https://orcid.org/0000-0003-4602-3710
Abstract
Bees are important pollinators and are essential for the reproduction of many plants in natural and agricultural ecosystems. However, bees can have adverse ecological effects when introduced to areas outside of their native geographic ranges. Dozens of non-native bee species are currently found in North America and have raised concerns about their potential role in the decline of native bee populations. Osmia taurus Smith (Hymenoptera: Megachilidae) is a mason bee native to eastern Asia that was first reported in the United States in 2002. Since then, this species has rapidly expanded throughout the eastern part of North America. Here, we present a comprehensive review of the natural history of O. taurus, document its recent history of spread through the United States and Canada, and discuss the evidence suggesting its potential for invasiveness. In addition, we compare the biology and history of colonization of O. taurus to O. cornifrons (Radoszkowski), another non-native mason bee species now widespread in North America. We highlight gaps of knowledge and future research directions to better characterize the role of O. taurus in the decline of native Osmia spp. Panzer and the facilitation of invasive plant-pollinator mutualisms.
Background
Bees comprise a group of over 20,000 species distributed in all continents except Antarctica, where they provide critical pollination services to natural and agricultural ecosystems (Klein et al. 2007, Ascher and Pickering 2020). At least 80 bee species around the world have been introduced to new areas outside of their native ranges (Russo 2016). While some species have naturally spread and extended their geographic ranges with the creation of human-dominated habitats [e.g., the squash bee Eucera pruinosa (Hymenoptera: Apidae); López-Uribe et al. 2016], most species have been moved around as a result of increased global international trade (Russo 2016, Sheffield et al. 2020, Lanner et al. 2022). Dozens of non-native species have been introduced accidentally across continents [e.g., the European wool carder bee, Anthidium manicatum (Hymenoptera: Megachilidae); Gibbs and Sheffield 2009, Strange et al. 2011], and others have been intentionally introduced into novel areas for crop pollination services [e.g., the alfalfa leafcutting bee, Megachile rotundata (Hymenoptera: Megachilidae); Pitts-Singer and Cane 2011]. However, only a few of these non-native bee species (e.g., Apis mellifera scutellata-hybrid a.k.a. African honey bees; Schneider et al. 2004) are considered invasive—defined here as species found in higher abundance outside of their native range and that negatively impact the environment or other native species (Keller et al. 2011, Russo 2016).
Whether or not a species is considered invasive, the introduction and establishment of non-native bees can lead to negative ecological interactions (Goulson 2003). For bee pollinators, these negative interactions can occur via three mechanisms: 1) competition for food and nesting resources with native bee species, 2) facilitation of the reproduction of other non-native species (e.g., pollination of invasive weedy plants), and/or 3) the spread of novel parasites and pathogens that can be detrimental to native bees (Mallinger et al. 2017). These potential ecological outcomes of the establishment of non-native bees have been most studied for species intentionally introduced to novel areas for crop pollination. Competition for floral resources between introduced and native bees has been demonstrated even though these negative effects can be highly context-dependent (Roubik and Wolda 2001, Thomson 2004, Cane and Tepedino 2017, Graham et al. 2019). There is also evidence for non-native bees facilitating the reproduction of non-native plant species. In North America, A. mellifera is the primary pollinator of the invasive yellow star thistle (Centaurea solstitialis, Asterales: Asteraceae) in what has been dubbed as an ‘invasive mutualism’ (Barthell et al. 2001). Evidence for pathogen spillover between introduced B. terrestris and native bumble bees in Chile has been demonstrated and linked to declines in several Chilean bumble bee species (Schmid-Hempel et al. 2014, Arismendi et al. 2021). In contrast, the potential for negative effects from unintentionally introduced bees has received less attention.
