First European record of the southern giant hornet Vespa soror

Translator’s preface by Davide Simeone

The natural world is constantly changing. Human activity, helped by the speed of modern transport, has accelerated the movement of organisms between ecosystems. Adaptation takes time, while the disruption can be immediate. The arrival in Spain of another hornet native to Asia recalls the introduction of Vespa velutina and the damage it has caused. Italy’s mild, temperate climate might favour the new species if it were introduced. Italian beekeeping already faces the northward spread of Vespa orientalis and the potential threat of V. velutina farther south. The prospect of Vespa soror adds to those concerns. Research and cooperation are essential for limiting the spread of introduced species and managing their impact over time. I share this translation of the first Spanish study of Vespa soror in the hope that it will help researchers and beekeepers working together.

Abstract

An insect monitoring programme in northern Spain, designed to assess the effects of yellow-legged hornet (Vespa velutina) trapping on insect diversity, yielded Europe’s first record of the southern giant hornet, Vespa soror (Hymenoptera: Vespidae). The authors combined morphology and genetic identification and explain how to distinguish the species from other Iberian hornets. They consider plausible introduction pathways and vectors, the potential for establishment and the implications for the receiving environment. The preliminary results raise concerns about the possible threat from Vespa soror to human health and ecosystem dynamics. The hornet preys on other insects and can also take small vertebrates. The study illustrates how examining trap catches can help detect alien species early and support a rapid response.

1 Introduction

The genus Vespa Linnaeus, 1758, in the family Vespidae and subfamily Vespinae, comprises eusocial hymenopterans commonly called hornets (Noll et al. 2020). Globalisation has increased the opportunities for social vespids, especially vespines, to be accidentally introduced into new regions and countries. Mated queens shelter during unfavourable weather and can go unnoticed in materials or containers that people move far from their original location (Arca et al. 2015; Castro 2019; Lioy, Bergamo and Porporato 2022; Otis, Taylor and Mattila 2023).

Another factor in successful introductions is the small number of individuals needed to establish a population. Where conditions are favourable, one or a few mated queens, depending on the species, can found colonies if they survive until the first workers emerge. This occurred in Europe with Vespa velutina Lepeletier, 1836 (Villemant et al. 2011; Monceau, Bonnard and Thiéry 2014; Arca et al. 2015; Otis, Taylor and Mattila 2023).

All species of Vespa are native to Asia. Only Vespa crabro Linnaeus, 1758, and Vespa orientalis Linnaeus, 1771, are also native to parts of Europe or North Africa (Archer 2012; Smith-Pardo, Carpenter and Kimsey 2020). The Iberian–Balearic fauna originally included only the European hornet, Vespa crabro, which occurs across much of the Iberian Peninsula (Vega et al. 2022). Further species arrived in the decade preceding the study, including the yellow-legged hornet Vespa velutina (Castro and Pagola-Carte 2010; López, González and Goldarazena 2011; Grosso-Silva and Maia 2012), the Oriental hornet Vespa orientalis (Hernández et al. 2013; Sánchez, Fajardo and Castro 2019; Fajardo and Sánchez 2020) and the black shield hornet Vespa bicolor Fabricius, 1787 (Castro 2019).

The southern giant hornet, Vespa soror du Buysson, 1905, is native to warmer parts of Asia, including northeastern India (Arunachal Pradesh, Meghalaya and Nagaland), northern Myanmar, northern Thailand, Laos, northern Vietnam and southern China (Yunnan, Fujian, Jiangxi, Zhejiang, Hunan, Guangdong, Hong Kong and Hainan Island) (Lee 2009; Daglio 2019; Smith-Pardo, Carpenter and Kimsey 2020).It is one of the largest hornets, comparable in size to its congener the northern giant hornet, Vespa mandarinia Smith, 1852 (Kurzenko 1995; Girish-Kumar and Srinivasan 2010; Lee 2010). The authors reported no known established non-native populations of V. soror, although a queen was found at Vancouver Harbour, British Columbia, in 2019 (Kozak and Otis 2020; Bass, Needham and Bennett 2022). V. mandarinia was detected in British Columbia and Washington State. The absence of detections in 2023 and 2024 led researchers to provisionally conclude that the introduction had failed, potentially due to eradication efforts, Allee effects, ecological factors and chance (Bass, Needham and Bennett 2022; Taylor et al. 2024).

