A New Species of Aleiodes Wesmael (Braconidae, Rogadinae) with Potential for Biological Control of Spodoptera spp. (Lepidoptera, Noctuidae), and Notes on the Definition of the gastritor, circumscriptus, and Related Species-Groups

A new species of Aleiodes Wesmael in the gastritor Thunberg species-group is described and illustrated. The new species, Aleiodes ceres Shimbori sp.n., occurs in Southern Brazil and Argentina. It is a potential biological control agent of caterpillars in the genus Spodoptera Guenée, namely S. cosmioides (Walker), S. eridania (Stoll), and S. frugiperda (J.E. Smith), which are serious pests on major crops including maize and soybean. The position of the new species within Aleiodes is discussed in relation to existing proposals of subdivisions of the genus into species-groups. In agreement with recent publications, the gastritor species-group is treated as separate from the circumscriptus Nees/bicolor Spinola group, as evidence suggests the latter is absent in the New World. We also propose the provisional shakirae Shimbori & Shaw species-group, to accommodate a clade of Neotropical species with geometrid hosts and morphological features similar to the gastritor and circumscriptus groups. A key to Neotropical species in the gastritor group and similar species is provided, except for species recently named without a morphological description. Supplementary Information The online version contains supplementary material available at 10.1007/s13744-023-01076-8.


Introduction
The genus Aleiodes Wesmael is the most diverse lineage within Rogadinae (Hymenoptera, Braconidae), and has a worldwide distribution.All rogadines are koinobiont endoparasitoids of lepidopteran larvae and mummify their host caterpillar while pupating inside it (Shaw and Huddleston 1991;Shaw 2006;Zaldívar-Riverón et al. 2008).Virtually, all Aleiodes species are solitary on exposed feeders in several lepidopteran families, with basal groups mainly associated with Noctuidae (Zaldívar-Riverón et al. 2008;van Achterberg and Shaw 2016).
The most comprehensive studies subdividing Aleiodes into species-groups form the foundation for studies on the systematics of the genus (Shaw 1997;Fortier and Shaw 1999).Two decades after that proposition, some groups had their monophyly contested, mostly based on molecular phylogenies (Zaldívar-Riverón et al. 2008;van Achterberg et al. 2020), reinterpreted in various ways (Townsend and Shaw 2009;van Achterberg and Shaw 2016), and additional subgroups have been proposed (Fortier 2006;Areekul-Butcher and Quicke 2011;Areekul-Butcher et al. 2012;Shimbori et al. 2016).These rearrangements and new species-groups are based on taxonomic studies, mainly of faunas of previously underrepresented biogeographical regions, such as the Oriental and Neotropical regions, with or without the inclusion of molecular data.
Currently, there are several isolated groups recovered as monophyletic in at least one molecular phylogeny (Quicke et al. 2006;Areekul-Butcher et al. 2012;van Achterberg et al. 2020), but the majority of the species in Aleiodes (precisely those that need major taxonomic efforts, such as the seriatus Herrich-Schaffer group sensu Fortier and Shaw 1999) are still laying outside these groups, especially in tropical faunas (Areekul-Butcher et al. 2012;Shaw et al. 2020).In this context, the Palearctic fauna is much better resolved, with revisions clarifying the systematics of the genus (van Achterberg and Shaw 2016;van Achterberg et al. 2020).Unfortunately, the subdivision system developed for the Palearctic is not entirely appropriate for the New World, and even for that fauna, there are several species not included in any group (van Achterberg and Shaw 2016).Because there is no recently published phylogeny testing monophyly of all subgroups of Aleiodes, or with enough coverage to propose a new subdivision to the worldwide fauna, the subdivisions based on morphology are still useful as a working framework, especially for those species-groups with unresolved phylogenetic relationships.
