Biodiversity loss is undoubtedly one of the most urgent problems of the present time. Key drivers of this crisis are rather well known - and as habitat change, pollution, climate change, invasive species and overexploitation of natural resources (Mazor et al. 2018). In order to counteract this biodiversity crisis, the UN declared the protection of biodiversity on land and in water as two of the 17 sustainable development goals (United Nations 2015). It is obvious that comprehensive knowledge of biodiversity is a fundamental basis for conservation approaches. Studies have shown that only between 10–20% of the true species diversity has been discovered and described up to today (~1.2 Million species) and that most undescribed species are expected within insects (Mora et al. 2011). In order to achieve a robust knowledge base about biodiversity and how to protect it, two fundamental steps are necessary. First, we should aim to complete taxon inventories with species occurrence data and speed up the description of new species. Second, we need to make taxonomic knowledge accessible to a wider audience, not only to experts for a specific group, but also for example to students, naturalists and ecologists. By closing this link between taxonomists and a wider audience, we get the opportunity to collect large amounts of distributional and ecological information and will obtain a powerful tool in the fight against biodiversity loss. One outstanding example in biodiversity research is the Swedish Taxonomy Initiative (STI), an all–taxa biodiversity inventory that has the goal to identify all multi-cellular organisms in Sweden (Karlsson et al. 2020b). The Swedish Malaise Trap Project (SMTP) is funded by the STI and is known as one of the most ambitious insect inventories in the world. The main focus of the SMTP are species-rich groups like Diptera and parasitoid Hymenoptera, which are known to be particularly under-researched (Karlsson et al. 2020a, Ronquist et al. 2020).

Darwin wasps (Ichneumonidae) are believed to have one of the largest gaps between described species (~25,000) (Yu et al. 2016) and expected species diversity (~60,000) (Townes 1969a; Klopfstein et al. 2019a). The enormous diversity of Darwin wasps is certainly connected with their parasitoid life history and corresponding adaptation to certain host species and the physiology and ecology of the latter. The reproductive strategies of Darwin wasps can be rather complex and sometimes include several levels of parasitism, e.g., Mesochorinae are invariably obligate hyperparasitoids (Broad et al. 2018). It can be assumed that their adaptation to specific hosts makes parasitoid wasps especially vulnerable to environmental changes which affect the latter, and it has been shown that biodiversity loss across trophic levels can have tremendous impacts on entire ecosystems (Cardinale et al. 2012). However, almost no attention has been paid to the conservation of parasitoid wasps due to the lack of taxonomic and ecological knowledge for this group (Shaw and Hochberg 2001).

Campopleginae belong to the clade of the higher Ophioniformes within Ichneumonidae (Bennett et al. 2019; Klopfstein et al. 2019b) and consist of 65 genera worldwide, of which 43 occur in the Western Palaearctic (Yu et al. 2016). The morphology of Campopleginae is considered as relatively homogenous. Hence, finding apomorphies to differentiate species and genera is often difficult. As a result, the subfamily as a whole and many of its genera are lacking modern taxonomic revisions (Broad et al. 2018). A study focusing on the ribosomal gene 28S retrieved unstable relationships between the different Campopleginae genera (Quicke et al. 2009). Common morphological traits of Campopleginae are: 1) propleuron with an elongated lobe reaching posteriorly to cover part of the fore coxa; 2) subapically notched dorsal valve of ovipositor; 3) petiolate first tergite; 4) clypeus weakly separated from the face; 5) laterally compressed metasoma 6) iridescent compound eyes (Broad et al. 2018).

Dusona Cameron, 1901 is the most species-rich genus within Campopleginae, with 442 species worldwide and 125 species in the Western Palaearctic (Yu et al. 2016), and one of the most species-rich genera among all Darwin wasps. Dusona species can be easily recognised on genus level due to some very distinct diagnostic characters: 1) elongated, oval or slit-shaped propodeal spiracle; 2) very strongly compressed metasoma; 3) closed areolet; 4) petiolar suture often obliterated or positioned very low anteriorly. As Dusona includes the Campopleginae species with the largest body sizes of up to more than 20 mm, e.g., in D. falcator (Fabricius, 1775), it is not surprising that they received the attention of entomologists early on in the 19^th century (e.g., Holmgren 1872; Thomson 1887) and the 20^th century (e.g., Teunissen 1947). Perhaps for the same reason, Dusona attracted the focus of Rolf Hinz and Klaus Horstmann, who were the first to prepare extensive species descriptions, conduct taxonomic revisions and compile keys for the Eastern and Western Palaearctic Dusona species (Hinz 1963, 1977, 1985, 1990, 1994; Horstmann 1995, 2009, 2011; Hinz and Horstmann 2004). Undoubtedly, the revisions of the Palaearctic Dusona set an excellent basis to work with the genus (Shaw et al. 2016). However, despite the keys, species identification remains rather challenging for Dusona, because there are hardly any drawings and no photographs included, and the morphological terminology is sometimes inadequately explained (Hinz and Horstmann 2004; Horstmann 2009). Some of the Palaearctic Dusona with occurrence in Korea were illustrated by Choi and Lee (2014), but this work only covers 25 species. The application of several characters that are hardly used for other genera of Darwin wasps, such as the shape of the antennal carina or the junction of the epicnemial and transverse carinae on the mesopleuron, further decreases the accessibility of the keys. Hence, it is not surprising that the quality of species determinations in natural history collections is very inconsistent, and misidentifications are rather frequent in the genus (pers. observation, N. Meier).

