Stator limbatus is a phytophagous beetle native to warm regions of North and Central America, feeding on Fabaceae seeds and one of the most polyphagous species within the subfamily Bruchinae, here reported for the first time in Europe and on new hosts. Adult beetles emerged from Acacia spp. seeds collected in the islands of Corsica (France), and Sardinia (Italy). The wide presence in Sardinia and Corsica supports the hypothesis that this alien species was introduced several years ago. In both islands, S. limbatus emerged from Acacia mearnsii seeds, with infestation rates of up to 74.2 and 90.8% in 2019 and 2020, respectively. This seed beetle also emerged from two previously unreported host species, Acacia saligna and A. pycnantha, showing highest infestation rates of 4.0 and 95.1%, respectively. Both Acacia species are reported as new host associations with S. limbatus. Overall, seed infestation rates recorded in 2019 and 2020 indicate that S. limbatus is well established and that Mediterranean bioclimatic conditions are suitable for its population increase in size. This study lays the foundations for further research on known and potential host species and the spread and distribution of S. limbatus in Europe.

Acacia mearnsii, Acacia pycnantha, Acacia saligna, alien species, bean weevil, biological invasion, Mediterranean islands

The global movement of people and goods and climate change are dramatically promoting the introduction of alien species in non-native environments in the Anthropocene (Kueffer 2017), resulting in a continuous accumulation of these species worldwide (Seebens et al. 2017; Venette and Hutchison 2021). This indicates that current measures to avoid new introductions of alien species are not always effective. Therefore, prevention, continuous monitoring in priority sites, early detection, and rapid intervention are of major importance for avoiding the establishment of new invasive alien species and agricultural or forestry pests and for reducing the spread of the existing ones, with special concern towards protected areas and natural ecosystems.

Among seed-feeding insects, the subfamily Bruchinae (Coleoptera, Chrysomelidae) beetles, renowned as bean weevils, is highly specific and likely the most important (van Klinken 2005). This family includes about 4,350 taxa distributed worldwide (Borowiec 1987). The beetle Stator limbatus (Horn, 1873) (Coleoptera, Chrysomelidae: Bruchinae) is an endophagous seed feeder of legumes (Fig. 1). Its native range spans from semiarid and xeric regions of southwestern United States and northern Mexico to dry tropical forests of Central America and northern South America. Stator limbatus has a generalist habit and a wide host range, as it has been collected from > 90 host plant species (de Jesús Parra-Gil et al. 2020), including many species of the genus Acacia s.l. In its native range, it affects mostly native species, but also about 20 non-native species (Stillwell et al. 2007). Despite that, host colonization of S. limbatus populations varies greatly among regions, and distinct populations exhibit host specialization at a local scale (Morse and Farrell 2005a, 2005b). Beetle populations are known to express phenotypic plasticity to host species by adapting pre-imaginal development time and body and egg size (Amarillo-Suarez and Fox 2006; Amarillo-Suarez et al. 2017).

Figure 1.

Habitus of adult Stator limbatus A dorsal and B lateral view.

Eggs are oviposited on mature seeds inside of dehiscent or partially dehiscent pods when they are still on the plant (Johnson 1981a; Kingsolver 2004). Females usually lay one egg per seed, and newly hatched larvae burrow into the seed integument beneath the egg, complete their development and pupate inside the same seed. In the case in which seeds are limiting, more eggs are deposited across a seed (Morse and Farrell 2005a). Beetles emerge from seeds as adults, mate and females start ovipositing within 24–48 hours, under laboratory conditions. Adults are facultatively aphagous, as they only require resources acquired during the pre-imaginal stage to complete development and reproduce (e.g. capital breeders) (Stillwell and Fox 2009). The generation time at 28 °C was determined to be 28–30 days (Amarillo-Suarez and Fox 2006).

Several species within the S. limbatus host range, such as Acacia mearnsii De Wild and Acacia saligna (Labill.) H.L.Wendl. native to Australia, have shown in Europe invasive potential and negative impacts on native species, to the extent that containment measures have been implemented (Lowe et al. 2000; European Union 2014; Tozzi et al. 2021). Therefore, monitoring the presence of seed beetles of invasive Acacia spp. in Europe is relevant in the perspective of finding and evaluating potential natural enemies able to slow the expansion and mitigate the adverse impacts of those species. Since Acacia in the broad sense have been grouped into distinct genera, e.g., Mariosousa, Vachellia, and Senegalia, and also other host species in the Leguminosae have been synonymized or renamed, a dedicated study would be required to define the current host range of the bruchid with valid plant names.