In the Neotropical region, the diversity of Aleiodes has proven to be one of the highest worldwide (Townsend and Shaw 2009;Shimbori et al. 2015Shimbori et al. , 2016;;Garro et al. 2017;Shaw et al. 2020), and the fauna is especially diverse in two species-groups: the seriatus and the circumscriptus/gastritor Townsend & Shaw groups.The first of these is presumed to be the most diverse group in the neotropics and has recently received a revision for one of the many morphologically cohesive subgroups, the bakeri Brues species-subgroup (Shaw et al. 2020).The second group is a combination of two groups proposed after a major study on Neotropical Aleiodes (Townsend and Shaw 2009).In that study, the authors highlighted the lack of a clear line defining the gastritor Thunberg and circumscriptus Nees groups in neotropical species, even though this line is clear in the Holarctic fauna (Shaw et al. 1997a, b;Fortier and Shaw 1999).The speciesgroup defined as circumscriptus/gastritor (Townsend and Shaw 2009) comprises many undescribed species and is one of the major taxonomic challenges in the genus.Its diversity could be explained by rapid recent radiation (van Achterberg et al. 2020), one of the reasons why this is the only speciesgroup not yet revised for the Nearctic region.
The moth genus Spodoptera Guenée (Lepidoptera, Noctuidae) comprises 31 species, of which 15 are considered pests in the Western and Eastern hemispheres (Kergoat et al. 2021).In Brazil, there are at least eight species of Spodoptera, namely S. albula (Walker, 1857), S. androgea (Stoll, 1782), S. cosmioides (Walker, 1858), S. dolichos (Fabricius, 1794), S. eridania (Stoll, 1782), S. evanida Schaus, 1914, S. frugiperda (J.E.Smith, 1797), and S. ornithogalli (Guenée, 1852) (Pogue 2002), and some of them are considered main pests on cotton, maize, and soybean fields (Martinelli et al. 2006).Since the years 2013/2014 soybean crop season, with the approval of the commercial release of genetically modified soybean expressing Cry1Ac protein from Bacillus thuringiensis Berliner (Bt) (CTNBio 2010), there has been an increase in the occurrence of S. cosmioides, S. eridania, and S. frugiperda.These outbreaks of Spodoptera species can be explained by the low susceptibility of these insects to genetically modified soybean expressing Cry1Ac protein from Bt (Bernardi et al. 2014).In maize, S. frugiperda is the most destructive and important pest in Brazil (Cruz et al. 2012;Blanco et al. 2016).This species has already developed resistance to 45 active ingredients which include pesticides and several Bt transgenic events (Mota-Sanchez and Wise 2023), which makes the control of this pest even more difficult.
Biological control plays an important role in integrated pest management in Bt crops because natural enemies can delay the evolution of insect resistance to this technology (Liu et al. 2014).Research interest in parasitoids of Spodoptera aiming at its control has been high for at least a century, especially for the fall armyworm (Molina-Ochoa et al. 2003).A few species of Aleiodes are recorded as parasitoids of Spodoptera, of which Aleiodes laphygmae (Viereck) is the most commonly reported, mainly in North America (USA and Mexico) (Yu et al. 2016).Overall, Aleiodes have been reported to have little impact on fall armyworm populations in crops, but its prevalence among parasitoids in grasses reached 92% (Braman et al. 2004).It is important to clarify that the record of A. laphygmae in Brazil, attributed to Cruz et al. (1997a, b) by Molina-Ochoa et al. (2003), is not correct, and other records from South America are sparse and may also be incorrect.Other species reared from Spodoptera include Aleiodes terminalis Creson in S. frugiperda and S. ornithogalli, in North America (Marsh and Shaw 2001), and Aleiodes vaughani (Muesebeck) in S. frugiperda and S. eridania, in Central and South America (Muesebeck 1960;Redolfi-Huiza and Marin-Loayza 1992;Shaw et al. 1997a, b).The discovery of a new species of Aleiodes parasitizing caterpillars of the Spodoptera complex in Brazil opens new prospects for the use of this parasitoid in biological control programs for Spodoptera species.
The objective of this paper is to describe a new species of the genus Aleiodes, with excellent potential for application in the biological control of Spodoptera.Because Aleiodes is a highly diverse genus, with hundreds of described species, we also discuss the placement of this new species in the context of the proposed subdivisions of the genus into species-groups, especially the circumscriptus/gastritor group and other similar or related groups.We also present an illustrated key to species based on morphological characters and provide DNA barcode sequences of the new species, aiming to facilitate easier identification of the new species, especially for applied entomologists (Shimbori et al. 2023).A summary of the biological information of the neotropical species previously included in the circumscriptus/gastritor group is also provided.