Compared to other Campopleginae, Dusona have an even more strongly laterally compressed metasoma. It is possible that this strong compression is an ecological adaptation to hot temperatures and arid habitats, as it minimizes the exposure to sunlight. Indeed, Darwin wasps with laterally compressed metasomata, such as Campopleginae and Anomaloninae, are more likely to be found in the middle of the day than those with dorso-ventral compression or cylindrical metasomal shape (pers. observation, S. Klopfstein). Another hypothesis suggests that the lateral compression of the metasoma is an adaptation for ovipositing in lepidopteran hosts with especially long setae (Broad et al. 2018). Admittedly, other subfamilies such as Ophioninae also share laterally compressed metasomata but are neither active during daytime nor specialized on lepidopteran larvae with long setae. Hence, further research is required to understand the significance of the various factors that contribute to this metasomal shape. Rearing studies have shown that lepidopterans (mostly Geometridae, Noctuidae, Nolidae, Drepanidae, Erebidae and Notodontidae) are the main hosts of Dusona (Horstmann 2011; Shaw et al. 2016); only one species, Dusona minor (Provancher, 1879), has been reported as a parasitoid of a sawfly of the family Diprionidae, Monoctenus juniperi (Linnaeus, 1758), although this record needs independent confirmation (Horstmann 2011). However, for many Dusona species, host records are still missing or presumably incomplete. Finding further host records would be a fundamental step towards understanding the ecology and host adaptation of the genus, but it is rarely done, as taxonomic knowledge for both the parasitoid and the host species is required. Collaborations between specialists for both hosts and parasitoids could be an efficient approach to obtain new host records. Host adaptations are often phylogenetically linked, as for instance in spider parasitoids in Pimplinae (Matsumoto 2016). Hence, a molecular phylogeny of Dusona could help to further understand the evolutionary processes behind ecology and speciation of the genus (Tschopp et al. 2013).

In addition, a phylogeny of Dusona could help to split the genus into subgenera that reflect its evolution and help to compartmentalize identification and potentially simplify the application of the species keys. Earlier attempts to divide Dusona into species groups were often based on homoplastic characters (Hinz 1977), such as the shape of the epicnemial carina or the presence of light yellow colouration on the metasoma or hind tibia. Hence, Hinz and Horstmann (2004) doubted that these groupings are monophyletic. So far, no molecular methods have been used to reconstruct the intrageneric phylogeny of Dusona, which could be very helpful to properly evaluate the significance of these species groups and to define subgenera. Numerous sequences attributed to this genus have been published both on GenBank (n = 850) and BOLD Systems (n = 1900). More than three quarters of these sequences are only determined to genus level, mainly involving specimens from Canada and Costa Rica. The remaining sequences cover COI barcodes for 44 mainly Nearctic and Palaearctic species and very few ribosomal 18S and 28S sequences (Quicke et al. 2009). In order to facilitate ecological studies with Dusona, a reliable reference barcode library would be needed (Lue et al. 2021).

The aim of this study is to make Dusona available to a wide audience of professional taxonomists, layman entomologists, and ecologists. In order to achieve this, we provide a detailed photographic guide to diagnostic characters and species and supplement the limited faunistic data of Dusona in Sweden and Switzerland. Furthermore, we lay the foundation for a reference barcode library for the Western Palaearctic, based on carefully identified voucher specimens deposited in public collections. Finally, we build a phylogenetic framework based on four standard markers, which can be easily expanded in the future.

In order to do phylogenetic analyses based on molecular data, the availability of recently collected material was crucial for this study. Hence, we mainly used material collected by the SMTP (collection period: 2003–2006). In total, we examined a subsample of ~6,000 Campopleginae (~1/3 of the available SMTP material) including 465 Dusona specimens originating from 55 traps dispersed over the country. In addition to the SMTP material, we examined material from various Swedish, Swiss, and Norwegian collections (Table 1).

We used the keys of Hinz and Horstmann (2004) and Horstmann (2009) for species determination. Furthermore, nearly all determined species were checked against either type material or specimens determined by R. Hinz or K. Horstmann from the MZL and ZSM. In this study, we generally follow the terminology of R. Hinz and K. Horstmann for morphological characters, in order to directly refer to their keys. However, in some cases, they deviate from commonly accepted terminology for Darwin wasp morphology (Broad et al. 2018). In these cases, we followed the terminology of Broad et al. (2018). Discrepancies in the keys for Dusona (Hinz and Horstmann 2004; Horstmann 2009) from the generally accepted morphological terminology (Broad et al. 2018) are listed in Table 2.

In order to improve the accessibility of the keys for Dusona (Hinz and Horstmann 2004; Horstmann 2009), we produced images of all species in our sample. The images were taken with the photosystem Keyence VHX 6000. Both stitching and stacking techniques were used to enhance image quality. Habitus images were taken with 100× magnification (150× for small specimens), while character close-ups were taken with 150× magnification (200× for small specimens). Images were then arranged in plates, each consisting of a habitus image (female, or if not available, male) and six character close-ups. Furthermore, we created a photographic guide for seven diagnostic characters (with four to six character states each). Image processing and figure alignment was made with Adobe Photoshop and InDesign. Additional elements, such as coloured line fragments and arrows, were used to show length ratios or mark hidden characters. In order to improve the understanding of the shape of the epicnemial and the transverse carina, line drawings were made with Adobe Photoshop and placed next to the photographs. Full resolution figure plates are uploaded to Zenodo, DOIs are given in the figure captions.

All specimen data from SMTP was digitized using a preformatted table provided by the Station Linné (Suppl. material 1, https://doi.org/10.5281/zenodo.6535195). Studied material that was recently collected from Switzerland was digitized in a separate table (Suppl. material 2, https://doi.org/10.5281/zenodo.6535211). Barplots and a species accumulation curve were drawn in R (R Core Team 2017) using the function “specaccum” from the R package VEGAN (Oksanen et al. 2020). We calculated the mean species accumulation curve and corresponding standard deviation based on random permutations of the data. Maps with the trap distribution, the species diversity and species distribution were created in the open-source software QGIS3 (v.3.4; QGIS.org 2022) using the OpenStreetMap as background layer. Where needed, points on the map which represent trap locations were slightly shifted in order to prevent overlapping.

