As part of an ongoing collaboration between the Université libre de Bruxelles (NJV, LM, JMM) and the Departamento Biología Animal, Edafología y Geología of the University of La Laguna (CR) on apple tree pollinators, as well as the biogeography of the Canary Islands bees and the development of new taxonomic tools on the bees of Europe, field surveys have been conducted on the islands of Tenerife, Fuerteventura and Lanzarote in April 2021.

On 18.iv.2021, a male Pseudoanthidium was photographed by NJV (Fig. 1A, B) and collected with a hand net at Haría (Lanzarote); a female was collected with a hand net at the same site on the same day by JMM. The following day, on 19.iv.2021, two male Pseudoanthidium specimens were collected at Teguise (Lanzarote). All specimens were first assumed to be new island records of P. canariense, until a closer examination of pinned specimens under the microscope challenged this view. The type specimens curated at the ULB Agroecology Lab entomological collection (Brussels, Belgium) will be transferred to the DZUL entomological collection curated by CR at the University of La Laguna (Tenerife) upon the publication of this manuscript.

Figure 1. 

Ecology and distribution of Pseudoanthidium jacobii in the Canary Islands A, B male nectaring on an inflorescence of Asteriscus intermedius (Asteraceae) (Photos NJ Vereecken) C female collecting pollen on an inflorescence of A. intermedius (Photo B Jacobi) D female nesting in a pre-existing cavity located in a volcanic lava rock (Photo B Jacobi) E distribution map showing all occurrence records available to us and relevant to both P. canariense and P. jacobii in the Canary Islands.

On 12.ii.2023, a female Pseudoanthidium was photographed at Bco. Valle del Palomo (Lanzarote) by MPG, and two female specimens were collected by MPG at Haría (Lanzarote), one on 21.ii.2023 (Bco. Valle del Palomo), and another one on 16.iv.2023 (Bco. de Elvira Sánchez). These specimens were sent to CR to be pinned, prepared for identification and deposited at the entomological collection curated by CR at the University of La Laguna (Tenerife). On 19.iv.2023, CR collected one female Pseudoanthidium species at Bco. de Elvira Sánchez (Haría, Lanzarote) now included in the DZUL entomological collection at the University of La Laguna (Tenerife). Publicly available occurrence data on wild bees across the Canary Islands reveal a significant disparity in sampling (GBIF 2023a): Tenerife (3,211 records), Gran Canaria (787), La Palma (560), Lanzarote (449), La Gomera, (383), Fuerteventura (263) and El Hierro (16). These figures are consistent with the historical patterns published by Hohmann et al. (1993). As a result, although additional collection efforts are needed across the entire island group, we can reasonably conclude that the range of the new Pseudoanthidium species does not include Gran Canaria and Tenerife.

On 12.iii.2023 and on 16.iii.2023, two female Pseudoanthidium specimens were photographed by BJ at Mácher (Lanzarote) (Fig. 1C, D).

Finally, G. Peña Tejera notified CR of another observation made and published on Flickr.com on 7.iii.2020 at Betancuria (Fuerteventura) by photographer L. Mullins.

DNA extractions were performed on single legs from one male and one female specimen of this new species using Nucleospin tissue DNA extraction kits (Macherey - Nagel). A 658 base pair fragment of the mitochondrial gene cytochrome c oxidase (CO1) was amplified using the primers Lep-F1 and Lep-R1 under the PCR conditions described in Hebert et al. (2004). PCR products were purified using a combination of exonuclease and FastAP thermosensitive alkaline phosphatase (Fermentas) and sequencing was performed using the same primers as those used for PCR reactions.

Sequences were edited using Geneious (Kearse et al. 2012) and aligned with eight other CO1 sequences representing the closely related species Pseudoanthidium canariense (Mavromoustakis, 1954), P. scapulare (Latreille, 1809), P. nanum (Mocsáry, 1880), P. palestinicum (Mavromoustakis, 1938), P. tenellum (Mocsáry, 1880), P. cribratum (Morawitz, 1875) and P. stigmaticorne (Dours, 1873), as well as the more distantly related P. reticulatum (Mocsáry, 1884) as an outgroup (see Suppl. material 1 for taxon list). Alignments were performed using MAFFT v7.520 (Katoh and Standley 2013) and were verified visually using Mesquite v3.81 (Maddison and Maddison 2023). Data were divided into two partitions, with first and third codon positions in one partition and second codon positions in another. Model testing and maximum likelihood analyses were performed using the IQTree web server (Trifinopoulos et al. 2016). One thousand bootstrap replicates were performed on the partitioned dataset using the models TIM2+F+I (first and third codon positions) and HKY+F+G4 (second codon positions). Calculations of genetic distance were performed under a K2P model using Mesquite v3.81 (Maddison and Maddison 2023).