In North America, a total of 55 non-native bee species have been introduced (Russo 2016). While most of them are not considered invasive, the ecological effects of their introduction remain poorly understood. Recent reviews about the biology and expansion of Megachile sculpturalis and A. manicatum have indicated numerous gaps of knowledge about these two unintentionally introduced non-native bees and have suggested a high potential for invasiveness for both (Strange et al. 2011, Dubaic and Lanner 2021). Here, we review the natural history and distribution of Osmia taurus (Hymenoptera: Megachilidae), a recently and accidentally introduced species to eastern North America, and discuss its potential for invasiveness. In addition, we compare the biology and history of colonization of O. taurus to its congener mason bee species Osmia cornifrons, which was intentionally introduced to the east coast of North America in 1978 for orchard pollination (Batra 1982).
Nesting Biology, Life Cycle, Floral Preferences, and Natural Enemies
O. taurus is a mason bee native to eastern Asia (Maeta 1978, Yong et al. 2014, Branstetter et al. 2021). Taxonomically, O. taurus is part of the subgenus Osmia group bicornis, which also contains several other important orchard pollinators such as O. cornifrons (Radoszkowski), O. lignaria Say, O. bicornis (L.), and O. cornuta (Latreille) (Branstetter et al. 2021). As indicated by their common name, these bees use mud to construct brood cell partitions and caps for their nests. Female mason bees use pre-existing cavities, such as reeds, twigs, or other hollowed-out holes to construct their cells, store pollen provisions, and lay their eggs (Fig. 1). The diameter of the cavities varies among species and scales with the bees’ body size (Rust 1998). Generally, both O. taurus and O. cornifrons females prefer to use cavities between 6 and 8 mm in diameter to lay their eggs (Maeta 1978) but males have been found developing in cavities as small as 4 mm in diameter (K.A.L., personal observations).
Nesting structure of Osmia taurus. Mud partitions separate the cocoons through the cavity. Yellow areas in nesting structures are residual pollen while specks inside and on the side of the blocks are frass. Note the mud plug at the entrance of the nest (right side). Photos courtesy of Dawson M. Little.
Morphologically, O. taurus and O. cornifrons can be easily distinguished from native North American Osmia by the presence of two large horns on the outside edges of the clypeus (Rust 1974). In contrast, differentiating O. taurus and O. cornifrons based on external morphology can be difficult, particularly for males. Both species have a brassy black body with whitish-yellow, tawny hair on their face, thorax, and abdomen (Yasumatsu and Hirashima 1950, Droege and Pickering 2022). The most commonly used characters to differentiate these two Osmia species are on their clypeus (Fig. 2; Yasumatsu and Hirashima 1950). For males, the main difference is that the lower edge of the clypeus in O. cornifrons is completely smooth and shiny while the lower edge of the clypeus in O. taurus is smooth and shiny in the center but is pitted further out (Fig. 2B and D). Females are easier to identify than males, as O. cornifrons females have a knob-like acute median apical projection on their clypeus while O. taurus have a clypeus that is markedly dented in the middle (Fig. 2A and C). Characters in the male genitalia can also be used for the differentiation of these two Asian Osmia. O. cornifrons have slightly larger genitalia than O. taurus and the gonocoxite is narrowly expanded in O. cornifrons compared to the widely expanded structure in O. taurus (Yasumatsu and Hirashima 1950). Both species exhibit a distinct sexual dimorphism, with females being typically larger than males (body length in O. taurus females: 10–12 mm, males: 8–10.5 mm; body length in O. cornifrons females: 8–11.5 mm, males 7.5–9 mm) (Yasumatsu and Hirashima 1950). Males also lack scopa—a structure with long branched hairs —under the abdomen, which is used by the females for pollen transportation (Fig. 3).
Clypeal morphological characters to differentiate Osmia taurus (A–B) and Osmia cornifrons (C–D). The clypeus of the O. taurus female is markedly dented in the middle (A), and in males is smooth and shiny in the center but is pitted on the edges (B). The clypeus of female O. cornifrons has a central knob-like protrusion (C), while in males is completely smooth and shiny from edge to edge (D). Note that male clypeal hairs were slicked back with water in order to obtain a better view of the clypeus. Photos by Grace Gutierrez.