The paper documents the first European record of V. soror, from northern Spain, and provides a genetic and morphological characterisation of the specimens. The authors describe and illustrate diagnostic features to distinguish this hornet from native and introduced Iberian vespines. They also provide a simplified pictorial key to the Iberian species of Vespa and discuss likely pathways and vectors of introduction, the status of the species and its possible impacts if it becomes invasive.

2 Materials and methods

2.1 Specimen collection, morphological analysis and taxonomic procedures

Four V. soror specimens were collected in March 2022 and October 2023 at El Campo, Granda, Siero (43°22′36″ N, 5°45′41″ W), in Asturias, northern Spain. The specimens were caught in a commercial VespaCatch trap made for V. velutina by Véto-pharma (Palaiseau, France) and baited with VespaCatch attractant (https://www.blog-veto-pharma.com/) (Figure 1). They were prepared for preservation at the site, fixed in 70% ethanol for genetic analysis and later pinned. They were examined with an Optika SZM-2 dissecting stereomicroscope at 0.7–4.5×. Photographs were taken with a Canon EOS 1200D digital SLR and an EF-S 18–55 mm f/3.5–5.6 III lens. Specimens were deposited in the collection of the University of Oviedo’s Department of Organisms and Systems (BOS, https://bos.uniovi.es/).

2.2 DNA extraction, PCR amplification and sequencing

DNA was extracted from 20–50 mg of ethanol-preserved leg tissue from one specimen using an E.Z.N.A. Tissue DNA Kit (Omega Bio-tek) according to the manufacturer’s protocol and stored at −20 °C. The mitochondrial cytochrome c oxidase subunit I barcode region (COI-5P) was amplified by polymerase chain reaction (PCR) in 40 µL using the universal primers LCO1490 and HCO2198 (Folmer et al. 1994). The mixture contained 2.5 µL template DNA, MgCl₂, 0.25 mM of each dNTP, 0.25 µM of each primer, 0.5 U Taq polymerase (GoTaq G2 Flexi DNA Polymerase) and 1× buffer. The published methods specify the MgCl₂ concentration. After initial denaturation at 95 °C for 4 min, 35 cycles were run at 95 °C for 45 s, 48 °C for 45 s and 72 °C for 30 s, followed by a final extension at 72 °C for 7 min. The PCR product was checked by horizontal electrophoresis in a 2% agarose gel and sent to Macrogen (Madrid, Spain) for forward and reverse Sanger sequencing (Sanger and Coulson 1975).

Figure 1

Location where Vespa soror was found in Asturias (a) and location of Asturias within Spain (b). View of the study plot (c) and detail of a VespaCatch trap on an apple branch at the site (d).

2.3 Genetic analysis

Forward and reverse Sanger reads were processed in Geneious Prime 2022.2.2 (https://www.geneious.com) to trim low-quality sequence and remove primer sequences. The reads were aligned and checked manually for base-calling errors. A consensus sequence was generated from the corrected alignment using the default settings.

Preliminary genetic identification used nucleotide BLAST in Geneious Prime with default settings to search GenBank. The authors also searched the BOLD Public Record Barcode Database using its identification engine (https://www.boldsystems.org/index.php/IDS_OpenIdEngine).

The phylogenetic analysis used 31 COI sequences (GenBank accessions in Supporting Information SX). The dataset included the new Spanish sequence, available V. soror sequences longer than 500 bp, at least one sequence for each other Vespa species represented in the database and four outgroups. Sequences were analysed with RAxML in Geneious Prime 2022.2.2. The authors used ModelTest in the PAUP pipeline to select the nucleotide substitution model by the Akaike information criterion (AIC). They searched for the best-scoring maximum-likelihood tree under GTR+G+I with 10,000 bootstrap replicates. They reported a consensus tree with a 10% burn-in, a 50% bootstrap support threshold for the consensus and a 70% threshold for considering a node supported.

3 Results

3.1 Morphological characterisation

Synonyms include Vespa ducalis var. soror du Buysson, 1905, and Vespa mandarinia soror van der Vecht, 1957. Archer (1991) first treated Vespa soror as a distinct species.

Material examined. Four female workers from El Campo, Granda (Siero, Asturias, northern Spain), collected by A. Arias. Two were collected on 29 March 2022, one on 3 October 2023 and one on 10 October 2023.