ULB Agroecology Lab, Brussels School of Bioengineering, Université libre de Bruxelles, Belgium

DZUL Departamento Biología Animal, Edafología y Geología of the University of La Laguna, Tenerife, Spain

FLR Private collection of F. La Roche, San Cristóbal de La Laguna, Tenerife, Spain

The morphological terminology used in the description follows Michener (2007), Litman et al. (2021), Niu et al. (2021) as well as Kasparek and Ebmer (2023). All absolute measurements are made in millimetres (mm) and are used for body length. For all other structures, relative measurements are used. Abbreviations used in the description and diagnosis section below are as follows:

BL (body length): measured as the shortest absolute distance from the base of the antennal socket to the apex of the metasoma (see Niu et al. 2021);

ITD (inter-tegular distance): measured as the shortest absolute distance between the tegulae (scale-like structure covering the insertion point of the wings on the thorax) in dorsal view;

OOD (ocellar-occiput distance): assessed in dorsal view, under a stereomicroscope with continuous LED light, as a ratio between the distance separating the lateral ocelli and the posterior occiput (dorsal margin of the vertex) on one hand, and the ocellar diameter on the other hand;

MPD (median punctuation density): assessed as the ratio between the distance separating neighbouring points in the median region of the tergites and the puncture diameter;

LPD (lateral punctuation density): measured as the ratio between the distance separating neighbouring points in the lateral region of the tergites, particularly on T2 (second tergite) and T3 (third tergite), and the puncture diameter;

SSPD (scutum and scutellum punctuation density): measured under a stereomicroscope with continuous LED light as the distance between neighbouring points on the scutum (dorsal side of the thorax/mesonotum) and on the scutellum (dorso-apical plate of the thorax/mesonotum) on one hand, and the puncture diameter;

TEG (tegulae): colour of the scale-like structures covering the insertion point of the wings on the thorax/mesonotum;

ProN (pronotum): colour of the dorsal lobe of the first thorax/mesonotum segment;

ProLo (pronotal lobe): colour of the protonal lobe (also known as “humeral tubercles”) located just next to the tegulae, towards the anterior site of the thorax/mesonotum;

TCP (tergite colour patches): colour, size and delineation between the orange colour patches on the tergites and the surrounding black cuticle;

FACE (face): colour of the clypeus, mandibles, lower part of paraoccular region;

LEGS (legs): colour of the coxae, femurs, tibiae on each pair of legs.

Photographs of the type material were taken using a Leica S8APO equipped with a Leica MC190 HD digital camera and a Leica LED3000 DI light dome. Series of shots were taken by manually adjusting the precision dial to cover the sharpness of the target body parts. The specialized hooked and waved hairs on the metasomal sterna (S3-S4-S5) in males were photographed using a Canon 5DS R equipped with a Canon MP-E 65mm lens at 5× and (set at f/3.2 and ISO 100), mounted on a StackShot macro rail (distance/step = 0.01mm) and lit with two custom diffused, IR-operated Godox 860vii cobra flashes. All resulting photos were stacked with Helicon Focus (version 8.2.6.) using the software’s “Depth Map mode (Method B)”. Resulting stacked shots were slightly edited in Adobe Lightroom and cleaned in Adobe Photoshop 2023.

Shapefiles and map data derived from OpenStreetMap (copyrighted OpenStreetMap contributors and available from https://www.openstreetmap.org) were downloaded from GeoFabrik (https://download.geofabrik.de). Species distribution points were plotted using QGIS 3.22 -Białowieża (QGIS Development Team 2023). We have included a series of 35 “research-grade” occurrence records of P. canariense from GBIF.org (GBIF 2023b) and we have estimated the coordinates of another 15 conspecific specimens cited by Hohmann et al. (1993) based on the approximate center of the most specific locality given from the island of Gran Canaria. All new Pseudoanthidium records relevant to the islands of Lanzarote and Fuerteventura resulting from specimens collected in the field or macrophotographs exhibiting enough detail to allow for an identification at the species level were included on the map too.

