Revision of the western Palaearctic species of Aleiodes Wesmael (Hymenoptera, Braconidae, Rogadinae). Part 2: Revision of the A. apicalis group

Abstract The West Palaearctic species of the Aleiodes apicalis group (Braconidae: Rogadinae) as defined by van Achterberg & Shaw (2016) are revised. Six new species of the genus Aleiodes Wesmael, 1838, are described and illustrated: A. carbonaroides van Achterberg & Shaw, sp. nov., A. coriaceus van Achterberg & Shaw, sp. nov., A. improvisus van Achterberg & Shaw, sp. nov., A. nigrifemur van Achterberg & Shaw, sp. nov., A. turcicus van Achterberg & Shaw, sp. nov., and A. zwakhalsi van Achterberg & Shaw, sp. nov. An illustrated key to 42 species is included. Hyperstemma Shestakov, 1940, is retained as subgenus to accommodate A. chloroticus (Shestakov, 1940) and similar species. Fourteen new synonyms are proposed: Rogas bicolor Lucas, 1849 (not Spinola, 1808), Rogas rufo-ater Wollaston, 1858, Rhogas bicolorinus Fahringer, 1932, Rhogas reticulator var. atripes Costa, 1884, and Rhogas similis Szépligeti, 1903, of Aleiodes apicalis (Brullé, 1832); Rogas (Rogas) vicinus Papp, 1977, of Aleiodes aterrimus (Ratzeburg, 1852); Rogas affinis Herrich-Schäffer, 1838, of Aleiodes cruentus (Nees, 1834); Bracon dimidiatus Spinola, 1808, and Rhogas (Rhogas) dimidiatus var. turkestanicus Telenga, 1941, of Aleiodes gasterator (Jurine, 1807); Rogas alpinus Thomson, 1892, of Aleiodes grassator (Thunberg, 1822); Rhogas jaroslawensis Kokujev, 1898, of Aleiodes periscelis (Reinhard, 1863); Rhogas carbonarius var. giraudi Telenga, 1941, of Aleiodes ruficornis (Herrich-Schäffer, 1838); Ichneumon ductor Thunberg, 1822, of Aleiodes unipunctator (Thunberg, 1822); Rogas heterostigma Stelfox, 1953, of Aleiodes pallidistigmus (Telenga, 1941). Neotypes are designated for Rogas affinis Herrich-Schäffer, 1838; Rogas nobilis Haliday (in Curtis), 1834; Rogas pallidicornis Herrich-Schäffer, 1838; Rogas ruficornis Herrich-Schäffer, 1838. Lectotypes are designated for Rhogas (Rhogas) dimidiatus var. turkestanicus Telenga, 1941, and Rhogas hemipterus Marshall, 1897.


Introduction
In this 2 nd part of a revision of western Palaearctic species of Aleiodes Wesmael we treat the group identified in Part 1 (van Achterberg and Shaw 2016) as the A. apicalis group. It should be noted that our A. apicalis group is constituted in a different (wider) way than of Shaw et al. (1998), , and Garro et al. (2017). The Aleiodes apicalis group as interpreted here is easily recognised from the key given by van Achterberg and Shaw (2016) and includes a majority of species that are rather large for the genus as a whole. From data presented in this paper, some of the commonest species, as well as being rather morphologically isolated, can now be said to be well-understood biologically and appear to be taxon-specialists (see Shaw, 1994Shaw, , 2017, which suggests to us that they have not been involved in recent speciation events (see Shaw, 2003). However, for others, including a substantial number of rare, or at any rate rarely collected species, there is practically no biological information, with the result that this generalisation cannot be extended: indeed, there are some groups of apparently closely related species that are much more difficult to separate and in these parts of the A. apicalis group speciation has probably been more recent. We deal here with 42 species, a few of which are included only because they have been recorded from the region by others (i.e., the relevant specimens not examined by us) and/or are considered likely to occur in the eastern part of the area. Several of the most seldom-collected species occur as adults early in the year and may not in reality be as rare as they seem.

Specimens, methods, and presentation of records
The biological data from rearings of wild-collected hosts is in some cases supplemented by experimentation, and the protocols and means of scoring results are as outlined in van Achterberg and Shaw (2016). The rather full introductory sections of that paper apply here and are not repeated except when not to do so would leave this paper difficult to use by itself. The term plurivoltine is used to indicate more than one generation in the year (very often this would be only two, but it could be more under favourable circumstances).
Overall, many of the species treated here have been widely misinterpreted in the literature and, as in Part 1 of our revision (van Achterberg and Shaw 2016), we have ignored published records when compiling host and distributional data, depending only on specimens we have actually seen ourselves. As previously, we have simply updated the nomenclature of hosts rather than transcribing obsolete names or obvious misspellings from data labels, and unless stated otherwise the reared material cited is in National Museum of Scotland, Edinburgh (NMS). From the host data we are able to give, indicating compact host ranges, in comparison with that expressed in Yu et al. (2016) it should be clear that this was a wise action. By similarly ignoring distribution data (e.g., from Yu et al. (2016)) we do not suggest that published distribution records are necessarily wrong: simply that we are unable to confirm them from the many thousands of specimens we have examined. The sheer number of these prevented us from listing specimen data in full, except for the few species of which we have seen only a very few specimens. The countries we list from the area under consideration (the western Palaearctic) are followed by a list of extralimital countries, in square brackets, from which we have also examined the species in question.
All available collections containing recently collected material of Aleiodes from the western Palaearctic region were used for our revision; collections with type material are separately listed under the description of the species. The following collections and acronyms are used: In addition, we have examined specimens from various smaller and private collections, which are cited in significant cases. Unless otherwise specified, reared material is in NMS.
The number of antennal (i.e., flagellar + 2) segments is frequently an important aid to species recognition and of interest also because in some species the female has more segments on average than the male (males have a greater number in other species, which is the normal condition seen in Braconidae). We give counts of antennal segments for the specimens we have examined, but for some species (especially when the segments did not need to be counted for determination) sometimes only for the first hundred or so of the specimens examined of each sex.
Attention has been paid to the apical tergites of males. The medial dorsal pores of A. fortipes (Reinhard), which are unique to this species within the A. apicalis group as treated here, are described and discussed in the entry for that species (note that the unknown male of A. caucasicus (Tobias) is likely to be similar). In the remainder of the species group there is either no evident modification, or a different development is evident to a greater or lesser extent. In some species specialised setae are present on tergites 4-6(7), presumably connected with pheromone dispersal from tergal glands. Broadly, two kinds of specialised setae can occur on these tergites. First, a fringe of short backwards-projecting setae (hereafter "fringe"), possibly associated with pores, originate from close to an apparent sulcus near the extreme base of the tergite (which is normally concealed). The presence and nature of the fringe varies between species, and even when present, it may not be visible in a given specimen owing to telescoping of the tergites. Second, there may be backwards-directed and more or less dense patches of longer setae (hereafter "setal patches") on each side of the mid-line, the setae to some extent being adpressed in their anterior part but tending to be raised posteriorly (in extreme cases giving the tergites a concave appearance) and appearing different from the arrangement of setae on the more anterior tergites. A median glabrous area is left between the paired setal patches on each tergite, which collectively present as a glabrous and often shiny dorsal stripe along the length of these tergites (hereafter "glabrous stripe"). There is considerable variation in the extent to which these features are developed in the species keyed here, and indeed in some species they are scarcely present or wholly absent. In the species accounts given below we attempt to give a score from 1 to 4 for the development of the setal patches and glabrous stripe in males, with minimal elaboration (but including also mention of the setal fringe in cases for which we have been able to observe it). Type 1 = not at all developed, setae as on anterior tergites and evenly distributed. Type 2 = setal patches hardly developed, but glabrous stripe evident to some extent. Type 3 = setal patches clearly developed but relatively weak or sparse, glabrous stripe strong. Type 4 = setal patches strongly developed, making the tergites appear concave, glabrous stripe also strong. It should be borne in mind that there is some intraspecific variation, much of which may be artefactual (i.e., the condition of the specimen may make it hard to assess and score accurately).
For the recognition of braconid subfamilies, see van Achterberg (1990Achterberg ( , 1993Achterberg ( , 1997, for the identification of Aleiodes Wesmael, see van  and Chen and He (1997). For additional references see Yu et al. (2016). For the terminology used in this paper see Figs 1-6 or van Achterberg (1988, 1993; note, however, that in the present work the distance between eye and lateral ocellus is measured differently). An asterisk indicates a new country record according to Yu et al. (2016).

Molecular methods
A molecular dataset of the barcode region of cytochrome oxidase c subunit 1 (CO1) was compiled for a total of 141 Aleiodes specimens and three of Heterogamus ( Fig. 1) which are the sister group of Aleiodes (Zaldivar-Riveron et al. 2008, Quicke et al. unpublished) and were used for rooting the trees. Most of the recent DNA extractions were carried out using normal procedures for 96-well plates (Ivanova et al. 2006), and PCR Figures 1-6. Terminology and measurements used in this paper 1 wing venation: pa = parastigma, pt = pterostigma, 1 = marginal cell, 2a, b, c = 1 st , 2 nd and 3 rd submarginal cell, respectively, 3a, b = 1 st and 2 nd discal cell, respectively, 4a = 1 st subdiscal cell, 5 = basal cell, 6 = subbasal cell 2 head, dorsal aspect: a = length of eye, b = length of temple 3 head, lateral aspect: c = width of temple, d = width of eye, e = height of eye, f = width of malar space (measured as actual true distance in its own plane) 4 head, anterior aspect: g = width of face, h = width of hypoclypeal depression 5 fore femur, lateral aspect: i = length, j = width 6 1 st metasomal tergite, dorsal aspect: k = length of tergite (measured from adductor), l = apical width of tergite. and sequencing reactions were carried out using standard protocols (Hajibabaei et al. 2005). Most sequences were obtained using the LCO-HCO primer pair combination (Folmer et al. 1994: LCO 5'-GGT CAA CAA ATC ATA AAG ATA TTG G-3', HCO 5' -TAA ACT TCA GGG TGA CCA AAA AAT CA-3') or, less often, LepF1-LepR1 (Smith et al. 2005: LepF1 5'-ATT CAA CCA ATC ATA AAG ATA TTG G-3', LepR1 5'-TAA ACT TCT GGA TGT CCA AAA AAT CA-3'). Sequence alignment was carried out manually and was largely trivial as there was no length variation apart from a three base pair deletion uniting most species of the A. risaae Quicke and Butcher species group as previously noted (Butcher et al. 2012) and its precise location determined by reference to amino acid identities and the known codon positions. Sequences were analysed using maximum likelihood with the programme RAxML (v.8) (Stamatakis 2014), using a GTR + G rate model with three data partitions corresponding to the three codon positions. Each analysis comprised 100 replicates with two threads. Trees were visualised using Figtree (1.4.3) (Rambaut 2016).
GenBank accessions numbers are given in Appendix 1. Specimens with an identifier code comprising MRS followed by a number are deposited in NMS, with the exception of the paratype of A. coriaceus (MRS311) which is in RMNH. The specimens indicated CollHH with a number are retained in the personal collection of Håkon Haraldseide (Norway); voucher locations of samples prefixed by BCLDQ are as follows: for Thai specimens depositories are given in Butcher et al. (2012), USA specimens are in University of Wyoming collection; others and that of A. mexicanus (BMNHE897778) are in the collection of the Natural History Museum, London; the voucher of A. cameroniiJanz01 (DHJPAR0021064) is in the collection of Prof. Dan Janzen (Philadelphia); the voucher of A. trianguliscleroma (CCDB27844-E03) is in the collection of Tel Aviv University, Tel Aviv, Israel; the voucher of A. aff. wyomingensis (BIOUG01036-F12) is in the collection of the Center for Biodiversity Genomics, University of Guelph, Canada.