Side photos of Osmia taurus showing sexual dimorphisms. (A) Side profile of O. taurus male shows long antennae and lack of scopal hairs under the abdomen. (B) Side profile of O. taurus female shows a more robust abdomen and the presence of a scopa. Note the color of the abdominal hairs with distinct red to orange hue. Photos by Grace Gutierrez
Both within and outside of its native range, O. taurus emerges in the spring, typically between late March to early April, with males emerging 2–4 days before females (Maeta 1978, LeCroy et al. 2020). Adult O. taurus are active for about 2–6 weeks and females are active about 2 weeks longer than males (Maeta 1978). Complete development from egg to adult occurs between late May and late August and is temperature-dependent (Maeta 1978). From data collected in Japan, it appears that O. taurus has a significantly shorter developmental duration compared to O. cornifrons when raised at constant temperatures between 18 and 26°C (Maeta 1978).
In its native range, O. taurus has been classified as a polylectic species with a preference for pollen from the Juglandaceae, Rosaceae, and Fabaceae families, which is similar but broader than the pollen preferences of O. cornifrons that prefers pollen from the Rosaceae and Fabaceae families (Maeta 1978, Haider et al. 2014, Vaudo et al. 2020). In North America, O. taurus has been observed visiting Cercis canadensis, the eastern redbud (Fabales: Fabaceae), and several species in the genus Prunus (Rosales: Rosaceae) (Potter and Mach 2022). However, O. taurus seems to collect pollen primarily from non-native shrubs species of the genera Aesculus (Sapindales: Sapindaceae), Viburnum (Dipsacales: Adoxaceae), and Ilex (Aquifoliales: Aquifoliaceae) (Potter and Mach 2022). This pattern of preference for non-native plants has also been reported for O. cornifrons (Vaudo et al. 2020). Despite their overlap in pollen preferences, the texture of the pollen provisions is markedly different between these two species. Pollen provisions in O. taurus are dry while pollen is wet in the provisions of O. cornifrons (Maeta 1978; Fig. 4).
Pollen provisions for offspring in Osmia taurus and O. cornifrons. (A) Nests of O. taurus have dry pollen provisions. Note that the eggs are located in the midst of the dry pollen. (B) O. cornifrons provision their cells with wet pollen. Eggs can be observed in the center of the pollen provisions. Photos by Kate LeCroy.
O. taurus is host to several natural enemies in its native range, which consume the pollen provisions or the larvae themselves (Maeta 1978). There are records of O. taurus being parasitized by several parasitoid wasp species including Leucospis japonica (Hymenoptera: Leucospidae), Monodontomerus osmiae (Hymenoptera: Chalcidoidea), and Melittobia acasta (Hymenoptera: Eulophidae). Other macroscopic parasites of O. taurus include the bee fly Anthrax jezoensis (Diptera: Bombyliidae), the skin beetle Trogoderma varium (Coleoptera: Dermestidae), the spider beetle Ptinus japonicus (Coleoptera: Ptinidae), the corn moth Nemapogon granella (Lepidoptera: Tineidae), the booklouse Liposcelis bostrychophila (Psocoptera: Liposcelididae), and parasitic mites in the genus Chaetodactylus spp. (Sarcoptiformes: Chaetodactylidae) (Kamijo 1963, Maeta 1978). Relatively less is known about microscopic nest parasites. An unknown fungus, potentially in the genus Ascosphaera (Onygenales: Ascosphaeraceae)—the causing agent of chalkbrood disease—has been detected in O. taurus’ nests in Japan (Maeta 1978, Skou 1988).