Sampling
Specimens of Aleiodes ceres Shimbori sp.n. were collected in 2016/2017 soybean (Glycine max (L.) Merrill) crop season in São José dos Pinhais (25°36′49.0″S49°08′01″W), Paraná, Brazil.In the 2017/2018 and 2018/2019 soybean crop seasons, between January and March, a survey of pest caterpillars was conducted weekly by the beat cloth method (Shepard et al. 1974) in 10 points of two areas of genetically modified soybeans.One of the areas (non-Bt area) was planted with the seed variety NA5909RG, expressing tolerance to the herbicide glyphosate, during the two crop seasons evaluated.The other area (Bt area) was planted with the variety Syn13671 IPRO in the first crop season, and in the second crop season with variety M5917 IPRO and Syn1561 IPRO.Varieties from the Bt area also provide tolerance to the herbicide glyphosate, besides expressing the Cry1Ac protein of Bacillus thuringiensis (Bt) which confers resistance to target lepidopteran species.
Collected caterpillars were separated individually in plastic containers, identified at species level, and fed with soybean leaves from the same variety in which they were collected.Within the species collected, we identified the most common species of the Spodoptera complex in Brazil, namely Spodoptera cosmioides (Walker), Spodoptera eridania (Stoll), and Spodoptera frugiperda (J.E.Smith).The caterpillars were observed daily for parasitism.After emergence of the adult parasitoids, lab-reared Spodoptera eridania caterpillars were provided aiming to establish a laboratory rearing of the parasitoids.Specimens of the parasitoid were killed at low temperature (− 20 °C) and preserved in 99% ethanol for identification.
Additional samplings were conducted in the 2019 maize (Zea mays L.) crop season, in Pinhais (25°24′01″ S 49°07′01″ W), Paraná, Brazil, in an experimental area with the variety 30F53 Pioneer.The methodology used was similar to the soybean sampling.

Biological data
Fifteen 48-h-old parasitoid females previously mated were kept individually in plastic containers (AE 98.5 × 210 mm) with a drop of honey in an incubator room (25 ± 2 °C, 70 ± 10% UR and 14:10 photoperiod).Second instar larvae of S. eridania (n = 15) were provided for each female, and after 24 h, caterpillars were removed and new specimens were offered.This procedure was repeated for 3 days.
Larvae were individualized in plastic containers (500 mL) and daily, soybean leaves from the variety BRS 1003IPRO which express tolerance to the herbicide glyphosate and the Cry1Ac protein of Bt were provided.We evaluated the total number of larvae parasitized (by observing the formation of mummies), development time (in days), and emergence rate.

Taxonomy
For identification of the subfamily Rogadinae (Braconidae), see van Achterberg (1993) or Sharkey (1997).For recognition of rogadine genera, refer to the identification keys of van Achterberg (1991) or Shaw (1997).The definition of Aleiodes adopted here follows that of van Achterberg (1991) and Shaw (2006).For identification of gastritor and circumscriptus species-groups, see Shaw et al. (1997a, b) and Fortier and Shaw (1999).The identification key presented here was adapted from the key to species of Aleiodes from Ecuador (Shimbori and Shaw 2014).Additional images for the majority of the species in the key can be found there.

Molecular methods
Genomic DNA was extracted from the whole wasp using DNeasy Blood & Tissue Kit (QIAGEN Inc., Valencia, California) following the manufacturer's instructions.Subsequent DNA purification was performed by Ethanol Precipitation (Sambrook and Russell 2001) and then DNA extracts were resuspended in 50 µL of TE Buffer.The COI fragment belonging to the barcoding locus was amplified using the universal LCO 1490 and HCO 2198 primers (Folmer et al. 1994).Polymerase chain reaction was carried in 25 µL final volume (2.5 mM MgCl 2 , 0.2 mM dNTP Mix, 0.2 µM each primer, 1 × HOT FIREPol® Buffer B2, and 1 U HOT FIREPol® DNA Polymerase, Solis Biodyne) and followed a cycling process of initial denaturation at 95°C for 15 min; 35 cycles of denaturation at 95°C for 45 s, annealing at 56°C for 30 s, and extension at 72°C for 1 min; and finally a final extension at 72°C for 5 min.PCR products were purified by polyethylene glycol precipitation (Lis and Schleif 1975) and sequenced at Macrogen, Seoul, South Korea.