Phylogeny
Three datasets were investigated with different levels of taxonomic and sequence inclusion.
Firstly, we conducted an overview analysis including representatives of a wide range of extra-limital species groups of Aleiodes, single representatives of the species treated in this paper for which molecular data were available (22 of the 42 species), and representatives of other West Palaearctic species groups, with three members of the genus Heterogamus used as outgroups (Fig. 7). In close agreement with the molecular tree presented for Thai Aleiodes (Butcher et al. 2012: fig. 5 loc. cit.), our results show that most of the A. apicalis species group sensu van Achterberg and Shaw (2016) (= Chelonorhogas auctt.) form a grade together with various generally large bodied extralimital species, notable among which are the Oriental and East Palaearctic A. coronarius group which are characterised by having a deep pronope. Immediately basal to this grade are two large bodied species (A. melanopterus (Erichson) and A. mexicanus Cresson) that had previously been included in a separate genus, Eucystomastax Brues, but which Shaw (1993) showed to be a distinctive monophyletic species group of Aleiodes within which he treated them as a subgenus. The sister to all other Aleiodes as recovered in this analysis are a group of species most members of which have males with metasomal tergal glands that open at a single medial subposterior pore on one or more of tergites four to six, although they are lacking, for example, in A. miniatus. This clade includes the West Palaearctic A. fortipes (Reinhard), the Nearctic A. cameronii (Dalla Torre) and some other species within the A. pulchripes group sensu Shaw et al. (1997), and the Palaeotropical subgenus Hemigyroneuron Baker , Delfin and Wharton 2000, Butcher and Quicke 2011. From within this large basal grade emerges, on a relatively long branch, a monophyletic group which includes the vast majority of Aleiodes species. The West Palaearctic and Nearctic members of this clade have been placed in various species groups including the A. bicolor, A. circumscriptus and A. gastritor complexes. However, many species even within the Palaearctic fauna fall outside of these as isolated groups, often more closely related to extralimital taxa.
Secondly, we analysed a matrix comprising the most complete available sequence for each West Palaearctic species and using A. fortipes as the outgroup based on the results of the first analysis (Fig. 8). The A. circumscriptus group in the sense of van Achterberg and Shaw (2016) (including the A. similis and A. gastritor subgroups and the rather isolated A. circumscriptus (Nees) itself, as well as some other species) and the A. bicolor group sensu van Achterberg and Shaw (2016) were recovered together as a monophyletic clade but without strong indication of its comprising two separate groups. Indeed, the previous concept of the A. circumscriptus group was challenged by its paraphyly with respect to the A. bicolor group. The clade comprising the A. bicolor  group and the A. similis subgroup (and a few extraneous species including A. circumscriptus) was not treated as a unit by van Achterberg and Shaw (2016) because of rather clear apparent differences: the clade indicated both in that work and here as the A. bicolor group comprises rather stocky and heavily sculptured species that have various morphological features (such as a long malar space and margined T4) in common and they, and several additional similar species, will be dealt with in Part 3 of this work. The species indicated as the A. similis subgroup and the A. gastritor subgroup will be treated together in Part 4. In practice, these two subgroups are less easy to separate on morphological grounds, and again there are many additional species. Throughout most of this large clade radiation seems to have been relatively recent and/or rapid with many morphologically and biologically clearly distinguishable species often having CO1 sequences that differ from one another by less than 1 %. The remainder of the species outside the A. apicalis grade are rather well characterised and indeed isolated, with little evidence of recent radiation, and the species for which host relations are known are taxon-specialists (cf. Shaw 2003). All of these species (with the exception of spM3) have already been treated in Part 1 of this work (van Achterberg and Shaw 2016).
Thirdly, we constructed a tree for the available barcodes for the species treated in this paper (Fig. 9). Each species represented by multiple sequences is recovered as a monophyletic cluster, mostly with relatively little intraspecific variation. Some of the differences observed are likely due to particular sequences being quite short compared to the full-length barcode, others no doubt due to reading errors particularly for those samples that were sequenced more than ten years ago with different methodologies. Re-examination of existing electropherograms has usually confirmed the generality of this and such reading errors are typically at the 5' or 3' ends of sequence reads. The most notable exception is provided by the three specimens of A. hirtus (Thompson), with the sequence from the British specimen differing at 15 positions (2.5 %) along the 606-base region of overlap with the two Romanian individuals (which were identical). These specimens are briefly discussed in the species entry for A. hirtus.
Notes. According to the 28S + COI analysis by Zaldivar- Riverón et al. (2008) the following former subgenera or genera belong to this group: Chelonorhogas Enderlein, [1 st Sept.] 1912 (worldwide), Eucystomastax Brues, [(end of?) Sept.] 1912 (Neotropical group with 2 nd and 3 rd maxillary palp segments enlarged), Hemigyroneuron Baker, 1917 (Old World group with distal half of subbasal cell of fore wing modified and glabrous), and Dimorphomastax Shenefelt, 1979 (males of this monotypic Neotropical group have a large curved tooth near the base of the mandible (an outgrowth of the condylar carina) and the hind tibial spurs are blunt apically; females have the tooth smaller and triangular, and the hind tibial spurs are acute). Butcher et al. (2012) indicate in their cladogram based on the analysis of COI sequences that A. fortipes (Reinhard) forms together with Hemigyroneuron Baker and Arcaleiodes Chen & He the most basal clade of Aleiodes Wesmael, and it is noteworthy that all known hosts of this clade are Geometridae (see species entry for A. fortipes). According to the same analysis the A. apicalis group consists of three clades: (i) the Hemigyroneuron clade (see above; likely also includes A. caucasicus (Tobias)), (ii) the A. rugulosus clade (including the Asian group with modified pronotum), and (iii) the A. gasterator clade. Since we do not have the COI sequences of all species, we unite these three clades in the A. apicalis group to allow identification based on their morphology.
Head. Antennal segments of ♀ 52, length of antenna 1.1 × fore wing, its subapical segments approx. as long as wide; frons with irregular curved rugae, shiny, and rugose behind antennal sockets; OOL 2.4 × diameter of posterior ocellus, and finely remotely punctate, interspaces much larger than diameter of punctures; vertex spaced punctate, shiny; clypeus short, coarsely and densely punctate; ventral margin of clypeus thick and rather protruding forwards (Fig. 36); width of hypoclypeal depression 0.65 × minimum width of face (Fig. 34); length of eye 0.8 × temple in dorsal view (Fig. 35); vertex behind stemmaticum sparsely punctate; clypeus near lower level of eyes; length of malar space 0.3 × length of eye in lateral view.

Metasoma.
First tergite rather flattened, as long as wide apically; 1 st and 2 nd tergites coarsely and densely rugose, robust, with distinct median carina; medio-basal area of 2 nd tergite wide and short; 2 nd suture deep medially and shallow laterally; basal half of 3 rd tergite finely rugose, remainder of metasoma largely smooth, punctulate; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath with mediumsized setae and apically rounded (Fig. 22).
Description. Paralectotype, ♀, length of fore wing 6.6 mm, of body 7.0 mm. Head. Antennal segments of ♀ 47, antenna as long as body and its subapical segments moderately slender; frons rugose, shiny; OOL 1.3 × diameter of posterior ocellus; OOL and vertex remotely punctate, with satin sheen, OOL also with some rugulae; anterior part of clypeus 9 × wider than high, coarsely punctate and rather convex; clypeus above lower level of eyes; ventral margin of clypeus thick and not protruding forwards (Fig. 48); width of hypoclypeal depression 0.8 × minimum width of face (Fig. 46); length of eye 1.8 × temple in dorsal view; vertex behind stemmaticum convex and sparsely punctate; length of malar space 0.19 × length of eye in lateral view; occipital carina nearly complete, fine and ventrally strongly curved; mandible massive triangular, coarsely punctate and with thick ventral lamella (Fig. 46).

Metasoma.
First tergite robust, as long as wide apically, distinctly narrowed anteriorly and rather flat posteriorly; 1 st and 2 nd tergites finely longitudinally striate-rugulose; medio-longitudinal carina of 1 st and 2 nd tergites indistinct; 2 nd tergite 0.7 × longer than its basal width; medio-basal area of 2 nd tergite wide triangular, rather short; 2 nd suture shallow and narrow; 3 rd tergite mainly smooth and with satin sheen; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath rather slender, with short setae and apically truncate (Fig. 39).
Colour. Black; pronotal side largely yellowish brown; mesoscutum medio-posteriorly and postero-laterally partly chestnut brown; tegulae, clypeus and antenna (except yellow scapus and pedicellus) yellowish brown; mandible, legs and palpi pale yellowish, but hind coxa and most of middle coxa dark brown; metasoma dark brown but with yellow patches (Fig. 38); pterostigma brown medially and dark brown laterally; ovipositor sheath dark brown; veins of fore wing (but pale yellow in basal 0.2 of fore wing) brown; wing membrane hyaline.
Distribution. Armenia, Iran. Included in this revision, because it may occur in Turkey. ( Biology. Time of flight varies according to harshness of summer. In its more temperate sites plurivoltine April-September(October), overwintering in the mummy, but in Cyprus (and presumably other places with extremely hot dry summers) it appears to be most active from autumn to spring (October-May), with a prolonged summer diapause (June-October or later) in the mummy (reared series ex "Plusia" in BMNH and NMS, W.R. Ingram, six with mummification dates recorded in May or June and adult emergence in the following October-December, further specimens in the series have only one date, which is ambiguous). Reared from Noctuidae: Autographa gamma (Linnaeus) (6 [4 ZISP/Moldova, 1 HSC/Germany, 1 NMS/Malta]; J.L. Gregory, H. Schnee), indet. Plusiinae (14). There is no reason to suppose that the hosts recorded as indet. Plusiinae are anything except A. gamma. A further specimen labelled as ex Peribroma [sic] saucia is accompanied by a clearly Plusiinae mummy (Sicily, NMS). Also, one labelled as from Anarsia lineatella Zeller (Gelechiidae) (Ukraine, ZISP), but without a mummy and clearly in error on grounds of size alone. Another specimen labelled as "ex Sesamia pupa" (Iran, BMNH) lacks its mummy but accompanies two individuals of A. aestuosus (q. v.) from the same source, and the remarks made under that species apply also to this record -but with the added objection that the small hypoclypeal opening and flat clypeus of A. apicalis strongly suggest that its hosts do not mummify in deep concealment. The mummy (Fig. 52) is of a pale chalky buff colour, and the cocoon occupies approx. abdominal segments 4-7 of the host larva. Several of the mummies examined, all of which seem to be penultimate instar, have been formed in a more or less curled leaf beneath a web (Fig. 53) that the host had been induced to spin before being mummified, and were weakly stuck to the substrate.
Head. Antennal segments of ♀ more than 40, but apical segments missing (length of antenna of ♀ from Lesbos 1.4 × fore wing and its subapical segments robust); frons with coarse curved rugae, shiny; OOL 1.5 × diameter of posterior ocellus, and distinctly striate; vertex transversely striate, rather weak; clypeus normal, punctulate and convex; ventral margin of clypeus thick and not protruding forwards; width of hypoclypeal depression 0.3 × minimum width of face (Fig. 60); length of eye 1.6 × temple in dorsal view (Fig. 61); vertex behind stemmaticum transversely striate; clypeus near lower level of eyes; length of malar space 0.4 × length of eye in lateral view; occipital carina complete, fine.