In its non-native range, reports have shown that O. taurus nests are parasitized by chalcid wasps (Monodontomerus spp.), as well as infected with Ascosphaera fungi (LeCroy pers. obs., LeCroy et al. 2020), and Wolbachia bacteria (Saeed and White 2015). Several other parasites and pathogens are likely to be found in O. taurus based on their detection in other closely related Osmia species. For example, two parasitoids from their native range, Chaetodactylus spp. and Melittobia acasta have been documented in O. cornifrons and O. lignaria, respectively, but not in O. taurus in North America (McKinney and Park 2013, Glasser and Farzan 2016). Other parasites and pathogens such as Vairimorpha spp. (Dissociodihaplophasida: Nosematidae; formerly known as Nosema) and Crithidia bombi (Trypanosomatida: Trypanosomatidae), which are important pathogens in some groups of bees, have been recorded in other species of Osmia but with few reports of negative fitness effects (Müller et al. 2019, Figueroa et al. 2021). Honey bee-associated viruses such as Deformed Wing Virus (DWV), Lake Sinai Virus (LSV), Apis mellifera filamentous virus (AmFV), Black Queen Cell Virus (BQCV), Varroa destructor Macula-like Virus (VdMLV) have similarly been detected in other Osmia species but not directly in O. taurus (Ravoet et al. 2014). Overall, little research has been done on the presence of these pathogens and the fitness effects these pathogens can have on O. taurus and other mason bees.
Species Distribution and History of Expansion into North America
O. taurus’ ancestral geographic range comprises Japan, Korea, China, and Eastern Russia (Maeta 1978, Wei and Wang 1992, Proshchalykin 2004, Yong et al. 2014). In Japan, Maeta (1978) reported that O. taurus has a widespread distribution and is found across the main five Japanese islands (Fig. 5A). This distribution contrasts with O. cornifrons, which was primarily found in the northern and central parts of Japan. However, in South Korea, O. taurus and O. cornifrons appear to have similar distributions. Little data is available on their ranges in North Korea, Russia, and China (Fig. 5A). In their non-native range of North America, O. taurus and O. cornifrons appear to have been introduced around the same area, despite their first records occurring decades apart (LeCroy et al. 2020). At present, both species have spread to similar areas except for O. taurus reaching south to Georgia and Florida and O. cornifrons establishing disjunct populations in western states, likely through anthropogenic transport (Fig. 5B; Supp Table S1 [online only]). Despite their similar distributions in North America, both O. taurus and O. cornifrons appear to have some differences in habitat use. In general, O. taurus is found in greater abundance in wooded areas, while O. cornifrons has a closer association with agricultural landscapes dominated by orchards (Batra 1982, Savoy-Burke 2017, Makino and Okabe 2019, Fuminori 2020).
Maps showing the distributions of Osmia taurus and O. cornifrons in their (A) native range in Eastern Asia, and (B) their non-native range in North America. Geographic boundaries indicate state and provinces. Colors indicate the presence of O. taurus (red), O. cornifrons (yellow), or both species (orange).
The introduction and establishment history of O. taurus in North America is uncertain. The first recorded occurrence of O. taurus was in Maryland in 2002 (Droege and Pickering 2022), and in only 20 years, it has been reported in 22 states in the United States and southern Ontario in Canada (Fig. 6). It has been suggested that O. taurus and O. cornifrons were introduced at the same time on the east coast and that O. taurus merely escaped detection for many years because it is morphologically similar to O. cornifrons (Gibbs et al. 2017). However, no specimens of O. taurus have been reported before 2002 despite efforts to find them mislabeled as O. cornifrons in public and private collections (K.A.L., personal observations). Furthermore, given how quickly it has spread since 2002, it is unlikely that O. taurus would have stayed undiscovered for over 20 years if it had been introduced concurrently with O. cornifrons.
Spread of Osmia taurus in its non-native range since first recorded presence in 2002. Stars represent the locations where O. taurus was first documented in North America in 2002 (West Virginia and Maryland, USA).