Phylogeny
The molecular dataset for phylogenetic analyses was mainly gathered from GenBank and BOLD, based on previous publications (Areekul-Butcher et al. 2012;Shimbori and Shaw 2014;van Achterberg et al. 2020;Sharkey et al. 2021), aiming to include representatives of most species-groups in the genus Aleiodes, and all biogeographic regions.A total of 286 sequences were used for the analyses, four of them of species of Heterogamus (Appendix 1 in Supplementary Information (SI)).Sequences were aligned using MAFFT v. 7.130b (Katoh and Standley 2013).Molecular phylogeny was performed using IQ-TREE v 1.6.12maximum-likelihood analyses (Nguyen et al. 2015), with ultrafast bootstrap (6000 replicates) (Hoang et al. 2018).The ModelFinder (Kalyaanamoorthy et al. 2017) was used to choose the best model for each of the three partitions, corresponding to each codon position (Chernomor et al. 2016).The best-fit models according to Bayesian information criterion were TIM2 + F + R6 for partition 1 and TIM + F + I + G4 for partition 2 + 3. A single branch SH-aLRT test was performed with 2000 replicates (Guindon et al. 2010).
The resulting best tree is presented with species-groups highlighted with the same color code used in van Achterberg et al. ( 2020 on the COI Barcode locus, although the subgenus Chelonorhogas (or the apicalis species-group) is recovered as a monophyletic clade rather than a grade leading to the subgenus Aleiodes (Fig. 1).
Phylogenies based on a single locus, such as the one presented here, have serious limitations and are to be taken as a tentative assessment of the phylogenetic relationships.The main objectives of the analysis are to aid species delimitation and support the discussion on the classification of the new species at the species-group level, including more elements of the neotropical fauna.Relevant results for the systematics of neotropical Aleiodes are the recognition of a relatively basal clade within subgenus Aleiodes, which is morphologically similar to the gastritor Thunberg speciesgroup, and the absence of the circumscriptus Nees/bicolor Spinola group in the New World.Additionally, several neotropical species are found scattered within the gastritor clade, including A. luteosicarius Shimbori & Shaw, here transferred from pallidator Thunberg to gastritor, and the new species A. ceres sp.n.

Comments on the gastritor, circumscriptus, and similar species-groups in the subgenus Aleiodes in the Neotropical region
As proposed by Shaw et al. (1997a, b) and Fortier and Shaw (1999), the gastritor species-group would be distinguished from the circumscriptus species-group in having relatively larger ocelli (OD > OOL), and metasomal terga usually yellow (although a pattern similar to that of circumscriptus is possible), as opposed to the smaller ocelli (OD < OOL) and metasomal terga always yellow medially and black laterally in circumscriptus.In addition, the circumscriptus group is broadly associated with Noctuoidea hosts, while gastritor group species are mostly associated and Geometridae (Table 1, Fig. 5 in Zaldívar-Riverón et al. 2008 andFig. 7 in van Achterberg et al. 2020).However, as noted by Townsend and Shaw (2009), several (borderline) Neotropical species do not precisely fit in the diagnoses of the gastritor or circumscriptus species-groups.These authors used this evidence to propose the circumscriptus/gastritor species-group, lumping the two groups.
Both gastritor and circumscriptus species-groups are recovered as derived groups in the subgenus Aleiodes (sensu Zaldívar-Riverón et al. 2008), sister to one another, in all molecular phylogenies published so far (Zaldívar-Riverón et al. 2008;Areekul-Butcher et al. 2012;van Achterberg et al. 2020).Thus, the classification proposed by Townsend and Shaw (2009) could reflect the evolutionary history of Aleiodes, comprising the most derived clade in the subgenus Aleiodes.In this broader sense, the circumscriptus/gastritor group would include species of other small groups such as pallidator, compressor Herrich-Schaffer, and coxalis Spinola (= bicolor and excluding Tetrasphaeropix Ashmead), and many unassigned species (see trees in Zaldívar-Riverón et al. 2008 andvan Achterberg et al. 2020).