Metasoma.
First tergite robust, evenly convex; 1 st and 2 nd tergites rather coarsely obliquely rugose; 1 st tergite and basal half of 2 nd tergite with median carina; 2 nd tergite robust and with striae diverging posteriorly; medio-basal area of 2 nd tergite wide triangular, rather short; 2 nd suture rather deep medially; 3 rd tergite largely smooth, except anteriorly with some striae; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath with rather long setae and apically rounded (Fig. 51).
Colour. Black; scapus, pedicellus, tegulae (but humeral plate brownish yellow), base of hind tibia narrowly, apical half of hind tibia, telotarsi, hind tarsus largely, ventral apical half of metasoma, pterostigma and veins (except C+SC+R of fore wing) dark brown; remainder of basal half of antenna and palpi yellowish brown; remainder of legs (but apical two-fifths of hind femur black), 1 st and 2 nd tergites, 3 rd tergite basally and laterally orange brown; remainder of hind tibia pale yellowish; apex of middle femur and wing membrane somewhat infuscate.
Variation. A. apicalis is very variable in colour and the colour patterns are not restricted to certain areas, but in general southern Palaearctic specimens are darker than northern ones (or specimens from high altitudes). The tegula is dark brown or black, and the humeral plate usually paler than the tegula or equally black, but both usually yellowish in southern specimens; the hind tarsus is dark brown or black, but sometimes 3 rd and 4 th segments yellowish; the hind tibia variably reddish to black, but palest at extreme base; the pronotum is very occasionally reddish. The extent of black colouration of the legs is especially variable, and sometimes all legs are entirely black (var. rufoater (Wollaston, 1858)). Antenna, especially in females, can be more or less light reddish brown, especially basally, or dark brown/black throughout. Antennal segments: Distribution. *Albania, Austria, *Bosnia & Herzegovina, *Bulgaria, *Croatia, Cyprus, *Czech Republic, *France (including Corsica), *Georgia, *Germany, Greece (including Chios, Corfu, Crete, Lesbos, Rhodes), *Hungary, Iran, *Iraq, *Israel, *Italy (including Sardinia, Sicily), *Kazakhstan, *Malta, *Moldova, *Montenegro, *Morocco, *North Macedonia, *Oman, *Portugal (including Madeira), *Romania, *Russia (including Dagestan), *Serbia, *Slovakia, Spain (including Mallorca and Canary Islands: Tenerife, Fuerteventura), *Syria, Switzerland, *Tunisia, Turkey, *Turkmenistan.
New synonymy. The synonymy of Rogas rufo-ater Wollaston, 1858, andRhogas similis Szépligeti, 1903, are based on examination of the types listed above. The lectotype of Rogas bicolor Lucas, 1849(not Spinola, 1808 and of Rhogas bicolorinus Fahringer, 1932, has been examined by Dr Jenö Papp and we agree with his opinion that it is a synonym of A. ductor auct. (= A. apicalis). The types of Rogas reticulator Nees, 1834, and Rhogas reticulator var. atripes Costa, 1884, are lost or unavailable and their synonymy is based on the original description and the interpretation by later authors.