Outside its native range, S. limbatus has been reported in Hawaii (Bridwell 1920), South America (Oliveira and Costa 2009; Romero Gomez et al. 2009; Meiado et al. 2013), South Africa (Rink 2013), Iran (Boroumand 2010; Ghahari and Borowiec 2017), and United Arab Emirates (Delobel 2011), whereas reports from Mauritius, Saudi Arabia, Yemen, and Oman were unconfirmed (Rink 2013).

In the framework of an international project assessing the risk of invasion of selected alien species (ALIEM) (Inghilesi et al. 2018), some Acacia spp. seeds were tested in a germination test during which several individuals of S. limbatus adults emerged from seed lots of A. mearnsii seeds collected in Corsica (France) and Sardinia (Italy) in 2018. This insect species has not been yet recorded in Europe, so that new field collections were planned and carried out in 2019 and 2020.

The main aim of the present study was to investigate the establishment of S. limbatus in Sardinia and Corsica according to the traits described by Yus-Ramos et al. (2014) for alien seed beetles, as well as its host association and infestation levels. In addition, a literature search analysis was carried out to provide an updated inventory of host species of S. limbatus with valid names, as understanding and predicting host shifts on other Acacia species is of pivotal importance in order to define its potential distribution in the Mediterranean Basin.

The extensive collection of S. limbatus during the field surveys in 2019 and 2020 in Sardinia and Corsica following the first record in 2018 indicates that the seed beetle has found suitable climatic conditions and has established in Europe. Stator limbatus can be considered established according to the definition of Yus-Ramos et al. (2014), i.e., a species able to reproduce successfully in natural ecosystems. Stator limbatus exhibits biological features that could support its further spread in Europe. At first, this species has a wide host range worldwide, with about 15 species reported in Europe (Euro+Med 2021; GBIF 2021). Furthermore, its native geographic range includes diverse climates, spanning from dry forests of northern South America to deserts of Central America and southwestern United States (Stillwell and Fox 2009). In addition, this bruchid developed under laboratory conditions also on non-native species, including Acacia cyclops A.Cunn. G.Don and S. ataxacantha (syn. A. ataxacantha) (native to Australia and South Africa, respectively) (Rink 2013), as well as non-host species, such as C. humilis, C. texanum, P. juliflora, P. velutina, V. constricta, and V. farnesiana (Bridwell 1920; Johnson 1981b; Fox et al. 1996). Finally, S. limbatus have shown adaptive oviposition phenotypic plasticity in response to host species, as fewer and bigger eggs are laid on exotic or unfavorable hosts (Amarillo-Suarez et al. 2017). Such maternal egg-size plasticity is suggested to be an ancestral trait influencing the evolution of the diet breadth (Amarillo-Suárez and Fox 2006). Overall, the wide presence of host species of S. limbatus in Europe, its strong host shift potential, and climate adaptation suggest its possible spread in Mediterranean environments, and its presence in unsampled areas cannot be ruled out.

This species was recovered from Acacia spp. seeds in Sardinia, in multiple sites distant up to 150 km, and Corsica, in four areas distant about 130 km. Even though the country of first introduction in Europe remains undetermined, the wide presence of this alien insect in distant areas supports the hypothesis that its introduction occurred several years ago. The introduction of S. limbatus in Europe was most likely accidental and its detection unexpected. The pathway of first introduction is presently unknown, as no specific custom interception has so far been reported. With regard to pathways of secondary spread, in view of its wide host range and endophytic behavior of larvae, we may assume that it was introduced through movement of contaminated commodities, i.e., plants for planting, as a parasite of seeds (CBD 2014; Faulkner et al. 2020). In fact, after its first introduction, a secondary spread pathway may have occurred as a result of movement of contaminated plants (with pods) or seeds of A. saligna, A. mearnsii, and A. pycnantha, which are commonly planted in southern Europe and significantly traded. In addition, the very large number of different host species should be taken into account (Table 1), as many are common ornamental, i.e., Albizia spp., Leucaena spp., Parkinsonia spp., and Glycine max (L.) Merr., or forestry and multipurpose species, i.e., Acacia spp., in the Mediterranean area. Therefore, in order to investigate the S. limbatus presence or intercept its introduction in areas nearby Sardinia and Corsica, specific monitoring plans on its host species should be set up in southern France and mainland Italy. Although the pathways of first introduction and secondary spread are generally not known for bruchid seed beetles, several authors suggest introductions through importation of seed or nursery stocks of host plant species for ornamental or forestry purposes, e.g., Bruchidius terrenus (Sharp, 1886) on Albizia julibrissin Durazz. and Amblycerus robiniae (Fabricius, 1781) on Gleditsia triacanthos L. in the United States (Kingsolver 2004; Hoebeke et al. 2009; Yus-Ramos et al. 2014).

