Eight new species and an annotated checklist of Microgastrinae (Hymenoptera, Braconidae) from Canada and Alaska

Abstract Based on the study of 12,000+ specimens, an annotated checklist of 28 genera and 225 species of Microgastrinae braconids from Canada and Alaska is provided, increasing by 50% the number of species for the region. The genera Distatrix, Iconella, Protomicroplitis and Pseudapanteles for Canada, and Diolcogaster for Alaska are recorded for the first time; all but Iconella and Protomicroplitis represent the northernmost extension of their known distribution. Eight new species are described: Apanteles huberi sp. n., Apanteles jenniferae sp. n., Apanteles masmithi sp. n., Apanteles roughleyi sp. n., Apanteles samarshalli sp. n., Distatrix carolinae sp. n., Pseudapanteles gouleti sp. n., and Venanus heberti sp. n. For the more diverse genera, especially Cotesia, Microplitis, Apanteles, Dolichogenidea and Glyptapanteles, many more species are expected to be found. DNA barcode sequences (cytochrome c oxidase I, or CO1) for 3,500+ specimens provided an additional layer of useful data. CO1 sequences were incorporated to the new species descriptions whenever possible, helped to clarify the limits of some species, and flagged cases where further study is needed. Preliminary results on the latitudinal gradient of species/genera richness (45–80° N); as well as biogeographical affinities of the Canadian/Alaska fauna, are discussed. Taking into account the number of specimens in collections still to be studied, data from the barcoded specimens, and extrapolations from Lepidoptera diversity (the host group of the subfamily) the actual diversity of Microgastrinae in the region is estimated to be at least twice that currently known.


Introduction
Microgastrinae are the single most important group of Lepidoptera parasitoids (Whitfi eld 1995(Whitfi eld , 1997, and with over 2,000 described species, rank as the second most diverse subfamily of Braconidae (Yu et al. 2005;Jones et al. 2009). Th e actual diversity of the group has been estimated at 4,000-10,000 species worldwide (e.g. Mason 1981;Dolphin and Quicke 2001;Jones et al. 2009).
Th e Catalogue of Nearctic Hymenoptera (Marsh 1979) recorded 124 species of Microgastrinae in Canada and Alaska, a number that 30 years later had increased to 150 (data compiled after Yu et al. 2005;Fernández-Triana et al. 2009b). However, those numbers represent just a fraction of the actual diversity of the group, a fact that has become more evident recently with the examination of extensive material collected throughout the region and the advent of new techniques (such as DNA barcoding) that have been made available for the study of the subfamily.
In this paper eight new species are described; and an updated checklist of the Canadian and Alaskan Microgastrinae is provided with known distribution, taxonomic and/or biological comments when necessary.

Methods
Th is study is based mostly on the study of the Microgastrinae housed in the Canadian National Collection of Insects (CNC). CNC is one of the largest collections of the group in the world with over 100,000 pinned specimens plus many thousands more in alcohol (Fernández 2007). Th e scope of the CNC is worldwide but the strongest representation is from the Nearctic, especially Canada. More than 11,000 Canadian specimens and around 1,000 from Alaska were reviewed, but a signifi cant amount of material still awaits study.