We collected data from the barcoding portion of Cytochrome Oxidase Subunit 1 (COI) as a reference library and from three additional standard markers for phylogenetic reconstruction (28S rRNA, ITS2 rRNA, and the carbamoyl phosphate synthase domain CAD). We used samples of 128 Dusona specimens for molecular analysis that were less than 20 years old. As Dusona might be the sister genus to the remaining Campopleginae or at least branches off early in the tree (Sharanowski et al. 2021), we chose eight outgroup species representing the entire diversity of the subfamily (Suppl. material 3, https://doi.org/10.5281/zenodo.6535230). Extractions and PCR reactions took place at the University of Basel (Section of Conservation Biology, NLU). DNA extracts are stored at NMB, vouchers are stored in the same place or at NHRS. Of each specimen, one leg with coxa was used to extract DNA. Extractions were done with the DNeasy blood and tissue kit from Qiagen according to the standard protocol, but with a prolonged digestion step over night at 56 °C and two elusion steps with only 50 ml each. The PCR conditions are reported in Suppl. material 4, https://doi.org/10.5281/zenodo.6535284. We used previously published primers for COI (Folmer et al. 1994), 28S (Belshaw and Quicke 1997; Mardulyn and Whitfield 1999), and ITS2 (Quicke et al. 2006). For CAD, we used the forward primer for Ichneumonidae from Klopfstein et al. (2013) (forward – CADfor9s: ATG AAR AGY GTN GGN GAR GTR ATG GC) but modified it by resolving ambiguous positions according to a preliminary alignment including several Campopleginae sequences (new primer for9sB: ATG AAG AGC GTT GGT GAA GTT ATG GC). As the reverse primer, we newly designed a primer (reverse – CADIchnRb1B: ACC AAT GAC CAT CGT GTA ATT TCC). The PCRs for CAD were run at an annealing temperature of 52 °C. Two additional internal primers for degenerated COI sequences were used; forward FA336 and reverse RA435 (Johansson and Klopfstein 2020), these primers were combined with LCO/HCO and run on an annealing temperature at 50 °C. The quality of the PCR products was then checked on a 2% agarose electrophoresis gel and they were sent for cleanup and sequencing to Macrogen Europe in the Netherlands. COI data from Norway was produced by the Barcoding of Life consortium in Guelph, following their standard protocols.

Sequence cleaning and alignments of the protein-coding genes were done with MEGA7 (Kumar et al. 2016), while the two ribosomal genes were aligned using the E-INS-i strategy on the MAFFT web server (Katoh et al. 2019). No gaps or stop codons were detected in COI and CAD. Uncorrected p-distances for COI were calculated in MEGA7, using pairwise deletions. For the phylogenetic analysis, we combined all four markers into a concatenated alignment, while only including specimens that had at least 1,000 bp of sequence data in total. Phylogenies were reconstructed under Bayesian inference with MrBayes (Ronquist et al. 2012), both for each gene individually and for the four genes concatenated. We partitioned the protein-coding genes into 1^st and 2^nd versus 3^rd codon positions, while no partitioning was applied to the ribosomal genes. We let MrBayes sample over the complete GTR model space and used a separate gamma distribution and proportion of invariant sites for each partition to model among-site rate variation. We ran 10 million generations, sampling every 1,000 generations and using a burn - in of 50% (for input files of MrBayes, see Suppl. material 5, https://doi.org/10.5281/zenodo.6535309). All phylogenetic analyses were run on the UBELIX Cluster of the University of Bern. Nodes with support values below 50% are indicated as polytomies. Phylogenetic trees were drawn with FigTree (v.1.4.4) and species groups were indicated on the trees using Adobe Illustrator. We plotted character states on the genus tree in order to find morphologically distinct species groups as candidates for subgenus divisions.

Horstmann (2009) used several characters not commonly used for identification in Darwin wasps. We here assess and illustrate the seven most important characters in detail (ordered from anterior to posterior), covering the shape of the antennal carina (I), the junction of the genal and hypostomal carinae (II), the shape of the epicnemial and transverse carinae (III), the sculpture of the mesopleuron (IV), the shape and sculpture of the propodeum (V), the shape and sculpture of the first tergite (VI) and the separation of the third laterotergite (VII). These characters were chosen because they are either especially important for the identification of Dusona species or their states are sometimes difficult to assess based on the key couplets (Hinz and Horstmann 2004; Horstmann 2009).

The antennal carina (sensu Hinz and Horstmann 2004) equals the rim surrounding the antennal sockets. This character is rarely used in other ichneumonid genera, in which the antennal carina is usually flat and narrow (Fig. 1a). This is also the case in most Dusona species; however, in several species, the antennal carina shows a large diversity of dorsal extensions (Fig. 1c–f). In some species they appear as nose-like bulges (Fig. 1c, d), while in other species, the rim of the antennal sockets is extended into an ear-like crescent-shaped plate (Fig. 1f), sometimes also called auricles (Broad et al. 2018), such as in Tryphon latrator (Fabricius, 1781), or just slightly raised (Fig. 1e). These dorsal extensions are often accompanied by some sort of transverse striation (Fig. 1d, e). Furthermore, some species show a strong longitudinal carina between the antennal sockets (Fig. 1b). In specimens, where the scape is positioned parallel to the frons pointing upwards, the examination of the antennal carina can be difficult. In certain cases, it is necessary to either relax the specimen in water and shift one antenna or to separate one antenna from the antennal socket in order to identify the species.

Figure 1. 

Shape of the antennal carina. a antennal carina narrow and flat (D. blanda) b antennal carina narrow and flat, frons with a strong median longitudinal keel (D. carinifrons) c antennal carina widened dorsally to a rather inconspicuous nose-like projection (D. stygia) d antennal carina widened dorsally to a large nose-like projection with transverse striae (D. pineticola) e antennal carina slightly raised dorsally and transverse striate (D. sobolicida) f antennal carina raised and strongly bent upwards forming an ear-like, crescent-shaped plate (D. infesta). Scale bars: 0.5 mm. https://doi.org/10.5281/zenodo.6340012.

This character is used in more than 20 couplets throughout the keys of Hinz and Horstmann (2004) and Horstmann (2009). Our molecular analysis reveals no phylogenetic relationship between species that show any of these dorsal extensions of the antennal carina (see section molecular analysis).

The genal carina (continuous with the more dorsal occipital carina) separates the gena from the occiput. The hypostomal carina separates the occiput from the hypostoma. The junction of the genal carina and the hypostomal carina is usually either at the base of the mandible or positioned more dorsally. The genal index describes the ratio of the distance between the junction of the genal and the hypostomal carinae and the base of the mandibles (green line in Fig. 2) to the basal width of the mandibles (blue line in Fig. 2). In some species the genal carina can be slightly obliterated ventrally, which makes it difficult to identify the position of the junction with the hypostomal carina (Fig. 2c). Furthermore, in some species the junction of the genal and the hypostomal carina can be distinctly raised, forming a lopsided triangular pyramid (Fig. 2d, e), sensu Hinz and Horstmann (2004). In specimens where the head is closely positioned to the mesosoma, the examination of this character can be tricky. In certain cases it is necessary to either relax the specimen in water and shift the head slightly in its position or separate the head from the rest of the body for proper examination of this character.

