Exploring the hidden riches: Recent remarkable faunistic records and range extensions in the bee fauna of Italy (Hymenoptera, Apoidea, Anthophila)

Abstract Background The area sourrounding the Mediterranean basin is recognised as a major biodiversity hotspot for bees, and Italy is amongst the European countries with the highest bee species richness. Detailed knowledge of bee distribution is crucial for understanding bee biology and designing tailored conservation strategies, but is still insufficient in southern European countries, especially in Italy. New information We report recent finds of 48 bee species that yield significant novelties for the Italian bee fauna. Eight species, namely Andrenaconfinis Stöckhert, Anthidiellumbreviusculum Pérez, Coelioxysalatus Foerster, Lasioglossumalgericolellum Strand, Megachilelapponica Thomson, Megachileopacifrons Pérez, Megachilesemicircularis auct. nec Zanden and Trachusaintegra Eversmann are reported as new for Italy. In addition, Andrenabinominata Smith, Andrenacompta Lepeletier, Colletesacutus Pérez, Lasioglossumstrictifrons Vachal, Rhodanthidiumsiculum Spinola and Rhodanthidiumsticticum Fabricius are newly recorded from mainland Italy, Osmiaheteracantha Pérez from Sardegna and Nomadaflavopicta Kirby from Sicilia. We also report significant range extensions for other bee species and recent records of species that had long gone unrecorded in Italy. The combination of morphology and DNA barcoding provided reliable identifications even for the most challenging specimens. As several of our records come from areas neglected by bee experts in the past, this study stands out as a key indicator of a bee faunistic richness still awaiting discovery and hopefully it will stimulate the interest of taxonomists and stakeholders in pursuing bee research in Italy in the near future.


Introduction
The Mediterranean Basin is one of the major hotspots for bee diversity (Michener 2007, Nieto et al. 2014, Orr et al. 2021).Italy, in particular, has one of the richest bee faunas in the world by land area, with well over 1000 confirmed species (Pagliano 1995, Comba 2019, Ghisbain et al. 2023, Reverté et al. 2023).However, this fauna remains insufficiently known in comparison with the ones of nearby Central European and Mediterranean regions, which have received considerably more attention, especially in recent years.The coverage of the Italian territory is markedly uneven, with large areas only occasionally investigated or still virtually unexplored.Moreover, published data are sometimes of poor quality and unreliable and recent ones are often still unpublished and difficult to access.This makes it hard to design and implement effective conservation actions.For instance, these factors significantly limited the evaluation of the conservation status of several potentially threatened Italian bee species (Quaranta et al. 2018).
It was primarily the awareness of the problems sketched above that led to the creation in 2022 of an informal network of Italian bee enthusiasts.The primary goal of the network is to contribute to a better knowledge of the Italian bee fauna, through the exchange of data, images and specimens and the collective discussion of critical determination cases.The present paper details the main results obtained by the network during its first year of activity.We report records of 368 specimens belonging to 48 bee species.Eight of the species are new for the bee fauna of Italy and, in addition to these, six are new for mainland Italy, one for Sardegna and one for Sicilia; our data form the basis for the inclusion of these species in the Italian, Sardinian and Sicilian checklists (Reverté et al. 2023).We also report biogeographically significant extensions of the Italian ranges of thirty species.Finally, we document the continuing occurrence in Italy of a few species which had gone unrecorded in the country for a long time.A central aspect of our work is that it combines morphological identifications with the use of DNA barcoding targeting the mitochondrial COI region, which played a significant role in our study in at least two ways.On the one hand, in several cases, barcoding provided additional confirmation of morphological identifications.On the other hand, for some species, the DNA sequences obtained here constitute the first COI barcodes ever produced.
Axio Zoom V16 microscope with a Plan Z 1.0/0 FWD 60 mm lens integrated with Axiocam 807.Stacking of pictures was performed with Helicon Focus (HeliconSoft) or CombineZP software.

