1860; Bingham 1896; Friese 1922; Lieftinck 1944, 1956). The RMNH collection was studied because it contains the collection of Lieftinck as well as part of the collection of R. Desmier de Chenon who collected in Indonesia, providing a nearly complete and comprehensive overview of the Glossamegilla known to the Indo-Australian Archipelago. Photographs of the specimens (including the types of the new species found in the RMNH collection) were taken with using an Olympus E-M1 Mark II with a 60 mm macro lens. Additional close-ups were taken with the addition of a Mitutoyo M Plan Apo 10X infinity corrected objective lens in combination with an Olympus M.Zuiko 2x 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) 2.10.32. Post-processing of some images was made in Photoshop Elements (Adobe Systems, USA) to improve lighting to highlight specific characters. Biogeography of the subgenus Glossamegilla Label information of Glossamegilla species of the RMNH (entire collection), NHMUK (partim, type of Anthophora amymone Bingham, 1896 and type of Anthophora himalajensis var. pahangensis Meade-Waldo, 1914), MSNG (partim, type of Anthophora proserpina Gribodo, 1893, Anthophora violacea var. anthracina Gribodo, 1894 and Anthophora tetrataeniata Gribodo, 1894), OÖLM (partim, a single female of Anthophora cyrtandrae Lieftinck, 1944), and OUMNH (partim, a single female of Anthophora himalajensis Radoszkowski, 7 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line 1882, the holotype of Anthophora vigilans Smith, 1860, the holotype of Anthophora insularis Smith, 1857, and three females of A. insularis) collections (see abbreviations above) were first collected and digitised (dataset available on RMNH repository of Bakker and Creuwels 2025). The specimens without coordinates provided on the labels were then georeferenced based on the location information available using Google Maps, Google Earth, OpenStreetMap, and FloodMap in order to identify the locations of as many specimens as possible. However, some localities could not be found, and were thus not georeferenced and do not appear in distribution maps. Once the specimens were georeferenced, the distribution maps were produced using Rstudio 2024.09.1 and R v. 4.4.2 as well as the packages cowplot, googleway, ggplot2, ggrepel, ggspatial, libwgeom, sf, rnaturalearth, and rnaturalearthdata. During the production of distribution maps, some species were grouped together based on their morphological similarities and thus their supposed affinities to have a better overview of the subgeneral biogeography. These groups are therefore: (i) A. pendleburyi group, composed of A. pendleburyi (Cockerell, 1929), A. feronia (Lieftinck, 1944), A. celineae sp. nov. and A. suzanneae sp. nov.; (ii) A. cinnyris group, composed of A. cinnyris, A. insularis and A. pagdeni; (iii) A. sumatrana group, composed of A. sumatrana and A. jacobi. Other species were not classified into groups as these are generally too dissimilar morphologically for this kind of grouping. Results Wing size analyses Regarding females, the six studied species significantly differ in wing size (p-value < 0.05; Table 2, Fig. 2A). When compared pair by pair, all the species significantly differ in wing size (all p-values < 0.001, Suppl. material 1), except for A. pendleburyi - A. feronia (p-value = 0.161) and A. sumatrana Lieftinck, 1956 – A. feronia (p-value = 0.100). For males, the six studied species significantly differ in wing size as well (p-value < 0.05; Table 3, Fig. 2B). When compared pair by pair, all the species significantly differ in wing size (all p-value < 0.05; Suppl. material 2) except for A. pendleburyi – A. feronia (p-value = 0.056), A. sumatrana – A. feronia (p-value = 0.127) and A. pendleburyi – A. insularis (Smith, 1857) (p-value = 0.992). Amegilla sumatrana – A. cyrtandrae (Lieftinck, 1944) (p-value = 0.023) and A. insularis – A. feronia (p-value = 0.013) show only weakly significant differences in terms of the size of the wings (Suppl. material 2). Table 2. Results of the ANOVA test for the differentiation of female’s centroid sizes from the subgenus Glossamegilla in Indonesia. Df is the Degree of Freedom. Sum Sq is the sum of squared differences between observed data and averages. Mean Sq is the sum of squared divided by the corresponding degree of freedom. F value is the ratio of variance explained to residual variance. ** indicates p < 0.01. Df Sum Sq Mean Sq F value P-value (>F) Species 5 30.725 6.145 147.3 <0.001 ** Residuals 97 4.046 0.042 8 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Wing shape analyses The PCA plot for the females (Fig. 3A) shows the presence of three main clusters based on wing shape. One group, located in top-left of the plot consists of A. feronia and A. pendleburyi, two species that are morphologically Figure 2. Distribution of the centroid sizes for the wings in both males and females. A. Violin plot for females; B. Violin plot for males. Letters above the boxplots indicate significant differences when the letters are different. Table 3. Results of the ANOVA test for the differentiation of male’s centroid sizes from the subgenus Glossamegilla in Indonesia. Df is the Degree of Freedom. Sum Sq is the sum of squared differences between observed data and averages. Mean Sq is the sum of squared divided by the corresponding degree of freedom. F value is the ratio of variance explained to residual variance. ** indicates p < 0.01. Df Sum Sq Mean Sq F value P-value (>F) Species 5 20.276 4.055 61.330 <0.001** Residuals 110 7.273 0.066 9 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line similar outside of wing shape and closely aligned in terms of wing shape. A second group, located in top-right of the plot, includes more morphologically distinct species - A. cinnyris (Lieftinck, 1944), A. cyrtandrae, and A. sumatrana with considerable overlap between them. Finally, a third group, positioned in the bottom-right, is composed of only A. insularis. The Procrustes ANOVA for females shows significant differences in wing shape among the groups (p-value = 0.001, Table 4). Pairwise comparisons show that all groups are significantly different (all p-value < 0.04; Fig. 4A), with the exception of A. cyrtandrae and A. sumatrana (p-value = 0.181; Fig. 4A; Suppl. material 3). Amegilla cinnyris – A. sumatrana (p-value = 0.025) and A. feronia – A. pendleburyi (p-value = 0.034) show weakly significant differences in wing Figure 3. Ordination of the studied taxa along the two first axes of the Principal Component Analyses (PCA) (explaining 42.53% and 14.28% of the variance respectively for females and 43.50% and 12.66% for males). A. PCA for the females; B. PCA for the males. 16 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line or less black hairs intermixed) (Fig. 7B) as well as terga with hair bands on the apical margins, basally covered by black pubescence on the discs, or entirely covered by pale pubescence of various colour on at least T2-4 (Fig. 7F) (while the two other subgenera have black hairs on almost all the terga except T3-5 that have portion of the apical margins with appressed white hairs). Finally, Glossamegilla can be distinguished from Micramegilla, Amegilla sensu stricto (partim), Zebramegilla, Megamegilla (partim) and Aframegilla (partim) by the Indo-Malayan distribution (this group of species are mainly distributed in India Figure 8. Holotype of A. celineae Carion, sp. nov. (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Labrum in ventral view; E. Protuberance of the clypeus in profile view; F. Terga in dorsal view. 17 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 9. Pictures of the male of Amegilla (Glossamegilla) vigilans (Smith, 1860) (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Labrum in ventral view; E. Protuberance of the clypeus in profile view; F. Terga in dorsal view; G. Genitalia in frontal view; H. Sterna 7 and 8 in dorsal view. 18 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 10. Morphological comparison between the male of both colour forms of Amegilla (Glossamegilla) vigilans (Smith, 1860) (RMNH). A., C, E, G Amegilla vigilans male pale form. B, D F, H Amegilla vigilans male dark form. A., B Habitus in profile view. C, D Face in frontal view. E, F Terga in dorsal view. G, H Genitalia in dorsal view. 19 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 11. Pictures of the male of Amegilla (Glossamegilla) gigas (Friese, 1922) (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Labrum in ventral view; E. Protuberance of the clypeus in profile view; F. Terga in dorsal view; G. Genitalia in frontal view; H. Sterna 7 and 8 in dorsal view. 20 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line while the other subgenera are distributed in the Palaearctic and in Africa) and by the clypeus entirely black except sometimes with a small mark basomedially (while the other subgenera have more extended pale clypeal marks compound of an inverted T-shape or a median longitudinal line, only the species from Cape Verde Islands have an entirely black clypeus). If pale paraocular marks are present (Figs 5C, 8C, 9C, 17C, 18C, 19C, 20C, 21C, 22C, 23C, 24C), the Glossamegilla can be separated from Zebramegilla (partim), the Micramegilla (partim) and the Dizonamegilla (partim) by their Indo-Malayan distribution while the other subgenera are distributed in Africa and Western Palaearctic. T1-2 can be either entirely black haired without pale hair bands on the apical margins nor appressed brown hairs (sometimes sides of the T2 can be covered by tuft of white hairs) (Fig. 21F) or banded with pale hairs on the apical margins (sometimes the T2 have poorly developed hair bands) (Figs 5F, 7F, 10F, 18F, 19F, 20F, 23F, 24F), sometimes most of T1-2 are covered by adpressed brown hairs that can be sometimes very sparse or absent medially. If T1-2 are entirely black haired without pale hair bands on the apical margins nor appressed brown hairs (sometimes sides of the T2 can be covered by tuft of white hairs), Glossamegilla can be separated from the Dizonamegilla (partim) by the terga being entirely black haired, except for tufts of white hairs on the sides of T5. Figure 12. Holotype of Amegilla amymone (Bingham, 1896), initially published as Anthophora amymone Bingham, 1896 (NHMUK). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Terga in dorsal view. 21 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line If T1-2 are banded with pale hairs on the apical margins (sometimes the T2 have poorly developed hair bands and sometimes most of T1-2 are covered by adpressed brown hairs that can be sometimes very sparse or absent medially), the Glossamegilla can be differentiated from the Zonamegilla (partim) and Zebramegilla (partim) by (i) a body length of at least 12 mm (while the two other subgenera have a length of ≤ 9 mm), (ii) the mesosoma with ochraceous to bright orange hairs with more or less or without black hairs intermixed (Figs 5B, 7B, 10B, 18B, 19B, 20B, 23B, 24B) (while the two others subgenera have a mainly white haired mesosoma with some black hairs intermixed) and (iii) with pale hair bands on the apical margins of the terga that are generally brown (but can also be ochraceous to bright orange) while the discs is at least partially black haired or terga entirely covered by ochraceous to fulvous or brown pubescence that is sometimes denser apically, leading to slightly contrasting hair bands (Figs 5F, 7F, 8F, 9F, 10F, 17F, 18F, 19F, 20F, 22D, 23F, 24F) (while the two others subgenera have pale hair bands on the apical margins of the terga). As previously said, the females are more difficult to determine at the subgenus-level as mainly colouration-based characters are used and these characters vary substantially. However, the biogeographic regions and the size of the galea (that reaches at least the middle of the hind coxa when resting, with the galea ~3 × as long as the foretibia when measured from the apex to the maxillary palpus) are powerful diagnostic characters. Figure 13. Holotype of Amegilla bouwmani (Lieftinck, 1944), initially published as Anthophora bouwmani Lieftinck, 1944 (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Terga in dorsal view. 22 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 14. Morphological comparison between the males of Amegilla (Glossamegilla) bouwmani (Lieftinck, 1944) and Amegilla (Glossamegilla) amymone (Bingham, 1896) (RMNH). A, C, E, G. Amegilla amymone male. B, D, F, H. Amegilla bouwmani male. A, B. Habitus in profile view. C, D. Face in frontal view. E, F. Terga in dorsal view. G, H. Genitalia in dorsal view. 23 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Amegilla (Glossamegilla) floresiana Carion & Wood, sp. nov. https://zoobank.org/8A452C0C-D945-44A6-9331-F0BF8EF40198 Figs 5, 6 Type material. Holotype: Indonesia • ♀; West Flores, Rana Mese; 1300 m a.s.l.; Apr. 1958; A.M.R. Wegner leg.; RMNH, RMNH.INS.1713957. Paratypes: Indonesia • 5♀♀, 3♂♂; same collection data as for holotype; RMNH, RMNH. INS.1713958 to RMNH.INS. 1713965 • 1♀; Ruteng; 1200 m a.s.l.; 17 Feb. 1961; J.M.A. v. Groenendael leg.; RMNH, ZMA.INS.5187867. Diagnosis. Male: The male of A. floresiana sp. nov. is a species with contrasting pale hair bands on the metasoma (sometimes a little obscured by the surrounding pale pilosity) and the tergal discs usually partially covered by black pilosity. This produces an effect of hair bands that are poorly defined but clearly present. This species differs from the group of A. jacobi/A. sumatrana by the wider and more ochraceous to fulvous hair bands, the tergal discs being less covered by black pilosity than in these comparison species. Amegilla floresiana sp. nov. can be differentiated from the group of A. feronia and A. cyrtandrae by the larger size (~18 mm for A. floresiana sp. nov. while A. feronia and A. cyrtandrae are ~15–17 mm) and a more protuberant clypeus which in profile view equals or exceeds the width of the compound eye (while the clypeus is less protuberant in A. feronia and A. cyrtandrae, with a protuberance smaller than the width of the compound eye); the pale pubescence of mesosoma and metasoma also varies from ochraceous to fulvous for A. floresiana sp. nov. (while it varies from fulvous to bright orange for A. feronia and A. cyrtandrae). Finally, this species differs from A. gigas and the dark form of A. vigilans by the labrum slightly wider than long (in the two comparison species with the labrum as wide as long), a fine (not coarse) punctation of the clypeus with large interspaces of at least the diameter of a puncture (while the two comparison species have a coarse and not very sparse punctation with rather smaller interspaces of at most the diameter of a puncture but generally less), wider and very poorly defined hair bands with a small parts of the tergal discs with black hairs or no black hairs at all as well as, sometimes, a brighter pilosity (from ochraceous to bright orange-fulvous) (while A. gigas and the dark form of A. vigilans have narrower and more sharply defined hair bands with a larger black haired area on the discs that is always present as well as a less bright pilosity (from ochraceous to very slightly orange)). Female: As for the male, the female of A. floresiana sp. nov. is a species with contrasting pale hair bands on the metasoma (sometimes a little obscured by the surrounding pale pilosity) and the tergal discs partially covered by black pilosity. This produces an effect of hair bands that are poorly defined but clearly present. This species differs from the group of A. jacobi/ A. sumatrana by the wider and more colourful hair bands, the tergal discs being less covered by black pilosity than in these species. Amegilla floresiana sp. nov. can be differentiated from the group of A. feronia, A. cyrtandrae and A. suzanneae Carion & Wood, sp. nov. by the larger body size (~17–18 mm for A. floresiana sp. nov. while A. feronia, A. cyrtandrae and A. suzanneae sp. nov. are ~13– 16 mm) and a more protuberant clypeus which in profile view equals or exceeds the width of the compound eye (while the clypeus is less protuberant in A. feronia and A. cyrtandrae, with a protuberance smaller than the width of the 24 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line compound eye). Amegilla floresiana sp. nov. differs from A. gigas by a sharper and more protruding clypeal carina that thus does not appear flat (while in A. gigas the clypeal carina is very flat, not sharp at all, and does not protrude very much or not at all), more extended and ivory-yellow to yellow clypeal marks and entirely punctuate paraocular areas (while A. gigas have less extended and yellowish, tending to dark yellow, pale clypeal marks, the face is mainly dark; A. gigas have restricted ivory-white spots on the paraocular areas as well as a clear shining area between the punctation of the paraocular areas and the transition with the clypeus). Finally, A. floresiana sp. nov. can be distinguished from the dark form of A. vigilans by rather poorly defined hair bands at the apex of the terga that are wider than for A. vigilans with pale pilosity covering the majority of the discs of the terga and almost the entirety of T4 (while A. vigilans have rather sharply defined and narrower hair bands with the discs of the terga more covered by black pubescence), the pale clypeal marks less extended and ivory-yellow to yellow (while A. vigilans have more extended yellow-orange pale clypeal marks), a finer punctation of the clypeus with larger interspaces approximately the diameter of the punctures (A. vigilans have a clypeus with a coarser, less defined punctation with smaller interspaces of generally ≤ 1/2 the diameter of a puncture), paraocular areas of the same colour than the clypeal marks but the colouration is less extended than in A. vigilans (A. vigilans also has the paraocular areas of the same colour than the clypeal marks but the colouration is more extended) and a pubescence usually brighter, more fulvous to orange but can sometimes also be ochraceous, similarly to A. vigilans which have a pubescence ochraceous to slightly fulvous. Description. Male: Length of 15.5–16 mm. Interalar width of ~3.3–3.8 mm (shorter distance between the base of the tegulae). Head: Labrum rectangular, slightly wider than long (17:16–19:18, 1.7 × 1.6 mm – 1.9 × 1.8 mm, thus sometimes hardly visible, can be considered square without measurements) with two slightly protruding brown areas at the base, remaining labrum entirely yellow except for the black transverse carina at the apex of the labrum and a brownish to black narrow bands at the base. Labrum entirely, evenly, and densely punctate with interspaces of generally ≤ 1/2 the size of a puncture but sometimes more. Labrum entirely covered by not very dense but very pale ochraceous pubescence approximately as long as the diameter of an ocellus. Mandibles bidentate, both teeth blunt, not strongly sharp, the secondary tooth sometimes less developed but still visible. Mandibles ivory-yellow on the basal 1/2 and deep brown on the apical 1/2. Clypeal marks variable, clypeus either more yellow than dark or more dark than yellow but the two subrectangular black clypeal marks always present. The pale clypeal marks on the clypeus are compound of a yellow inverted T-shape that is more extended on the area between the paraocular area and the labrum. Clypeus entirely, finely, and relatively densely punctate with interspaces generally of approximately the diameter of a puncture. Clypeal carina slightly protruding but not very much, rather sharp and impunctate) vertical bar of the yellow T). Clypeus entirely, evenly, and sparsely covered by a mixture of ochraceous and black pubescence with a length of at least the diameter of an ocellus. Paraocular areas slightly protruding and yellow. Paraocular areas evenly and relatively densely punctate with interspaces of at most the diameter of a puncture, presence of an impunctate area on the side of the area, near 25 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line the margin with the clypeus (area of the protrusion) but separated from this margin by some punctures. Clypeus not strongly protuberant, in lateral view protuberance of the apical margin of the clypeus of approximately the diameter of the eye (sometimes less). Scape anteriorly yellow and posteriorly black except a narrow orange band at the apex of the scape (sometimes restricted to a small paler area or even absent). Scape entirely covered by an ochraceous to slightly fulvous pubescence approximately as long as the width of the scape and that is denser on the sides. A2 basally black and apically reddish. Anterior surface of A3 basally black (2/3) and apically reddish (1/3), posterior surface black. A4-12 paler on the anterior surface (reddish or brownish) and black posteriorly, sometimes the difference of colouration is not well marked to absent. A3-12 entirely and evenly covered by very small, hardly visible, white pilosity. A2 sparsely covered on the sides by ochraceous-fulvous or black pubescence that is longer than the pubescence of the next articles. A3 longer than wide, approximately as long as 4+5. A4 shorter than wide. A5-12 square (as long as wide) or slightly rectangular (slightly longer than wide). A13 not cylindrical, obliquely truncated at the apex. Frons to gena black (except a yellow triangle or band at the margin of the frons) and rather not evenly covered by an ochraceous to fulvous pubescence of various length that is denser near the apical insertions but not very dense in general. Some long black hairs intermixed in the pale pubescence near the ocelli. Presence of a relatively deep depression that goes from the middle ocellus to the pale mark of the frons, width slightly more than the diameter of a surrounding puncture, the carina hardly showing trough the pubescence. Genae with very long ochraceous hairs. Mesosoma: Mesosoma entirely covered by ochraceous to fulvous pubescence, with some to many black hairs intermixed in the pilosity of the dorsum. Scutum and scutellum finely and closely but not entirely punctate with interspaces generally ≤ 1/2 the diameter of a puncture. Centre of the scutum with a wide impunctate area with only a few punctures in it. Centre of the scutellum with a sharp carina that goes from the base to ~2/3 of the scutellum, base of the carina surrounded by a small impunctate area. Cuticle of the mesosoma deep black under the pilosity. Tegulae hyaline deep brown, sparsely and shallowly punctate with interspaces of at least one diameter of a puncture. Tegulae relatively densely and entirely or partially covered by ochraceous to fulvous pubescence. Wings relatively sparsely covered by dark hairs on the veins and inside the cells with a denser pilosity on the veins (most parts of the cells covered by pubescence). Apex of the wings covered with very sparse and dark brown protuberances that are not similar to the hairs. Wings translucent but slightly yellow. Cuticle of the legs dark brown to black. All legs exteriorly covered by an ochraceous to fulvous pubescence and interiorly covered with a slightly darker to black pubescence, generally the hind basitarsi exteriorly have a mix of pale and dark hairs (proportions of both variable). Tarsal claws of all the legs bidentate with subapical tooth located medially. Tarsal claws very sharp, the secondary claws are very long, 2/3 the length of the main one, sometimes as long. Main claws arched, without any tubercles on its ventral surface. Arolium between claws absent. Apex of mid tibia with one long and thick deep brown spine which is crenulate on two sides; the spine is slightly curved at the apex. Apex of hind tibia with two long and thick deep brown spines which are crenulate on both sides; spines slightly