Figure 2. 

Junction of the genal and the hypostomal carinae. a, b genal carina meets the hypostomal carina clearly above the base of the mandible (a D. blanda; b D. rubidatae) c genal carina meets hypostomal carina clearly above the base of the mandible, occipital carina obliterated close to the junction (D. tenuis) d, e genal and hypostomal carina distinctly raised, their junction forming a lopsided triangular pyramid (d D. mercator e D. aurita) f genal carina meets the hypostomal carina at the base of the mandibles (D. flagellator). Genal index = distance between the base of the mandibles and the junction of the genal and hypostomal carina (green bars) / the basal length of the mandibles (blue bars). Scale bars: 0.25 mm. https://doi.org/10.5281/zenodo.6362512.

Hinz and Horstmann (2004) and Horstmann (2009) both use this character in more than a dozen couplets of their keys. Our phylogenetic analysis reveals that the species that possess a lopsided triangular pyramid constitute the monophyletic mercator-group as it was previously proposed by Hinz and Horstmann (2004), see also section molecular analysis. In addition, a genal index of zero is one of the diagnostic characters for the flagellator group and the angustifrons group; however, it also occurs in several other species, such as D. thomsoni Hinz, 1963 and D. pulchripes (Holmgren, 1872).

The epicnemial carina separates the mesopleuron from the epicnemium, while the transverse carina separates the ventral part of the epicnemium from its pleural part. The former is present in nearly all ichneumonids, while the latter is very rare outside Dusona. The epicnemial carina is divided into a ventral part and pleural part. These two parts are separated at the junction of the epicnemial carina with the transverse carina. In some Dusona, the pleural part of the epicnemial carina is complete and merges dorsally with the anterior edge of the mesopleuron (Fig. 3c), such as in D. xenocampta (Förster, 1868). However, in many species the pleural part of the epicnemial carina can be replaced or covered by wrinkles (Fig. 3a, b) and be obliterated dorsally (Fig. 3b), such as in D. tenuis (Förster, 1868) or even completely obliterated (Fig. 3d), such as in D. juvenilis (Förster, 1868). In the latter case, the pleural part of the epicnemium is completely fused with the mesopleuron (Fig. 3d).

Figure 3. 

Shape of the epicnemial and transverse carinae. a ventral part of the epicnemial carina merging with the pleural part, both carinae low, transverse carina indistinct (D. bicoloripes) b ventral part of the epicnemial carina distinctly raised, merging with the transverse carina, pleural part of the epicnemial carina low and dorsally obliterated (D. tenuis) c ventral part of the epicnemial carina only slightly raised, merging with the pleural part or the transverse carina, pleural part of the epicnemial carina dorsally complete and merging with anterior edge of the mesopleuron (D. xenocampta) d ventral part of the epicnemial carina slightly raised, merging with the transverse carina, pleural part of the epicnemial carina completely obliterated (D. juvenilis). red = ventral part of the epicnemial carina. blue = pleural part of the epicnemial carina. purple = transverse carina. Structures in grey are for orientation purposes. Scale bars: 0.5 mm. https://doi.org/10.5281/zenodo.6362741.

Hinz and Horstmann (2004) and Horstmann (2009) both use this character in more than a dozen couplets of their keys. They propose that the ventral part of the epicnemial carina can either merge with the transverse carina or the pleural part of the epicnemial carina. This description can be misleading, as the ventral part of the epicnemial carina merges usually with both the pleural part and the transverse carina (except when one of these is obliterated). Nevertheless, it is possible to judge which half of these couplets to follow by comparing the height of the transverse carina and the pleural part of the epicnemial carina, which is usually described as secondary characters in these couplets. Our molecular analysis suggests that individual character states of epicnemial and transverse carina do not reflect phylogenetic relationships in Dusona. On the contrary, this character appears highly homoplastic; for instance, the strongly raised transverse carina occurs in the whole flagellator group, the mercator group, some members of the angustifrons group, and in D. tenuis (Fig. 3b).

The sculpture of the mesopleuron shows a rather broad range of variability within Dusona. It can be very shiny in some species (Fig. 4a) or appear rather dull due to a coriaceous background (Fig. 4d–f). Many species have a strong punctation on the mesopleuron, which can be dense (puncture diameter > interspace between punctures; Fig. 4b), rather dense (puncture diameter = interspace between punctures; Fig. 4d), or dispersed (puncture diameter < interspace between punctures; Fig. 4e). Only a few species have an irregularly wrinkled mesopleuron, instead of the punctation (Fig. 4f). The sculpture of the depression in front of the speculum can either be wrinkled (Fig. 4a), finely striate (Fig. 4b), granulate strigose or coriaceous (Fig. 4c). There are several couplets in Horstmann (2009) asking whether the longitudinal striae merge anteriorly with the punctation or with an area without punctation. This character is difficult to assess in our opinion. We rather suggest checking whether the punctation is distinct on the anterior dorsal part of the mesopleuron (Fig. 4d), such as in D. blanda (Förster, 1868) or if the punctures are very small and dispersed or obliterated (Fig. 4e), such as in D. humilis (Förster, 1868).

Figure 4. 

Sculpture of the mesopleuron. a mesopleuron with a distinct and dense or rather dense punctation on a smooth background, depression in front of speculum wrinkled (D. bucculenta) b mesopleuron with a dense punctation, background slightly coriaceous and shining, depression in front of speculum with fine longitudinal striae (D. falcator) c mesopleuron with a rather dense to dense punctation, background coriaceous and slightly shining, depression in front of the speculum with a few longitudinal striae dorsally and granulate-strigose ventrally (D. obliterata) d mesopleuron with a distinct and rather dense punctation, background distinctly coriaceous, slightly shining, depression in front of the speculum with longitudinal striae merging with the punctation anteriorly (D. blanda) e mesopleuron with a rather dispersed punctation, background distinctly coriaceous and dull, depression in front of the speculum granulate-strigose or with some longitudinal striae that merge with an area without punctation (D. humilis) f mesopleuron with wrinkles centrally, rugose punctate dorsally and ventrally, rather dull (D. leptogaster). Scale bars: 0.5 mm. https://doi.org/10.5281/zenodo.6368561.