DNA barcoding
In several cases, particularly for difficult-to-determine species, the morphological identifications were confirmed by DNA barcoding.To this end, total DNA was extracted from a leg of one or more specimens using a DNeasy Blood & Tissue Kit (Qiagen, Milan, Italy).The COI standard barcode region was targeted to obtain DNA barcodes (i.e.658 bp at the 5' end of the mitochondrial COI).To do so, the standard barcoding primers LCO1490/HCO2198 were used.The taxonomic identity of the processed bees was tested by comparing the COI sequences thus obtained with the reference DNA barcode bank stored in the Barcode of Life Database (BOLD System) using the Identification Engine tool (IDS) (http://www.boldsystems.org/index.php/IDS_IdentificationRequest;Species Level Barcode Records database).To obtain reliable identifications, the species lists, the matching rates and the neighbour-joining tree returned by IDS were scrutinised for each submitted sequence, in particular by verifying the presence of multiple concordant identification outputs of very high similarity scores and/or sequences forming well defined clades with a high majority of concordant sequences names in the neighbour-joining tree.All specimens subjected to DNA barcoding are listed in Suppl.material 1.Sequences are deposited in BOLD Systems (project "RApiTI -Italian Rare Bees" [ZPLRP]).Notes: Prior to our find, C. acutus was known in Italy only from Sardegna and Sicilia (Comba 2019 and references therein).The Puglia records are by far the easternmost known for the species, whose range they significantly extend (Fig. 1A).Notes: According to Le Divelec (2021), Hylaeus nigrifacies is the correct name for the species that has generally been known as Hylaeus moricei (Friese, 1898).The only published record of H. nigrifacies from Italy is that of a specimen from Naples, one of the paratypes of Prosopis nigrifacies rhenana Warncke, 1986(Warncke 1986).Notes: Previously known in Italy only from the Eastern Alps in Südtirol and from the Maritime Alps in Piemonte (Dathe 2000).Our records of H. nivaliformis are the first for the Apennines (Fig. 2B).Notes: One specimen confirmed by DNA barcoding: the match with reference sequences is between 100% and 99.84% (mean: 99.96%).All specimens were collecting pollen on Pimpinella major in a hay meadow.Previously recorded in Italy in Trentino-Alto Adige (Schmid-Egger 2005) and in the Maritime Alps of Piemonte (Schmid-Egger 2011).The occurrence of the species in these two regions and, more specifically, in the Ligurian Alps, is mentioned also by Stoeckhert (1942).A. alutacea has recently been found to occur, or have occurred, also on Mount Etna in Sicilia (Wood et al. 2023).Notes: In Italy, Andrena amieti had previously been recorded throughout the Alps and on Monte Pollino in Calabria (Praz et al. 2019).Our find shows that it also occurs in the Central Apennines, as was to be expected.It is shown by Praz et al. (2019) that two divergent mitochondrial lineages are found in A. amieti.One lineage, called Group 1, is reported from Calabria and, sparingly, from the Alps.The other lineage occurs widely in the Alps.DNA barcoding of one of our specimens confirmed the morphological identification and showed that the specimen belongs to Group 1.The match with the two sequences from Calabria in BOLD (ID: HYMAA051-18, HYMAA052-18) is 100% and 99.61%.Notes: In Italy, previously known only from Sicilia and Sardegna (Comba 2019 and references therein).This is the first record for mainland Italy (Fig. 1A).Notes: All specimens were collected on Lythrum salicaria, on the banks of the River Tiber.Our records extend the range of the species to continental Italy (Fig. 3A).

Andrena confinis
Previously known in Italy from a single record from south-western Sardegna (Ebmer 1986).Elsewhere, the species range includes Iberia, southern France and the Maghreb (Ebmer 1986).We sequenced two of our specimens.Only one sequence, labelled Lasioglossum strictifrons, from southern Portugal, was available in BOLD for comparison, matched at 99.02 and 99.16% by our sequences.Matches with BOLD sequences labelled as other species were all below 92%.Images: Fig. 4. Exploring the hidden riches: Recent remarkable faunistic records and range ...