In the future, O. taurus is likely to become widespread in North America given its rapid spread in the eastern United States and the potential for human-mediated long-distance movements. In Asia, O. taurus occupies a latitudinal gradient spanning 29°N to 43°N, which corresponds to the latitudes of its current naturalized range, from Florida to New Hampshire, in the United States. At the moment, only O. cornifrons has naturalized in climatically different parts of the western United States characterized by cool wet winters and warm dry summers (Cbs on Köopen-Geiger climate classification system) (Taleghani et al. 2014, Beck et al. 2018). However, it has been hypothesized that O. taurus exhibits tolerance to a wider range of climatic conditions than O. cornifrons in its native range (Maeta 1978), therefore it is possible that O. taurus will exceed the distribution of O. cornifrons in the western and southern ranges of North America.
Biological Traits and Data Suggesting Potential for Invasiveness
Non-native species often exhibit traits that make them suitable for habitats outside of their native ranges and contribute to their potential for invasiveness. Traits such as phenological plasticity (Sakai et al. 2001), generalist habitat requirements (Rosecchi et al. 2001, Sultan 2001), and broad thermal tolerance breadth (Kelley 2014) have been associated with species with a greater potential for invasiveness. For bees, another common trait among non-native species is the use of above-ground cavities like stems and twigs or man-made cavities such as bricks or plastic straws (Russo 2016). Even though O. taurus is a cavity nester and uses floral resources from a broad range of hosts, this species does not have several of the biological traits associated with insect invasions. For example, O. taurus has a relatively narrow activity period and the adult stage is limited to about 6 weeks in the spring. It has been hypothesized that differences in physiological thermal tolerance between O. taurus and O. cornifrons explain why O. taurus is distributed throughout the entire island of Japan while O. cornifrons is more restricted to the center and north of the island. Maeta (1978) speculated that adults of O. cornifrons cannot survive the warm temperatures of summers in southern Japan and that the winters are not cold enough to induce diapause. However, empirical evidence for these hypotheses is lacking. Future work should investigate the role of physiological traits in predicting the potential distribution of these bee species in their introduced ranges.
Still, one remaining question is whether O. taurus is likely to become invasive. Generally, evidence for negative impacts of non-native species in the introduced ranges is used as the determining factor in naming a species as either ‘non-native’ or ‘invasive’ (Russo 2016). Among mason bees, the native species in the eastern United States that has the greatest ecological overlap with O. taurus is the blue orchard bee, Osmia lignaria lignaria. Both species emerge at about the same time, use nest cavities of similar size (Rust 1998, LeCroy et al. 2020), and collect pollen from similar sources. O. lignaria typically collects pollen from Fabaceae and Rosaceae (Kraemer et al. 2014, Pinilla-Gallego and Isaacs 2018), some of the same plant families that O. taurus uses in its native range. Although there is evidence of strong population decline by O. lignaria lignaria in eastern North America (LeCroy et al. 2020), the decline apparently started prior to the arrival of the two Asian species O. taurus and O. cornifrons (Centrella 2019). Thus, it is unclear if O. taurus is further hastening the decline of O. lignaria lignaria or is taking over the niche that this native Osmia is relinquishing. Such mechanisms are not always easy to disentangle (Gurevitch and Padilla 2004).