This character is used in more than 20 couplets throughout the keys of Hinz and Horstmann (2004) and Horstmann (2009). Our molecular analysis suggests that the sculpture of the mesopleuron is not useful to separate species groups at basal nodes in the phylogeny of Dusona. However, the sculpture of the mesopleuron usually shows less intraspecific variability than other characters and is thus useful for differentiation at species level.

The propodeum of Darwin wasps carries many systematically relevant characters. In Dusona, mainly three types of characters of the propodeum are used for species identification: the general shape (e.g., central depression), the carination, and the sculpture of the surface. The propodeum in Dusona can either be weakly depressed (Fig. 5a), broadly and shallowly depressed (Fig. 5b, c), narrowly but deeply depressed to a longitudinal furrow (Fig. 5d, e), or broadly and deeply depressed (Fig. 5f). The propodeal carination can be obliterated to a variable extent (Fig. 5a, b), or almost complete with a distinct anterior transverse carina and lateromedian longitudinal carinae (Fig. 5c, less distinct in Fig. 5f). The surface sculpture of the propodeum is usually either wrinkled or coriaceous (often anteriorly). Especially in species with a deep and narrow median longitudinal furrow, the furrow can be expected to have distinct short (Fig. 5d) or long (Fig. 5e) transverse wrinkles. In some species, in which the propodeum is often only weakly depressed, the propodeal sculpture is very irregularly wrinkled (Fig. 5a).

Figure 5. 

Shape and sculpture of the propodeum. a propodeum hardly depressed with irregular wrinkles (D. bicoloripes) b propodeum broadly and shallowly depressed, irregularly wrinkled anteriorly and with transverse wrinkles medially and posteriorly (D. stygia) c propodeum broadly and shallowly depressed, rugose or irregularly wrinkled with some short transverse wrinkles posteriorly and a rather complete carination (D. terebrator) d propodeum narrowly and deeply depressed with short transverse wrinkles in the sulcus (D. angustifrons) e propodeum narrowly and deeply depressed with long transverse wrinkles in the sulcus (D. flagellator) f propodeum rather broadly and deeply depressed with distinct transverse wrinkles posteriorly and medially (D. stragifex). Scale bars: 0.5 mm. https://doi.org/10.5281/zenodo.6368594.

The propodeum is described in nearly all terminal couplets in Hinz and Horstmann (2004) and Horstmann (2009). Our molecular analysis suggests that the shape and sculpture of the propodeum is not very useful to separate species groups at basal nodes in the phylogeny of Dusona. Even though the characters of the propodeum often show some intraspecific variability, they are in many cases the best or even only character to differentiate closely related species. One such example are males of D. flagellator (Fabricius, 1793), which are best separated from D. notabilis (Förster, 1868) by the distinctly depressed propodeum with long transverse wrinkles (Figs 5e, 47e). While the degree of depression of the propodeum as it is used in Horstmann (2009) can be difficult to judge, determining the presence and prevalence of transverse wrinkles is usually much easier.

The first tergite is commonly used to characterize Darwin wasps and can even be diagnostic at subfamily-level, for example the presence of the glymma or the position of the spiracle (Broad et al. 2018). In Dusona, however, the presence and size of the glymma is highly variable between different species, ranging from large glymma (Fig. 6a, b), to medium-sized or small glymma (Fig. 6c–d), to glymma absent (Fig. 6e, f). In front of the glymma, some Dusona species show distinct sculpture in the form of short, transverse grooves (Fig. 6a) such as D. insignita (Förster, 1868), or long transverse wrinkles (Fig. 6b) such as D. nidulator (Fabricius, 1804). The lateral sides of the first tergite can be bordered by variably distinct longitudinal carinae. The area between these carinae (if present) often has a different surface sculpture from the rest of the tergite, e.g., coriaceous instead of smooth (Fig. 6a–e).

Figure 6. 

Shape and sculpture of the first tergite. a petiole with long row of short transverse wrinkles in front of a large glymma (D. insignita) b petiole with long row of long transverse wrinkles in front of a large glymma (D. nidulator) c petiole very fine sculpture in front of a large glymma (D. bicoloripes) d petiole with a small glymma at the anterior edge of narrow lateral areas (D. angustifrons) e petiole without a glymma but narrow lateral areas (D. limnobia) f petiole with indistinct glymma or lateral areas (D. pineticola). Scale bars: 0.5 mm. https://doi.org/10.5281/zenodo.6368627.

This character is used in many couplets throughout the keys of Hinz and Horstmann (2004) and Horstmann (2009). Our molecular analysis suggests that characters concerning the first tergite, such as the presence of the glymma, lateral areas and sculpture, are hardly conserved in any of the groupings within the phylogeny of Dusona. Furthermore, these characters are sometimes also subject to intraspecific variability, e.g., the presence or absence of small glymmae within species of the flagellator group (see also Horstmann 2009).

In the Western Palaearctic Dusona species, the second laterotergite is always completely separated from the tergite by a crease, while the fourth laterotergite is always completely fused with the tergite. In contrast, the third laterotergite shows a broad range of interspecific variability, from completely fused with the tergite to separated from the tergite by a crease of variable length. This crease can be seen as a thickened straight line and is often marked with black (Fig. 7e). However, also species with a completely fused third laterotergite can have a black longitudinal mark on this tergite, called “black lateral stripe” in Hinz and Horstmann (2004), which is then well separated from the thin and curved ventrolateral edge of the tergite (Fig. 7a). The epipleural index describes the ratio of the length of the crease to the lateral length of the third tergite. In some species, such as D. insignita (Fig. 7c), the third laterotergite is separated from the tergite by a crease but secondarily folded back. In such cases, only the length of the crease anterior to the secondary fold back is measured for the epipleural index.

Figure 7. 

Separation of third laterotergite. a laterotergite completely fused with the tergite, its ventral edge thin and curved (D. sobolicida) b–f laterotergite folded in anteriorly, separated from the tergite by a crease, mind that the crease is secondarily folded back in D. insignita b D. pugillator c D. insignita d D. stragifex e D. nidulator f D. leptogaster. Blue bars indicate the lateral length of the third tergite. Green bars indicate the length of the crease separating the laterotergite from the tergite. Epipleural index = length of the crease / lateral length of the tergite. Scale bars: 0.5 mm. https://doi.org/10.5281/zenodo.6368663.