There are two problems in using this broader concept.First, the higher morphological and biological variation   2006, 2009;Quicke et al. 2006;Zaldívar-Riverón et al. 2008;Areekul-Butcher et al. 2012), and are desirable as a way to define smaller monophyletic groups, facilitating identifications and revisions.However, considering the morphological variation in the Neotropical species, it is clear that the main characters used to distinguish gastritor and circumscriptus are homoplasious.In fact, nearly all Neotropical species in the subgenus Aleiodes,1 excluding the species in the seriatus group, will fit in the broader morphological concept of circumscriptus/gastritor.For instance, A. kingmani Shimbori & Shaw is not distinct from other gastritor group species, but the morphology of its mummy resembles the suspended mummies of the buzurae group (Quicke et al. 2006;Shimbori and Shaw 2014).Besides mummy shape, species in the buzurae group are morphologically similar to gastritor, being distinguished from it by the strongly sculptured fourth metasomal terga (Quicke et al. 2006), a feature absent in described Neotropical species.Our phylogeny recovers neotropical species of the subgenus Aleiodes scattered in several parts of the tree, for most of which we do not have morphological information (Sharkey et al. 2021).Therefore, further proposals for subdivisions are difficult and of limited value without a refined and comprehensive phylogenetic study of the genus.
Based on the evidence discussed above, we summarize the following conclusions on the systematics of the Aleiodes species-groups: - The homoplasious nature of the characters defining pallidator seems corroborated by all available evidence, and by the fact that the only described neotropical species in the pallidator group, A. luteosicarius, is recovered in the gastritor group in our phylogeny (Fig. 1).Therefore, A. luteosicarius is here transferred to the gastritor speciesgroup.-The gastritor group is represented by many species in the New World, most of them undescribed.Currently, it is not possible to recognize this group using morphology alone, but our phylogeny indicates that neotropical species with morphological features of gastritor or circumscriptus (as originally recognized by Shaw et al. 1997a, b) in most cases belong to this clade.-A relatively basal clade within the subgenus Aleiodes, comprising neotropical species with geometrid hosts and "mixed" characters of gastritor and circumscriptus (i.e., variable color and relatively large ocelli) was recovered and should be considered for a possible sub-division of the subgenus.Here, we provisionally name this clade the shakirae group, and include species with similar morphology or DNA barcode.This group is not yet distinguishable from neotropical members of the gastritor group; however, all species have larger eyes and ocelli, the ocelli-ocular distance shorter than the diameter of lateral ocelli, and are parasitoids on Geometridae.The metasoma is at least partly dark brown or black in described species.Species without morphological descriptions (Sharkey et al. 2021) are included based on our molecular phylogeny.According to images provided by Sharkey et al. (2021), some of those species have the body entirely honey yellow and therefore would not be keyed correctly to the shakirae group in the key provided below.The group is proposed to acknowledge the existence of this clade, even though morphological diagnosis and monophyly based on molecular data are preliminary.

Key scope
The key presented below includes only neotropical species in the subgenus Aleiodes, previously in the circumscriptus/ gastritor (Townsend and Shaw 2009) or pallidator speciesgroups (Shaw et al. 2013), and currently in one of the three groups listed below ((1) gastritor, (2) shakirae, or (3) unassigned).These are neotropical species that would key in the gastritor and/or circumscriptus groups following the key provided by Shaw et al. (1997a, b) (couplet 16), or in the pallidator group (couplet 10), herein included in the gastritor group.These species do not comprise a monophyletic group but are recognizable by the following set of characters: Diagnosis.Apex of hind tibia without comb of specialized adpressed setae; hind wing vein RS slightly sinuate and enclosing a marginal cell which is narrowest around its middle length; metasomal tergites 1-3 with finely rugose or rugose costate sculpturing, never smooth or strongly costate; tarsal claws without distinct blackish pecten, but sometimes a yellowish pecten with thinner spines present; metasomal tergite 2 without a distinct smooth triangular area medio-basally; ocelli size variable.