Metasoma.
First tergite rather flattened, as long as wide apically; 1 st and 2 nd tergites coarsely longitudinally and densely rugose, robust and posterior corners of 1 st protruding outside base of 2 nd tergite, with distinct median carina; medio-basal area of 2 nd tergite wide and short; 2 nd suture moderately deep and crenulate; basal half of 3 rd tergite longitudinally striate, remainder of metasoma largely smooth, punctulate; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath wide, setose and apically truncate (Fig. 73).
Distribution. Azerbaijan, *Turkey. Notes. Easily confused with A. ruficornis (Herrich-Schäffer); the relative size of the clypeus (wider and somewhat shorter in A. arnoldii than in A. ruficornis) seems to be the main difference in both sexes. In addition, the female of A. arnoldii has the temple ventrally and the malar space yellowish brown (dark brown in A. ruficornis). The male has darker legs and 1 st metasomal tergite than the female (the sexes more similar in A. ruficornis). Also reported from Uzbekistan (Yuldashev, 2006); the record from Poland (Huflejt, 1997) most likely concerns A. ruficornis (Herrich-Schäffer). Aleiodes arnoldii sensu Farahani et al. (2015) concerns a species closely related to A. gasterator (Jurine) but has basal half of 3 rd tergite coarsely longitudinally rugose, antenna of ♀ with 30-35 segments (of ♂ 36), head linearly narrowed ventrally and subbasal antennal segments of ♀ slightly slenderer.
Before becoming mummified the host moves to a narrow twig, to which the mummy will be very strongly glued. In the early stage of the mummification process ( Fig.  99), in which the anterior end of the host is particularly contracted, the parasitoid larva strongly protrudes anteriorly to spread the necessary glue (Fig. 101). The resultant almost semi-circularly domed and hard mummy (Fig. 100), in which the parasitoid occupies approximately abdominal segments 4-7 of the host, forms in ca May-June and persists through the remainder of the summer and the following winter until the adult emerges in ca April-May. (The univoltine hosts overwinter in the egg stage.) The swollen part of the mummy, which is moderately densely lined with silk, is externally usually matt chalky buff in colour, but dark brown diamond-shaped patches centred dorsally on intersegmental areas tend to remain (Fig. 97), and sometimes (perhaps especially when the mummy is unable to dry as it forms) these are coalesced to leave a single shiny dark brown patch covering most of the dorsal surface. Some of the mummies examined might be of somewhat stunted final instar hosts, but others are more clearly penultimate instar. The outcomes of an experiment involving six females and cultured A. pyramidea larvae were unfortunately marred by unavoidably high temperatures and then disease overcoming the cultures so that no mummies resulted, but the following observations were made: (i) 2 nd to 5 th instar hosts were potentially attractive; (ii) 2 nd instar hosts were, however, often ignored or else tended to be abandoned after being paralysed with a single jab (i.e., without oviposition subsequently taking place); (iii) 3 rd instar hosts were often ignored, but when attacked seemed the most smoothly parasitised, sometimes with a single paralysing jab being followed, after a short pause, by a single insertion of the ovipositor for presumed oviposition, although the pattern observed for 4 th and 5 th instars also occurred with 3 rd instars; (iv) 4 th and (2 only) 5 th instar hosts were embraced the most enthusiastically, but it required several (3-5) injections to subdue them, and then there were usually several (3-4) separate sequential and lengthy (often as long as 80 seconds) insertions of the ovipositor (which may or may not all have been actual ovipositions), the parasitoid turning between insertions and always grasping the host Figure 101. Larva of Aleiodes aterrimus (Ratzeburg) mummifying Amphipyra pyramidea (Linnaeus), with its anterior (indicated by the arrow) projecting from the ventral opening in the host to spread adhesive over a wide area. with all six legs during the insertion; (v) antennation of the host was minimal, and there was no post-oviposition association; (vi) all temporarily paralysed hosts hung from the substrate by one or usually more prolegs until they recovered, presumably preventing their falling from their pabulum; (vii) there is no long-term physiological venom effect. The behaviour of adults observed toward the different instars is intriguing, and the experiment would be well worth repeating under better circumstances.
Although the above is a consistent pattern for this species, it does not account for a small number of specimens (14 ♀, 4♂ in BZL, MRC, MTMA, NMS, SDEI, ZSSM) examined from various localities in central Europe (Czech Republic, Germany, Hungary) and S. Russia. These specimens share small but rather consistent morphological differences from the usual form, in particular tendencies towards: more intense sculpture on the metasomal tergites (T3 being more or less strongly punctate or even rugose-punctate); the hind wing marginal cell parallel-sided in basal three fifths; shorter 3-SR in fore wing; basal cell of fore wing with more, and more evenly distributed, setae; fewer antennal segments; wing membrane slightly brownish. These differences are not absolutely consistent and would be easy to let pass without comment were it not for the fact that they are correlated with an apparently different phenology, as (of the ten specimens with dates recorded) five ♀ were collected in July and one in August, with only three ♀ in May and one in June (none in April). This is in marked contrast with the earlier flight time of the usual form, and the usual hosts (arboreal Amphipyra spp.) are not available after early June. A further ♀ specimen (MTMA) examined and returned in 1997 by MRS but apparently no longer in the main MTMA collection was labelled "Hungaria, Fót, Somlyó-hegy, 30.vii.1958, Ehik"; "Ex Panchrysia deaurata Esp [J. Papp's handscript]"; "ex Pytometra deautate [sic]". Unfortunately, no mummy had been preserved, but this plusiine noctuid feeds on Thalictrum (a low plant, not a tree) and it is unlikely for an arboreal Amphipyra, even if fallen from a tree above, to have been mistaken for it. The date, whether referring to collection of the host larva or emergence of the adult parasitoid, is also out of step with arboreal Amphipyra species. We considered but rejected the possibility that these specimens belong to a separate species, and instead conclude that under certain circumstances A. aterrimus can have a partial 2 nd brood (in the southern part of its range) which uses different hosts, and that the morphological variation is merely seasonal. The material (which does not conform to A. sapporensis (Watanabe), see below) is being returned to holding institutions determined as A. aterrimus but with "var: T3 sculpture etc." appended to facilitate recall if necessary. It should be added that this form has (on account of its heavy metasomal sculpture and extensively parallel-sided marginal cell in the hind wing) sometimes been misidentified as A. rugulosus, but the two species are always easily separated by the sculpture of the mesoscutum and scutellum, as well as by leg colour.
Dr K. Samartsev (in litt.) kindly brought to the first author's attention that the East Palaearctic A. sapporensis (Watanabe, 1937) occurs in southern European Russia (Middle and Lower Volga territories). Aleiodes aterrimus and A. sapporensis differ only slightly, mainly by the colour of the extreme base of the hind tibia (completely dark brown in A. sapporensis and usually narrowly pale yellowish in A. aterrimus) and by the shape of temple in dorsal view (roundly narrowed in A. sapporensis and rather linearly narrowed in A. aterrimus). There is also a slight difference in the proportions of the face (A. sapporensis has facial width 1.50-1.60 × medial height including clypeus and A. aterrimus 1.65-1.75 ×). A. sapporensis seems to have the lateral carinae of propodeum more protruding and has 58-66 antennal segments.
Metasoma. First tergite evenly convex, as long as wide apically; 1 st and 2 nd tergites with medio-longitudinal carina and coarsely longitudinally rugose, but posterior quarter of 2 nd tergite irregularly rugose and no median carina; medio-basal area of 2 nd tergite triangular and rather distinct (Fig. 106); 2 nd suture deep and narrow; basal half of 3 rd tergite finely punctate-rugose, remainder of metasoma superficially micro-sculptured; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath wide, with long setae and apically truncate (Fig. 95).
Distribution. Austria, Czech Republic, Hungary, *Russia (Lake Baikal). Notes. Very similar to A. grassator (Thunberg), and especially A. carbonaroides; males of A. carbonarius and carbonaroides are normally black but males with partly orange brown metasoma occur. The three species exhibit sexual dimorphism of the 2 nd submarginal cell (less robust (and also longer in A. carbonarius) in female than in male). Giraud (1857) gave an incomplete description of the only two males he possessed, but clearly indicated that the antenna is slightly shorter than the body. The female of this species is reported for the first time.  (Morley, 1937, misidentified as A. carbonarius). Reared from the grass-feeding noctuid Cerapteryx graminis (Linnaeus) (6 [2 are RMNH]; K.P. Carl/Netherlands). If it is a specialist, it is presumably univoltine and overwinters in the mummy (the univoltine known host overwinters in the egg stage). Mummy similar to that of the closely related A. carbonarius and A. grassator, but slightly smaller.
Metasoma. First tergite rather flattened, 0.7 × as long as wide apically; 1 st and 2 nd tergites with medio-longitudinal carina and coarsely longitudinally rugose, but posterior quarter of 2 nd tergite without medio-longitudinal carina; medio-basal area of 2 nd tergite triangular and short; 2 nd suture deep and crenulate; basal half of 3 rd tergite finely longitudinally rugose, remainder of metasoma superficially micro-sculptured; 4 th and apical third of 3 rd tergite without sharp lateral crease; ovipositor sheath wide, with long setae and apically truncate (Fig. 139).
Distribution. Germany, Netherlands, U.K. Etymology. The suffix "-oides" indicates similar to; in this case the high similarity to A. carbonarius Giraud.
Variation. Head black or mainly dark brown, specimen from Bulgaria also anterior half of mesosoma; antenna of ♀ with 38 or 41 segments according to the original description; 11 th and 12 th antennal segments of ♀ dark brown or brownish yellow; hind femur 3.6-3.8 × as long as wide. The male is unknown, or possibly has not been distinguished from that of A. fortipes.
Distribution. *Bulgaria, Russia (SW). Notes. It remains unclear whether this predominantly rather yellowish orange species is distinct from A. fortipes, which in its more western localities is a much darker insect. Females intermediate in colour (and included in A. fortipes) seem to predominate in eastern Europe. More material (preferably with biological data) is needed to clarify the status of A. caucasicus. Biology. Unknown. The available specimens were collected in July, and it is almost certainly univoltine, but we have not seen reared material.
Legs. Tarsal claws with rather conspicuous and medium-sized dark brown pecten (Fig. 189); hind coxa (except depression) coriaceous and with some rugulae dorsally;  hind trochantellus robust and with long setae; length of hind femur and basitarsus 4.5 and 5.8 × their width, respectively; length of inner hind spur 0.5 × hind basitarsus.
Metasoma. First tergite convex and basally rather steep, as long as wide apically; 1 st and 2 nd tergites with medio-longitudinal carina and longitudinally rugose; maximum width of 2 nd tergite 1.5 × its median length; medio-basal area of 2 nd tergite mediumsized triangular and rather short (Fig. 183); 2 nd suture distinct and moderately crenulate; basal half of 3 rd tergite finely rugulose and rugulae diverging posteriorly, remainder of metasoma nearly smooth; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath wide, with long setae and apically truncate (Fig. 179).
Distribution. Sweden. Etymology. Coriaceus is Latin for leathery, because of the coriaceous sculpture of vertex and mesoscutum.
Biology. Probably univoltine, certainly overwintering as a mummy. Collected June-August, often at light and including around Dianthus barbatus harbouring larvae of the noctuid Hadena confusa (Hufnagel) (H. Schnee/Germany). In Austria it has been collected up to 2000 m. Only one reared specimen seen, from H. confusa [FMNH], the adult emerging in June in the year following host mummification. Extensive rearings of this host in various parts of Britain in recent years by one of us (MRS) has not produced A. cruentus, strengthening the view that it does not occur in Britain. The predominantly dark mummy seen (Fig. 198) is stout, rather short and weakly swollen dorsally, and has a paler and moderately strong lateral keel. The cocoon is substantially silk-lined and occupies most of the host's abdomen (approx. 2 nd -7 th abdominal segments). The mummy probably forms underground, albeit from penultimate instar hosts, and the somewhat reflexed and sideways twisted head suggests that it is not or scarcely stuck down; the caudal segments are also somewhat recurved ventrally. Although oviposition has not been witnessed, the somewhat laterally compressed apex of the female's metasoma appears to be an adaptation for attacking the host at rest or feeding within the seed capsules of its food plants (Dianthus, Silene, etc.). lobes of mesoscutum densely finely punctate, with interspaces approx. equal to diameter of punctures; precoxal area with some rugae medially; vein cu-a of fore wing vertical; surroundings of veins M+CU1 and 1-+2-CU1 largely glabrous; vein r of fore wing 0.3-0.4 × vein 3-SR (Fig. 199); vein 1-CU1 of fore wing 0.8-1.1 × vein 2-CU1 (Fig. 199), rarely shorter; hind tarsal claws with conspicuous dark brown pecten (Fig. 205); 1 st tergite widened apically; 2 nd tergite 0.7-0.9 × as long as wide (Fig. 202), its colour variable, often reddish; head black; vein 1-M of fore wing brownish; wing membrane subhyaline; 4 th -6 th tergites of ♂ with long setae, but flattened and narrowly glabrous medially.
Metasoma. First tergite rather flattened, as long as wide apically; 1 st and 2 nd tergites with medio-longitudinal carina and largely coarsely longitudinally rugose, but posterior quarter of 2 nd tergite irregularly rugose and no median carina; medio-basal area of 2 nd tergite triangular and rather distinct (Fig. 202); 2 nd suture deep medially, shallow laterally and crenulate; 2 nd tergite 0.7 × as long as wide (Fig. 202); anterior 0.7 of 3 rd tergite densely and finely punctate, remainder of metasoma largely smooth; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath wide, with rather long setae and apically rather rounded (Fig. 197).
Colour. Black; posterior half of mesoscutum, scutellum largely, apical rim of 1 st tergite and basal rim of 2 nd tergite reddish brown; fore coxa, bases of middle and hind coxae blackish; apex of hind tibia, telotarsi, hind tarsus, palpi, veins and pterostigma dark brown; tegulae and remainder of hind tibia pale yellowish; remainder of legs reddish brown; wing membrane subhyaline.
Notes. An examined female (NMS) from Albania (Mt Mali me Gropa, above Shengiergi, 1400 m, 13.viii.2019, MV light, C.W. Plant) has a CO1 sequence (MRS940) 3 % different from A. cruentus (19 differences in 626 bp of overlap) and although superficially similar in colour is clearly distinct in having OOL shorter (0.5 × lateral ocellus), a smaller hypoclypeal depression (0.5 × width of face), slenderer hind femur (5 × as long as wide), and several other differences. It may be A. parvicauda (Tobias, 1985) described from Afghanistan, but it has more (64; 58-60 in type series) and somewhat more elongate antennal segments than described for A. parvicauda, as well as other small deviations. Additional material as well as comparison with the type series of A. parvicauda are needed to settle the status of the Albanian species.
Variation. Length of body 7.0-8.2 mm, of fore wing 7.5-7.9 mm; temple punctate to smooth; precoxal sulcus area finely to rather coarsely rugose; pronotal side largely black (except ventrally) black or brownish yellow; lateral lobes of mesoscutum entirely dark chestnut brown or only posteriorly so, or mesoscutum largely yellowish brown posteriorly and prolonged to base of notauli; first tergite usually entirely brownish yellow, but sometimes dark brown and only posteriorly and laterally yellowish; pterostigma dark brown or brown. Antennal segments: ♀ 63(1).
Distribution. Turkmenistan, Uzbekistan. Notes. We have included this extralimital species from Central Asia because we suspect it may occur in Turkey. It should not be confused with Rogas aestuosus var. desertus , described from China in the same paper. The latter is an unavailable name (a primary homonym) and most likely a colour variety of R. aestuosus.
The moderately large hypoclypeal opening and protruding sharp-rimmed clypeus of A. dissector is seen in some other species (e.g., A. modestus (Reinhard), treated in part 1 of this work) whose hosts also pupate in shallow soil. In culture experiments A. dissector was found to prefer hosts in the early to middle part of the 3 rd instar, although late 2 nd instar host were often also acceptable. Oviposition into suitable hosts was rapid (1-2 seconds) and accomplished with a single insertion of the ovipositor, following only brief antennation and no use of the legs. There was no clear temporary paralysis. Experimental rearings from O. incerta (6:107\85\\75+10) and O. gothica (6:61\49\\34+15) were comparable (given that some insertions of less than a full second might have been scored as ovipositions incorrectly; and furthermore that some failures to oviposit into these hosts might be ascribed to temporary egg depletion, as the protocol of normally ceasing Diagnosis. Maximum width of hypoclypeal depression 0.6-0.7 × minimum width of face (Fig. 251); OOL of ♀ 0.6-0.7 × diameter of posterior ocellus (Fig. 252) and sparsely punctate; ventral margin of anterior part of clypeus comparatively sharp and more or less protruding outwards (Fig. 253); length of malar space 0.2 × length of eye in lateral view (Fig. 253); head transverse in dorsal view and eye 1.5-2.0 × as long as temple in dorsal view (Fig. 252); lobes of mesoscutum punctulate, with interspaces smooth to superficially micro-sculptured; precoxal area completely smooth or nearly so; vein 1-CU1 of fore wing 0.2-0.3 × vein 2-CU1 and horizontal (Fig. 243); hind tarsal claws with conspicuous dark brown pecten close to apical tooth (Fig. 250); 1 st tergite rounded antero-laterally and 1.0-1.1 × as long as wide apically; basal half of metasoma black and weakly sculptured; 3 rd tergite smooth; head black; palpi yellowish; basal half of hind tibia pale yellowish, but in some males almost uniformly dark; 4 th -6 th tergites of males depressed medially and conspicuously setose (Fig. 258).
Metasoma. First tergite flattened, basally narrowed, as long as wide apically; 1 st and 2 nd tergites with medio-longitudinal carina and largely finely punctate-rugose, but posterior quarter of 2 nd tergite irregularly rugose and no median carina; medio-basal area of 2 nd tergite wide and triangular, distinct (Fig. 247); 2 nd suture rather deep and micro-sculptured; 3 rd and subsequent tergites largely smooth; apical half of 3 rd and 4 th tergites without sharp lateral crease; ovipositor sheath wide, with long and mediumsized setae and apically truncate (Fig. 241).
Colour. Black; apical half of hind tibia and hind tarsus blackish; basal half of hind tibia pale yellowish; remainder of legs, palpi and tegulae yellowish brown; most veins and pterostigma dark brown; wing membrane slightly yellowish basally and remainder slightly infuscate.
Notes. Close to A. cruentus which, however, almost always has much or all of 1 st and 2 nd metasomal tergites orange-red (usually wholly black or dark brown in A. diversus). In addition to characters given in the key A. diversus is a more robust insect, and females have broader antennal segments (distinctly transverse near middle of flagellum) and on average they are fewer in number (although with overlap).
Diagnosis. Maximum width of hypoclypeal depression 0.5-0.6 × minimum width of face (Fig. 298); OOL of ♀ approx. as long as diameter of posterior ocellus (Fig. 299) and densely rugose; clypeus rather thin apically and rather protruding anteriorly (Fig.  300); eyes prominent (Fig. 299); lobes of mesoscutum distinctly punctate-granulate and rather matt; precoxal area more or less rugose and comparatively wide medially, and posteriorly punctate; vein 1-CU1 of fore wing 0.3 × vein 2-CU1 (Fig. 292); hind tarsal claws slender, brownish setose and without pecten (Fig. 302); basal half of 3 rd tergite striate; 3 rd antennal segment of ♀ dark brown; basal half of hind tibia pale yellowish or ivory, at least inner side contrasting with reddish or dark brown colour of basal half of hind femur (usually less pronounced in ♂). Head. Antennal segments of ♀ 51 remaining, but apical segments missing, length of antenna 1.2 × fore wing; frons with coarse curved rugae and dorsally coarsely rugose; OOL equal to diameter of posterior ocellus, and densely rugose; vertex spaced rugose, rather dull; clypeus medium-sized and coarsely rugose (as face); ventral margin of clypeus rather thin and rather protruding forwards (Fig. 300); width of hypoclypeal depression 0.5 × minimum width of face (Fig. 298); length of malar space in anterior view 0.7-1.0 × maximum width of hypoclypeal depression (Fig. 298); head in anterior view trapezoid; length of eye 1.3 × temple in dorsal view and temples directly narrowed behind eyes (Fig. 299); vertex behind stemmaticum densely rugose; clypeus near lower level of eyes; length of malar space 0.4 × length of eye in lateral view.
Colour. Yellowish brown; antenna (except dark brown scapus and pedicellus), stemmaticum and ovipositor sheath black; tarsi, medio-posterior patch of propodeum, basal patch of 1 st tergite and apex of hind tibia dark brown; veins rather dark brown at medial third of fore wing, remainder of veins pale brown or yellowish; pterostigma brownish yellow; wing membrane subhyaline.
Notes. This Asian species is included here because it was reported from Poland (Huflejt, 1997). The record needs confirmation to rule out confusion with a similar European species.