This character is used in both keys for Eastern (Hinz and Horstmann 2004) and Western Palaearctic Dusona (Horstmann 2009) as the first couplet, thus dividing Dusona species into 2 large groups. Our molecular analysis (see below) reveals that this character is in part correlated to phylogenetic relationships (see section molecular analysis).

Among the ~6,000 examined Campopleginae from the SMTP material, we have found 465 Dusona specimens (Suppl. material 1, https://doi.org/10.5281/zenodo.6535195). This means that Dusona, the most species rich genus within Campopleginae, comprised almost 8% of the studied sample. In total, the 465 specimens were identified to 42 different species, of which 9 new for Sweden and 31 among the 54 Dusona species previously recorded for Sweden. 23 of the previously recorded species for Sweden have been found in our sample. In 41 out of 55 Malaise traps, at least one Dusona species was collected (Fig. 8). Habitat and species diversity correlations could not be analysed quantitatively because the habitat of each trap was described individually. However, the three traps with the highest species diversity are described as follows: Trap 7: Uppland, Älvkarleby kommun, Båtfors; dry meadow with birch, field of fire (16 species). Trap 22: Öland, Mörbylånga kommun, Gamla Skogsby (Kalkstad); nemoral grove (12 species). Trap 1008: Småland, Nybro kommun, Alsterbro/Alsterån; mixed forest (10 species). D. bicoloripes (Ashmead, 1906) was the species that was collected by the highest number of traps (n = 19), followed by D. blanda (n = 15) and D. circumspectans (Förster, 1868) (n = 9). In contrast to these apparently common and widespread species, there were also 10 species for which only a single specimen was found in the SMTP sample. We have found no clear trends in the number of Dusona species and individuals on the latitudinal axis (Fig. 10a, b). However, the seven northern-most traps (>63°N) have each only caught very few specimens and species. Apparently, some of the new species for Sweden were more likely to be found in southern or coastal areas (Fig. 9). The sample-based species accumulation curve is not yet saturated (Fig. 10). Hence, it is very likely that more species would be found, if a larger sample were examined.

Figure 8. 

Map with number of Dusona species per trap. In 41 out of 55 traps, one or more Dusona species were found. Some points were slightly shifted in order to prevent overlap. Background map: OpenStreetMap.

Figure 9. 

The distribution of the three most abundant Dusona species in the SMTP sample without earlier records in Sweden. blue dots = traps where the species was found. red dots = traps where the species was not found a D. subimpressa b D. rubidatae c D. flagellator. background map: OpenStreetMap.

Figure 10. 

Faunistic analyses. a number of Dusona individuals in the SMTP material on the latitudinal axis (axis not numerical) b number of Dusona individuals in the SMTP material on the latitudinal axis (axis not numerical) c species accumulation curve (sites = Malaise traps, the curved line indicates the mean species accumulation curve, error bars indicate the SD).

We report 11 Dusona species new to the faunistic records of Sweden and 3 Dusona species new to the faunistic records of Switzerland (Yu et al. 2016; Klopfstein et al. 2019c; Dyntaxa 2022). While most of the newly recorded species for Sweden were found in the extensive collection of the SMTP, several species were already deposited in the collection of the NHRS, e.g., Dusona libertatis (Teunissen, 1947) or were found in the collection NJ, e.g., Dusona stygia (Förster, 1868). An up-to-date checklist of Swedish Dusona will be available on www.artfakta.se, the follow-up website of www.dyntaxa.se. We provided the website operators with the according information.

In the following, we report locality data for newly recorded species from both Sweden and Switzerland, observations regarding unknown intraspecific variation and their possible taxonomic implications. For each of the newly recorded species, only one location is reported here, the remaining locality data being given in Suppl. materials 1, 2 (Suppl. material 1, https://doi.org/10.5281/zenodo.6535195 and Suppl. material 2, https://doi.org/10.5281/zenodo.6535211). For species without new records, no locality data is listed below, but morphological variations are sometimes discussed. In the end of this section, illustrative plates for all species available to our study (n = 57) are shown (Figs 11–67).

Dusona admontina (Speiser, 1908) (Fig. 11)

Sweden • 1♂; Skåne, Simrishamns kommun, Stenshuvuds nationalpark, Svabeholmsskog (malaise trap ID 40); hornbeam forest; 55.6613830°N, 14.2687330°E; 2005.iv.22–2005.v.22.

The species has not previously been recorded for Sweden (Dyntaxa 2022). The recorded distribution range mainly covers countries in northern and eastern Europe: Austria, Belgium, Germany, Netherlands, Norway, Poland, Romania, Russia and United Kingdom (Horstmann 2011; Yu et al. 2016).

Figure 11. 

D. admontina (Speiser, 1908), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6046671.

Dusona aemula (Förster, 1868) (Fig. 12)

Switzerland • 1 ♀; GR, Sur, NE Sur; forest (pine, larch), river stream; 46.52419°N, 9.63425°E; 2006.vii.19–27; malaise trap; S. Klopfstein, H. Baur leg.

A single specimen of this species from Graubünden (CH) was included for this study. This specimen indicates that the species is slightly more variable than described by Hinz and Horstmann (2004) in terms of the leg colouration: front legs yellow from the trochanter, mid leg yellow from the apex of the trochanter, hind tibia yellowish medially, narrowly marked with black basally, rather broadly marked with black apically. All tarsi are darkened apically. Specimens with this colour pattern resemble Dusona carpathica (Szépligeti 1916) in many characters. However, they differ from D. carpathica by having the pleural part of the epicnemial carina complete dorsally and merging with the anterior margin of the mesopleuron (obliterated dorsally in carpathica) and the propodeum being rather deeply depressed and transverse striate (slightly depressed and irregularly wrinkled but with a median longitudinal keel posteriorly in carpathica). The species has not previously been recorded for Switzerland (Klopfstein et al. 2019c). However, the current records suggest that the distribution covers the entire Palaearctic: Austria, Azerbaijan, Belgium, Bulgaria, Croatia, Czech Republic, Finland, France, Germany, Hungary, Italy, Kazakhstan, Moldova, Netherlands, Norway, Poland, Romania, Russia, Spain, Sweden, Tunisia, Turkey, Ukraine and United Kingdom (Horstmann 2011; Yu et al. 2016).

Figure 12. 