The introduction of alien seed beetles in Europe shows an increasing trend in the last 20 years, in accordance with the worldwide trend described by Seebens et al. (2017). Beenen and Roques (2010) reported 14 Bruchinae alien species in Europe, seven of which introduced before 1900, three species in the period 1901–1950, two in 1951–2000, and finally two species reported from 2001 to 2010. Yus-Ramos et al. (2014) further extended the list of alien bruchids in Europe to a total of 42 species, including four recent introductions, namely Bruchidius radiannae Anton & Delobel, 2003 and Caryedon acaciae (Gyllenhal, 1833) on Vachellia karroo (Hayne) Banfi & Galasso (syn. Acacia karroo Hayne) in 2007 in Spain (Yus Ramos and Coello García 2007, 2008), Acanthoscelides macrophthalmus (Schaeffer, 1907) on Leucaena leucocephala (Lam.) de Wit in Cyprus in 2007 (Vassiliou and Papadoulis 2008), and B. terrenus on A. julibrissin in Bulgaria in 2009 (Stojanova 2010). Furthermore, A. robiniae was reported on G. triacanthos in Romania in 2018 following an unconfirmed report in Hungary in 1986 (Rădac et al. 2021). Therefore, according to literature reports, seven species of bruchids have been reported in Europe in the last 20 years. In both Corsica and Sardinia, S. limbatus larvae developed on seeds of A. mearnsii, a tree native to Australia which has shown to be invasive in Europe, South America, and Africa. This insect-host association has been previously reported in Brazil, where an infestation rate of 44.3% was observed (Oliveira and Costa 2009), and South Africa (Rink 2013). Acacia mearnsii is cultivated in Brazil for tannins, cellulose, and charcoal production (Garlet et al. 2011), whereas in Europe and in South Africa, presently, this species has a lower significant economic importance and is rather invasive (Souza-Alonso et al. 2017; Railoun et al. 2021).

In Sardinia, beetle adults emerged abundantly also from A. pycnantha seeds, and, interestingly, 45% of sampled seeds showed two exit holes, differently from A. mearnsii seeds which showed a single exit hole. This brings evidence that A. pycnantha seeds support the development of more than one larva of S. limbatus, most likely because of the bigger size of its seeds compared to those of A. mearnsii. In central eastern Sardinia, the infestation rate was more homogeneous among sampling sites in 2020 than in 2019, as the range decreased from 49.9% (24.3–74.2%) in 2019 to 5.4% (85.4–90.8%) in 2020. Moreover, infestation rates increased significantly on both A. mearnsii and A. pycnantha. However, the seed production of trees in the sampling sites was not quantitatively estimated being beyond the aims of the study. Estimates of seed infestation rates with no assessment of tree seed production and over such a short period, i.e. two years, prevent to infer on spatio-temporal population trends of S. limbatus. The same insect abundance can, in fact, cause high infestation rates in the event of low seed production or low rates when seed production is high. Nonetheless, although Acacia spp. seed production and accumulation may vary widely, Australian and African species usually produce large or very large quantities of seed and may have large soil-stored seed banks (Gibson et al. 2011). High production of seeds for the three investigated species has been observed both in the native and in the invaded ranges, e.g., A. mearnsii in South Africa (Impson et al. 2021), being one of the drivers of invasiveness at the global level. Indeed, large amounts of pods were observed on Acacia spp. trees as well as seeds in the topsoil in both 2019 and 2020 (A. Cocco, Y. Petit, pers. obs.). Furthermore, high numbers of seedlings were observed in the sampling sites with A. mearnsii.