Notes:
The only published record of L. subaenescens from Italy known to us is that of two females from Molise, one of them determined by Ebmer (Quaranta et al. 2004).Notes: Two specimens were confirmed by DNA barcoding with sequences matching reference sequences between 99.34% and 99.01%(mean 99.10%, excluding sequences labelled "subfulvicorne" appearing as single representatives with this name in clades dominated by other species in the neighbour-joining tree).The species had been reported in Italy only from the Alps in Trentino-Alto Adige/Südtirol (Ebmer 1988) and Lombardia (Praz et al. 2022), so our records appear to be the first for the Apennines (Fig. 2B).Exploring the hidden riches: Recent remarkable faunistic records and range ...  Stöckl (2000), where C. alatus is mentioned as (doubtfully) occurring in Südtirol, based on Biegeleben (1929).However, Biegeleben's paper, which is a general exposition of parasitism in bees, does not contain records of C. alatus or, for that matter, of any bee species.The only place where C. alatus is mentioned is a list of host bees and their parasites in the last two pages of the paper, preceded by a sentence which, in English translation, reads: "We owe to Prof. Dr. Bischoff, who excels in the study of the biology of Hymenoptera, a detailed table of the different species of host bees and their parasites.Some of them, namely the ones that are most widespread in our region, are listed here according to this table".In the case at hand, "widespread" likely does not refer to C. alatus, which is rare everywhere in Europe (Devalez 2010), but to its host Megachile ligniseca which is ubiquitous, though uncommon, in Südtirol.For good measure, we inspected the Biegeleben collection, now housed in the Museo Civico di Zoologia in Rome, but could not locate any specimens of C. alatus.Images: Fig. 5. Notes: The specimen from Veneto was caught in a pan trap and the ones from Abruzzo on Epilobium sp, in agreement with the stated oligolecty of M. lapponica on Epilobium ( Amiet et al. 2004, Scheuchl 2006).These are the first records of the species from Italy.The one from Abruzzo is particularly noteworthy because, excepting isolated records from Greece, it is by far the most southerly in Europe.Even in the Alps, the records of M. lapponica are sporadic, although this may be due more to the difficulty of intercepting the species than to its actual rarity (C.Praz, personal communication 2022).Images: Fig. 6.Exploring the hidden riches: Recent remarkable faunistic records and range ... Notes: This appears to be the first record of R. siculum for continental Italy (Fig. 1B).

Megachile opacifrons
The species was previously known in Italy only from Sicilia (Kasparek 2019).Notes: The specimens from Campotosto, Barrea, Camporotondo and Fano Adriano were collected in grassy clearings at the edge of beech forest.The male from Cappadocia was collected on Cirsium sp., along a road running through orchards and open mixed woodland.Prior to our records, B. hypnorum was known to occur throughout the Italian Alps and adjacent foothills, in the Northern Apennines in the Provinces of Genova (verified records in GBIF), Pavia, Piacenza (M.Cornalba personal observations), Massa Carrara (Intoppa et al. 1995), Lucca (P.Biella, personal observations) and Forlì-Cesena (Quaranta et al. 2004), perhaps in the Provinces of Pisa and Firenze (Rossi 1790) and, furthermore, in scattered localities in the plains of Lombardia, Veneto and Friuli Venezia Giulia (Pensa 1832, Intoppa et al. 1995, Barbattini et al. 2006 Notes: On flowers of Lysimachia vulgaris, in a clearing along the banks of the River Aniene, flying together with Macropis europaea.Our records significantly extend the range of the species, which had previously been recorded in Italy -very sporadicallyonly in the north (Comba 2019 and references therein).Notes: Found resting on grass and flowers at sunset, together with Colletes mlokossewiczi.Prior to our find, E. productulus was documented in Italy from Valle d'Aosta (Amiet et al. 2007), Piemonte (Schwarz et al. 1999, Bogusch andHadrava 2018) and Emilia-Romagna (Bogusch and Hadrava 2018).We sequenced one male and one female and the resulting sequences seem to be the first for E. productulus.Notes: At the Cecima site, T. inulae co-occurs with Tetralonia fulvescens Giraud, 1863 and Tetralonia alticincta (Lepeletier, 1841), the latter flying on average about one month later than T. inulae.We obtained a barcode sequence from one of the specimens of T. inulae from the Cecima site.Only two short (< 400 bp) sequences labelled Eucera inulae, both from Canton Ticino in Switzerland, were available in the BOLD data bank for comparison.They turned out to match our sequence at 99.74%.Aside from them, the nearest match for our sequence, at 97.36%, is a GenBank COI sequence (voucher ad98, accession MG251111) pertaining to a specimen from Erzurum, Turkey, determined as Eucera alticincta (Dorchin et al. 2018).Interestingly, the barcode sequences of two T. alticincta specimens from the Cecima site turned out to be almost identical to the Turkish one mentioned above (match 99.65%), but distant from the few T. alticincta sequences from western Europe available in the BOLD data bank, with matches below 97.5% (Suppl.material 1, Table S1).