Other collections (curator names provided between brackets), were partially studied and their data were used to compile the distribution of species by provinces.: Whitfi eld (1995) provided a much needed updated list of the Nearctic Microgastrinae, and assigned to genus all species from the region not treated by Mason (1981). Van Achterberg (2002b) proposed a radical reduction in the number of Microgastrinae genera, and re-arranged all western Palearctic species accordingly (some of those species are also found in Canada and/or Alaska). His modifi cations were incorporated in the Ichneumonoidea section of Taxapad (Yu et al. 2005), and also the Fauna Europaea website ( . Although certainly valid in some regards, van Achterberg (2002b) was based mostly on Holarctic species and contains some decisions not fully supported by additional data (as stated in his paper, more details were intended to be provided later, but there are no published developments to date). Until a study of the microgastrine fauna at world level is available, it seems premature to adopt van Achterberg's classifi cation (Broad et al. 2009). Th erefore, here I am following Whitfi eld (1995) as the standard for generic and species limits for the Nearctic. Th e only exceptions are: Dolichogenidea breviventris (Ratzeburg, 1848), where I am following Papp (1978); and Glyptapanteles pallipes (Reinhard, 1880), where I am following Papp (1983). Th ose two cases are further explained in the annotated checklist. Th e new species described in this paper are of importance in biological control efforts (3 species of Apanteles (Fernández-Triana and Huber 2010)), represent the northernmost record of two genera (one species each of Distatrix and Pseudapanteles), are bizarre species (one Apanteles) or illustrate the potential of integrating barcoding with traditional taxonomy (Venanus and one species of Apanteles). Morphologial terms follow those of Huber and Sharkey (1993), and Sharkey and Wharton (1997), with some additional measurements following Mason (1981) and Valerio et al. (2009). When providing measurements, the fi rst fi gure is that of the holotype, followed by the range for the rest of the specimens if diff erent. For the holotypes a detailed transcription of all labels is provided. All types are deposited in the CNC.
Whenever possible, DNA barcoding (henceforth referred as "barcoding") data for the new species were added to the descriptions. DNA extraction, PCR and sequencing were done at the Canadian Centre for DNA Barcoding (University of Guelph, ON). DNA extracts were prepared from small pieces of legs using a glass fi bre protocol. Extracts were resuspended in 30 μl of dH 2 O, and a 658-bp region near the 5' terminus of the COI gene was amplifi ed using primers (LepF1-LepR1) following standard protocols (Ivanova et al. 2006). Composite sequences or CO1 fragments smaller than the barcode standard were generated using internal primers when initial amplifi cation was not successful. Sequence divergences were calculated using the K2P distance model (Kimura 1980) and a NJ tree of distances was generated using the MEGA software (Tamura et al. 2007) to provide a graphic representation of the species divergences. Full details of methodology are as in Smith et al. (2008).
For barcoded specimens, the Supplementary Appendices 1-3 show their Sample ID and Process ID from BOLD (Barcoding of Life Data systems, www.barcodinglife. org). Sample IDs allow retrieval of all information associated with a particular specimen from the BOLD database, while Process IDs provide information about the sequence, trace fi les, laboratory processing, etc. Genbank accession numbers for the type material correspond to records HQ200902-HQ200929.
All genera, and species within each genus, are ordered alphabetically in the annotated checklist. General comments about species diversity, both reported here and estimated, availability of taxonomical reviews, and specimens in collections are provided for every genus. A detailed distribution within Canadian provinces and territories is provided for every species; acronyms follow the Canada Post standard (http://www. canadapost.ca/tools/pg/manual/PGaddress-e.asp).
Distribution outside of Alaska/Canada, based on data from Yu et al. (2005), is also briefl y mentioned, using the following acronyms: ENA, CNA and WNA (eastern, central and western North America), PAL (Palearctic), HOL (Holarctic), and NEO (Neotropical).
Biological information is provided only when new or relevant. No intent has been made here to comprehensively deal with the hosts of Microgastrinae in the region. More than 10,000 reared but unidentifi ed specimens in the CNC are currently under study; those results, when available, will be published elsewhere.
It was not possible to establish the specifi c identity of 29 species (13%) with certainty. Th ey are recorded here only to genus followed by a number (e.g. Cotesia sp. 1) and information on the specimens examined is provided. In most cases, study of the Holarctic fauna will be needed before determining their status.

Results and discussion
A total of 28 genera and at least 225 species are recorded for Canada and Alaska, representing a 50% increase in the number of known species (Table 1). Th e genera Distatrix, Iconella, Protomicroplitis and Pseudapanteles for Canada, and Diolcogaster for Alaska are recorded for the fi rst time. Except for Iconella and Protomicroplitis, these records also represent the northernmost extension of their known distribution.