Finally, we used the elevatr package (version 0.4.2.) (Hollister 2022) to compute the elevation associated with each occurrence record of both Pseudoanthidium species. Boxplots of all records were prepared with the “ggplot2” package (Wickham 2016). All records, including their latitude/longitude coordinates in decimal degrees (WGS84), their elevation, the date of each record and their source are compiled in Suppl. material 2. All analyses were performed with RStudio (RStudio Team 2020) for R (version 4.2.2; R Core Team 2022).

To evaluate the extent of shared environmental niche space among the two Pseudoanthidium species, we conducted an analysis of ecological niche characteristics. Significant niche differentiation is anticipated owing to the contrasting habitats between the eastern islands of Lanzarote and Fuerteventura and the western islands. The comparison is intended to illustrate the prevailing climatic conditions for the two species. For each occurrence record of the two species, we extracted environmental data from a 200m buffer. BIOCLIM data was obtained from CHELSA (Climatologies at High resolution for the Earth’s Land Surface Areas) climate dataset at 30 arc seconds resolution (Karger et al. 2017). We selected 4 bioclim variables to cover precipitation and temperature range, and variation (mean annual temperature/precipitation, temperature/precipitation seasonality). Elevation data was obtained from the “elevatr” package as described above. The background niche space was calculated based on 2,000 randomly generated points within the Canary Islands. These data were then used to classify the ecological niche space occupied by the two Pseudoanthidium species. Following Broennimann et al. (2012), we used a principal component analysis (PCA) that was calibrated based on the complete environmental space encompassing the Canary Islands, that applies smoothers to the species presences in environmental space for the purpose of selecting and weighting the environmental variables. We then computed niche overlap between the two species with Schoener’s D statistic (Schoener 1968; Warren et al. 2008). Finally, we tested whether the niche overlap of the two species is less equivalent than random by means of a niche equivalence test with 1,000 repetitions (Broennimann et al. 2012). These analyses were conducted with RStudio (RStudio Team 2020) for R (version 4.2.2; R Core Team 2022) using the “ecospat” package (version 3.5.1; Broennimann et al. 2023).

We used the “red” package (version 1.5.0) (Cardoso 2017) and all occurrence records to compute the extent of occurrence (EOO) and area of occupancy (AOO) of P. canariense and the newly discovered Pseudoanthidium species described below. EOO encompasses the total geographic range of a species, while AOO focuses on the current occupied area of a species within its known habitat; both metrics are vital for assessing a species’ conservation status and can influence its IUCN Red List categorization. We then used the rCAT package (version 0.1.6) (Moat 2020) to calculate the IUCN rating of each species based on their EOO in km².

Assessing the extinction risk of a species, including of hitherto overlooked, or newly described taxa, requires using a series of criteria listed by the International Union for the Conservation of Nature (IUCN) Red List (IUCN Standards and Petitions Committee 2022). These criteria are based on indicators of extinction risk and ultimately help assign a ranked threat category, such as critically endangered (CR), endangered (EN) and vulnerable (VU) (Mace et al. 2008; Nieto et al. 2014). In a nutshell, the five key components of an IUCN extinction risk assessment are: population reduction (Criterion A), restricted geographic range (Criterion B), small population size and decline (Criterion C), very small or restricted population size (abundance) (Criterion D) and a quantitative analysis of decline (Criterion E). For each criterion, threshold values are defined and associated with different threat categories; we performed the assessment for the new Pseudoanthidium species following practical guidelines (Rodríguez et al. 2015) and using as much direct (and to some extent, indirect) evidence as possible (Le Breton et al. 2019).

Twenty-two years after the last description of two new Megachilidae bee species in the CI archipelago, Osmia palmae Tkalçů (Tkalçů 2001a) and the cuckoo Dioxys lanzarotensis Tkalçů (Tkalçů 2001b), we provide evidence for the presence of a hitherto overlooked species of Pseudoanthidium on the islands of Fuerteventura and Lanzarote, the archipelago’s “Eastern Islands”. We describe this new species as Pseudoanthidium (Pseudoanthidium) jacobii Vereecken & Litman spec. nov., and we illustrate its close evolutionary relatedness to the Canarian endemic P. canariense. We have highlighted key diagnostic morphological traits to discriminate between the two Canarian Pseudoanthidium for each sex (Figs 1–5) and a degree of genetic divergence of 2.7%, lower than, yet still in keeping with, the genetic distances observed between other major clades in the P. scapulare complex (Litman et al. 2021). Last, our results illustrate that these two small carder bee species are allopatric in their contemporary distribution (Fig. 1), they exhibit a significant differentiation in elevation range (Fig. 6), as well as a very low overlap (1%) in their respective ecological niches (Figs 7, 8), due in large part to the different climate of the islands on which they occur. Collectively, and in light of the criteria generally used to delineate species in this genus (Litman et al. 2021; Niu et al. 2021), our observations therefore suggest that P. jacobii deserves its own species status.