Analysis
Overall, we report records of 368 specimens of 48 bee species, belonging to six families and 23 genera, coming from 14 of the 20 administrative regions of Italy.Most represent first records for Italy or for some of its regions (Table 1).Overall, eight of the species are previously unrecorded in Italy, an additional six previously unrecorded in continental Italy, one in Sicilia and one in Sardegna.A further thirty species are newly reported from at least one administrative region of Italy and we document the continuing occurrence in Italy of five species which had long gone unrecorded in the country (see notes in the "Annotated list of significant records" section).We retrieved long sequences of the COI Folmer region for all the specimens subjected to DNA analyses (see Suppl.material 1).The identification by DNA barcoding was particularly successful, as the resulting sequences yielded high similarity scores to existing sequences of trustworthy species identity (see the notes in the "Annotated list of significant records" section).Lasioglossum subfulvicorne (Blüthgen, 1934) ER Table 1.
Exploring the hidden riches: Recent remarkable faunistic records and range ...

Discussion
In this study, we report eight species of bees which are new for the Italian fauna and, in addition to these, eight which are new for the fauna of mainland Italy, Sardegna or Sicilia.
Our results are an important indication of a faunistic richness still awaiting discovery, especially considering that they were obtained by a small group of people, mostly nonspecialists, in just a few months.Our results confirm the effectiveness of the collaborative approach that we followed.It must also be stressed that virtually all the records presented here are recent.Only five of them date from before 2013 and the bulk of the remaining ones date from 2020-2023.Thus, they yield information on the present, as opposed to historical, bee fauna of Italy.
Various recent additions to the bee fauna of Italy come from western Liguria or southwestern Piemonte and concern species with mostly west Mediterranean or west European distribution, such as Andrena asperrima Pérez, 1895 ( Carisio et al. 2018), Andrena rhenana Stoeckhert, 1930 (Gamba andCarta 2020) and Dasypoda crassicornis Friese, 1896(Bonifacino 2021).This pattern is repeated in the present study with the discovery of Megachile opacifrons in western Liguria and of Anthidiellum breviusculum in the Val di Susa in Piemonte.Many other western species show a roughly similar distribution, occurring in France almost up to the Italian border.Several of them are likely to occur also in Italian territory.One could ask whether a similar situation might hold at the eastern end of northern Italy with species from south-eastern Europe.Unlike Liguria which has been surprisingly neglected, the area around Trieste has been intensively studied in the past by Austrian and German entomologists.Still, faunistic surprises might well occur here too.
Several records resulted from areas previously neglected by bee experts.Two of the species newly reported here, Coelioxys alatus and Megachile lapponica, were recorded in the Alpine portion of Veneto.This is an area that has been little explored in the past, as is true, surprisingly, for most of the Italian Alps.Here few areas, mostly Trentino-Alto Adige/ Südtirol and Valle d'Aosta, have seen significant melittological activity, especially by German, Austrian and Swiss researchers (Hellrigl 2006and references therein, Kopf 2008, Steinmann 2002).Similarly, several of the species reported in the present study were caught in riverine or humid inland areas.This seems to indicate that these areas have been partly ignored in the past and deserve particular attention.