Color (Holotype-female).Head dark brown, except palpi light pale yellow and mandibles pale brown.Antenna with flagellum brown; scape and pedicel mostly honey yellow, light brown dorsally.Mesosoma mostly honey yellow; Male.Body length 3.7-4.5;fore wing length 2.9-3.7.Similar to females in most aspects, except for the antenna which is longer than body (about 1.2 ×), compared with the Biology.Specimens of A. ceres collected in 2019/2020 soybean crop season have been kept in the laboratory for over 64 generations without losing their biological traits.In the laboratory, A. ceres can develop in the hosts: S. cosmioides, S. eridania, and S. frugiperda with their hosts being reared on a natural or artificial diet.Females of A. ceres can parasitize caterpillars from the first to third instar; however, better parasitism and emergence rates were observed in caterpillars of the first and second larval instar.When offered the second instar of S. eridania, A. ceres parasitized on average 9.07 ± 0.78 caterpillars in 24 h.In 3 days, a total of 26.0 ± 1.47 caterpillars were parasitized.The development time (egg-adult) was 16.58 ± 0.09 days and the emergence rate was 87.44 ± 2.59 at 25 ± 2 °C, 70 ± 10% RH.
Etymology.The choice for the name of the new species was made in a public contest, with the objective of raising awareness and interest of the general public, especially middle and high school students, for taxonomic research and its relevance to biological control.The contest was held during a community outreach fair "ESALQSHOW," on the 6th and 7th of October, 2022, on the campus of the "Luiz de Queiroz College of Agriculture (ESALQ)," of the University of São Paulo.For all participants, a brief explanation of the rules of zoological nomenclature was given, in addition to the economic and ecologic importance of the new species and the consequences of the gap in taxonomic knowledge.The name chosen by one of the public contest participants was given in reference to the goddess of agriculture, Ceres, of Roman mythology.Diagnosis.Very small ocelli, ocell-ocular distance more than 2 × longer than diameter of lateral ocellus in females (about 2 × in males); body color variable, mostly honey yellow with dark brown marking at mesoscutum, propodeum and metasoma, pronotum and mesopleuron always honey yellow, head nearly always black; stigma entirely pale brown; longitudinal carina at metasomal tergum 3 absent or shortly indicated anteriorly, complete at terga 1 and 2.
Comments.The new species differs from A. laphygmae in having much smaller ocelli.The ocell-ocular distance is more than two times longer than the diameter of the lateral ocellus (in contrast to an ocell-ocular distance equal to or slightly shorter than the diameter of the lateral ocellus in A. laphygmae).The metasomal tergum 3 has a nearly complete longitudinal carina in A. laphygmae, but in the new species, the longitudinal carina is absent or only weakly indicated anteriorly.The new species also has a much more extensive  The recent record of A. laphygmae from Argentina (Valverde et al. 2012) has to be revisited in view of our findings.Considering geographic distribution, host and plant food associations, and morphological features, the identity of the specimens recorded from there are likely to be Aleiodes ceres sp.n.Sequences of Aleiodes ceres sp.n.from Argentina included in the phylogeny were mined from BOLD, and there is no evident connection with specimens reported by Valverde et al. (2012).The images of one of the specimens corroborates identification as A. ceres sp.n.
Records from Argentina are based on DNA Barcode sequences deposited in public databases (BOLD record IDs: GMAGR178-15; GMARM1238-14; GMAGP1961-15; GMAGR182-15).The similarity between sequences from Argentinian and Brazilian specimens ranges from 98.3 to 99.1%.Our ML phylogeny also supports the conspecificity of Argentinian and Brazilian specimens.All Argentinian specimens are deposited at Museo Argentino de Ciencias Naturales Bernardino Rivadavia, in Buenos Aires, Argentina, and are assigned a single BIN (Barcode Index Number), BOLD:ACN2401 (available at: http:// bolds ystems.org/ index.php/ Public_ Barco deClu ster?clust eruri= BOLD: ACN24 01).Aleiodes laphygmae (Viereck, 1912) Figure 17(A-H).Rogas laphygmae Viereck 1912, 43: 581.Diagnosis Ocelli moderate-sized, ocell-ocular distance about as long as the diameter of lateral ocellus; body color entirely honey yellow; stigma brown with yellow spots at base and apex; longitudinal carina at metasomal tergum 3 present and extending nearly over the entire length of tergite; body length 4.0-5.0mm; antenna with 33-35 antennomeres; the angle between fore wing vein r and basal ventral line of stigma about 160 degrees (Fig. 17), both appearing nearly in the same line.