Previous studies on infestation by S. limbatus on Fabaceae species reported seed damages of 15% on E. timbouva (Meiado et al. 2013), 19% on Acacia mangium Willd. (Mojena et al. 2018), and 70% on Acacia podalyriifolia A.Cunn. ex G.Don (Garlet et al. 2011) in Brazil. In Mexico, seed infestation rates of 16.8% were observed on Painteria leptophylla (DC.) Britton & Rose (de Jesús Parra-Gil et al. 2020) and 33.6% on Mariosousa coulteri (Benth.) Seigler & Ebinger by both S. limbatus and Merobruchus santarosae Kingsolver, 1989 (Coleoptera, Chrysomelidae) (Romero Gomez et al. 2009). Susceptibility to S. limbatus widely varied among hosts and areas; however, comparisons are difficult, as seed infestation rates are influenced by a number of abiotic and biotic factors, including seed availability and environmental conditions. Despite its recent report in South Africa, S. limbatus has not been reported infesting A. pycnantha seeds (Rink 2013; Magona et al. 2018).

A word of caution is in order with regard to A. saligna as a host species for S. limbatus. In fact, infestation rates were very low in both years and countries, and the highest values (4% in 2019 and 2.6% 2020) were observed in the same site. Nonetheless, infestation by S. limbatus on A. saligna seeds was not limited to a single site, as infested plants were observed in both Sardinia and Corsica. Moreover, beetle eggs were observed on up to 80% of A. saligna seeds, especially on plants near to infested Acacia spp. trees. This could be due to an opportunistic egg-laying behavior on the nearest alternative hosts. Furthermore, oviposition on A. saligna indicates that seeds had no antixenotic effect on female oviposition and oviposition is promoted by suitable hosts nearby. Chemical or physical barriers on A. saligna seeds preventing larval development cannot be ruled out and would require further investigations. Laboratory tests carried out in South Africa investigating the oviposition preference showed that S. limbatus females accepted A. saligna seeds for oviposition, together with seeds of A. cyclops, A. mearnsii, Paraserianthes lophantha (Willd.) I.C.Nielsen (invasive non-native species in South Africa), and Vachellia tortilis (Forssk.) Galasso & Banfi [syn. Acacia tortilis (Forssk.) Hayne], S. ataxacantha, Senegalia caffra (Thunb.) P.J.Hurter & Mabb. [syn. A. caffra (Thunb.) Willd.], Senegalia nigrescens (Oliv.) P.J.Hurter [syn. A. nigrescens (Oliv.)] and Vachellia sieberiana var. woodii (Burtt Davy) Kyal. & Boatwr. [syn. A. sieberiana var. woodii (Burtt Davy) Keay & Brenan] (native species to South Africa). However, adults emerged only from A. mearnsii, A. cyclops, and S. ataxacantha, indicating that food availability may not be the only factor limiting the larval development (Rink 2013).

In view of its high seed infestation rates, S. limbatus has been suggested to play a role as biocontrol agent of invasive non-native Acacia species (Rink 2013). In South Africa, extensive biological control programs have been developed against invasive tree species, as, for example, the release of A. macrophthalmus for biological control of L. leucocephala in 1999 (Olckers 2004). Five seed-weevil Melanterius spp. (Colepotera, Curculionidae) were introduced from Australia in different periods to reduce the invasiveness of P. lophantha and ten Acacia spp., including the three species investigated in the present paper, i.e., A. mearnsii, A. saligna, and A. pycnantha (Impson et al. 2011). Seed damage caused by weevils varied largely among sites and years from 4% to over 90%. Such variability was explained by a specific 4-year study on Melanterius–Acacia relationship and was mostly due to variable seed quality that resulted in low larval and pupal survival rates (Impson and Hoffmann 2019). Overall, seed-feeders are unlikely to effectively reduce the Acacia spp. density as a stand-alone control agent due to the extraordinarily high prolificacy of plants resulting in huge accumulation of long-lived seeds in the soil. In fact, effective results were obtained through the release of the flower-galling midge, Dasineura rubiformis Kolesik (Diptera, Cecidomyiidae) complemented by a seed-feeding weevil, Melanterius maculatus Lea (Coleoptera, Curculionidae), which caused a strong reduction of seed production of A. mearnsii (Impson et al. 2021). This reduction is expected to curb the accumulation rate of the seed banks and, in the medium-long term, reduce the spread of the invasive species. Besides a potential biocontrol agent of invasive plant species, further beneficial environmental effects by S. limbatus may be represented by the promotion of seed germination, e.g., on Enterolobium contortisiliquum (Vell.) Morong and E. timbouva Mart. (Meiado et al. 2013).