Although the increase in species numbers is signifi cant, many gaps still remain in our understanding of the group in Canada/Alaska. For example, the list of species for the northern areas (AK, NT, NU and YT), the Atlantic provinces and the Prairies are far from complete; and studies currently underway should increase signifi cantly the numbers provided in this paper. Similarly, the examination of specimens housed in western Canadian collections will be necessary if progress is to be made in BC, AB and SK.
Based on this paper and work in progress, the latitudinal gradient of species and genera richness within the studied area show a marked increase towards south ( Fig. 1), as would be expected. North of 80° N (northern tip of Ellesmere Island) there are only 4-5 species in 3 genera of Microgastrinae. Between 70-80° N (most of the Canadian Arctic Archipelago with a few areas from the mainland, comprising almost exclusively tundra) there are 20-25 species from about 5 genera. Within the latitudinal range of 60-70° N (most of Alaska and the three Canadian territories, comprising mostly boreal forest with some tundra) there are at least 150 species and 15 genera (e.g. Fernández-Triana et al. 2009a). Th e southernmost range considered here (45-60° N, comprising the rest of Canada with many ecoregions represented) has over 250 species in 26 genera, but these fi gures are less conclusive because many more species await to be recorded -and thus should be seen as an underestimate.
Th e biogeographical affi nities of the fauna can be analyzed from the distributional data detailed in the checklist below. If only the described species (197 in total) are accounted for, 67% are widely distributed in the Nearctic, especially in Eastern North America (the latter could be an artefact due to the more intensive studies and eff orts done in that area); 15% are Holarctic species, many of them intentionally introduced for Biological Control programs; 10% are strict Canada/Alaska endemics (which is equivalent to say that they are restricted to the northern part of the Nearctic region); 4% of the species are also found in the Neotropics; and 4% are cosmopolitan.
Th e most diverse genera are Cotesia, Apanteles, Microplitis, Pholetesor and Dolichogenidea, while Microgaster, Glyptapanteles and Diolcogaster also have signifi cant, though smaller, number of species. Of these, only Pholetesor has been recently revised (Whitfi eld 2006) and its fi gure should be close to the actual number of species expected in the region (but see the checklist below). For the other genera (and especially for Cotesia, Microplitis, Apanteles, Dolichogenidea and Glyptapanteles), the fi gures provided here represent just a fraction of the actual diversity; with many more undescribed species among the CNC holdings as well as some that have been recently listed (e.g. Fernández-Triana et al. 2009b). Th ese records are not considered in this paper because comprehensive taxonomic reviews are needed to unravel the true magnitude of Microgastrinae in the region, a daunting task that will require years of work.  Yu et al. (2005) and Fernández-Triana et al. (2009b).
(2) Number of species recorded in the present paper. (3) Increase in the species number (%). (1) 25  13  45  10  26  9  22  6  1  80  3  47  8  1  151  (2) 38  16  73  57  46  20  38  10  4  136  8  97  16  3  225  (3) 52  23  62 470 77 112 73  67 300 70 167 106 100 200 50 It is diffi cult to provide accurate estimates of the actual diversity of the subfamily when so many species await study. However; the analysis of the available DNA barcoding data, the revision of the collections made so far, and the information of well studied areas (see below) suggest that the actual diversity of Microgastrinae in Alaska/Canada will be at least twice the number recorded in the present paper.  Th ere are currently over 3,500 specimens of microgastrine wasps in BOLD with CO1 sequences, collected from localities all over Canada and Alaska ( Fig. 2A). In spite of the relatively small proportion of specimens barcoded (compared to the more than 30,000 specimens from Canada and Alaska available in the collections studied) they represent over 240 species (Fig. 2B), an astonishing fi gure that surpasses the total of species listed in the present paper. DNA barcoding has proven to be a reliable tool to separate species of Microgastrinae (e.g. Smith et al. 2008), especially when supplemented by critical natural history data, and has tremendous potential to help reveal cryptic species for such a diverse subfamily.