DNA barcode of the holotype and two paratypes
Biology.The species is a parasitoid of the early stages of a few caterpillar species in the family Noctuidae, namely Mythmna unipuncta (Haw.),Spodoptera exigua (Hbn.), S. frugiperda (Smith), and S. orthogonali (Guen.).
Distribution  Fortier (2000) described the new species Aleiodes leptocarina from Costa Rica, which was notable (at that time) in being the only known Aleiodes from Costa Rica to develop gregariously inside large caterpillars.The identity of the host caterpillar was not known at that time, but Fortier (2000) did include a black and white photograph of the large, densely setose host mummy.More recently, Sharkey (2021) in Sharkey et al. (2021) described a similar "new species," which also exhibited gregarious development (80 specimens emerged from one large caterpillar mummy of Dysschema vidua (Erebidae)).Sharkey's "minimalist" species description was based on a DNA sequence, one photograph, the host association with Dysschema vidua, and only one stated morphological difference: the antennal flagellum was stated to be "melanic" (black) in A. normwoodleyi, as opposed to being "honey yellow" in A. leptocarina.During this study, we reexamined the holotype and paratype series for A. leptocarina Fortier, and observed that the flagellum is not yellow but instead is entirely black in all the type specimens.We believe that confusion was created by an error in Fortier's (2000) description of A. leptocarina, where the flagellum color was incorrectly listed as being "honey yellow."The images in Fortier's paper were all done by scanning electron microscopy, so the resulting photos are in gray tones and it is impossible to tell the antenna color from the images.While the caterpillar host for A. leptocarina was never determined, its densely hairy appearance is consistent with it possibly being a caterpillar of Dysschema vidua.Additionally, one of us (SRS) has previously identified in 2005 a long series of gregarious specimens reared from a mummified caterpillar of Dysschema vidua as being A. leptocarina (by direct comparison with the type series).This host association was reported to Dan Janzen and was posted on the Caterpillars of the ACG website for many years (until the species association was later changed to A. normwoodleyi), presumably on the basis of the wasps having a black flagellum (and the erroneous assumption that A. leptocarina specimens have a yellow flagellum).Therefore, we regard Dysschema vidua to be a previously known host record for A. leptocarina (in fact, the only known host for it so far established).We regard the species concept of A. normwoodleyi to be based on an (understandable) misidentification; therefore, the molecular sequence presented by Sharkey (2021) is presumed to actually be the barcode sequence for A. leptocarina Fortier.No other barcode data for A. leptocarina were found on the BOLD website for comparison.