D. aemula (Förster, 1868), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6333923.

Dusona alpina (Strobl, 1904) (Fig. 13)

Sweden • 1♀; Värmland, Munkfors kommun, Ransäter, Rudstorp (malaise trap ID 1002); 59.7729560°N, 13.4737140°E; sandy railway embankment through pastureland; 2005.vii.23–2005.viii.12.

The species has not previously been recorded for Sweden (Dyntaxa 2022). The recorded distribution range mainly covers countries in central and eastern Europe: Austria, Czech Republic, Germany, Norway, Poland, Romania and Ukraine (Horstmann 2011; Yu et al. 2016).

Figure 13. 

D. alpina (Strobl, 1904), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6368714.

Dusona subimpressa (Förster, 1868) (Fig. 62)

Sweden • 2 ♂♂; Skåne, Klippans kommun, Skäralid, dal (malaise trap ID 37); 56.027217°N, 13.223433°E; rich beech forest; 2005.vii.07–2005.viii.09.

This species has not previously been recorded for Sweden (Dyntaxa 2022). The current records suggest that the distribution covers the entire Western Palaearctic: Austria, Belarus, Belgium, Bulgaria, Czech Republic, Finland, France, Germany, Italy, Japan, Moldova, Netherlands, Norway, Poland, Romania, Russia, Slovenia, Spain, Switzerland, Ukraine and United Kingdom (Horstmann 2011; Yu et al. 2016; Varga 2021). In addition to more than a dozen specimens in the SMTP material, D. subimpressa was also found in the material of the collection NJ.

Figure 14. 

D. alticola (Gravenhorst, 1829), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6368756.

Figure 15. 

D. angustata (Thomson, 1887), ♀ Norway. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6368773.

Figure 16. 

D. angustifrons (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6368793.

Figure 17. 

D. annexa (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b – g). https://doi.org/10.5281/zenodo.6368862.

Figure 18. 

D. aurita (Kriechbaumer, 1883), ♂ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6368911.

Figure 19. 

D. baueri Hinz, 1973, ♂ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 1 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6369124.

Figure 20. 

D. bellipes (Holmgren, 1872), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6369136.

Figure 21. 

D. bicoloripes (Ashmead, 1906), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370015.

Figure 22. 

D. blanda (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370029.

Figure 23. 

D. bucculenta (Holmgren, 1860), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370034.

Figure 24. 

D. calceata (Brauns, 1885), ♂ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370039.

Figure 25. 

D. carinifrons (Holmgren, 1860), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370042.

Figure 26. 

D. carpathica (Szépligeti, 1916), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370050.

Figure 27. 

D. circumcinctus (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370053.

Figure 28. 

D. circumspectans (Förster, 1868), ♂ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370091.

Figure 29. 

D. confusa (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370096.

Figure 30. 

D. cultrator (Gravenhorst, 1829), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370126.

Figure 31. 

D. dubitor Hinz, 1977, ♂ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370139.

Figure 32. 

D. erythrogaster (Förster, 1868), ♀ Switzerland. a habitus. b antennal carina. c junction of genal and hypostomal carina. d mesopleuron. e propodeum. f first gastral tergite. g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370145.

Figure 33. 

D. falcator (Fabricius, 1775), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370158.

Figure 34. 

D. flagellator (Fabricius, 1793), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370164.

Figure 35. 

D. humilis (Förster, 1868), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370179.

Figure 36. 

D. inermis (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370190.

Figure 37. 

D. infesta (Förster, 1868), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370198.

Figure 38. 

D. insignita (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370208.

Figure 39. 

D. juvenilis (Förster, 1868), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370232.

Figure 40. 

D. leptogaster (Holmgren, 1860), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370357.

Figure 41. 

D. libertatis (Teunissen, 1947), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370363.

Figure 42. 

D. limnobia (Thomson, 1887), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370379.

Figure 43. 

D. mercator (Fabricius, 1793), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370394.

Figure 44. 

D. minor (Provancher, 1879), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370397.

Figure 45. 

D. cf. nebulosa Horstmann, 2009, ♂ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370415.

Figure 46. 

D. nidulator (Fabricius, 1804), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370425.

Figure 47. 

D. notabilis (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370489.

Figure 48. 

D. obliterata (Holmgren, 1872), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370522.

Figure 49. 

D. opaca (Thomson, 1887), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370545.

Figure 50. 

D. petiolator (Fabricius, 1804), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370570.

Figure 51. 

D. pineticola (Holmgren, 1872), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370593.

Figure 52. 

D. polita (Förster, 1868), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370620.

Figure 53. 

D. prominula (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370633.

Figure 54. 

D. pugillator (Linnaeus, 1758), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370641.

Figure 55. 

D. pulchripes (Holmgren, 1872), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370652.

Figure 56. 

D. recta (Thomson, 1887), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370667.

Figure 57. 

D. rubidatae Horstmann, 2009, ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370676.

Figure 58. 

D. sobolicida (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370684.

Figure 59. 

D. spinipes (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370695.

Figure 60. 

D. stragifex (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370705.

Figure 61. 

D. stygia (Förster, 1868), ♀ Switzerland. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370719.

Figure 62. 

D. subimpressa (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370729.

Figure 63. 

D. tenuis (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370741.

Figure 64. 

D. terebrator (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370750.

Figure 65. 

D. thomsoni Hinz, 1963, ♀ Norway. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370754.

Figure 66. 

D. vidua (Gravenhorst, 1828), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370761.

Figure 67. 

D. xenocampta (Förster, 1868), ♀ Sweden. a habitus b antennal carina c junction of genal and hypostomal carina d mesopleuron e propodeum f first gastral tergite g 2^nd and 3^rd gastral tergites. Scale bars: 3 mm (a); 0.5 mm (b–g). https://doi.org/10.5281/zenodo.6370765.