curved at the apex. 32 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line deep brown to ochraceous in fresh specimen (similarly to A. pendleburyi and A. feronia) but the integument decoloured with the time. All sterna with hair bands on the apical margin, pilosity ochraceous on S1-5, black on S6 and brownish and very dense on S7. Sides of S3-5 with ochraceous hair tufts. Integument of S1-2 ochraceous-yellow, S3-5 brown, basal 1/2 of S6 brown, apical 1/2 and S7 black. All the sterna can also be deep brown to black with apical margin pale (ochraceous) or not. S1-2 mainly impunctate with only some small areas with very close and shallow punctation on the centre of the sterna. Sides and apex of S3-4 densely and evenly punctuate with interspaces of approximately the diameter of a puncture (but sometimes more and sometimes less), basal 1/2 of these sterna very sparsely punctuate with interspaces generally several times the diameter of a puncture. S5 densely punctuate but with some interspaces larger than the diameter of a puncture. S6-7 more densely, entirely, and evenly punctuate with interspaces of at most the diameter of a puncture. Remarks. This description is based on three specimens, workers should therefore consider that intraspecific variation could occur for this species, especially for some characteristics like the clypeal marks and the colouration, or even the morphology of the mandibles. These three specimens show some variations (especially in colouration, which mainly differs for the two first antennal articles and slightly differs for the pale mark of the clypeus) but also very slight differences in punctation (mainly on the clypeus, slight changes in term of density of punctation). The male of this species is currently unknown but this species is, based on the currently recorded specimens, restricted to northern Borneo. As A. suzanneae sp. nov. is closely related to A. pendleburyi, workers should be careful concerning the identification of this group of closely related species in this region. Amegilla suzanneae sp. nov. is currently considered to be closely related to A. celineae sp. nov. and A. pendleburyi due to the morphological and biogeographical proximity (Fig. 26). New collection and genetic studies should be performed in order to characterise the relationship between these three species and maybe also with A. feronia which is also morphologically closely related to A. pendleburyi. Moreover, new expeditions would allow us to gain a better understanding of A. suzanneae sp. nov., either biogeographically or ecologically as little is known due to the small number of specimens (only the three type specimens) currently available. Etymology. Based on the name of FC’s goddaughter, Suzanne. Distribution. Amegilla suzanneae sp. nov. is distributed in the Malaysian part of Borneo (northern Borneo), in the regions of Sabah and Sarawak. Amegilla (Glossamegilla) celineae Carion, sp. nov. https://zoobank.org/B36CA2C7-75F7-4E70-91C9-258B1A8C4870 Fig. 8 Type material. Holotype: Indonesia • ♀; S. Borneo [East Kalimantan], Gn. Pandjang, Tanggarang [Gunung Panjang]; 4 Jul. 1937; Mrs. M.E. Walsh leg.; RMNH, RMNH.INS.1713956. Diagnosis. Male: Unknown. Female: Amegilla celineae sp. nov. is a species without typically contrasting hair bands on T2-4, except on T3-4 where the hair bands are slightly more 33 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line contrasting (but not as much as in the banded species). Amegilla celineae sp. nov. can be differentiated from both colour forms of A. amymone by the pale clypeal mark more extensive, not restricted to a small triangular paler mark at the apex (while A. amymone have pale clypeal marks restricted to a small triangular paler mark at the apex, sometime with a narrow line of the same colour just above) and terga without apricot-orange pubescence (while A. amymone have apricot-orange pubescence on at least T4-6, sometimes only apically on T4). Amegilla celineae sp. nov. differs from A. himalajensis (Radoszkowski, 1882) by the pale yellow clypeal marks (while the marks are dark brown in A. himalajensis) as well as the presence of black hairs intermixed in the pubescence of the mesosoma (while A. himalajensis does not have any black hairs intermixed on the mesosoma). Amegilla celineae sp. nov. can be separated from the trio A. insularis, A. pagdeni Lieftinck, 1956 and A. cinnyris by a more protuberant clypeus, in profile view the clypeus equals or slightly exceeds the width of the compound eye (while the trio have a less protuberant clypeus that in profile view is smaller than the width of the compound eye, sometimes only a bit smaller) and the outside of the basitarsus III without black hairs, entirely covered by pale pubescence (while the trio have a basitarsus III at least partially covered by black pubescence). Amegilla celineae sp. nov. differs from A. vigilans by less extended yellow clypeal marks (while A. vigilans have more extended and momre orangish pale clypeal marks), a brighter pilosity (more bright fulvous to bright orange while A. vigilans is more ochraceous to slightly fulvous), terga entirely and more evenly clothed by fulvous to orange pubescence leading to not very contrasting hair bands at the apex of the terga, except on T3-4 where the hair bands are more visible (while A. vigilans have the terga entirely clothed with pale ochraceous pubescence that is slightly denser apically, giving slightly contrasting and poorly-defined hair bands, the hair bands on T3-4 do not contrast more than these on the other terga, hair bands more consistent) as well as hind tibiae orange to pale brown (while A. vigilans have usually dark hind tibiae, neither orange nor pale brown). Amegilla celineae sp. nov. is therefore morphologically close to A. pendleburyi but differs by the mainly impunctate paraocular areas (only a very punctures visible while A. pendleburyi have entirely punctate paraocular areas), the labrum not entirely and evenly punctate with the presence of two impunctate areas on the sides of the labrum (below the two protuberances) (A. pendleburyi have a labrum entirely and evenly punctate), punctation of the clypeus coarser with smaller and hardly visible interspaces (while A. pendleburyi have a clypeus with finer punctation and broader well-visible interspaces), a metasoma with a more uneven pubescence and hair bands at the apex of the terga more contrasting (T3-4 basally covered by a small black haired area) (A. pendleburyi have a more even pubescence, the terga usually do not have any apical contrasting hair bands) as well as a larger size of ~19 mm (while A. pendleburyi is smaller, ~18 mm maximum). Description. Male: Unknown Female: Length of 19 mm. Interalar width of ~3.8 mm (shorter distance between the base of the tegulae). Head: Labrum rectangular, wider than long (8:7, 2.4 × 2.1 mm), with two slightly protruding brownish areas at the base, remaining labrum entirely brownish yellow except for the brown transverse carina at the apex of the labrum. Centre of the labrum punctate with a dense and deep punctation (interspaces of at 34 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line most the diameter of a punctures but generally smaller). Sides of the labrum less punctate with impunctate areas just below the protuberances of the labrum as wells as a less dense punctation on remaining sides with interspaces of at least the diameter of a puncture (generally larger than this diameter). A band on the centre of the labrum covered by long fulvous setae of approximately the size of the diameter of an ocellus, apex of labrum also covered by a band of (denser) hairs that are of the same colour but smaller. Mandibles not bidentate (bi-dentation not visible) and very blunt, almost square, basal 1/2 darkened yellow, other 1/2 dark brown, and the apex black. Clypeus mainly reddish brown with dark clypeal marks compound of two large sub-rectangular areas, pale clypeal marks on the clypeus resumed to a yellow-orangish inverted T-shape that is a more extended on the area between the paraocular area and the labrum. Clypeus densely and entirely punctate with a coarse and deep punctation (interspaces of ≤ 1/2 the diameter of a puncture but generally smaller). Clypeal carina protruding, slightly blunt but not very blunt and impunctate (vertical bar of the yellow T). Clypeus entirely, evenly but sparsely covered by fulvous pubescence with the hairs slightly longer than the diameter of an ocellus. Paraocular areas protruding and yellow, mainly impunctate except at the base, near the insertions of the antennae (punctation similar to the clypeus). Protuberance of the apical margin of the clypeus of approximately the diameter of the eye (or slightly less). Scape anteriorly dark orange or slightly reddish, posteriorly dark brown to black and covered by fulvous to slightly orange pubescence that is longer than the width of the scape. A2 dark brown to slightly reddish. Anterior surface of A3 mainly orangish brown, basally with a paler orange area and apically with a narrow ivory-white line; posterior surface dark brown to black. A4-5 basally dark brown and apically orange, ending in a narrow ivory-white line on the anterior surface, posterior surface dark brown to black. A6-12 brown-orange on anterior surface and posterior surface dark brown to black. A3-12 entirely covered by very small, hardly visible, white pilosity. A2 anteriorly and apically with a small tuft of very small (but larger than the pilosity on the other articles) fulvous hairs. A3 longer than wide, slightly longer than 4+5+6. A4-7 shorter than wide. A8-11 squared, as long as wide or very slightly shorter. A12 not cylindrical, obliquely truncated at the apex. Frons to gena black (except a yellow triangle at the margin of the frons) and not evenly covered by fulvous pubescence of various length that is denser near the apical insertions. Presence of some long black hairs near the ocellus. Genae with very long white to slightly yellow hairs. Mesosoma: Mesosoma entirely covered by bright fulvous-orange pubescence, without any black hairs intermixed in the pilosity of the dorsum. Scutum entirely, finely, and closely punctate with interspaces generally ≤ 1/2 the diameter of a puncture. Presence of a depression, the width less than the diameter of a surrounding puncture, on the centre of the scutum that does not go to the end of the scutum. Presence of two small but deep depression on the scutum, on each side, near the tegulae, symmetrical. Scutellum finely and densely punctate with interspaces of generally ≤ 1/2the diameter of a puncture. Presence of a small sharp carina at the centre of the scutellum that does not go on all the length of the scutellum. Cuticle of the mesosoma deep black under the pilosity. Tegulae hyaline brown-fulvous, sparsely and shallowly punctate with interspaces of twice (sometimes more) the diameter of a puncture. Wings sparsely covered by dark hairs on the veins and inside the cells with a denser pilosity on the 35 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line veins. Apex of the wings covered with very sparse and dark brown protuberances that are not similar to the hairs. Wings translucent but slightly yellow. All legs with an brownish orange cuticle, except for the hind tibiae that are exteriorly a bit darker and the basitarsi that are deep brown. All legs exteriorly covered by bright orange-fulvous pubescence and interiorly covered with a slightly darker pubescence. Tarsal claws of all the basitarsi bidentate with the second tooth situated on the middle of the main claw. Apex of mid tibia with one long and thick almost hyaline brown spine which is crenulate on one side and smooth on the other side; the spine is curved at the apex. Apex of hind tibia with two long and thick almost hyaline brown spines which are crenulate on one side and smooth on the other side; spines curved at the