Aleiodes fortipes (Reinhard, 1863) Figs 322-342
Rogas fortipes Reinhard, 1863: 272;Shenefelt, : 1229. Aleiodes (Neorhogas) fortipes; : 158, 1987a: 333, 1987b: 35, 1991aPapp, 2005: 176;Lozan et al. 2010: 17;Butcher et al., 2012: 14. Rhogas freyi Hellén, 1927Papp, 1985a: 158 (unnecessary Shenefelt, : 1229Shenefelt, -1230Tobias, 1986: 75 (transl.: 121). Biology. The flight time of this univoltine species is (April)May-June, and ca 10 months of the year is spent as an exposed mummy. The only mummy seen (Fig. 324) formed in captivity firmly attached beneath a thin stem and would have been positioned low down in the vegetation, but probably aerially. It is light brown, moderately slender, and the parasitoid occupied approximately abdominal segments 3-8. The host was Idaea sp. (Geometridae), either I. aversata (Linnaeus) or I. straminata (Borkhausen), and the rearing arose when a few larvae of the foregoing were collected (MRS) along with an adult female of A. fortipes at the same site in Poland (22.v.2016) and offered to the parasitoid, which had been fed honey water, on 24.v.2016. Although two of the caterpillars were well-grown, in their final instars and at least twice as long as the parasitoid, one was accepted avidly. This host was first pricked several times, at intervals. Paralysis was rather slow to take effect and not complete until after the host was revisited for oviposition: a single insertion of ca 30 seconds duration, with no postoviposition association (the parasitoid simply walked away after oviposition). The host mummified on 9.vi.2016 and an adult female emerged on 22.v.2017. The other host was rejected after being pricked just once, and later died. A penultimate instar caterpillar of the same host aggregate was also parasitised but died after an ecdysis. Subsequent barcoding (through the kindness of Axel Hausmann, ZSSM) of the dead caterpillars revealed one specimen each of I. aversata and I. straminata, leaving the precise determination of the successful host unclear. It is possible that the parasitised host had already been attacked before it was collected, but the rather long time before mummification occurred suggests not. In any case, at least one Idaea species in the aversata/ straminata group clearly serves as host. Some individuals of the long and slender, morphologically very different, larvae of Idaea muricata (Hufnagel) were also offered. Although possibly of less interest to the parasitoid, one penultimate instar larva (1.7 times the length of the female parasitoid) was immediately parasitised (a single prick for eventual paralysis, followed after an interval by a single insertion for oviposition lasting just more than a minute), but this larva later produced a moth. Final instars of this very elongate species of caterpillar were generally ignored, but one did elicit a downwards curl of the metasoma without, however, being stung.
There are two particularly significant aspects to the successful rearing. The first is that these Idaea species overwinter as quite well-grown larvae, so during the flight period of the parasitoid they are in late instars, and attacking hosts at this stage is an unusual strategy for Aleiodes (but see A. aterrimus and A. sibiricus). The second is that we know of no other Aleiodes species apart from A. sibiricus (q. v.) among those whose host overwinters as a larva that fails to take advantage of that to overwinter as an early instar larva within it. The apparently riskier strategy taken by A. fortipes, in both respects, may be plesiomorphic.
Aleiodes fortipes is the only known West Palaearctic species in which males have small, subapical setose pore (probably associated with pheromone release) situated middorsally on each of the 4 th -6 th metasomal tergites (Fig. 340). We also expect these pores to be present in A. caucasicus, which is only doubtfully distinct from A. fortipes, but we have not seen the male of A. caucasicus. Similar, probably homologous, pores are also a feature of males of Aleiodes (Hemigyroneuron) species which are found in the near East, Oriental and Afrotropical regions (Butcher & Quicke, 2015). Outside of Hemigyroneuron, metasomal pores are also found the New World Aleiodes cameronii (Dalla Torre) species complex and in a number of undescribed Madagascan Aleiodes. In Hemigyroneuron the pores have been shown to connect with large sub-tergal glands (Butcher & Quicke, 2011). Collectively these taxa form a basal clade in our molecular phylogeny (Fig. 1).
Aleiodes fortipes is the only species among those treated in this part of our revision with known hosts outside the Noctuidae and, although no host is known for rather a lot of these species, the apparently basal position of A. fortipes in the group is noteworthy and using geometrid hosts may also be plesiomorphic. The rather slender ovipositor sheath (Fig. 322) is another indication for its basal position. It is interesting that the known hosts of both A. fortipes and of the subgenus Hemigyroneuron are all Geometridae (two species of Hemigyroneuron with examined mummies, India and S. Africa, cited by Butcher & Quicke, 2011 [a label record indicating a pierid host of a 3 rd species is also cited in that paper but is discounted here because no mummy was present]). An Australian species described under Hemigyroneuron with examined mummy reported to be that of a geometrid by Butcher & Quicke (2016) is probably (a) actually not a member of A. (Hemigyroneuron) and (b) may be from a lasiocampid (W. Moore in litt.). The hosts of members of the Aleiodes cameronii complex, based on multiple rearings in both North America and Costa Rica include both Geometridae and Erebidae (Eiseman & Charney, 2010; http://v4.boldsystems.org).
Metasoma. First tergite evenly convex, 1.3 × as long as wide apically; 1 st and 2 nd tergites with weak medio-longitudinal carina and together with basal half of 3 rd tergite densely and finely longitudinally rugose; medio-basal area of 2 nd tergite narrow but rather distinct (Fig. 329); 2 nd suture rather shallow; remainder of metasoma superficially micro-sculptured; 4 th and apical half of 3 rd tergite without sharp lateral crease; 4 th -6 th tergites with a setose medio-dorsal depression.
Colour. Dark brown or blackish; palpi dark brown; mesopleuron with reddish brown streak; legs yellowish brown but tarsi, apex of hind femur (and indistinctly api- ces of fore and middle femora, and of tibiae) and base of hind coxa infuscate; tegulae and pterostigma brown; wing membrane slightly infuscate.
Rogas gasterator; Shenefelt, : 1230Shenefelt, -1231Zaykov, 1980a: 112. Rogas (Rogas) gasterator; Tobias, : 86, 1986 (Figs 345,346), though it may be that neither overwintered; a small, pale and relatively slender one from Agrotis sp. produced a small male, while a more normal sized individual emerged from the larger, dark and stout mummy whose host was (even) less certain. Even in the latter case, the mummy is less keeled, less lined with silk and much more in relation to the size of the emerging adult than is the situation with the overwintering mummies of the A. grassator/ carbonarius/ carbonaroides/ ruficornis complex. The appearance of both mummies suggest that they would normally form below ground.
New synonymy. The new synonymy of Rhogas dimidiatus var. turkestanicus , is based on direct comparison of the types of both taxa. The identity of Bracon dimidiatus Spinola, 1808, is problematic because the holotype from Italy (Genoa) is lost and the original description is far too incomplete for an easy identification. Its colour pattern (head completely yellowish, hind tibia and 3 rd tergite reddish) does not fit with A. ruficornis (Herrich-Schäffer); if the head is largely reddish brown then the temple ventrally and malar space remain blackish. This pattern agrees better with that of pale specimens of A. gasterator (named as A. dimidiatus var. turkestanicus . Aleiodes ruficornis occurs also in Italy, but its females have the head partly black ventrally, the apex of the hind tibia dark brown and most of the 3 rd metasomal tergite black. Therefore, we synonymise Bracon dimidiatus with A. gasterator (syn. nov.). The holotype of Bracon gasterator Jurine, 1807, has the 3 rd metasomal tergite finely curved (nearly circular) aciculate or striate basally, palpi (as far as present) pale brownish, maximum with of hypoclypeal depression 0.45 × minimum width of face, vein 1-CU1 of fore wing half as long as vein 2-CU1 and 4 th antennal segment 1.3 × as long as wide. Aleiodes arnoldii sensu Farahani et al. (2015) concerns a species closely related to A. gasterator (Jurine) having basal half of 3 rd tergite coarsely longitudinally rugose, antenna of ♀ with 30-35 segments (of ♂ 36), head linearly narrowed ventrally in anterior view and subbasal antennal segments of ♀ slightly slenderer.
Mesosoma. Mesoscutal lobes moderately punctate, laterally interspaces mainly smooth, medially superficially granulate and rather shiny; precoxal area of mesopleuron coarsely rugose medially and largely smooth posteriorly; remainder of mesopleuron mainly punctate, but dorsally coarsely rugose; scutellum flat, sparsely finely punctate and only anteriorly with lateral carina; propodeum coarsely rugose, medio-longitudinal carina present on anterior half, rounded posteriorly and dorsal part approx. as long as posterior part.
Colour. Orange brown; apical two thirds of antenna, labial palp, patch on hind femur dorso-apically, posterior patch of 2 nd tergite and telotarsi, dark brown; head, mesosoma (except side of pronotum postero-dorsally and pair of latero-posterior patches of propodeum), 3 rd -7 th tergites (except antero-lateral corners of 3 rd tergite) black; maxillary palp, basal third of antenna, tegulae and remainder of legs rather pale yellowish brown; veins and pterostigma dark brown; wings distinctly infuscate but hind wing less than fore wing.
New synonymy. The synonymy of Rogas alpinus Thomson, 1892, with Aleiodes grassator (Thunberg, 1822) is based on direct comparison of the types listed above.
Notes. Although males of A. carbonaroides are generally easily distinguished from A. grassator through being black, it is possible that lighter forms occur which would be difficult to recognise. Also, females of A. carbonaroides are similar in colour to those of A. grassator. Therefore, specimens collected at low altitude away from northern areas that appear, on other characters, to be A. grassator might well really be A. carbonariodes. See also remarks under A. carbonarius and A. ruficornis.
Description. Lectotype, ♀, length of hind wing 1.7 mm (fore wing missing, but in other specimens ca one-third longer than hind wing and 2.2 mm, brachypterous), of body 7.8 mm.
Distribution. Morocco, Tunisia. Notes. Marshall (1897) based his description on three females from Tunisia. Papp (2003) listed a female in BMNH as lectotype, but this was not accepted as a designation by Belokobylskij & Kula (2012). Therefore, the redescribed female above is here designated formally as lectotype and is the same specimen intended to become lectotype by Papp (2003).
Metasoma. First tergite evenly convex, approx. as long as wide apically; 1 st and 2 nd tergites with medio-longitudinal carina and densely longitudinally rugose, but posterior quarter of 2 nd tergite irregularly rugose and no median carina; medio-basal area of 2 nd tergite minute (Fig. 415); 2 nd suture deep and moderately crenulate; basal half of 3 rd tergite finely rugose, remainder of metasoma largely smooth; 4 th and apical half of 3 rd tergite without sharp lateral crease; 4 th -6 th tergites with long setae and flat.
Colour. Black; legs (except black coxae, trochanters and trochantelli, 1 st and 2 nd metasomal tergites (but base of 1 st tergite partly infuscate) and base of 3 rd tergite orange brown; vaguely near base of femora, telotarsi, apex of hind femur, apical half of hind tibia, hind tarsus largely black or blackish; basal half of hind tibia pale yellow; palpi, tegulae, veins and pterostigma dark brown; wing membrane slightly infuscate. Variation. Hind femur varies from apically black to entirely orange. Propodeum can be partly orangish in posterior part. Usually both sexes have hind trochanter (often also trochantellus) more or less infuscate and darker than the orange part of the hind femur, but this is scarcely evident in a series from S. Russia (MRC, NMS). Hind coxa varies from orange to black. Female is similar to the more distinctive male but is less conspicuously setose (Figs 419-421) and its ovipositor sheath is wide, with long setae and apically truncate (Fig. 410). Precoxal sulcus smooth to superficially rugulose medially. A female and a male from Romania (NMS) are slightly different from British ones; ocelli approx. 1/5 larger and frons coarsely rugose posteriorly. This appears to be reflected by a small divergence in CO1 (2.75 %), but for the moment we treat them as belonging to A. hirtus. Antennal segments of ♀ 54(1), 55(2), 56(6), 57(6), 59(2), 60(2), 61(1), of ♂ 56(1), 58(1), 59(3), 60(3), 61(1). Apical tergites of male type (1-)2, setae rather sparse but long and glabrous stripe consequently not always evident and fringe present but poorly differentiated. A female from Mongolia (BZL) with completely black hind femur and base of hind tibia pale yellowish and with dark basal ring may be another very similar species.