This study reveals the presence of 11 previously unrecorded Dusona species for Sweden. However, so far, we have examined only a third of the Campopleginae material from the SMTP and some of the individuals in the collections of the NHRS and of Niklas Johansson. There is certainly much more Swedish material present in other natural history museum and private collections, such as at the Lund Museum of Zoology. The presence of multiple species represented only by a single specimen in our data set and the clearly unsaturated species accumulation curve (Fig. 10c) indicate that the occurrence of additional Dusona species is certainly to be expected for Sweden. Furthermore, there might be habitat types that have not been covered extensively by SMTP (Johansson 2020), hence additional sampling might be necessary to complete the faunistic records for this genus. This result lends support to other studies that showed major knowledge gaps for parasitoid wasps and Diptera, even in the faunistic record of well-studied countries such as Sweden (Johansson and Cederberg 2019; Johansson 2020; Johansson and Klopfstein 2020; Riedel 2020; Ronquist et al. 2020). Several of the newly recorded species were restricted to southern and coastal areas (Fig. 9). Hence, it is plausible that in addition to previously overlooked species, some of the newly discovered species could represent recent faunistic changes as a by-product of climate change. Effects of climate change on faunistic assemblages are already observable (Cerini 2020) and expected to lead to immigration of species into higher latitudinal (Dortel et al. 2013) and altitudinal (Gilgado et al. 2022) regions. To further evaluate this aspect, additional sampling is needed in diverse habitats and main collections need to be more thoroughly revised. Even though Dusona is the largest genus of Campopleginae, it experienced rather recent revisions in the Palaearctic (Hinz and Horstmann 2004; Horstmann 2009). Interestingly, the proportion of Dusona specimens in our sample (8%) was distinctly lower than the proportion of Dusona species in the Western Palaearctic fauna of Campopleginae (12%). Maybe Dusona species are less abundant or less likely to be caught by Malaise traps than other Campopleginae, but more likely, this indicates that other Campopleginae genera, such as the four next largest genera, Diadegma Förster, 1869, Hyposoter Förster, 1869, Olesicampe Förster, 1869, and Campoplex Gravenhorst, 1829, are still much less intensively investigated than Dusona. Hence, even more faunistic and taxonomic novelties can be expected for these genera.

Since the identification of Dusona species and Campopleginae in general has been considered a challenge in taxonomy (Townes 1969b; Broad et al. 2018), one major goal of this study was to improve the applicability of existing keys and to facilitate species identifications for a wide target audience. The photographic guide to diagnostic characters we provide (Figs 1–7) will allow for a better understanding of the different character states mentioned in the keys of Hinz and Horstmann (2004) and Horstmann (2009). Furthermore, the illustrations of almost half of the 125 Western Palaearctic Dusona species (n = 57, Figs 11–67) with highly resolved colour plates provides a solid, though incomplete basis for a reference-based species identification. We consider these species plates a highly valuable supplement to the dichotomous keys, as many species have highly distinctive traits in coloration, which are only partly addressed in the existing keys (Hinz and Horstmann 2004; Horstmann 2009). However, in order to be aware of the intraspecific variability of characters and especially of colour traits, the species descriptions in Hinz and Horstmann (2004) and Horstmann (2009) should also be checked. Some species need to be considered as even more variable than previously assumed. In the future, the figures used for the character guide and species plates could even be used to create an interactive key, which again would simplify the species identification compared to strictly dichotomous keys. However, in ecological studies, time and expertise for morphological species determination is often limited. Even though the species treated in this study are generally morphologically distinct, cryptic species also occur in Darwin wasps (Smith et al. 2011; Veijalainen et al. 2011; Johansson and Cederberg 2019). Hence, we have also provided a first extensive reference set of barcodes for 46 Dusona species based on reliably identified specimens that are deposited in public collections (Suppl. material 3). Barcodes allow for ecological studies, such as food-web, pollination, or community analysis, with species-level data (Smith et al. 2011; Joly et al. 2014). But also systematics benefits from species-level ecological studies as they can contribute ecological data to strengthen species boundaries (Van Valen 1976; Smith et al. 2011). The completeness of reference barcode libraries is crucial for their value in applied studies (Pinna et al. 2021). In the future, these resources should both be expanded to cover an increasing proportion of the diversity of the genus.

We have proposed a first phylogeny of Western Palaearctic Dusona species, which is based on a multilocus dataset with 45 Dusona species. Many morphological characters were found to be highly homoplastic, and it was thus not possible to define well-supported subgenera. Instead, the evolution of Dusona appears to have led to a plethora of parallel evolution of several traits, which means that the genus has to be treated in its entirety in identification keys. Nevertheless, we have recovered several closely related species groups with distinct, unique characters or combinations thereof, such as the mercator group or the subimpressa group that correspond to species groups suggested already by Hinz (1977). Interestingly, the basal “nidulator group” shares the same characters differentiating it from the rest of the examined Dusona species as the former genus Kartika from India (Gupta and Gupta 1976): complete separation of the third laterotergite from the tergite (epipleural index = 0) and a complete petiolar suture. Because these character states are plesiomorphic for Campopleginae, Wahl (1991) declared Kartika as a synonym of Dusona. Further molecular analyses of Indian Dusona species belonging to the former Kartika are necessary to evaluate their relatedness to the “nidulator group”. Even though our data does not suggest to split Dusona into subgenera, studying these species groups can help to understand the evolution of characters and maybe host adaptations.

Based on the p-distances, we found that the barcodes were able to differentiate most of the Dusona species well (<1% intraspecifically, > 4% interspecifically). However, there were some notable exceptions, with closely related species that showed distances of about 1% or that even turned out as paraphyletic. As these species are morphologically distinct, we suggest this might be evidence either for rapid evolution or mtDNA introgression within Dusona, and thus a limitation of barcoding in separating these species. Indeed, barcoding has been shown in the past to have a limited discriminatory power in some Darwin wasp groups, such as Diplazontinae, Ichneumon Linnaeus, 1758 and Enicospilus Stevens, 1835 (Tschopp et al. 2013; Klopfstein et al. 2016; Shimizu et al. 2020).

Future taxonomic work on the genus Dusona should aim to complete the barcode library for the well-established Palaearctic fauna, but also focus on the genus diversity in regions lacking recent revisions. Since many Dusona species are described from the Nearctic and the Oriental (Yu et al. 2016), these regions appear especially relevant for future revisions. However, the southern hemisphere is still massively under-researched and its diversity of Dusona most likely heavily underestimated (Vas and Di Giovanni 2021). Further research is needed to understand the complex evolutionary history of Dusona, the processes leading to the rapid radiation within this genus, and the factors driving its host adaptation. Complementing our phylogenetic framework with an extended taxon sampling, including the global diversity of Dusona, would be a very ambitious approach but could certainly deliver answers to some of these questions. Based on our study, we hope that the determination of Dusona specimens has become more accessible and that more people will be encouraged to work with the genus.