Most females in the subgenus Pseudoanthidium (Pseudoanthidium), and particularly in the P. scapulare species complex, appear to have a strong preference for host plants in the family Asteraceae (Litman et al. 2021; Niu et al. 2021; Kasparek and Ebmer 2023). A strong preference of P. jacobii for host plants in the family Asteraceae might therefore reflect a phylogenetic conservatism of pollen diet, a phenomenon already described in several other groups of wild bees in Europe (Müller and Kuhlmann 2008; Sedivy et al. 2008; Dötterl and Vereecken 2010; Wood et al. 2021; Dorchin et al. 2022).

The Pseudoanthidium scapulare complex of species is distributed throughout the Palaearctic region. Certain members of the complex represent closely related species pairs that are sympatric throughout a part of their distributions, including P. nanum - P. scapulare and P. tenellum - P. cribratum. Both of these pairs are morphologically distinct but exhibit relatively low levels of genetic differentiation in analyses of CO1 (0.35% and 0.59%, respectively) (Litman et al. 2021). In recent analyses, however, neither P. nanum - P. scapulare nor P. tenellum - P. cribratum showed evidence of barcode-sharing and in both cases, historical mitochondrial introgression was deemed the most likely explanation for the low levels of genetic differentiation observed in these species pairs (Litman et al. 2021). This argument was further supported by a UCE analysis of P. nanum and P. scapulare that provided strong evidence of two genetically distinct lineages (Litman et al. 2021). In comparison, P. canariense and P. jacobii show consistent morphological differences but a K2P-corrected genetic distance of 2.7% at the CO1 locus, somewhat higher than the distances separating P. nanum - P. scapulare and P. tenellum - P. cribratum. Pseudoanthidium canariense and P. jacobii may thus represent species whose genetic divergence was facilitated by the reproductive barrier imposed by their distributions on different islands. Further analyses of genomic-level data are needed to better understand the evolutionary history of these taxa.

Our morphological analysis and that of Litman et al. (2021) illustrate that both P. canariense and P. jacobii have gonostyli that are approximately parallel-sided and exhibit a rounded (i.e., unnotched) apex. This is in marked contrast to other species in the P. scapulare complex (except P. tropicum (Warncke, 1982) known only from Iran so far, see Litman et al. (2021)), which have an obvious U-shaped notch at the apex of their gonostyli). Interestingly, the absence of a notch is also a feature shared with non-scapulare complex Pseudoanthidium species (see Niu et al. 2021; Kasparek and Ebmer 2023), which calls into question the ancestral or derived nature of the gonostylus notch in small carder bees.

Given the lack of resolution in our analyses of CO1, namely regarding the phylogenetic relationships among different clades within the complex, we can only propose hypotheses to explain the presence of a rounded gonostylus in P. jacobii and P. canariense. If the clade represented by these two species is the sister group to all other members of the P. scapulare complex, then one possible explanation is that the common ancestor of P. jacobii and P. canariense colonised the Canary Islands prior to the origin of other members of the P. scapulare complex (Litman et al. 2021), i.e., the presence of the rounded gonostylus may represent the plesiomorphic state for the complex. If this clade, however, turns out to be nested within the complex, another possible explanation is that the gonostylus in the common ancestor of P. jacobii and P. canariense may have undergone a reversion to an ancestral state. If speciation may be driven, at least partially, by morphological barriers to reproduction (Oneal and Knowles 2013; Huang et al. 2020), perhaps the marked differences in the shape of the gonostylus in other, closely related members of the P. scapulare complex (i.e., P. nanum and P. scapulare) may have been significant drivers of speciation, especially in sympatric populations. On the Canary Islands, where the diversity of closely related species is considerably lower than on the mainland, selective pressure on the shape of the gonostylus may be less intense, thus facilitating a reversion to an ancestral state. A future phylogenomic approach to an analysis of this genus using more samples of each species from across their distribution should shed light on the evolution of this and other traits in the genus Pseudoanthidium and contribute to improving our knowledge on the diversification and historical biogeography of small carder bees.