apex. Metasoma: T1 entirely and evenly covered by short fulvous pubescence except on the side where there are longer hairs of the same colour. T2 almost entirely covered by short fulvous pubescence, except a little patch of short black hairs on the centre. Discs of T3-4 at least partly covered by short black pubescence, the pale pubescence on the apical part forms hair bands that are wider at the centre of the terga. Some fulvous hairs are intermixed with the black hairs of T3-4, the hairs become longer when going towards the sides of T4. T5 mainly black-haired with short black hairs, laterally with two fulvous hair tufts composed of sparse hairs. T1-5 entirely, evenly, and sparsely punctate with shallow punctation and interspaces several times the diameter of one puncture. Integument of the terga mainly black but some parts of some apical margins are brown to ochraceous it is therefore highly probable that the apical margins were originally deep brown to ochraceous in fresh specimen (similarly to A. pendleburyi and A. feronia) but the integument decoloured with the time. All sterna mainly glabrous, hairs limited to sparse band at the apex of some sterna (2,4–6). Integument of all the sterna dark brown to orangish but not dark. S1-3 with only a few sparse punctures on the discs, the apical margins and sometimes the sides more densely punctate but with interspaces of approximately the diameter of a puncture. S4 more densely punctate at the apex and less densely at the base, interspaces become wider towards the base with only a few sparse punctures. S5 densely punctate at the apex and on the sides but interspaces become wider towards the base with only a few sparse punctures. S6-7 entirely, evenly, and densely punctate with interspaces of at most the diameter of a puncture (generally less). Remarks. This description is based on only one specimen, determinators should therefore consider that intraspecific variation could occur for this species, especially on some characteristics like the clypeal marks, the colourations, or even the morphology of the mandibles. The holotype seems to have lost pilosity on the mesosoma: the description about the pilosity of the mesosoma may therefore not be representative of the species. Consider the colour of the cuticle of the apical margin of the terga in fresh and collection specimens as decolouration can occur in collection specimens. The male of this species is currently unknown but this species seems, based on our available specimen, to be distributed in South Borneo. However, as only one specimen is known, the distribution of this species on the island of Borneo cannot be known precisely. As A. celineae sp. nov. is morphologically close to A. pendleburyi, determinators should be careful regarding the identification of this group of closely related species in this region. 36 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Amegilla celineae sp. nov. is currently considered as closely related to A. suzanneae sp. nov. and A. pendleburyi due to the morphological and biogeographical proximity (Fig. 26). New collection and genetic studies should be performed to better characterise the relationship between these three species and maybe also with A. feronia which is also morphologically closely related to A. pendleburyi. Moreover, new collections will permit to have a better understanding of A. celineae sp. nov., either biogeographically or ecologically as little is known due to a small number of specimens (only the holotype) currently available. Etymology. Based on the name of FC’s partner in life, Céline, who provides much support. Distribution. Amegilla celineae sp. nov. seems, based on our available specimen, distributed only on the island of Borneo: the collector of the holotype wrote “S. Borneo” on the label, but the name of the location provided leads to eastern Borneo, in the Indonesian part of the island. “S. Borneo” was therefore interpreted as being the Indonesian part of Borneo as a whole. Amegilla (Glossamegilla) vigilans (Smith, 1860) Fig. 9 Material examined. Indonesia • 1♂; South East Sulawesi, nr Sanggona, 1 km W of Base Camp Gn Watuwila [Sanggona]; 200 m a.s.l.; 12–15 Oct. 1989; C.v. Achterberg leg.; Malaise trap; RMNH, RMNH.INS.1713942 − 1♂; [labels unreadable]; RMNH, RMNH.INS.1689433. Diagnosis. Male: The male of A. vigilans usually do not have very contrasting hair bands at the apex of the terga, the tergal discs are entirely covered by pale ochraceous pubescence intermixed with sparse black hairs, the pale pubescence a little bit more dense at the apex of the terga but not forming a very contrasting hair bands (integument normally ochraceous at the apex, accentuating the pubescence, giving an impression of hair bands). Amegilla vigilans can be separated from A. himalajensis by the deep black dark clypeal marks that are sharply defined compared to the paler ivory-yellow to slightly salmon pale clypeal marks (while A. himalajensis have dark brown clypeal marks that does not contrast from each other), a dorsal pubescence of the mesosoma with black hairs intermixed (while A. himalajensis does not have any black hairs intermixed on the mesosoma) and all the terga entirely covered by ochraceous pubescence (while A. himalajensis only have T1-2 partly or wholly and sides of T3 covered by orange pubescence). Amegilla vigilans can be distinguished from both colour forms of A. amymone by the less extended pale clypeal marks that are either ivory-yellow or orange (but never ivory-white like in A. amymone), the two sub-rectangular dark clypeal marks (absent in A. amymone, dark clypeal marks different than two sub rectangular areas), an ochraceous pubescence of the mesosoma (black to dark brown with many black hairs intermixed in A. amymone) and an absence of apricot-orange pubescence on the terga (while A. amymone have apricot-orange pubescence at least on T46). Amegilla vigilans can be distinguished from the trio A. insularis, A. pagdeni, and A. cinnyris by a more protuberant clypeus of approximately the width of the compound eye (while the trio have a clypeus smaller than the width of the compound eye) and a basitarsus without any black pilosity, the pilosity entirely 37 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line ochraceous (while the trio have at least some black hairs on the basitarsus III). Finally, A. vigilans can be separated from A. pendleburyi by the more extended, ivory-yellow to slightly salmon, pale clypeal marks (A. pendleburyi with clypeal mark less extended, especially the central line which is narrow, unlike in A. vigilans where the central mark is too thick to be labelled as a line), the pilosity that is paler, rather ochraceous to slightly yellow (while A. pendleburyi have a more fulvous-orange pilosity) as well as the terga entirely covered with a pale ochraceous pubescence that is denser apically, giving slightly contrasting and poorly defined hair bands (A. pendleburyi have terga entirely and more evenly covered by fulvous-orange pilosity, leading to the absence of contrasting hair bands). The dark form of A. vigilans is a banded form of Amegilla close to A. gigas concerning the habitus but can be separated from this species by a less protuberant clypeus of, in profile view, approximately the width the compound eye (while A. gigas have, in profile view, a clypeus that exceeds the width of the compound eye), pale orangish clypeal mark that are less extended leading to two sub-rectangular dark clypeal marks (while A. gigas have yellow, very expanded, pale clypeal marks and dark clypeal marks restricted to two small black to slightly reddish marks and two thin brown lines), a pilosity less bright in general (more ochraceous-yellow for A. vigilans while the pilosity is more fulvous to slightly orange in A. gigas) as well as a dorsal pubescence of the dorsum of the mesosoma with more black hairs intermixed (fewer black hairs intermixed in the pilosity of the dorsum of the mesosoma in A. gigas). Description. Male: Length of 17.5–18.5 mm. Interalar width of ~3.3–3.8 mm (shorter distance between the base of the tegulae). Head: Labrum rectangular, slightly wider than long (18:17–21:20, 1.8 × 1.7 mm – 2.1 × 2 mm, thus sometimes hardly visible, can be considered squared without measurements) with a protruding base (with two brown or yellow areas, one on each side of the base) giving a curved appearance to the labrum in lateral view. Remaining labrum yellow or brownish orange (dark form), except the black transverse carina at the apex of the labrum and a narrow deep brown band at the base, brown band at the base absent in the dark form. Labrum entirely, evenly, and densely punctate with interspaces of the size of a puncture or less. Labrum entirely covered by a not very dense and ochraceous pubescence that is equal or shorter than the diameter of an ocellus. Mandibles bidentate and blunt, giving teeth a rounded apex, yellow on the basal 2/3 and deep brown to black on the apical 1/3 (darkened yellow to brown on the base for the dark form). Clypeus mainly yellow or orange with two subrectangular dark marks that can be black to slightly reddish. Clypeus entirely, finely (not very coarsely but can be coarse sometimes) and densely punctate with interspace of at most the diameter of a puncture but generally less. Clypeal carina slightly protruding, not reduced, punctate and rather blunt. Clypeus entirely, evenly, and sparsely covered by ochraceous hairs of various length with some long darker hairs intermixed. Paraocular areas protruding, yellow or orange. Paraocular areas evenly and densely punctate with interspaces ~1/2 the diameter of a puncture (sometimes hardly visible due to the pale colouration of the integument). Clypeus not strongly protuberant, in lateral view, apical margin of the clypeus of approximately the size of the diameter of the eye. Scape anteriorly yellow or darkened orange and posteriorly black. Scape entirely covered by ochraceous pubescence of various length that is denser on the sides and 38 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line less dense anteriorly and posteriorly. A2 black on the basal 1/2 and slightly orange on the apical 1/2. A3 black on the basal 2/3 and orange on the apical 1/3 anteriorly, posteriorly black. A4-13 paler (orange to brown) anteriorly and black posteriorly. A3-13 entirely and evenly covered by very small, hardly visible, white pilosity. A2 sparsely covered by ochraceous or black hair tuft that is longer than the pubescence of the next articles. A3 longer than wide, longer than 4+5 but shorter than 4+5+6. A4 shorter than wide. A5-10 square, as long as wide. A1112 rectangular, slightly longer than wide (can be considered squared without measurements). A13 not cylindrical, obliquely truncated at the apex. Frons to gena black (except a yellow or orange triangle at the margin of the frons) and rather not evenly covered by ochraceous pubescence of various length that is denser near the antennal insertions. Long black hairs intermixed in the pale pubescence near the ocelli. Presence of a relatively deep depression that goes from the middle ocellus to the pale mark of the frons, width variable. The depression hardly showing through the pubescence. Genae with very long white to ochraceous hairs (especially long under the eyes). Mesosoma: Mesosoma entirely covered by ochraceous and dense pubescence with a variable number of black hairs intermixed (but always with at least some black hairs present) in the pilosity of the dorsum. Scutum and scutellum finely and closely punctate with interspaces of generally at most the diameter of a punctures (but can be wider on some areas of scutum and scutellum). Cuticle of the mesosoma deep black under the pilosity. Tegulae hyaline deep brown, sparsely and shallowly punctate with interspaces of at least one diameter of a puncture. Tegulae relatively densely and entirely or partially covered