Description. Vespa soror has a distinctive colour pattern shared by very few Asian species in the genus and unique among European hornets. Both sexes are tricoloured, with black to blackish-brown, light brown and yellow areas (Figure 2). Four alternating light and dark blocks are apparent. The exceptionally large head is mostly yellow, followed by a black to blackish-brown section comprising most of the mesosoma. A long section from the scutellum to the end of the second tergum (T2) is mainly yellow and light brown, generally with dark bands or marks. The terminal metasomal segments, roughly the rear half, are black. The antennae and legs are mostly brown and the wings are yellowish. The metasomal colour pattern distinguishes V. soror from the other Vespa species and queens of Dolichovespula media (Retzius, 1783) in Iberia and elsewhere in Europe despite some variation within species (Figure 3).In V. soror the posterior metasomal segments are entirely black. Other native or introduced hornets have substantial yellow or orange areas in the rear half (Figure 3).

Figure 2

Vespa soror specimens trapped in Asturias, northern Spain. Ethanol-preserved specimens are shown in dorsal view (A), dorsal detail of the head and thorax (B), lateral detail of the head and thorax (C), full lateral view (D), ventral view (E) and lateral view (F). A pinned specimen is shown in head detail (G) and in lateral (H), ventral (I) and dorsal (J) views.

Figure 3

Typical colour patterns of Iberian Vespa and Dolichovespula. Vespa velutina (A), Vespa orientalis (B), Vespa crabro (C), Vespa bicolor (D), Vespa soror (E) and a Dolichovespula media queen (F). Drawings adapted from Perrard et al. (2014).

Notes. Archer (2012) distinguished females of V. mandarinia and V. soror from other Vespa by their unusually broad genae. In lateral view, at the level of the posterior margin of the clypeus, genal width (GW) exceeds 1.7 times eye width (EW), compared with less than 1.3 times in other species. In V. ducalis the ratio is about 1 (L. Castro, unpublished data). Males of V. mandarinia and V. soror also have broad genae (GW over 1.4 times EW at the same level) and share a distinctive combination of characters in the sword-shaped aedeagus, the two apical sterna, the eyes and the antennae (Archer 2012). Characters distinguishing the two species include the median pronotal impression and the proportions of the first gastral tergum in dorsal view. Its length is about half its width or more in V. soror and less than half its width in V. mandarinia. T3–T6 are usually entirely black in V. soror. In V. mandarinia, T3–T5 have orange bands and T6 is orange (Nguyen et al. 2006). Vespa ducalis Smith, 1852 has a similar colour pattern, but its genae in lateral view are about as wide as the eyes, whereas those of V. soror are visibly wider.

Identifying female caste in social vespids can be important. Body length depends on specimen posture and shrinkage, and wing span and wing length can be difficult to measure. Weight, cuticular chemistry and ovarian development have also been used to identify Vespa castes but present practical difficulties (Pérez de Heredia et al. 2017; Castro and del Pico 2021). Pérez de Heredia et al. (2017) proposed measuring mesoscutal width between the tegulae. The mesosoma is rigid, remains stable after death and can be measured precisely. The ranges reported for gynes and workers of the species examined did not overlap. Mattila et al. (2023) applied six measures to V. soror: head width, forewing length, mesosomal width and the width of the first three terga. The authors of the Spanish study considered T2 width the most reliable, as the observed ranges of gynes and workers showed almost no overlap. They identified the four Asturian specimens as workers from intertegular measurements of reference specimens from five Asian countries (L. Castro, unpublished data) and maximum T2 width following Mattila et al. (2023).

3.2 Genetic results

BLAST identified the sequence from the one sequenced specimen (GenBank OP735561, BOLD INVA001-22) as V. soror, with 97.7–98.1% pairwise identity to four reference specimens. The remaining matches to V. crabro and V. bicolor had no more than 87.8% similarity. The BOLD identification procedure yielded 100% and 99.44% similarity to two records identified as V. soror. BOLD marked those sequences as private and the authors reported that they were part of unpublished research (Table 1).

Table 1

BLAST and BOLD identification results for the consensus sequence from the collected specimen.

The phylogenetic analysis placed the Spanish sequence with GenBank sequences identified as V. soror. The node grouping them had 99.97% bootstrap support (Figure 4). The sequence distance matrix is available in Supporting Information SY.

Figure 4

Best-scoring maximum-likelihood tree from the COI sequence analysis under GTR+G+I with 10,000 bootstrap replicates. The analysis used 31 nucleotide sequences and 658 positions. GenBank accession numbers are listed in Supporting Information SX.