Ever since Antiquity, historians, traders and (bio)geographers have acknowledged the peculiar nature of Lanzarote and Fuerteventura, these “Eastern Islands” of the Canary archipelago. Although they form two islands today, they are, geologically speaking, the oldest emerging part of the archipelago that used to be merged during the Pleistocene glacial cycles, forming the paleo-island of Mahan (Rijsdijk et al. 2014). These two islands were also historically referred to as the “Islas Purpurarias” (Garcia-Talavera 2016). The origin of this name supposedly traces back to the peak trade period of Roccella canariensis Darb. (Roccellaceae), a lichen species endemic to the Canary Islands and locally known as orchilla. This lichen grew on the cliffs of Lanzarote and Fuerteventura at the seashore and was unique as a natural source of the red-purple pigment orcein, a highly sought-after resource in the textile industry up until the end of the 19^th century.

By describing P. jacobii as an endemic species from the Purpurarias, we provide further evidence for the peculiar environmental conditions met on the islands of Lanzarote and Fuerteventura within the Canary Islands archipelago, and how original life forms have evolved solely on these two islands. Single-island endemics are reported in the eastern CI, such as Tetralonia lanzarotensis Tkalçů, 1993 (Apidae) and Dioxys (Dioxys) lanzarotensis Tkalçů, 2001 (Megachilidae) from Lanzarote, or Megachile (Eutricharaea) hohmanni Tkalçů, 1993 (Megachilidae) and Dufourea fortunata Ebmer, 1993 (Halictidae), both recorded exclusively from Fuerteventura within the CI archipelago (Hohmann et al. 1993).

Examples of other wild bee species sharing a distribution restricted to these two islands include Anthophora purpuraria Westrich, 1993 (Apidae) (whose specific epithet derives directly from the “Purpurarias”), as well as other Anthophorini species like A. (Heliophila) lanzarotensis (Tkalçů, 1993), A. (Heliophila) lieftincki (Tkalçů, 1993), A. (Pyganthophora) porphyrea Westrich, 1993 (all endemic to Lanzarote and Fuerteventura) and some of their associated cuckoos such as Melecta (Melecta) caroli Lieftinck, 1958, and M. (Melecta) prophanta Lieftinck, 1980 (on Lanzarote only) (Hohmann et al. 1993). Other cuckoos of these narrow endemic and other more ubiquitous Anthophorini species might include Thyreus histrionicus (Illiger, 1806) and T. ramosus (Lepeletier, 1841) that have a Circum-Mediterranean distribution (Michez et al. 2019; Leclercq et al. 2022). Likewise, the family Megachilidae has a few endemic representatives, such as Hoplitis (Tkalcua) zandeni (Teunissen & van Achterberg, 1992) that nests in empty snail shells (Müller and Mauss 2016), Megachile (Chalicodoma) fuerteventurae (Tkalçů, 1993) and M. (Eutricharaea) binominata Smith, 1853 (Hohmann et al. 1993). The Andrenidae fauna of the Canary Islands also includes narrow endemics to Lanzarote and Fuerteventura, such as Andrena (Chlorandrena) damara Warncke, 1968 or A. (Aciandrena) hillana Warncke, 1968 (Hohmann et al. 1993).

It is important to note that some species originally described from a single island (such as A. (H.) lanzarotensis or M. (C.) fuerteventurae, as their specific epithet suggests) turned out to be discovered on both islands after a few decades of field surveys. These recent records contribute to the emergence of distribution patterns among closely related, endemic species in different groups of wild bees, with some restricted to the eastern islands and others present only in the central (and western) islands. For example, the three Megachile (Chalicodoma) species recorded in the archipelago exhibit such a distribution pattern similar to the Pseudoanthidium species discussed here, with M. (C.) canescens (Brullé, 1840) restricted to the central and western islands, whereas M. (C.) sicula (Rossi, 1794) and M. (C.) fuerteventurae are found only on the eastern islands (Lanzarote and Fuerteventura) (Hohmann et al. 1993). Likewise, Megachile (Eutricharaea) canariensis Pérez, 1902 is restricted to the central and western islands, whereas M. (E.) binominata is endemic to Lanzarote and Fuerteventura.