by ochraceous pubescence. Wings relatively sparsely covered by dark hairs on the veins and inside the cells with a denser pilosity on the veins (most parts of the cells covered by pubescence). Apex of the wings covered with very sparse and dark brown protuberances that are not similar to the hairs. Wings translucent but slightly yellow. Cuticle of all the legs deep brown to dark brown, never pale. All legs exteriorly covered by ochraceous to slightly fulvous pubescence and interiorly covered with a slightly darker to black pubescence, hind basitarsi without black hairs exteriorly. Tarsal claws of all the basitarsi bidentate with the base of the second tooth situated on the middle of the main claw. Tarsal claws very sharp, the secondary claws are very long, 2/3 the length of the main one. The main claws arched but without any tubercles. Apex of mid tibia with one long and thick deep brown (dark form) to black (pale form) spine which is crenulate on both sides; the spine is slightly curved at the apex. Apex of hind tibia with two long and thick deep brown (dark form) to black (pale form) spines which are crenulate on both sides; spines slightly curved at the apex. Metasoma: For the pale form, T1-6 entirely covered by short ochraceous pubescence that is denser on the apical margin for T1-5, leading to slightly contrasting hair bands on these terga. Absence of dark hairs on discs of T1-6 for the pale form. For the dark form, disc of T1 basally and centrally covered by short black pubescence, laterally and apically covered by ochraceous short pubescence that is denser apically creating a wide but poorly contrasted hair band. T2-5 covered by black pubescence on the disc and covered by ochraceous pubescence on the apical margin, leading to a strongly contrasted hair band. T6 entirely covered by brown pubescence. T1-6 entirely, evenly, and sparsely punctate with shallow punctation and interspaces of at least the diameter of 39 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line one puncture. T7 (pygidial plate) with two spine-like structures. Integument of T1-6 deep brown to dark brown, sometimes the apical margins is paler (brown to ochraceous) it is therefore probable that the apical margins were originally deep brown to ochraceous in fresh specimen (similarly to A. pendleburyi and A. feronia) but the integument decoloured with the time. All the sterna entirely covered by short and relatively dense ochraceous pilosity (sometimes the S1-3 not entirely covered). Integument of sterna brown to deep brown with ochraceous apical margin (except sometimes on S1-2). S1-2 mainly impunctate with only some small areas with very close and shallow punctation on the centre of the sterna. Sides and apex of S3-4 densely and evenly punctuate with interspaces of approximately the diameter of a puncture (but sometimes more and sometimes less), basal 1/2 of these sterna very sparsely punctuate with interspaces generally several times the diameter of a puncture. S5-7 entirely, densely, and shallowly punctate with interspaces of approximately the diameter of a puncture. S8 with some very small yellow hairs on the base, spine of the spiculum rounded and not curved (Fig. 9H). Gonocoxae mainly glabrous with only a few sparse white hairs. Intern face of the apex of gonocoxae curved. Genitalia yellow to orange, not very dark (Fig. 9G). Remarks. This description is based on only two specimens; determinators should therefore consider that intraspecific variation could occur for this species, especially on some characteristics like the clypeal marks, the colourations, or even the morphology of the mandibles. However, two forms exist inside the species A. vigilans, namely the dark and the pale forms. These two forms do not show significant structural differences but differs on colourations and the hair bands of the metasoma (Fig. 10). Moreover, the genitalia of both forms are the same (Fig. 10G, H) and both forms are therefore considered as conspecific. The colour variation observed in this species fits a broad pattern of variation in pubescence seen across other bees on Sulawesi, such as Xylocopa Latreille, 1802 (van der Vecht 1953), though this would benefit from focused study. A few specimens of this species (female of both forms and male of the dark form), hosted in the RMNH collection, were determined by Lieftinck in 1955 and 1956. It is therefore surprising that Lieftinck (1956) did not consider A. vigilans, even if the species was already known at the time; the reasons for the absence of this species in his work remain unknown. Distribution. Amegilla vigilans still appears to be restricted to the island of Sulawesi. However, the specimens studied in this paper extend the distribution of the species to the north of the island while the type was collected in the south. Amegilla (Glossamegilla) gigas Friese, 1922 Fig. 11 Type material. Lectotype: Asia Arch. [Indonesia] • 1♀; Wetter [Pulau Wetar]; 3 Apr. 1901; Kühn leg.; ZMHB, http://coll.mfn-berlin.de/u/837c94. [images examined only] Additional material examined. Timor [Timor-Leste] • 1♂; Same; 27 Dec. 1965–14 Jan. 1966; R.N. Ierreira leg.;RMNH.INS.1713966. Remarks on the type series of Friese. Friese (1922: 61) mentioned two females in his type series, one from Wetar (Lesser Sunda islands) and one from 40 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Sumatra (Deli, eastern Sumatra). However, these two specimens are on each side of the Wallace Line which is, with the current information on the distribution of Glossamegilla, doubtful. Moreover, the ZMHB seems to only host the specimen from Wetar while the specimen from Sumatra was not found (P. Rasmont, pers. comm., 2 December 2024). Therefore, without further information, the specimen from Sumatra cannot be considered as conspecific with A. gigas, especially because of its doubtful locality compared to everything that is currently known on the subgenus Glossamegilla. The specimen from Wetar (March 1901, leg. Kühn) conserved in the ZMHB (examined by photograph) is considered to be the lectotype following an unpublished designation by P. Rasmont in 2024. This designation decisively fixes the locus typicus as the island of Wetar. Diagnosis. Male: The male of A. gigas shows strongly contrasting hair bands, with only black hairs on the discs T2-4 while T1 is entirely covered by pale fulvous pubescence. Amegilla gigas differs from A. jacobi and A. sumatrana by the wider and more colourful (fulvous in A. gigas, white to slightly ochraceous in A. jacobi and A. sumatrana) hair bands at the apex of the terga as well as a slightly smaller black-haired area. Amegilla gigas also differs from A. feronia and A. cyrtandrae by the more strongly protuberant clypeus of, in profile view, ~1.3 × bigger than the width of the compound eye (A. feronia and A. cyrtandrae have a less protuberant clypeus that is, in profile view, smaller than the width of the compound eye) and a larger body size of ~18 mm (while A. feronia and A. cyrtandrae are ~15–17 mm maximum). Amegilla gigas can be distinguished from A. floresiana sp. nov. by the labrum as long as wide (1:1, while A. floresiana sp. nov. have a labrum slightly wider than long; ~17:16–19:18), a coarser punctation of the clypeus that is not very sparse with interspaces of at most the diameter of a puncture (A. floresiana sp. nov. have a fine punctation that is also very sparse with large interspaces of at least the diameter of a puncture) as well as the tergal hair bands narrower and more sharply defined with the blackhaired area on the tergal discs larger and more consistently present (A. floresiana sp. nov. have wider and poorly defined tergal hair bands at the apex of the T2-4 with a small black-haired basal part the discs, sometimes some or all the terga do not have a black-haired area and are entirely covered by the pale pubescence). Finally, A. gigas can be distinguished from the dark form of A. vigilans by a more protuberant clypeus of ~1.3 × bigger than the diameter of the eyes (1:1 in A. vigilans), yellow clypeal marks that are very extended with only a small dark mark and one narrow brown line on each side of the clypeal carina (A. vigilans have pale yellow to orange less extensive pale marks and two sub-rectangular black marks), a pilosity in general more bright (rather fulvous to slightly orange in A. gigas, ochraceous in A. vigilans) as well as a dorsal pubescence of the mesosoma without or with only few black hairs intermixed (while A. vigilans have more black hairs intermixed in the dorsal pubescence of the mesosoma). Description. Male: Length of 18 mm. Interalar width of ~3.6 mm (shorter distance between the base of the tegulae). Head: Labrum square, as wide as long (1:1, 2 × 2 mm), with two slightly protruding pale brownish areas at the base, remaining labrum entirely lemon-yellow except for the dark brown to black transverse carina at the apex of the labrum. Labrum entirely but sparsely punctate with interspaces wider than the diameter of a puncture, the interspaces become narrower while going towards the apex of the labrum (interspaces of ~1/2 the diameter of the punctures). 41 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Labrum entirely covered by yellow pubescence longer than the diameter of an ocellus. Mandibles bidentate and slightly blunt (not very sharp) at the apex, the bases lemon-yellow, similarly to the labrum. Face mostly lemon-yellow, only the clypeus with four dark brown to black mark (two lines and two subrectangular areas) that resume the dark clypeal marks. Clypeus entirely and evenly punctate with relatively small interspaces of approximately the diameter of a puncture. Clypeal carina slightly protruding, blunt and impunctate. Clypeus mainly covered by black and very sparse hairs, except at the apex, near the labrum where the pubescence is yellow and denser. Paraocular areas protruding and yellow, almost entirely punctate (same punctation as on the clypeus) except some wider clear interspaces on the margin of the paraocular areas and on their centres. Clypeus protuberant, protuberance of the margin of the clypeus of ~1.3 × bigger than the diameter of the eyes (while A. vigilans have a clypeus of approximately the size of the eye). Scape anteriorly orangish yellow and posteriorly black, covered entirely by fulvous pilosity. A2-13 black or at least very dark and covered by very small white hairs that are hardly visible. A3 longer than wide, longer than 4+5 but slightly shorter than 4+5+6. A4-12 squared, approximately as long as wide or a little bit longer than wide. A13 not cylindrical, obliquely truncated at the apex. Frons to gena black (except a yellow triangle at the margin of the frons) and covered (but not evenly) by fulvous pubescence. Mesosoma: Mesosoma entirely and evenly covered by fulvous to slightly orange pubescence, without any black hairs intermixed in the pilosity of the dorsum. As the pubescence is dense, the punctation underneath can sometimes not be seen. Scutum and scutellum finely and closely punctate (deep punctures) with very small interspaces (≤1/2 the diameter of a punctures). Cuticle under the pilosity deep black. Tegulae hyaline deep brown. Wings sparsely covered by dark hairs on the veins and inside the cells with a denser pilosity on the veins. Apex of the wings covered with very spars and dark brown protuberances that are not similar to the hairs. Wings translucent but slightly yellow. All legs with a dark brown cuticle, exteriorly covered by fulvous to slightly orange pubescence and interiorly covered by dark brown to black pubescence. Tarsal claws of all the basitarsi bidentate. Apex of mid tibia with one long and thick brown spine which is crenulate on one side and smooth on the other side. Apex of hind tibia with two long and thick brown spines which are crenulate on one side and smooth on the other side. Metasoma: T1 entirely covered with long fulvous