Finally, peninsular and insular Italy are probably the most interesting parts of the country from a melittological point of view.Here, large areas are almost unexplored: Basilicata, much of Calabria, southern Campania, much of Molise and Puglia, parts of Abruzzo and much of Sardegna.Various records discussed in the present paper come from these areas and one may add to them the recent discovery of Eucera breviceps Friese in Abruzzo (Aubert et al. in press), of Andrena freygessneri Alfken and A. probata Warncke in Abruzzo and of A. oralis Morawitz in Puglia (Wood et al. 2023).As indicated by our finds of Hylaeus glacialis, H. nivaliformis, Lasioglossum subfulvicorne, Megachile lapponica and other species, the Apennines, particularly in their montane and alpine stages, probably harbour a faunistic richness that we have only begun to explore.
We have hinted at the effectiveness of the collaborative methods adopted by our network.Reliable identifications were achieved via accurate morphological evaluation with exchange of photographic documentation and specimens and with the help of up-to-date literature, discussion with the involvement of specialists and DNA barcoding when possible.
As in other similar contexts (Cornalba et al. 2020, Biella et al. 2022), DNA-based identification techniques proved to be a very useful means of confirming the identity of challenging specimens, taking advantage of a well populated European reference dataset in BOLD.In turn, our study contributed to the BOLD reference dataset by producing regionally important sequences (e.g.first Italian reference barcodes) or first reference sequences for species that had never been barcoded before.
As we mentioned in the Introduction, precise data on the bee fauna of Italy are scattered and often of difficult access.It would be highly desirable to collect and organise as many as possible of the available data in a publicly accessible database, patterned for instance after the Swiss National Apoidea Databank (Praz et al. 2022), other possible examples being the Database of Iberian bees (Bartomeus et al. 2022) or the Wild Bees of Chile project (López-Aliste et al. 2021).This would be important not only for research purposes, but also because it would provide an essential tool for the evaluation of conservation needs and for the planning of conservation actions.A central node should be created that would coordinate the databasing of public and private collections.The node should be provided by a public or private institution that can guarantee a robust IT infrastructure and a continuity of effort over the years.Arrangements should be made for continuing the process of digitising entomological collections to populate the database.Experience, including the present paper, indicates that much valuable material is held in private and research collection.The owners of these collections should be encouraged to confer their data to the central node.The most difficult part will probably be the quality control of the data in order to validate the identifications and the metadata, as experience indicates that misidentifications are particularly frequent in various Italian collections.Correcting identification errors would require a major effort, for which Italy is ill-equipped, given the scarcity of taxonomic expertise available in the country.However, with this study, we hope to stimulate the interest of taxonomists, collection owners and curators, administrators and conservationists in building a reliable source of data of bee occurrence in Italy in the near future.
CEUSS = Entomological Collection of the University of Sassari, Sassari, Italy CPC = private collection of Christophe Praz, Neuchâtel, Switzerland MBC = private collection of Marco Bonifacino, Vado Ligure MCC = private collection of Maurizio Cornalba, Pavia, Italy MEC = private collection of Maurizio Mei, Rome, Italy MIB:ZPL = Collection of the ZooPlantLab of the University of Milano-Bicocca, Milan, Italy MKC = private collection of Michael Kuhlmann, Kiel, Germany MSC = private collection of Marco Selis, Viterbo, Italy MZUR = Museum of Zoology of the Sapienza University of Rome, Rome, Italy RCC = private collection of Roberto Catania, Catania, Italy SFC = private collection of Simone Flaminio, Bologna, Italy SGC = private collection of Sirio Gamba, San Biagio della Cima, Italy