Aleiodes improvisus van
Description. Holotype, ♀, length of fore wing 5.5 mm, of body 7.7 mm. Head. Antennal segments of ♀ 41, antenna as long as fore wing, its subbasal and subapical segments rather robust (Fig. 444); frons with curved rugae; OOL 1.8 × diameter of posterior ocellus, densely rugose and with satin sheen; vertex densely rugose (also behind stemmaticum), with satin sheen; clypeus transversely rugulose; ventral margin of clypeus thick and not protruding forwards (Fig. 441); width of hypoclypeal depression 0.4 × minimum width of face (Fig. 439); eye as long as temple in dorsal view (Fig. 440); clypeus below lower level of eyes; length of malar space 0.55 × length of eye in lateral view; temple striate near eye, and remainder rugose; head with long setae.
Colour. Black; antenna (but only scapus partly yellowish), right fore coxa, trochanter, trochantellus, and femur (but left all yellowish brown except dark base of coxa and infuscated apex of femur), middle femur dorso-apically, middle coxa basally, hind trochanter, trochantellus and femur (but dorso-basally yellowish and left femur also ventrally), apical third of hind tibia (but left tibia yellowish ventrally), tegulae, pterostigma, veins largely, and metasoma ventrally largely dark brown; dorsal part of scutellum, 1 st tergite laterally and narrowly medially and posteriorly, 2 nd tergite and antero-laterally 3 rd tergite orange brown; right fore tibia (except basally and left one yellowish brown) and tarsi more or less infuscate (but left fore tarsus only telotarsus dark brown); fore wing membrane somewhat infuscate, but hind wing nearly subhyaline.
Variation. Eye of ♀ as long as temple in dorsal view (of ♂ 1.0-1.4 ×); length of malar space 0.5-0.6 × length of eye in lateral view; palpi black or largely dark brown; 1-CU1 0.3-0.6 × 2-CU1; length of fore wing 4.0-6.5 mm. Antennal segments: ♀ 41(1); ♂ 44(2), 49(2), 51(2). Male often has much darker legs (largely dark brown with coxae black as right legs of holotype, but legs are more extensively orange, including basal half  of hind femur, in two paratypes) than female and scutellum black; metasoma similarly sculptured and coloured or also basal half of 3 rd tergite orange brown or 1 st tergite only posteriorly orange or only 2 nd and 3 rd tergites (except posteriorly) dark reddish brown; in the largest male paratype (Winterstallen) traces of inwardly curved sculpture are discernible posteriorly on the almost completely longitudinally rugose 3 rd tergite; marginal cell of fore wing similar to ♀, with apical tergites type 1 and fringe not observed (Figs 450, 452).
Distribution. Austria, Switzerland. Etymology. Improvisus is Latin for unexpected, unforeseen, because at first sight the specimens were expected to belong to A. gasterator or A. ruficornis.
Notes. As suggested by its name this species can be easily confused with A. gasterator or A. ruficornis. It differs from A. gasterator mainly by being darker (subbasal antennal segments of ♀, hind trochanter and trochantellus, inner and dorsal side of hind tibia, parastigma) and somewhat higher number of antennal segments of ♀ (41 vs 29-39). Aleiodes ruficornis has an inflated fore femur (hardly or not inflated in A. improvisus), antenna of ♀ medium-sized (1.0-1.2 × fore wing vs 0.8-0.9 ×) and head of ♀ at least partly reddish brown.
Colour. Black; orbit near ocelli reddish brown; fore and middle legs (except blackish or dark brown coxae, trochanters and trochantelli), apex of hind trochantellus and basal third of hind tibia brownish yellow; tarsi darkened and remainder of legs dark brown; palpi (except basally) pale yellowish; mandible yellowish but basally largely dark brown; propleuron and tegula anteriorly dark brown and tegula posteriorly brown; 1 st and 2 nd metasomal tergites and metasoma ventrally (except apically) orange brown; pterostigma dark brown; veins brown; wing membrane subhyaline.
Variation. Orbit near hind ocellus sometimes only very slightly lighter in colour. Antennal segments: ♂ 68(1); according to the original descriptions of A. krulikowskii and A. csikii, the female types have 60 and 62 antennal segments, respectively. Apical tergites of ♂ type 2 and no fringe observed.
Molecular data. MRS950 (Sweden), MRS951 (Sweden). Biology. Unknown, but it seems to inhabit herb-rich calcareous steppe grasslands. Collected in (May)June-August; presumably univoltine, but we have not examined reared material of this large and distinctive species and there is no indication of how it may overwinter. A series in BZL (one now in NMS) is labelled "Wien D. Au" which can be interpreted as [? wet] woodland near the Danube (M. Schwarz, pers. comm.), which would probably be well under 200 m a.s.l. In contrast, a recent specimen (also in BZL) from Kyrgyzstan was collected higher at 2550 m.
Diagnosis. Maximum width of hypoclypeal depression approx. 0.5 × minimum width of face (Fig. 469); OOL of ♀ approx. twice as long as diameter of posterior ocellus and punctate (Fig. 477); ventral margin of clypeus thin and distinctly protruding in lateral view; length of malar space approx. equal to height of eye in lateral view (Fig.  478); mesoscutal lobes densely punctate; area of precoxal sulcus wide and coarsely rugose; length of vein 1-CU1 of fore wing 0.4 × vein 2-CU1; 2 nd submarginal cell of fore wing short and square (Fig. 469); vein 1r-m of hind wing longer than vein 1-M; vein 2-SC+R of hind wing subquadrate; 3 rd tergite densely punctate (Fig. 473); head and mesoscutum orange or brownish yellow; basal half of hind tibia (largely) pale yellowish; metasoma (except part of 1 st tergite) orange or brownish yellow.
Head. Antennal segments of ♀ 65, length of antenna 1.1 × fore wing, its subapical segments somewhat longer than wide; frons with coarse curved rugae; OOL 2.3 × diameter of posterior ocellus, and punctate; vertex densely punctate and shiny; clypeus densely punctate; ventral margin of clypeus thin and distinctly protruding forwards (Fig. 478); width of hypoclypeal depression 0.5 × minimum width of face (Fig. 476); length of eye as long as temple in dorsal view (Fig.477); vertex behind stemmaticum densely punctate; clypeus just below lower level of eyes; malar space 0.5 × length of eye in lateral view.
Diagnosis. Maximum width of hypoclypeal depression 0.5-0.6 × minimum width of face; OOL of ♀ 0.8 × (of ♂ 0.9 ×) diameter of posterior ocellus and rugose; ventral margin of anterior part of clypeus comparatively sharp and more or less protruding in lateral view (Fig. 493); head rather transverse (Fig. 492); mesoscutal lobes coriaceous; precoxal area of mesopleuron rugose medially; vein 1-CU1 0.3-0.4 × vein 2-CU1 and 0.3 × vein m-cu; hind tarsal claws with conspicuous and robust brownish pecten (Figs 494, 503); posterior orbit black; pterostigma of ♀ pale brown medially, of ♂ dark brown; coxae and femora completely black or dark brown; hind tibia usually ivory or pale yellowish basally; 1 st and 2 nd metasomal tergites of both sexes black. According to Papp (1985) most closely related to A. sibiricus (Kokujev), but that species does not have all black females and has the shape of the clypeus different. According to the original description the pterostigma is yellowish and laterally darkened, 1 st subdiscal cell of the fore wing rather short, because vein cu-a distinctly more postfurcal than its own length (ca twice its own length) and meaning vein 1-CU1 of intermediate [approx. 0.6 ×] length of 2-CU1 and 0.8 × vein m-cu] [= "discoidali posterior brevior" as indicated for A. pallidicornis], precoxal sulcus area rugose medially, 1 st tergite twice wider posteriorly than basally, hind leg black, except pale yellowish dorso-basal area of hind tibia and palpi dark brown. Here we accept the interpretation of the first reviser  despite the difference in the shape of the 1 st subdiscal cell, because it may be part of intraspecific variation.  Description. Redescribed ♀ (MTMA) from Hungary (Budapest). Length of fore wing 7.5 mm, of body 9.4 mm.
Notes. The lost lectotype from Germany had hyaline wings (which separates it from the A. carbonarius/ grassator/ carbonaroides complex), the pterostigma paler medially than laterally (entirely dark brown), base of the hind tibia pale yellow (black in ♂) and the body of ♀ entirely black (more or less yellowish or reddish). Molecular data. None. Biology. Unknown; the only known specimen was collected at the end of May which gives no clue of voltinism or how the winter is passed.

Aleiodes nobilis (Haliday [in
Type material. Neotype of A. nobilis here designated: ♀ (NMS), "[Scotland:] W. Ross, Coppachy, Letterewe Estate, ix.2007, Mal. trap, P. Tinsley-Marshall", "BCLDQR _00123". Lectotype of A. medianus, ♀ (ZIL), "[Sweden:] Scan", "medianus m.", "Funnen vid Esperöd I Skåne, teste Papp J., 1983", "Lectotypus Rogas medianus Thoms., 1891, ♀, Papp, 1983", "Aleiodes medianus Th., ♀, det. Papp J., 1983". The lectotype designation for A. nobilis is necessary for nomenclatural stability, because the type series is lost (van Achterberg, 1997) and the species has been confused with similar species in the past. The specimen from Scotland is selected neotype because it fits well the original description, Scotland is relatively close to both type localities (near Holywood in Ireland and Monk's Wood in England) and it is in good condition.   Biology. Collected predominantly in grassy places, June-October. Reared from the noctuid Autographa gamma (Linnaeus) (4 [1 NRS, 2 HSC], Germany, Sweden; H. Schnee) but, in view of its moderately northern areas of occurrence, it seems very likely that other plusiine noctuids would play an important part in its host range. The rearing data indicate that it is plurivoltine, and adult emergence in November from mummies forming in October suggests that it overwinters in the host larva.

Metasoma.
First tergite rather flattened, as long as wide apically; 1 st and 2 nd tergites rather regularly sublongitudinally striate, without medio-longitudinal carina on 2 nd tergite; medio-basal area of 2 nd tergite wide triangular and rather distinct (Fig. 525); 2 nd suture rather deep and narrow; basal quarter of 3 rd tergite finely striate, remainder of metasoma smooth; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath rather long and slender, with long setae and apically rounded (Fig. 520).