) and Areekul-Butcher et al. (2012) for easier comparison.Our results are in great conformity with molecular trees for Thai (Areekul-Butcher et al. 2012) and Palaearctic Aleiodes (van Achterberg et al. 2020), all based

Fig. 1
Fig. 1 a, b Maximum likelihood tree based on DNA barcode sequence data for the genus Aleiodes.Clades corresponding to species-groups and subgenera proposed by several authors are highlighted.Terminal taxa with the comb of specialized flattened setae at the apex of the hind tibia are colored purple, and of the new species, Aleiodes ceres, are colored blue.Terminal text shows species name, voucher code, and country of provenance.Support values, SH-aLRT support (%)/ultrafast bootstrap support (%), are shown in relevant clades Aleiodes in the gastritor and shakirae species-groups (Aleiodes bonariensis is not included due to poor condition of type specimen; species named by Sharkey et al. (2021) based only on molecular data are not included) 1. Ocelli small, ocelli-ocular distance longer than width of lateral ocellus (Fig 2A, B) ………………………… 2 Ocelli moderate-sized, ocelli-ocular distance equal to or shorter than width of lateral ocellus (Fig 2C, D) …… 14 2(1).First and/or second metasomal terga with median carina present (Fig 3A); ovipositor sheaths at most 2/3 length of hind basitarsus ……………………………… 3 -First and second metasomal terga with median carina absent (Fig 3B); ovipositor about 2× length of hind basitarsus; Geometridae hosts …… albiterminus Townsend

Table 1
Yu et al. (2016), andSharkey et al. (2021) species-groups of the Neotropical species of the subgenus Aleiodes Wesmael, 1838, related to the gastritor species-group.The biological information is based onTownsend and Shaw (2009),Abreu et al. (2014),Shimbori and Shaw (2014),Yu et al. (2016), andSharkey et al. (2021)DHJPAR0042782: "encontrado con momia de parásito silvestre en estadío intermedio entre PU y U, tome fotos de la momia").Images of the mummy (available at http:// janzen.sas.upenn.edu/caterpillars/datab ase.lasso) indicate it is most likely from a Geometridae caterpillar to diagnose the group.The circumscriptus group, for instance, is morphologically distinct and nearly entirely associated with one host family, the Noctuidae.This group is also absent in the New World, whereas the gastritor species-group is widespread.Second, the Neotropical species that would fit in the broader morphological concept do not form a monophyletic clade.Instead, at least two clades comprise Neotropical species, one being the gastritor and a second and relatively basal clade within the subgenus Aleiodes, which we refer to as the shakirae Shimbori & Shaw clade.This basal clade comprises Neotropical species with geometrid hosts and "mixed" characters, of both the gastritor and circumscriptus groups (Fig.1).Since our results are preliminary with respect to the phylogenetic history of the group, and the key point of separating species-groups is to facilitate the identification of this massive genus, we consider these subdivisions useful for their purpose of separating the genus into manageable parts, as long as each part is identifiable, either using morphology or DNA.Subdivisions of this large clade already exist (e.g., groups bicolor/circumscriptus, gastritor, and similis Curtis in the Palearctic-van Achterberg et al. 2020), as well as within the subgenus Aleiodes as a whole (e.g., the buzurae He & Chen, Tetrasphaeropyx and risaaeQuicke  & Butcher groups-van Achterberg et al. 2020; Fortier * Host record for A. michelledsouzae is based on a mummy collected on the referred plant (Voucher Code: 11-SRNP-31492; Parasite code: The circumscriptus species-group (including the bicolor or coxalis species-group and A. pallidator species-group) is likely a monophyletic group absent in the New World (it may also be called bicolor or coxalis-priority of Spinola 1808 over Nees 1834).It should be treated as its own group, as proposed by van Achterberg et al. (2020).-Regarding the pallidator group, the type species Aleiodes pallidator is recovered within the Palearctic circumscriptus + bicolor + coxalis clade (van Achterberg et al. 2020; Fig.1).All species in this clade, including A. pallidator, are native to the Old World(Shaw et al.  2006), even though A. pallidator was introduced from Europe to North America for biological control(Shaw  2006).In addition, A. pallidator is the only known species in the pallidator species-group, sensu Shaw et al.
(Zaldívar-Riverón et al. 2008)ten in the tarsal claws.A closer relationship with A. coxalis was also recovered in a phylogeny using COI + 28S(Zaldívar-Riverón et al. 2008).If the phylogenetic inferences are correct, pallidator should be part of the circumscriptus (= bicolor) group, and if the biogeographical distribution of circumscriptus is indeed restricted to the Old World, the species from the New World are better placed in another group.
. Southern USA, Cuba, and Nicaragua.Recent records from Argentina are most likely of Aleiodes ceres sp.n.TAXONOMIC NOTES Aleiodes leptocarina Fortier 2000 Aleiodes normwoodleyi Sharkey, 2021 NEW SYNONYMY During this study, we established that Aleiodes normwoodleyi Sharkey, 2021 is a junior synonym of the previously described species Aleiodes leptocarina Fortier 2000.