The present findings indicate the adaptability of S. limbatus to new host species when established in new areas. Stator limbatus also showed phenotypic plasticity in response to seed size or seed quality (Amarillo-Suárez and Fox 2006), in accordance with findings in other species (Hardy et al. 1992; Tsai et al. 2001). Moreover, this is consistent with results from studies showing that development time decreased and adult mass increased when insects developed on high quality hosts (Lindroth et al. 1991; Stockhoff 1993). Therefore, host shifts on local plants and new host associations cannot be ruled out in Europe in view of its ability to accept and adapt to local hosts. Adaptation to new or non-preferred host species has been observed on other coleopteran alien species, such as the red palm weevil Rhynchophorus ferrugineus (Olivier, 1790) (Coleoptera, Dryophthoridae) on the dwarf palm, Chamaerops humilis L. (Cocco et al. 2019). Importantly, S. limbatus has been reported on > 90 host species and ≥ 20 genera (de Jesús Parra-Gil et al. 2020), which is one of the widest host ranges within the Bruchinae. In view of its tropic spectrum, it has been classified as polyphagous, i.e., feeding in the seeds of various plant genera of different subfamilies (Ribeiro-Costa and Almeida 2012; Yus-Ramos 2018). However, its host use is widely variable and it shows local specialization depending on the diversity of available host species (Morse and Farrell 2005a, 2005b). The establishment of S. limbatus in Europe and new associations with A. pycnantha and A. saligna required a redefinition and update of the bruchid host range to facilitate further research on its potential adaptation and spread in Europe. The exact definition of the host range of S. limbatus is not trivial due to nomenclatural issues within the family Fabaceae which have not been resolved (LPWG 2017). In addition, in a number of cases, the literature reported incorrect or partial names for the host plants. The bibliographic search analysis allowed to extend the global host range of S. limbatus to 111 species, in most part belonging to the mimosoid clade of the subfamily Caesalpinioideae (Fabaceae) (LPWG 2017). Synonym issues were resolved, e.g., Acacia diversifolia and Leucaena diversifolia both mentioned by Romero Gomez et al (2009) and synonymized in Leucaena diversifolia (Schltdl.) Benth, and up-to-date nomenclature provide the current and comprehensive overview of the feeding spectrum of S. limbatus. However, some old or unconfirmed reports would require further investigations, e.g., G. max, Wallaceodendron celebicum Koord., and Arachis hypogaea L. (Brian 1932; Kingsolver 2004). Since no previous records were found in literature, A. pycnantha and A. saligna are included in the present paper for the first time in the host range of S. limbatus.

This report of establishment of S. limbatus in Europe contributes to updating the insect worldwide distribution, which now includes North and Central America (native region), South America, South Africa, the Middle East, and southern Europe. Future research is required on known and potential host species in order to investigate its potential distribution and new host associations with native or non-native plant species (Parry et al. 2013). Studies on suitable climatic conditions for S. limbatus development will further assess the risks of spread in the Mediterranean Basin. Such surveys should include also urban habitats, in which seed feeders are frequently found (Branco et al. 2019).

The authors gratefully acknowledge Gianluigi Bacchetta (Biodiversity Conservation Centre, University of Cagliari, Italy) for fruitful discussions and technical support, and Roberto Mannu (University of Sassari) for statistical advice. This study was financially supported, in part, by the Project ALIEM “Action pour Limiter les risques de diffusion des espèces Introduites Envahissantes en Méditerranée” PC IFM 2014–2020 and by RESTART-UNINUORO Project “Azioni per la valorizzazione delle risorse agroforestali della Sardegna centrale/Actions for the valorisation of agroforestry resources in central Sardinia” Regione Autonoma della Sardegna, D.G.R. N. 29/1 del 7 June 2018—fondi FSC 2014–2020. AS, GB, and IF gratefully acknowledge University of Sassari for the financial support through “Fondo di Ateneo per la Ricerca 2020”. The authors have declared that no competing interests exist.