Aleiodes pallidicornis (Herrich-Schäffer, 1838) Figs 543-557
Rogas pallidicornis Herrich-Schäffer, 1838: 156;Shenefelt, : 1241Zaykov, 1980b: 87. Rhogas pallidicornis;Fahringer, 1932: 266. Rogas (Rogas) pallidicornis; Tobias, : 84, 1986. Aleiodes (Neorhogas) pallidicornis; Papp, 1987bPapp, : 36, 1991a: 70 (as senior synonym of A. hirtus). Aleiodes (Chelonorhogas) pallidicornis; Belokobylskij, 2000: 42;Ku et al., 2001: 236;Belokobylskij et al., 2003: 398 Type material. Neotype of A. pallidicornis here designated, ♀ (RMNH), "[Netherlands], [Zuid-]Holland, Asperen, 6.viii.1972, C.J. Zwakhals". The neotype designation is necessary for nomenclatorial stability, because the types of Braconidae described by Herrich-Schäffer are lost (Horn and Kahle 1935-37; no specimens could be found by the first author in ZMB), and the species has been confused with similar species in the past. The specimen is selected because it fits well the original description, Netherlands is relatively close to the probable type location in Germany and it is in excellent condition. Biology. Very little is known. Specimens collected in (May) June-August (September), the great majority in June-July strongly suggesting that it is at least largely univoltine. The Dutch specimens were collected in fairly humid coppice woods. The single British specimen (BMNH; G.T. Lyle) was reared (emergence 20.vi.1926) from a "noctua" caterpillar collected by E.A. Cockayne in Aberdeenshire. The mummy is lost. At that time, the term "noctua" was used generally for Noctuidae rather than in the restricted sense of the genus of that name, and it would appear (as Cockayne was by then a distinguished amateur lepidopterist) that the host larva did not belong to an obviously identifiable species. Otherwise we have not seen reared material, and there is no indication of how the winter is passed.
Diagnosis. Maximum width of hypoclypeal depression approx. 0.3 × minimum width of face (Fig. 552); OOL of ♀ approx. as long as diameter of posterior ocellus and remotely punctate with interspaces superficially granulate (Fig. 553); ventral margin of clypeus thick and not protruding in lateral view (Fig. 554); mesoscutal lobes and vertex very finely and densely granulate, with satin sheen; precoxal area smooth medially, but sometimes some rugae below it; vein 1-CU1 0.4-0.6 × vein 2-CU1 and equally slender (Fig.  545); tarsal claws with distinct dark brown pecten (Fig. 557); hind femur and basitarsus slender (Fig. 543); basal quarter of 3 rd tergite largely finely striate; at least basal half of 4 th -6 th tergites of ♂ usually with long dense setosity; head and pronotum black; both tegula and humeral plate equally yellowish; base of hind tibia with narrow dark brown band; hind femur and tibia at least partly black or dark brown; 2 nd tergite yellowish or reddish.
Metasoma. First tergite rather flattened, as long as wide apically; 1 st and 2 nd tergites with medio-longitudinal carina and coarsely irregularly rugose, but posteriorly 2 nd tergite largely smooth and no median carina; medio-basal area of 2 nd tergite triangular and rather large (Fig. 549); 2 nd suture rather deep and finely crenulate; basal half of 3 rd tergite smooth (except for punctuation) and shiny as remainder of metasoma; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath moderately wide, with long setae and apically truncate (Fig. 544).
Colour. Black; hind tarsus largely infuscate, but 3 rd and 4 th segments paler than other segments; apices of fore and middle tibiae slightly infuscate, base of middle and hind tibiae and telotarsi dark brown; apical two-fifths of hind femur and hind tibia (except a pale yellowish band subbasally) black; remainder of legs, 1 st and 2 nd tergites, and 3 rd tergite antero-laterally orange brown; palpi and tegulae brownish yellow; most of veins and pterostigma dark brown; wing membrane subhyaline.
Notes. The type of Rogas pallidicornis Herrich-Schäffer, 1838, has been lost. Traditionally, it has been considered to belong to Aleiodes ductor (Thunberg, 1822), but the latter species is a synonym (see under A. unipunctator). The inadequate original description indicates that the 2 nd tergite has diverging rugae, which excludes part of A. ductor auctt. Female specimens with yellowish or brownish palpi, basal half of the antenna yellowish and blackish hind tibia (except its pale yellowish base) fit well the original description of A. pallidicornis.
Metasoma. First tergite rather flattened, as long as wide apically; 1 st and 2 nd tergites and base of 3 rd tergite finely and irregularly longitudinally rugose, with mediolongitudinal carina weak; medio-basal area of 2 nd tergite triangular and short (Fig.  564); 2 nd suture rather deep and crenulate; apical half of 3 rd tergite punctate-granulate, remainder of metasoma smooth except for some superficial micro-sculpture; 4 th and apical half of 3 rd tergite without sharp lateral crease; ovipositor sheath wide, with long setae and apically truncate (Fig. 559).
Colour. Black; telotarsi largely and basal quarter of antenna dark brown; palpi, tegulae and pterostigma pale yellow; remainder of legs, 1 st and 2 nd tergites, basal half of 3 rd tergite largely and pronotum orange brown; veins brown; wing membrane subhyaline.
New synonymy. We tried to separate the East Palaearctic A. pallidistigmus from the West Palaearctic A. heterostigma, but efforts were in vain. The differences such as the colour of the basal half of the antenna (dark brown in A. heterostigma and usually yellowish or brown in A. pallidistigmus), the eyes and ocelli often smaller, OOL 1.1-1.8 × diameter of ocellus (0.7-1.4 ×), malar space 1.2-1.3 × basal width of mandible (1.0-1.4 ×) and metapleuron with a shiny and more or less punctate area (less shiny and rugulose-coriaceous or rugose) are too variable to justify separation of A. heterostigma. Therefore, we synonymise A. heterostigma with A. pallidistigmus (syn. nov.).
Molecular data. None. Biology. Unknown but presumably univoltine. Specimens of both sexes collected in April and May suggest that the winter is passed in the mummy. We have not seen reared material, but several Hungarian specimens appear to have been collected in Quercus-dominated woodland, but without indication of any association with Quercus as such.
Metasoma. First tergite rather flat posteriorly, wide subbasally and 0.9 × longer than wide apically; 1 st and 2 nd tergites with coarse medio-longitudinal carina and coarsely longitudinally rugose, but posterior quarter of 2 nd tergite rather finely rugose; medio-basal area of 2 nd tergite triangular and wide (Fig. 581); 2 nd suture moderately deep, finely crenulate and narrow; basal two-thirds of 3 rd tergite finely longitudinally rugose, remainder of metasoma superficially micro-sculptured and with satin sheen; 4 th and apical half of 3 rd tergites without sharp lateral crease; ovipositor sheath wide, with rather long setae and apically truncate (Figs 574, 576).
Colour. Black; palpi dark brown basally and remainder pale brown; antenna (except dark brown scapus and pedicellus), tegulae (but anteriorly dark brown), middle and hind trochanters and trochantelli brownish yellow; fore coxa, trochanter and femur dark brown; basal 0.4 of hind tibia ivory and remainder black; remainder of legs (but hind femur with a blackish patch dorso-apically), 1 st and 2 nd tergites and basal two thirds of 3 rd tergite, largely dark reddish brown; pterostigma dark brown; veins mainly yellowish brown, but medially brown (Figs 577, 578); wing membrane subhyaline.
Distribution. Austria, Czech Republic, Germany, Hungary, Russia. New synonymy. The ♀ holotype of Rhogas jaroslawensis lacks the antennae, but according to the original description the antenna was 42-segmented, distinctly shorter than the body, reddish brown, except for the darkened apex and the black scapus. This and the other characters still visible agree well with our interpretation of A. periscelis (except that the hind femur is yellowish brown apically); therefore, we synonymise R. jaroslawensis with A. periscelis (syn. nov.).
Type material. Holotype of A. pulchripes, ♂ (KBIN), "A. pulchripes ♂ mihi", "A. pulchripes mihi, dét. C. Wesmael", "Coll. Wesmael", "Belgique, Charleroi/ teste Papp J., 1983", "Holotypus", "Aleiodes pulchripes Wesm., 1838 M.R. Shaw), Acronicta psi or tridens (2), indet. Acronictinae (1). A quantitative account of rearing this species at its only known English site, comprising old hedges rich in Sorbus aucuparia bordering a largely reclaimed peat bog, is given by . Experimental rearings were unfortunately limited to unobserved exposures of multiple hosts in closed boxes; extremely hot weather marred the results, but from one box containing 15 of each of Subacronicta megacephala (Dennis & Schiffermüller) and A. tridens, the surviving 13 S. megacephala were dissected after three days of exposure and contained no hosts, while at least eight of the A. tridens contained parasitoids (two found by dissection + six mummies formed; of the other seven, one contained no parasitoid on dissection + six resulted in moths). This suggests that S. megacephala is outside the host range. Similar but less well quantified experiments also excluded the low-feeding Acronicta rumicis (Linnaeus) and the arboreal lymantriine Erebidae Euproctis similis (Fuessly). It is worth adding that the rather frequent citation of lymantriine hosts in the literature can undoubtedly be explained by misidentification of the setose and rather colourful larvae of most species of arboreal acronictine noctuids. The mummy is dark grey in colour, leaving only little evidence of the patches of bright colour that had been a feature of the host larva. It forms in the caudal part of the host, the anterior segments of which strongly contract towards the extensive point of attachment, and the cocoon occupies approx. 4 th -7 th abdominal segments. As mummification approaches, the host aligns itself on a narrow aerial twig to which the mummy becomes ventrally adpressed, thus leaving a weakly arched dorsal profile bearing a strong resemblance to an overwintering lateral bud (e.g., of Sorbus aucuparia: Fig. 605). Diagnosis. Maximum width of hypoclypeal depression 0.3-0.4 × minimum width of face (Fig. 616); OOL distinctly less than diameter of posterior ocellus, largely smooth but micro-sculptured near eyes; ventral margin of clypeus thick and not protruding in lateral view (Fig. 618); mesoscutal lobes coriaceous; mesopleuron (including precoxal sulcus area) nearly or completely smooth; propodeum with pair of crest-like protuberances laterally; vein 1-CU1 of fore wing much shorter than vein 2-CU1; basal half of marginal cell of hind wing parallel-sided (Fig. 609); tarsal claws with large dark brown pecten up to apical tooth of claw (Fig. 621); hind spurs dark brown; hind tibial spurs of ♂ obtuse apically (Fig. 624); head black; pterostigma pale yellowish or light brown; mesopleuron, mesosternum and scutellum brownish yellow; apex of hind femur yellowish or reddish; basal half of hind tibia pale yellowish.
Legs. Tarsal claws with conspicuous and robust dark brown pecten up to apical tooth of claw (Fig. 621); hind coxa dorsally largely smooth and remainder remotely punctate; hind trochantellus robust; length of hind femur and basitarsus 4.3 and 5.0 × their width, respectively; length of inner hind spur 0.45 × hind basitarsus.
Metasoma. First tergite evenly convex, 1.3 × longer than wide apically; 1 st and 2 nd tergites with medio-longitudinal carina and finely longitudinally rugose, but posterior quarter of 2 nd tergite smooth and no median carina; medio-basal area of 2 nd tergite triangular and wide (Fig. 632); 2 nd tergite as long as wide basally and with shallow transverse impression; 2 nd suture shallow and narrowly crenulate; 3 rd and subsequent tergites finely punctulate and strongly shiny; apical half of 3 rd and 4 th tergites without sharp lateral crease; ovipositor sheath widened apically, with medium-sized setae and apically truncate (Fig. 628).
Notes. The holotype of A. quadrum is a male and it is less reliable to identify this species from it than from the holotype female of A. illustris; nevertheless, we accept the synonymy proposed by . The figured male from Bulgaria (NMS; Figs 642-651) is considered to be this species (initially through its CO1 sequence); it is morphologically very similar to A. cruentus and there is a possibility that some similar males have been returned to depositories determined as A. cruentus with no recognition that they might belong to A. quadrum. However, the matter remains unresolved until more males of A. quadrum become available.
Resembles A. grassator because of the robust antennal segments and dark wings, but A. ruficeps has frons, OOL, vertex, malar space, and third tergite less sculptured, apex of hind tibia and palpi dark brown, basal antennal segments of ♀ somewhat less robust, hypoclypeal depression wider, marginal cell of fore wing slenderer and vein r of fore wing shorter. Differs from the similar A. ruficornis by having hypoclypeal depression wider, clypeus wider and lower, apical antennal segments of ♀ slenderer, OOL less sculptured and more antennal segments (♀: 45-47 vs 35-39(-41) of A. ruficornis).