Seladonia gavarnica (Pérez, 1903) Materials
a. countryCode: IT; stateProvince: Lazio; county: Rieti; municipality: Lonessa; locality: Sella di Leonessa; verbatimElevation: 1700-1870 m; verbatimLatitude: 42.48; (Warncke 1980, Schmid-Egger 2011)e 2006) in the Sibillini Mountains( Piatti and Ricciardelli D'Albore 2006).There is also an indication of its occurrence in Piemonte(Comba 2019).Sequencing of one of our specimens yielded a 658 bp sequence matching the single one in BOLD labelled Halictus gavarnicus, from the French Maritime Alps, at 99.39%.All matches with other BOLD sequences were well below 96.5%.On flowers of Lysimachia vulgaris, in clearings along the banks of the Rivers Aniene and Tiber.Our records significantly extend the range of the species, which had previously been recorded in Italy only in the north (Comba 2019 and references therein).These records were communicated to us by Christophe Praz.Anthidiellum breviusculum s.l. is split byKasparek et al. (2023)into three species, A. breviusculum s.s.from Iberia and France, A. africanum Kasparek, 2023 from the Maghreb and A. troodicum(Mavromoustakis, 1949)which ranges from the Aegean through Anatolia and the Levant to Iran. A. breviusculum is here first reported from Italy.Its occurrence close to the French border is not unexpected, given its presence in the Var and Roia drainages just across the border(Warncke 1980, Schmid-Egger 2011).

Pérez, 1897
This is the first Italian record of M. opacifrons.The specimen was found in a clearing surrounded by deciduous woodland, vineyards and an abandoned olive grove.M. opacifrons has a West Mediterranean distribution and was known to occur in southern France east to the upper Roia Valley (Schmid-Egger 2011).Hence, it was conceivable that it might also occur in neighbouring parts of Italy.Images: Fig.7.

Trachusa integra (Eversmann, 1852)
Rasmont and Dehon (2015)hes: Recent remarkable faunistic records and range ... Its range, as described inKasparek (2020), is discontinuous and includes parts of southern France, the southern Balkans, inner Anatolia, Crimea and parts of southern Russia.Trachusa integra is here reported for the first time from Italy.Eight specimens, two from Sardegna and six from Lazio, were confirmed by DNA barcoding: the match with reference sequences is between 99.84% and 98.32% (mean 99.05%).Images: Suppl.material 2, Figs.S5 and S6.All specimens were collected on Centaurea solstitialis, flying together with Tetralonia graja and T. julliani.The Italian literature records of A. vinctus are all from northern Italy, more precisely from Piemonte and Trentino Alto Adige/Südtirol (Comba 2019 and references therein).Our records appear to be the first from Lazio and central Italy.Images: Suppl.material 2, Figs.S8-S10.The occurrence of A. affinis in Italy was for a long time unclear, mostly due to persistent confusion with other species, chiefly A. mucida Gribodo, 1873 and A. agama Radoszkowski, 1869.The species is said to be of uncertain presence in Italy byRasmont and Dehon (2015).Images: Suppl.material 2, Figs.S19-S22.We are not aware of any previous solid records of A. calcarata from Sicilia.The species is mentioned in a catalogue of Sicilian bees by De Stefani (1895), without details.The absence from this same list of the similar Anthophora crassipes Lepeletier, 1841, which occurs widely in Sicilia, is curious and might hint at a misidentification.However, since De Stefani's collection is thought to be lost, there is no way of verifying the identity of De Stefani's specimens.Our record confirms the occurrence of A. calcarata, if not on the mainland of Sicilia, at least on the surrounding islands.Images: Suppl.material 2, Figs.S11 and S12.

Anthophora femorata (Olivier, 1789)
(Comba 2019)re records of A. femorata from Italy(Comba 2019)are few, all very old and somewhat questionable.Our records are the only recent ones known to us.

Bischoff, 1930
(Pagliano 1993)cies for which sequences are deposited in BOLD is Epeolus variegatus (average match 96.89%, range 97.99%-92.56%).We are not aware of any published Italian records of E. furfurea.The species has been recorded from Puglia and Sicilia (S.Risch, personal communication 2022).E.pannonica has been recorded from Mount Gargano in Puglia and from Sicilia (S.Risch, personal communication 2022).Our records seem to be the first from Lazio and Abruzzo.Exploring the hidden riches: Recent remarkable faunistic records and range ... Ours seem to be the first records of N. hungarica from Liguria, Lazio and Abruzzo.The only Italian literature records of N. hungarica known to us are from Piemonte(Pagliano 1993)and Sicilia(Nobile 1990, as Nomada lagrecai Nobile).