pubescence that is denser and shorter on the apical margin, forming a hair band that does not contrast very much in the surrounding pilosity. Discs of T2-4 dark haired with whitish to fulvous hair bands on the apical margins. Disc of T2 laterally with fulvous hair tufts covering all the length of the disc but that does not expand very much transversally. Discs of T3-4 laterally with small, not very expanded, fulvous hair tuft. T5 and following entirely black haired. Discs of T1-5 sparsely punctate with shallow punctations and interspaces generally between 1.5 to twice the diameter of a puncture. Apical margin of T1-5 impunctate. T6-7 entirely and less sparsely punctate, punctation of T6 as shallow as previous terga but punctation of T7 deeper. T7 ending in two protruding spines at the apex with a curve between the two spines. Integument of the terga entirely black except for T6-7 that have a dark brown integument. S1-6 mainly black haired. S1-4 with tuft of not very expanded ochraceous hairs on the sides of the apical margin, S5 black haired and S6 with a band of ochraceous hairs on the apical margin. S1-3 with a deep brown 48 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line specimens (RMNH) collected in 1996 by R. Desmier de Chenon. Moreover, this species was considered to be coming from Siam (Thailand) in the summary table of the biogeography of this group by Lieftinck (1956). However, the materials examined by Lieftinck in this same paper all come from continental Malaysia and Lieftinck’s listing is therefore erroneous. In the group of A. sumatrana and A. jacobi, A. sumatrana seems to be restricted to north Sumatra while A. jacobi seems to be restricted to Java (mainly distributed in west Java (Fig. 28). Figure 17. Holotype of A. cinnyris (Lieftinck, 1944) (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Labrum in ventral view; E. Protuberance of the clypeus in profile view; F. Terga in dorsal view. 49 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Fig. 29 shows the distribution of A. himalajensis a widespread species distributed from Laos and Thailand to northern Sumatra and including continental Malaysia. Amegilla hanitschi is distributed in continental Malaysia as well as in Sumatra (across all the island) (Fig. 29). Concerning A. cyrtandrae, this species is distributed mainly in Java but can also be found in Bali (to the east of Java) (Fig. 29). Amegilla anthracina seems, with the current information, restricted on the islands of Tanimbar while Amegilla tetrataeniata seems restricted to the island of Timor. Finally, A. vigilans is distributed in the island of Sulawesi, the terra typica, on the eastern side of the Wallace Line and seems to be found all over Figure 18. Holotype of A. cyrtandrae (Lieftinck, 1944) (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Labrum in ventral view; E. Protuberance of the clypeus in profile view; F. Terga in dorsal view. 50 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line the island (Fig. 29; Smith 1860). Indeed, the type is known from Makassar, in the south-west of the island while the species seems to be more widespread overall based on the new data available (Fig. 29; Smith 1860). With the current biogeographic information of the species, A. vigilans can be considered as restricted to Sulawesi while also being the only known Glossamegilla of the island. The known distribution of A. himalajensis is enlarged, similarly to A. pagdeni, to the island of Sumatra compared to the distribution given by Lieftinck Figure 19. Holotype of A. elephas (Lieftinck, 1944) (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Labrum in ventral view; E. Protuberance of the clypeus in profile view; F. Terga in dorsal view. 51 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line (1956) thanks to one specimen (RMNH) coming from north-east Sumatra, near the Malaysian peninsula collected by R. Desmier de Chenon in 1996. Moreover, two other specimens, collected by Hagen on an unknown date were determined by Lieftinck in 1955 and 1956 (Lieftinck 1956: 21). The locality labels of the two specimens stated “?Sumatra”, re-enforcing the possibility of the presence of this species on the island and, therefore, in Indonesia. Amegilla himalajensis, as well as A. pagdeni, may have been overlooked in the Figure 20. Holotype of A. feronia (Lieftinck, 1944) (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Labrum in ventral view; E. Protuberance of the clypeus in profile view; F. Terga in dorsal view. 52 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line past due to a lack of recording on Sumatra, the northern mountainous areas of which can be challenging to sample. Amegilla proboscidea is restricted to the island of Simalur, on the western side of Sumatra (Fig. 30) but A. elephas is distributed across Sumatra and in continental Malaysia (Fig. 30). On the eastern side of the Wallace Line, two more species can be found, A. floresiana sp. nov. which is mainly distributed on the island of Flores as well as A. gigas that can be found on the island of Timor (suspected male, Fig. 30) but also on the island of Wetar (for the lectotype previously designated; Friese 1922). Figure 21. Holotype of A. jacobi (Lieftinck, 1944) (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Labrum in ventral view; E. Protuberance of the clypeus in profile view; F. Terga in dorsal view. 53 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Revision of Lieftinck’s (1956) key 13 antennal articles, 7 terga…Males 12 antennal articles, 6 terga…Females Key to the males 1 Body length of 19 mm or more. First 3 terga basally with black tomentum/ pilosity and fulvous hair bands along the posterior margin (Fig. 19F). In profile view, anterior margin of clypeus projecting beyond anterior margin of compound eye by a distance greater than the diameter of the compound eye (Fig. 19E) ........................................ A. elephas (Lieftinck, 1944) – Body length not exceeding 18 mm. Pubescence variable. Face variable, anterior margin of clypeus either projecting or clearly projecting by less than the diameter of the compound eye ......................................................2 2 Dorsal part of the metasoma with short emerald to turquoise pilosity evenly distributed on the terga. Dorsal pubescence of the mesosoma with black, grey, and pale green hairs. Labrum basally with a brown stripe, the remainder ivory white while the clypeal and mandibular markings are yellow ..................................................... A. hanitschi (Meade-Waldo, 1914) – Dorsal part of the metasoma without any green pilosity, the pubescence is generally pale (white to fulvous) to dark (dark brown to black) or a mix of both (Figs 6F, 9F, 10F, 11F, 12D, 13D, 17, 18F, 20F, 21F, 22F, 23F, 24F)........3 3 At least terga 2–4 with poorly to strongly defined pale hair bands on their apical margins, these contrasting the darker hairs of the tergal discs ; for some species the bands are incomplete in the centre (Fig. 21F); the hair bands can be wide or narrow; the tergal discs are generally clothed with black or darker pubescence and in some species, T1 mostly to entirely covered with pale pubescence (Figs 6F, 10F, 11F, 18F, 20F, 21F, 24F) .......4 – Terga without any hair bands on their marginal areas or if pubescent then this not strongly contrasting in the surrounding pilosity; either with an entirely (sometimes evenly) distributed pale pubescence on the first 4 terga or with other types of pubescence (Figs 14E, F, 17F, 22D) .......................10 4 Metasoma mainly black-haired with very narrow sharply defined white to slightly fulvous hair bands; the hair bands can be interrupted in the centre or not (Figs 21F, 24F) ....................................................................................5 – Metasoma with broader hair bands and fewer dark hairs on the tergal discs; hair bands generally with darker colourations (fulvous to ochraceous-orange but never white) (Figs 6F, 10F, 11F, 18F, 20F) ....................... 6 5 Terga 1–4 apically with complete, narrow, and sharply defined bands of short hairs that can be whitish to slightly fulvous; T1 with longer hairs of the same colours that expand from the sides of the tergum to the centre (pubescence more dense on the side than on the centre); sides of T2-3 with tuft of whitish to fulvous hairs that do not expand (Fig. 24F) ... A. sumatrana Lieftinck, 1956 – T2-4 apically with incomplete hair bands of whitish to fulvous hairs, except sometimes complete only on T4; hair band on T1 limited to only a few short pale hairs that are hardly visible; T1 with longer whitish to slightly fulvous hairs on the sides that do not expand very much; T2-3 without tuft of paler hairs (Fig. 21F) ....................................... A. jacobi (Lieftinck, 1944) 54 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line 6 Larger species, ~18 mm. Clypeus strongly protuberant; in profile view, anterior margin of clypeus projecting beyond anterior margin of compound eye by a distance of approximately the diameter of the compound eye or more (Figs 6E, 9E, 11E). Apical hair bands usually paler in colours, ochraceous, but sometimes fulvous to bright orange (Figs 6F, 10F, 11F) ...........7 – Smaller species, with a maximum of 15–17 mm. Clypeus less protuberant; in profile view, anterior margin of clypeus projecting beyond anterior margin of compound eye by a distance smaller than the diameter of the compound eye (Figs 18E, 20E). Apical hair bands darker, fulvous to bright orange but never ochraceous (Figs 18F, 20F) .............................................9 7 Labrum slightly wider than long (17:16–19:18); punctation of the clypeus fine, not coarse, and very sparse with large interspaces (at least the diameter of a puncture) (Fig. 6D). Hair bands on the apex of metasoma wider and very poorly defined with a small basal part with black hairs or black hairs absent (Fig. 6F). Pilosity variable, ochraceous to bright orange (Fig. 6B) ...............................................A. floresiana Carion & Wood, sp nov. – Labrum as long as wide (1:1); punctation of the clypeus coarse and not very sparse, rather with smaller interspaces (at most the diameter of a puncture) (Figs 9D, 11D). Hair bands on the apex of metasoma narrower and more sharply defined with a larger basal part with black hairs that are always present (Figs 10F, 11F). Pilosity ochraceous to very slightly orange (Figs 10B, 11B) ..............................................................................................8 8 Clypeus more protuberant; in profile view, anterior margin of clypeus projecting beyond anterior margin of compound eye by a distance greater than the diameter of the compound eye (~1.3 compound eye diameter) (Fig. 11E). Pale clypeal mark yellow and very expanded, 1 small black to slightly reddish mark and 1 thin brown line on each side of the carina that resume the black clypeal marks (Fig. 11C). Pilosity in general more bright, fulvous to slightly orange (Fig. 11B, F). Dorsal pubescence of the mesosoma without or with only few black hairs intermixed (Fig. 31A) ..... A. gigas (Friese, 1922) – Clypeus less protuberant; in profile view, anterior margin of clypeus projecting beyond anterior margin of compound eye by a distance of approximately the diameter of the compound eye (Fig. 9E). Pale clypeal marks more orangish and less expanded with the presence of 2 subrectangular dark clypeal mark (1 on each side of the carina) (Fig. 10D). Pilosity in general less bright, rather ochraceous (Fig. 10B, F). Dorsal pubescence of the mesosoma with more black hairs intermixed (Fig. 31B)… (partim, dark form) ................................................ A. vigilans (Smith, 1860) 9 Pale clypeal marks ivory-white and covering a larger surface of the clypeus (Fig. 18C). Paraocular areas less protruding (Fig. 18E). Labrum square-shaped, as long as wide (Fig. 18D). Tergal hair bands broad, thus dark hairs occupying relatively small area of terga (Fig. 18F). Tergal margins predominantly black with the rim narrowly paler hyaline-brown, never extensively coloured orange or pale brown, any colouration obscured by and barely showing through the pubescence. Dorsal pubescence of the mesosoma paler and with many black hairs intermixed (Fig. 32A) ............. ...................................................................... A. cyrtandrae (Lieftinck, 1944) – Pale clypeal marks yellow and covering a smaller surface of the clypeus (Fig. 20C). Paraocular areas more protruding (Fig. 20E). Labrum rectangular-shaped, 55 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line clearly longer than wide (Fig. 20D). Tergal hair bands relatively narrow, thus dark hairs appearing abundant on the tergal discs (Fig. 20F). Tergal margins normally extensively paler orangish to pale brown, visible through and around the pubescence (the integument can be decoloured in collection specimen). Dorsal pubescence of the mesosoma less pale, more orange, with fewer black hairs intermixed (Fig. 32B) .............................A. feronia (Lieftinck, 1944) 10 Dark clypeal marks reddish to orangish, not very sharply defined compared to the orangish yellow pale marks and never deep black (Fig. 33A). Dorsal pubescence of mesosoma bright orange-rufous, without black hairs intermixed (Fig. 33B). First 2 terga partly or wholly covered with fulvous hairs (sometimes the sides of the T3 with small tuft), the remaining terga black-haired .................................... A. himalajensis (Radoszkowski, 1914) – Dark clypeal marks neither reddish nor orangish, rather deep black to only very slightly reddish, with their outer margins sharply defined (Figs 14C, D, 17C, 22C). Dorsal pubescence of mesosoma generally less bright with at least some black hairs intermixed (Figs 14A, B, 17B, 22B) ......................11 11 Metasoma with the apical terga (from T4 onwards) entirely with apricot-orange pubescence (sometimes T4 only 1/2 covered by this pubescence), T1-3 variable in pubescence but always abundant black hairs on the discs of T1-3 (Fig. 14E, F). Dorsal pubescence of the mesosoma black to very dark brown or even slightly ochraceous (Fig. 14A, B). Clypeus mainly ivory-white and with the pale median marking large, without defined sub-rectangular dark marks (Fig. 14C. D). Restricted to Sumatra ........... 12 – Terga typically without distinctive apricot-orange pubescence on T4-6, either because all terga are uniformly covered with orange pubescence, or pale pubescence is more sombre or pale (Figs 9F, 17F, 22D). Dorsal pubescence of the mesosoma paler, never black or dark brown and generally brighter than ochraceous (Figs 9B, 17B, 22B, 31B). Clypeus with well-defined sub-rectangular dark marks, the central medial marking therefore smaller and more restricted (Figs 9C, 17C, 22C). Distribution variable ...13 12 T1-3 predominantly black-haired, with only hints of apricot-orange pubescence (Fig. 14E). Dorsal pubescence of mesosoma rather black to very dark brown (Fig. 14A). Generally T4 basally clothed with black pilosity and apically with apricot-orange pubescence (Fig. 14E)… dark form ................. ........................................................................A. amymone (Bingham, 1896) – T1-3 apically with apricot-orange pilosity and generally all basally covered with black pubescence (Fig. 14F). Dorsal pubescence of mesosoma brighter, mixture of black, brown, and ochraceous hairs (Fig. 14B). T4 and after entirely covered with apricot-orange pubescence (Fig. 14F)… pale form (bouwmani) ...........................................A. amymone (Bingham, 1896) 13 Clypeus more protuberant; in profile view, anterior margin of clypeus projecting beyond anterior margin of compound eye by a distance of approximately the diameter of the compound eye (Fig. 9E). Basitarsus III without any black pilosity, entirely clothed with pale pubescence that can be either ochraceous/slightly yellow or fulvous/slightly orange (Figs 9B, 22B) .....14 – Clypeus less protuberant; in profile view, anterior margin of clypeus projecting beyond anterior margin of compound eye by a distance smaller (at least slightly) than the diameter of the compound eye (Fig. 17E). Basitarsus III at least partly, if not entirely, covered by black pubescence (Fig. 17B) .........15 56 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line 14 Pale clypeal marks more extended and slightly salmon compared to the next species (Fig. 9C). Pilosity in general paler, rather ochraceous to slightly yellow (including the outside of basitarsus III) (Fig. 9B). Terga entirely clothed with pale ochraceous pilosity that is denser apically, giving slightly contrasting poorly-defined hair bands (Fig. 9F). Restricted to Sulawesi... (partim, pale form) ................................................ A. vigilans (Smith, 1960) – Pale clypeal marks less extended and yellow (Fig. 22C). Pilosity brighter, more fulvous-orange (including the outside of basitarsus III) (Fig. 22B). Terga entirely and more evenly clothed with a bright fulvous-orange pilosity, leading to not really contrasting hair bands (Fig. 22D). Distributed on Sumatra, continental Malaysia, and all across Borneo .... A. pendleburyi (Cockerell, 1929) (NB, take care, as specimens on Borneo may belong to 1 of the 2 newly described species morphologically close to A. pendleburyi) 15 Clypeus less protuberant, anterior margin of clypeus projecting by ~1/2 the diameter of the eye (Fig. 34A). Pale clypeal marks ivory-yellow and covering a slightly smaller surface of the clypeus compared to the next 2 species (Fig. 34B). Transverse carina at the apex of the labrum narrowly emarginate medially, laterally with 2 small protruding tubercles (Fig. 34C). Pale pubescence covering entirety if T1-5, hairs on remaining terga deep black. General appearance wider and more robust ........................... ............................................................................A. insularis (Smith, 1857) – Clypeus more protuberant, anterior margin of clypeus projecting by slightly less than the diameter of the eye but more than 1/2 the diameter (Fig. 17E). Pale clypeal marks ivory-white or maize-yellow to buff-yellow and covering a slightly larger surface of the clypeus compared to insularis (Figs 17C, 34F). Transverse carina at the apex of the labrum without a central depression, if 1, not surrounded by 2 protruding teeth (Fig. 34D, E) ....................16 16 Pale face marks maize-yellow or buff-yellow (Fig. 34F). T3-4 partly, the succeeding terga entirely, covered with black hairs (sometimes hardly visible on T3 but generally well visible on T4) ........A. pagdeni Lieftinck, 1956 – Pale face marks ivory-white (Fig. 17C). T1-5 entirely covered with orangish pubescence, the succeeding terga, occasionally also part of the T5, black-haired (generally T6 and succeeding hardly visible because they are hidden by T5) (Fig. 17F) ................................... A. cinnyris (Lieftinck, 1944) Key to the females 1 Body length of 21 mm or more .....................................................................2 – Body length not exceeding 19 mm ...............................................................3 2 Pale clypeal marks lemon-chrome, sometimes slightly orangish, dark clypeal marks slightly brownish/reddish black to only black and sharply defined (Fig. 19C). T1-4 basally clothed with black hairs and apically clothed with wide yellow to slightly orange hair bands (more extended at the centre on T3-4), remaining terga entirely covered by the yellow-orange pubescence (Fig. 19F). Distributed in Malaysia and Sumatra ...................... ........................................................................... A. elephas (Lieftinck, 1944) – Pale clypeal marks cinnamon-rufous, dark clypeal marks brownish reddish and poorly defined (Fig. 23C). Terga mainly black-haired, except for narrow orange hair bands at the apex of the first 4 terga and an apical 57 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line orange hair fringe at the apex of T5 (Fig. 23F). Restricted to Sumatra (only known from Lasikin on the island of Pulau Simeulue).................................. ...................................................................... A. proboscidea Lieftinck, 1956 3 Dorsal surface of metasoma evenly covered by short bright green pilosity (emerald to golden green) that tend to yellow in collection due to decolouration. Pale clypeal mark restricted to only an inverted T-shaped yellow mark. Pubescence of the tibia III 1/2 ochraceous-orange (upper part) and 1/2 black (lower part) on the outer surface of the tibia. Green hairs are present on the dorsal part of the mesosoma and on the head, mixed with hairs of other colours ............................ A. hanitschi (Meade-Waldo, 1914) – Dorsal part of the metasoma without any green pilosity, the pubescence is generally pale (white to fulvous) to dark (dark brown to black) or a mixture of both type of colouration; hair bands present or not (Figs 5F, 7F, 8F, 9F, 10F, 12D, 13D, 17F, 18F, 20F, 21F, 22D, 24F) .................................................4 4 At least T2-4 with very contrasting poorly-defined to sharply-defined hair bands on the posterior margin; for some species the bands are incomplete in the centre; the hair bands can be wide or thin; the terga are generally basally clothed with black or darker pubescence and sometime T1 is entirely clothed with pale pubescence (Figs 5F, 7F, 8F, 10F, 18F, 20F, 21F, 24F) ...................................................................................5 – Terga without any hair bands on their posterior margins, sometimes most terga entirely covered by pale pubescence that is slightly denser apically, leading to hair bands that are not very contrasted compared to the surrounding pilosity; metasoma either with an entirely (sometimes evenly) distributed pale pubescence on the first 4 terga or with other type of pubescence (Figs 8F, 9F, 12D, 13D, 17F, 22F) ................................................. 12 5 Metasoma mainly black-haired with very thin sharply defined white to slightly fulvous hair bands; hair bands can be interrupted in the centre or not depending on the species (Figs 21F, 24F) ............................................. 