Distribution. *Armenia, Bulgaria, *Iran, Russia, *Turkey.  (Horn and Kahle 1935-37; the first author could not find any specimen in ZMB) and the species has been confused with similar species in the past. The specimen is selected because it fits well the original description, the probable type location was in Germany, it has been reared and it is in good condition. Additional material. Andorra, Austria, British Isles (England (V.C. 5,11,13,15,17,19,20,22,23,24,28,29,31,33,38,39,60,63) Shaw), but was no doubt of this species. The above hosts belong to three different subfamilies of Noctuidae, all feeding and resting close to ground level. In addition, we have seen a specimen labelled as ex the nymphalid Brenthis ino (Rottemburg) but accompanied by a mummy of a noctuid, probably Hoplodrina sp. (det. M.R. Shaw). In experiments a range of hosts recorded for this species in the literature (several arctiine and lymantriine Erebidae and the lasiocampid Lasiocampa quercus (Linnaeus)) that are actually hosts of superficially similar species such as A. alternator (Nees) were offered to the female reared ex H. octogenaria but, as expected, they were firmly rejected. However, this female readily accepted late 1 st instar larvae of Agrotis exclamationis (Linnaeus), from which adult progeny resulted very smoothly (1:16\13\\12\12+0. The few failures to oviposit were almost certainly due to egg depletion). Searching in the vicinity of hosts included antennal drumming (the tips curled downwards) and indeed the antennae seemed to be the only proximal means of locating and assessing the host. Once the host was found it was immediately accepted, rapidly jabbed and stood over or often withdrawn from (1.0-1.5 cm) while the venom took affect (20-40 secs), then relocated via antennal searching (when it had been withdrawn from this might take up to a minute, but it was always eventually successful) scooped in with the fore legs (the antennae only slightly involved), positioned and held between the mid legs for the duration of oviposition (20-30 secs). Frequently the host larva was kicked free of the ovipositor by the parasitoid's hind leg(s) and the parasitoid rapidly left without any period of post-oviposition association. Recovery from the venom was rather protracted (up to 20 mins), during which time hosts were rejected if rediscovered. Towards the end of successful oviposition runs it was evident that venom depletion ran ahead of egg depletion, resulting in erratic (but nevertheless successful) oviposition sequences. No host feeding took place. In this experimental series oviposition took place in mid-July with mummification at the end of August and adult emergence in late May of the following year. However, although the winter was passed in the mummy in this entire series (and probably also the case for the other, natural, Agrotis hosts), it is clear that the rearings from Hoplodrina and probably also Mythimna involved overwintering in the host larva with adult emergence in the year of mummification. This host-related difference in overwintering is not inconsistent with univoltinism in Britain (where the experiments and other observations were done) but it is certainly an interesting quirk of its host range and might be of significance in suggesting one way in which a temporal isolating mechanism could potentially arise as a forerunner to speciation (cf. Shaw, 2003). The mummy is formed in the soil and is not strongly (if at all) glued to the substrate. It is predominantly dark brown in colour, very large in relation to the size of the insect that will emerge and, although basically cylindrical, somewhat flattened in appearance owing to a pronounced but blunt lateral keel (Fig. 669). It is more or less strongly contracted at the anterior end, markedly less so caudally, and the copiously silken lining typically occupies 3 rd -8 th abdominal segments.
Aleiodes ruficornis is the commonest and most widespread of a small group of related species parasitising grassland and "cutworm" hosts, exhibiting strong sexual dimorphism with unremarkable males but the more extensively orange females having a stronger build and much shorter antennae. The least extreme in these respects is A. gasterator, which largely (but not completely) replaces A. ruficornis in the Mediterranean region. Aleiodes grassator is similar to A. ruficornis, but it appears to be restricted to montane and northern habitats where it might be thought to replace A. ruficornis. However, some males that morphologically agree best with A. ruficornis have been collected at high altitude in the Alps (up to 2550 m), where A. improvisus also occurs, but whether these highaltitude A. ruficornis males are parts of breeding populations or have simply been carried up in thermals is impossible to say. The females in this group (excluding A. gasterator) are scarcer in collections than males, as they fly very little and rarely enter Malaise traps.
Variation. Legs usually largely reddish, but telotarsi, apices of hind femur and tibia frequently dark brown and sometimes most of hind tibia and apical half of hind femur black; clypeus blunt to rather acute ventrally; depression near posterior ocelli smooth or finely crenulate; mesoscutum of ♀ usually partly reddish brown, but sometimes largely or entirely black; ventral third of mesopleuron regularly and finely punctate.  . These species all feed on low plants. The mummy is moderately arched, very strongly glued down (usually to a narrow twig or stem low in the vegetation), and persists through the winter. The pupation chamber, occupying approximately abdominal segments 3-7 of the host, is rather densely lined with silk which is laid down after the mummy has hardened suggesting that the larva within can turn easily. Rearing experiments, undertaken using stock originally reared from A. menyanthidis, suggested that this host and A. rumicis were equally suitable, but most experiments were not conducted in a way to provide clear data in this respect. The behaviour of the adult females towards these hosts indicated some adaptation to use of highly aggregated species (i.e., that lay large batches of eggs) as, firstly there was a habituation process whereby repeated contacts with hosts generally preceded oviposition, and secondly there was only weak displacement following oviposition (resulting in rather frequent super-parasitism). The antennae were used to locate hosts with wide sweeping motions, and usually the host curled up and was manipulated backwards against the hind tarsi before the ovipositor was inserted and the egg was laid. Generally, there was no pre-oviposition sting and post-oviposition association with the only slightly subdued host was minimal, but the oviposition process was variable and occasionally there was a brief jab, but no subsequent waiting period, before oviposition. Less enthusiasm for sub-active hosts, such as those oviposited into a few seconds or minutes earlier, provided a short-lived impediment to super-parasitism, although sometimes two (on one occasion four, confirmed by dissection) eggs were laid into a single host in separate consecutive bouts without the parasitoid really relinquishing the host. First instar hosts were less easy than 2 nd or 3 rd instars for the parasitoid to deal with, and although oviposition into 2 nd instar hosts was somewhat more successful than into 3 rd instars, occasionally successful oviposition into early 4 th instar hosts occurred. Mean development times from oviposition to mummification in different instar hosts (A. rumicis) under the same ambient conditions (Reading, S. England, July) varied as might be expected given that mummification was always at essentially the same (penultimate instar) stage of the host's larval life: for 1 st (N = 23), 2 nd (N = 40) and 3 rd (N = 7) instars, 27.0, 25.5 and 20.7 days, respectively. There is no venom effect to influence successful host development. Opportunities to offer other hosts were limited but it was clear that, although oviposition into larvae of the closely related arboreal species Subacronicta megacephala (Dennis & Schiffermüller) was fairly readily obtained (N = 15), though slightly inhibited by the host's adherence to its silken pad rather than curling up, the parasitoid was always encapsulated (as a 1 st instar larva in observed cases) and no progeny resulted. No rearings of A. rugulosus from arboreal Acronictinae have been seen, although these conspicuous larvae are often collected and reared.
Colour. Black (including fore and middle telotarsi, apical half of hind tibia and hind tarsus); basal half of hind tibia pale yellowish; palpi (but basally somewhat infuscate) and remainder of legs reddish brown; tegulae yellowish brown; hind tibial spurs and pterostigma dark brown; veins brown; mesopleuron with broad dark reddish longitudinal band; wing membrane slightly infuscate.
Aleiodes rugulosus is a very colour-variable species; the 1 st and 2 nd metasomal tergites are quite frequently mostly or entirely dark red or orange brown (Fig. 701) in populations in which entirely black females (Fig. 717) also often occur. The variation is not geographical, since most populations definitely have both forms. In rearing experiments, a strong genetic basis for this feature became evident: from a virgin female with completely black metasoma, all five males reared were black, and a cross between one of them and a (wild-reared) black female produced eleven female progeny, all black. In contrast, a lineage from a pairing between a weakly red male and a red female comprised three red males and eight red females, separate individuals of which produced a single red and a single black male as progeny. When sufficient material from single sites is available it is usual to see a clear predominance of one form or the other. Sometimes the lower part of mesopleuron is pale (often looking unpigmented there, but narrow- ly), and in extreme reddish specimens the scutellum, much of the mesopleuron (but usually the mesosternum remains darkish), the metanotum dorsally, the metapleuron in part and much of the propodeum are also reddish.
Aleiodes sibiricus   (Kokujev), ♀, lectotype 758 habitus lateral 759 head anterior. Photographs: K. Samartsev. an adult. This is corroborated by the data with the single reared specimen examined (MTMA), from the noctuid Noctua comes Hübner collected 7.iv.1961 and emerging on 3.iv.1962 (Germany; [R.] Hinz). The rearer was widely experienced with caterpillars, and the host determination is unlikely to be wrong (the other caterpillar species with which it might conceivably be confused all have similar biology and phenology in any case). This host initiates its overwintering as a small larva, feeding in mild weather through the winter and normally being well-grown by April, by then in its penultimate or final instar. The rearing is of great interest because it shows that A. sibiricus, like A. fortipes (q. v.), not only parasitises a host that has overwintered as a larva, but also must habitually attack late instar hosts. The reared specimen is accompanied by a stout mummy, large but not unduly so for the size of the adult that emerged, lacking its anterior portion to leave a partitioned chamber comprising abdominal segments 4 onwards, which is well lined with silk and would presumably normally form in the soil (Fig. 762).
Colour. Black; palpi, base of middle coxa, apical 0.2 (dorsally)-0.3 (inner side) of hind tibia, apex of hind femur and telotarsi blackish; remainder of tarsi more or less darkened and base of hind tibia dark brown; basal seven segments of antenna (remainder more or less dark brown), 1 st and 2 nd tergites and antero-lateral corners of 3 rd tergite, and remainder of legs orange brown; humeral plate pale yellowish but partly darkened; pterostigma blackish; veins dark brown; wing membrane blackish infuscate.
Description. Holotype of A. venustulus, ♀, length of fore wing 6.7 mm, of body 8.6 mm.
Head. Antenna incomplete, with eight segments remaining; frons rugose and shiny; OOL twice diameter of posterior ocellus, mainly rugose and shiny; stemmaticum densely punctate; vertex remotely punctate and shiny; clypeus punctate and slightly convex; ventral margin of clypeus thick and anterior part weakly protruding (Fig. 824); width of hypoclypeal depression 0.6 × minimum width of face (Fig. 822); length of eye 1.3 × temple in dorsal view (Fig. 823); vertex behind stemmaticum convex and remotely punctate; clypeus near lower level of eyes; length of malar space 0.4 × length of eye in lateral view and temple as wide as eye.
Variation. No specimens with intact antennae examined; 2 nd tergite coarsely rugose-reticulate or coarsely longitudinally rugose. Male is very similar and with apical tergites type 3, setae quite dense, glabrous stripe narrow, and fringe very weak.

Erratum for Part 1
In the key given by van Achterberg and Shaw (2016) the newly described species A. carminatus van Achterberg & Shaw was inserted at a late stage, which led to an error in couplet 14, as there is also a weak apical comb at the apex of the inner side of the hind tibia present in A. angustipterus van Achterberg & Shaw. The other characters provided should easily separate the two species.