Interestingly, no species in the families Colletidae or Melittidae is endemic to Lanzarote and Fuerteventura: these families encompass CI endemic species distributed across the eastern, central and western islands, but none are restricted to eastern islands. Those that are found on these islands have a much wider distribution encompassing Morocco and sometimes extending to the Levant and even the Arabian Peninsula (e.g., for Melitta schmiedeknechti Friese, 1898 (Melittidae); see Shebl et al. (2016)).

At their nearest point, Lanzarote and Fuerteventura are located just under 100 km (60 miles) off the coasts of Morocco (García Talavera 1999; Florencio et al. 2021). This relative proximity is therefore likely to have favoured faunal exchanges across families of bees between the Canary Islands archipelago and the African continent. For example, Andrena (Distandrena) mariana s.str. Warncke, 1968 (Andrenidae) was described from the island of Fuerteventura in the Canary Islands, and its description was associated with a remark that the species could potentially be found in Morocco as well (Warncke 1968). A similar distribution bridging the Canary Islands archipelago and the south-western coasts of Morocco has already been found in Lasioglossum (Evylaeus) phoenicurum (Warncke, 1975) (Pauly 2016), in Nomioides (Nomioides) fortunatus Blüthgen, 1937 (Pauly 2017) (both Halictidae), as well as in Haetosmia circumventa (Peters, 1974) (Müller 2022), and these findings could probably be echoed in different genera of the diverse family Apidae among others (A. Dorchin, pers. comm. July 2023; P. Rasmont, pers. comm. July 2023). More surveys along the coasts of south-western Morocco are needed to determine if other species considered as endemic to the Canary Islands, including P. jacobii, show similar patterns “bridging” the distribution gap with the African continent.

Are newly discovered species at a higher risk of extinction than those first described long ago? According to Liu et al. (2022), the trend is positive and significant across major vertebrate groups (amphibians, birds, fish, mammals, and reptiles), and it is driven by several factors, including their often-smaller population numbers and restricted distribution range making them vulnerable to habitat loss and fragmentation. Whether this applies to less intensively investigated groups of organisms like insects, and wild bees in particular, has not been adequately tested so far. Yet, these results suggest that extinction risk assessments that are based on overall threat status of all species combined may seriously underestimate the true number of species threatened with extinction (McKinney 1999).

Here, we classified the new species P. jacobii as Endangered (EN) according to IUCN Criteria B1ab(i,ii,iii,iv,v), B2ab(i,ii,iii,iv,v) and Criterion C2a(i), as well as based on the parallel calculation of the IUCN rating based on EOO Area (in km²). In other words, and practically speaking, this means that P. jacobii is threatened with extinction and might experience continuous decline unless conservation efforts are made. The results of our extinction risk assessment are also motivated by the fact that P. jacobii, similarly to other threatened wild bee species, typically occurs within a few small patches rather than a spatially continuous area of uniform presence. This patchy distribution implies that P. jacobii is exposed to comparatively higher extinction risks, because there is a greater chance that one or several of the identified threats will act in concert and will affect all or most of the distribution within a certain time frame. How imminent the identified risks are has remained elusive for a long time, but in recent years the Canary Islands and the “Purpurarias” in particular have seen droughts and heatwaves, just like large parts of the drought-stricken mainland. These changing weather patterns are the direct outcome of climate change operating in real time, and in the worst-case scenario they have even combined with environmental hazards like the 2021 Cumbre Vieja volcano eruption on the island of La Palma, followed in 2023 by massive wildfires breaking out and threatening all terrestrial forms of life on several of the Western Canary Islands. Such wildfires are less likely on Fuerteventura and Lanzarote due to their scarce vegetation and lack of forested areas (Carillo et al. 2022).

The Canary Islands archipelago, like many other insular ecosystems renowned for their breath-taking natural beauty, faces major challenges in achieving a delicate balance between promoting recreational activities and preserving its fragile biodiversity as the effects of climate change become increasingly pronounced. Our study illustrates that our understanding of biodiversity in the archipelago is far from complete, and that international scientific collaborations with local experts and citizen science projects can help gain significant insights into species distribution and ecological interactions. This paper also confirms the vivid interest of the international scientific community towards the Canary Islands bee fauna and its conservation (see also Monasterio León et al. 2023), and highlights that more targeted field surveys and increased sampling efforts, primarily in the eastern and less sampled western islands, but also slightly before or after the typical “bee season” of January-April, and in high altitude ecosystems found on Gran Canaria and Tenerife still have the potential to reveal other unique taxa and contribute to refine our understanding of the spatial taxonomic, functional and phylogenetic patterns of diversity and endemism of the CI wild bees.