6 – Metasoma with fewer black hair, with wider and more poorly-defined hair bands at the apex; hair bands generally with darker colours (fulvous to ochraceous-orange but never white) (Figs 5F, 7F, 10F, 18F, 20F) ...............7 6 First 4 terga with complete white to slightly fulvous hair bands; T1 entirely covered with sparse and long fulvous hairs; sides of T2 (and occasionally T3) with a fulvous hair tuft (Fig. 24F). Sometimes the habitus looks wider than in A. jacobi but both species are generally similar in width ................. ......................................................................... A. sumatrana Lieftinck, 1956 – T1-4 with incomplete white to slightly fulvous hair bands at the apex, sometimes complete to nearly complete on the T4 and restricted to only few sparse hairs on the T1; T1 with sparse long fulvous hairs that are interrupted at the centre of the tergum; absence of fulvous hair tuft on the side of T2-3 (Fig. 21F). Sometimes the habitus is less wide than in A. sumatrana but both species are generally similar in width ... A. jacobi (Lieftinck, 1944) 7 Larger species, ~17–18 mm. Clypeus a bit more strongly protuberant; in profile view, anterior margin of clypeus projecting beyond anterior margin of compound eye by a distance of approximately or slightly greater than the diameter of the compound eye (Figs 5E, 9E) ........................................8 – Smaller species, restricted ~13–16 mm. Clypeus less strongly protuberant; in profile view, anterior margin of clypeus projecting beyond anterior 64 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Based on the results obtained previously, wing morphometrics seems to be an efficient method that can reliably determine the studied species of the subgenus Glossamegilla in Indonesia. However, with an accuracy of ~87.1% for the males and ~94.2% for the females, the males having a higher chance of being misassigned than the females solely based on wing morphometrics, the method is currently not perfect and needs to be used in combination with other morphological or biogeographical features in order to present better results, as suggested for other groups (Dehon et al. 2019; Gérard et al. 2020). For example, in both males and females, it is the pair A. cyrtandrae-A. sumatrana that cannot be significantly separated from one another using morphometrics, Figure 24. Holotype of A. sumatrana Lieftinck, 1956 (RMNH). A. Labels of the specimen; B. Habitus in profile view; C. Face in frontal view; D. Labrum in ventral view; E. Protuberance of the clypeus in profile view; F. Terga in dorsal view. 65 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 26. Distribution map for four morphologically close species in the Indo-Australian Archipelago: A. celineae Carion, sp. nov. (blue points), A. feronia (Lieftinck, 1944) (red points), A. pendleburyi (Cockerell, 1929) (green points), and A. suzanneae Carion & Wood, sp. nov. (orange points). The empty points represent specimens collected before and in 1950 while the full points represent specimens collected in 1951 and after. Figure 25. Distribution map for A. amymone (Bingham, 1896) in the Indo-Australian Archipelago. This map contains both colour forms of A. amymone with the dark form (true amymone) represented by the blue points and the pale form (A. bouwmani (Lieftinck, 1944) syn. nov.) represented by the red points. The empty points represent specimens collected before and in 1950 while the full points represent specimens collected in 1951 and thereafter. 66 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 27. Distribution map for three morphologically close species in the Indo-Australian Archipelago: A. cinnyris (Lieftinck, 1944) (blue points), A. insularis (Smith, 1857) (red points) and A. pagdeni (Lieftinck, 1944) (green points). The empty points represent specimens collected before and in 1950 while the full points represent specimens collected in 1951 and thereafter. Figure 28. Distribution map for two morphologically close species in Indo-Australian Archipelago: A. jacobi (Lieftinck, 1944) (blue points) and A. sumatrana Lieftinck, 1956 (red points). The empty points represent specimens collected in 1950 and before, the full points represent specimens collected before and in 1951 and squares represent specimens with an unknown collection date. 67 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 29. Distribution map for six species in the Indo-Australian Archipelago: A. anthracina (Gribodo, 1894) (blue points), A. cyrtandrae (Lieftinck, 1944) (red points), A. hanitschi (Meade-Waldo, 1914) (green points), A. himalajensis (Radoszkowski, 1882) (orange points), A. tetrataeniata (Gribodo, 1894) (grey points) and A. vigilans (Smith, 1860) (rose points). The empty points represent specimens collected before and in 1950 the full points represent specimens collected in 1951 and after, and squares represent specimens with an unknown collection date. Figure 30. Distribution map for four species in the Indo-Australian Archipelago: A. elephas (Lieftinck, 1944) (blue points), A. floresiana Carion & Wood, sp. nov. (red points), A. gigas (Friese, 1922) (green points) and A. proboscidea Lieftinck, 1956 (orange points). The empty points represent specimens collected before and in 1950 while the full points represent specimens collected in 1951 and after. 68 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 31. Dorsal pubescence of the mesosoma seen in profile view. A. Mesosoma of one Amegilla gigas (Friese, 1922) male (RMNH); B. Mesosoma of one Amegilla vigilans (Smith, 1860) male (RMNH). Figure 32. Dorsal pubescence of the mesosoma seen in profile view. A. Mesosoma of one Amegilla cyrtandrae (Lieftinck, 1944) female (RMNH); B. Mesosoma of one Amegilla feronia (Lieftinck, 1944) female (RMNH). Figure 33. Two general views of one Amegilla himalajensis (Radoszkowski, 1882) male (RMNH) to illustrate key characteristics of the species. A. Face of the specimen seen in frontal view; B. Dorsal pubescence of the mesosoma seen in profile view. 69 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 34. Several views of one Amegilla insularis (Smith, 1857) male (RMNH), one Amegilla cinnyris (Lieftinck, 1944) female (RMNH), and one Amegilla pagdeni Lieftinck, 1956 male (RMNH) to illustrate key characteristics of these species. A, B, C. Images of A. insularis; D. Image of A. cinnyris. E, F. Images of A. pagdeni; A. Head in profile view showing the protuberance of the clypeus. B, F. Face in frontal view showing the clypeal marks. C, D, E. Labrum in ventral view showing the difference of emargination on the carina at the apex. but these two species show strong morphological differences (the metasomal hair bands are clearly visually distinct) and they do not overlap in term of distribution, with A. sumatrana restricted to the island of Sumatra and A. cyrtandrae restricted to the islands of Java and Bali (Figs 28, 29). The same combination can be used on other overlapping group such as A. feronia-A. pendleburyi where A. feronia have contrasting hair bands on the metasoma and is restricted to 70 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Figure 35. Face of a female of Amegilla vigilans (Smith, 1860) dark form (RMNH) showing the punctuation of the face. Figure 36. Face of two female species in frontal view. A. Face of Amegilla cyrtandrae (Lieftinck, 1944) (RMNH); B. Face of Amegilla feronia (Lieftinck, 1944) (RMNH). Figure 37. Two views of one Amegilla pendleburyi (Cockerell, 1929) female (RMNH) to illustrate key characteristics of the species. A. Face of the specimen with the punctuation of the clypeus and the paraocular areas; B. Labrum of the specimen with its punctuation. 71 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Java while A. pendleburyi does not have contrasting hair bands on the metasoma and is widely distributed in Sumatra, continental Malaysia, and Borneo (Figs 20, 22, 26). On the other hand, A. insularis can be significantly differentiated from A. cinnyris (either males or females) based on wing morphometrics, these two species are very close morphologically, the males are almost cryptic and can be mistaken. Moreover, they can be both found in sympatry on the island of Sumatra (Fig. 27; Lieftinck 1956). These encouraging results can lead to a cheaper and less time-consuming approach for broadly supporting the species-level classification of Glossamegilla compared to genetic approaches that require more expensive equipment and consumables (Kozmus et al. 2011). Moreover, the study of centroids size unveiled that the wings of the males are more likely to have similar sizes than for the females (Fig. 2, Tables 2, 3, Figure 38. Face of one Amegilla insularis (Smith, 1857) (RMNH) to illustrate the clypeal marks of the females. Figure 39. Face of two female species in frontal view. A. Face of Amegilla cinnyris (Lieftinck, 1944); B. Face of Amegilla pagdeni Lieftinck, 1956. 72 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line Suppl. materials 1, 2). This tendency seems to relatively match with body size of the tested species, the males being in general more similar in terms of body size compared to the females (Lieftinck 1944; 1956). Overall, the number of specimens assessed ranged from 13 to 20 per species and per sex, which means that some groups were below an optimal minimum of 20 but still above the 10 specimens, which is the threshold under which inaccuracies begin to be especially impactful (Cardini et al. 2015). Despite our encouraging results, further collections and collaborations with other museums should be conducted in order to increase the number of specimens studied to limit even more the inaccuracies as well as to expand the number of species studied. Similarly to A. cyrtandrae and A. sumatrana, both forms of A. amymone are not significantly different based on wing morphometrics (Table 7). We were able to evaluate nine males of the dark form, though this is low for geometric morphometrics (Cardini et al. 2015). The females were not assessed as their number was even lower. Therefore, new collections and work on other museum collections should be done in the future in order to have stronger statistical results on the wing shape of both forms of this species, even if the current tests support the hypothesis of a synonymy. Modern collections of specimens and genetic studies would be ideal to gain a better understanding of this species and its colour variations. To conclude with the morphology of A. amymone, the males appear to have more colour variation compared to females, which is the opposite of that observed in most Anthophorine bees species (e.g., Brooks 1983; Wood and Praz Figure 40. Distribution map with all the occurrences of the studied specimens from the subgenus Glossamegilla. This map allows one to have a better understanding of the distribution limits of the subgenus as well as the potential sampling biases. The empty points represent specimens collected before and in 1950, the full points represent specimens collected in 1951 and after, and squares represent specimens with an unknown collection date. 73 ZooKeys 1256: 1–79 (2025), DOI: 10.3897/zookeys.1256.162903 Frédéric Carion et al.: Glossamegilla across the Wallace Line 2024). Indeed, just one female of the pale form has been found (the holotype of A. bouwmani, Lieftinck 1944; 1956), while numerous males of the pale form were found with females and males of the dark form. Conclusions Geometric morphometrics can help to determine the tested species of Glossamegilla in the Indo-Australian Archipelago but needs to be coupled with other characters such as the other morphological features or the biogeography. This paper expands the list of known species in the Indo-Australian Archipelago by seven, permitting the elaboration of a new revised and currently complete key. We conclude that the Wallace Line is not a biogeographic border for the subgenus, contrarily to the Lydekker Line, but is a border for the species individually, increasing the knowledge about these “lines of separation” in this archipelago. Acknowledgements We would like to thank Pierre Rasmont for the access to photographs of the lectotype of Amegilla gigas hosted in the ZMHB and his comment about the absence of the Sumatran specimen in this same collection. We also thank Joseph Monks (NHMUK), Esther Ockermüller and Martin Schwarz (OÖLM), James Hogan (OUMNH), Roberto Poggi and Maria Tavano (MSNG), and Frederique Bakker (RMNH) for access to their respective collections. We also thank the anonymous reviewer for comments which substantially improved the manuscript. Additional information Conflict of interest The authors have declared that no competing interests exist. Ethical statement No ethical statement was reported. Use of AI No use of AI was reported. Funding Travel to Oxford and Genoa was supported by a grant from the Stichting Funds Pontium (RSIN 862206546). Maxence Gérard was supported by a postdoctoral grant “Chargé de Recherches” from the Fond de la Recherche Scientifique (F.R.S.-FNRS). Author contributions Conceptualization: TJW, GG, FC. Formal analysis: MG, FC, TJW. Methodology: MG, TJW, FC. Supervision: TJW, GG. Validation: TJW. Visualization: FC, MG. Writing - original draft: FC, TJW. Writing - review and editing: TJW, GG, MG. Author ORCIDs Frédéric Carion https://orcid.org/0009-0001-4944-4353 Maxence Gérard https://orcid.org/0000-0002-2485-0662