4 Discussion

Vespa soror preys on invertebrates of different sizes, including butterflies, dragonflies, mantises, grasshoppers and other wasps. It also takes small vertebrates such as geckos (Lee 2009). It has been observed raiding nests of Vespa bicolor, Polistes and Parapolybia (Pa. indica, Pa. nodosa and Pa. varia) to feed the captured larvae and pupae to its own brood (Lee 2009). Reports also describe how V. soror attacks colonies of the Asian honey bee, Apis cerana Fabricius, 1793, in groups. These attacks can destroy colonies and cause losses to beekeepers (Lee 2009; Mattila et al. 2020; Cappa et al. 2021). A similar risk could arise for Apis mellifera Linnaeus, 1758, in Asturias if V. soror becomes established, adding to losses already associated with invasive V. velutina (Requier et al. 2019; Laurino et al. 2020; Cappa et al. 2021).

V. soror stings are reported to cause intense pain and lasting effects. The authors hypothesise that its venom may be potent, as in its close relatives V. tropica (Linnaeus, 1758) and V. mandarinia, both of which have been described as hazardous stinging insects (Schmidt et al. 1986; Lee 2009).Stings could become a public health concern if the species establishes. In its native range, the authors highlight September to December, when large colonies can be particularly defensive before males and new queens emerge (Lee 2009).

In the native range, most V. soror colonies occur in wooded areas from sea level to over 700 m. The nest is usually underground, often in well-drained soil on a slope, with an angled entrance hole often situated among the roots of large trees (Lee 2009; Mattila et al. 2023). Some nests are only a few centimetres below the surface, while others are deep underground and connected to the entrance by one or more tunnels at least 60 cm long (Lee 2009). Nesting behaviour resembles that of V. mandarinia (Archer 1995; Lee 2009; Smith-Pardo, Carpenter and Kimsey 2020; Bass, Needham and Bennett 2022; Mattila et al. 2023). As with V. velutina, detecting and removing nests would be an important part of control if V. soror spreads in northern Spain. In that case, underground nesting could make nests harder for the responsible authorities to locate and remove.

In Hong Kong, queens usually emerge between late March and April each year. Nest initiation probably begins in early May, with the first workers generally emerging in June (Lee 2009). Mattila et al. (2023) give preliminary information on the colony cycle in Vietnam. These studies help assess when one or more colonies may have been active in Asturias. The two workers collected in late March 2022 were detected earlier in the year than would be expected from those native-range observations. The authors discuss whether they could have come from a nest founded unusually early in 2022 or from a nest founded in 2021 that survived winter. They note similarities between native-range conditions and the weather in Asturias in spring 2022 and suggest that the mild climate might have influenced nesting timing, as reported locally for V. velutina. The two workers collected in October 2023 could have come from a colony founded by a queen produced in 2022 that survived winter. A separate introduction is another possibility. The captures alone cannot establish the history of the colonies.

The authors consider accidental transport of an overwintering queen from the native range a plausible introduction pathway, as proposed for V. mandarinia and V. velutina (Otis, Taylor and Mattila 2023).

The Entomological Society of America proposed “southern giant hornet” as the common name for V. soror. Spanish renderings include “avispón gigante del sur” and “avispón gigante meridional”. Both are precise, but in everyday use they may be shortened to “avispón gigante”, which could cause confusion with V. mandarinia, the northern giant hornet. Generic or sensational names can likewise obscure the identity of the species.

The authors propose the shorter Spanish name “avispón sóror”, which has a direct connection to the scientific name, along the lines of “avispa velutina”, “avispón bicolor” and “avispón oriental”.

5 Conclusions

The first European records of V. soror are supported by morphological and genetic analyses. This brings the number of introduced Vespa species recorded on the Iberian Peninsula to four.

Accidental transport from the native range is a likely explanation. If V. soror becomes established in northern Iberia, where V. velutina is already widespread, their combined effects could increase environmental, economic and public health impacts.

The authors call for early surveillance and a rapid response to locate any nests and assess potential spread and impacts around Siero and neighbouring areas. The dedicated Asturian trapping programme reported no V. soror captures in its 2025 survey. The four original specimens therefore document an occurrence, while establishment remains unconfirmed.

Omar Sánchez, Leopoldo Castro, Álvaro Fueyo, Yaisel J. Borrell, Andrés Arias

First published 9 November 2024 · https://doi.org/10.1002/ece3.70502