Another piece of evidence suggesting invasiveness potential is the rapid population growth observed in O. taurus compared to O. cornifrons. In Maryland and Virginia, O. taurus went from undetected before 2002 to comprising 22% of all Osmia individuals captured using pan traps in 2003–2009, and 43% of all Osmia individuals collected in 2010–2017 (LeCroy et al. 2020). Overall, O. taurus’ population in this region has increased by 17% per year, dramatically different from the native Osmia spp. whose populations have declined annually over the same time span, and the non-native congener O. cornifrons that exhibits stable populations (LeCroy et al. 2020). However, O. taurus has not become the dominant Osmia species everywhere it has reached. In Pennsylvania, a state where it was found shortly after the first records, O. cornifrons remains more abundant than O. taurus. In a recent study of 6 years of bee monitoring using Blue Vane traps in southern Pennsylvania, O. cornifrons accounted for 23% of all Osmia species while O. taurus accounted for about 10% (Turley et al. 2022). However, this study took place around apple orchards, the preferred habitat of O. cornifrons, which may explain its higher relative abundance compared to O. taurus in southern Pennsylvania (USA). It is still unclear what is contributing to the rapid local population growth of O. taurus relative to O. cornifrons in some parts of the distribution in their non-native ranges. The two species overlap in the same ways that O. taurus and O. lignaria lignaria overlap, and O. cornifrons became established in eastern North America two decades before O. taurus. Additionally, O. cornifrons has been routinely supplemented from commercial sources for orchard pollination, yet O. taurus is regionally more common in some areas and shows a greater rate of population growth at least in the states of Maryland and Virginia (USA) (LeCroy et al. 2020).
Additional evidence indicates that O. taurus may have adverse effects on the environment through the facilitation of other non-native species. Seemingly, O. taurus has been reported pollinating non-native plant species in North America suggesting a possible indirect negative effect on native plants that can be the preferred floral hosts of native Osmia bees (Potter and Mach 2022). However, future studies should investigate to what extent native and non-native Osmia share the same native and non-native floral resources in different habitats (e.g., Vaudo et al. 2020). There is a need for studies on whether the preference of non-native plants by non-native bees promotes an invasive mutualism by which both species facilitate the others’ spread to the detriment of native species.
The potential spread of non-native pathogens from O. taurus to native Osmia species should also be investigated. The presence of fungal pathogens from Japan (i.e., Ascosphaera naganensis and Ascosphaera fusiformis) has been reported in the United States in several native Osmia spp. (Hedtke et al. 2015, LeCroy et al. 2022, in press). While the presence of these Ascosphaera spp. in North America has been linked to the introduction of O. cornifrons (Hedtke et al. 2015), it is plausible that this pathogen could have been introduced or re-introduced by other species of Japanese origin (e.g., O. taurus). Similarly, LeCroy and colleagues found that native bees show a higher prevalence of non-native Ascophaera fungus at sites that contained larger numbers of O. taurus and O. cornifrons (LeCroy et al. 2022, in press). More comprehensive pathogens and parasite surveys in the native and non-native range of O. taurus and O. cornifrons will be necessary to fully characterize to what degree novel parasites and pathogens have been introduced, if they differentially influence population growth, and to what degree, their spread could be detrimental to native bee species.
Future Directions
O. taurus was first recorded in the United States two decades ago and since then its population has rapidly expanded in size and range. Gaps of knowledge about this species’ ecology in its native and introduced range include information about pollen use, habitat requirements, and thermal tolerance range to better determine its potential spread and competition with native species. With a broad trend of decline among native Osmia species in the eastern United States and no clear cause of those declines (LeCroy et al. 2020), there is a critical need to examine local Osmia species richness, abundance, floral preference, and incidence of disease, parasitism, and predation before and after the arrival of O. taurus and O. cornifrons. Such studies will help discern the broader ecological effects of the introduction of non-native bee species (Mooney and Cleland 2001) as well as the mechanisms that give them an advantage over closely related native species (Catford et al. 2018).
Acknowledgments
We would like to thank Shelby Fleischer and the López-Uribe lab for their comments on earlier versions of the manuscript. We would also like to thank Julie Urban and Tanya Renner at Penn State University for the use of their camera microscopes, and Z. Wang and C. Tern for their translation help.
Funding
GMG was supported by a Huck Graduate Fellowship from the Pennsylvania State University. MML-U was funded through the USDA National Institute of Food and Agriculture (NIFA) Appropriations under Projects PEN04716 Accession No. 102052.
Conflict of Interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