AB AK BC MB NB NL NS NT NU ON PE QC SK YT ALL
Th e proportion of Lepidoptera to Microgastrinae from three well known areas within the region was also calculated (Table 2) and then the average was extrapolated to estimate the total of Microgastrinae for Alaska/Canada. Choosing Lepidoptera makes sense because they are a much better known group and, most importantly, they are the hosts of Microgastrinae, which parasitizes almost all of the lepidopteran families (Whitfi eld 1997). Th e proportion of host/parasitoid species was between 7 and 17, with an average of 12. Interestingly, the same proportion (10-12) is found in other well studied areas around the world such as temperate British Isles and tropical Area de Conservacion de Guanacaste, Costa Rica -data calculated from Fauna Europaea (van Achterberg 2004) and Janzen et al. (2009) respectively. If a proportion of 12 Lepidoptera to each Microgastrinae is extrapolated to the all Canada/Alaska fauna -with over 6700 estimated species of Lepidoptera (Biological Survey of Canada 2010), the results show an estimated diversity of about 550 species of Microgastrinae for the region.
Regardless of the approach used, even the most conservative scenarios show an unexpected and unprecedented level of species diversity in a region of the planet supposed to have a rather low diversity. Th e results reported here, as well as previous papers from other areas (e.g. Smith et al. 2008) suggest that indeed the Microgastrinae might be much more diverse than anticipated.  Danks (1981) for the Arctic Archipelago; Lafontaine and Wood (1997) for the Yukon; and Lafontaine (1997)  Metasoma. Mediotergite 1 almost parallel sided, just slightly widening posteriorly; basal width/apical width 1.1×; length/apical width 1.4×; mediotergite 1 with smooth, basal depression; apical 2/3 sculptured with longitudinal striae, except for a median, sub-apical depressed area which is mostly smooth and a polished knob centrally in the apical margin. Mediotergite 2 transverse, trapezoidal in shape; basal width/apical width 0.7×; length/apical width 0.3×; with longitudinal striae covering most of the surface. Mediotergite 3 twice the length of mediotergite 2. Mediotergite 3 and following unsculptured, polished and uniformly covered by sparse setae. Hypopygium striate, with acute tip slightly protruding beyond apical tergites. Ovipositor sheaths fully setose, 0.9-1.0× as long as metatibia length.
Wings. Forewing vein R1a 1.1× as long as stigma length; length of R1a about 2.0× as long as the distance between its end and the end of 3RSb. Vein r 0.8× the maximum width of stigma. Join of veins r and 2RS angulated, sometimes with small knob at their junction; vein 2M 1.0-1.1× as long as vein (RS+M)b. Edge of vannal lobe of hindwing medially straight to slightly convex and with setae of uniform length which are shorter than those at base and apex of lobe.
Colour: Maxillary and labial palps, and two fi rst pairs of legs (except for coxae), yellow; head, meso and metasoma, and all coxae dark-brown or black; apex of metatibia and part (sometimes most) of the metafemur and metatarsus orange-red or light brown. Most of veins light brown, stigma borders light brown, centrally pale.
Male. As females, except for slightly smaller size (2.3-2.4 mm), legs with brighter yellow coloration, and width of mediotergite 1 slightly less than in females. Molecular data. Partial barcodes (144 bp) from the holotype and three paratypes of A. huberi were obtained and compared with two paratypes of A. fumiferanae with a similar sequence length (Fig. 21). In spite of the relatively short sequences available for analysis (about one fi fth of the barcoding region) the two species consistently diff ered between 1-4 base pairs (0.8-2.8%).
Distribution and biology. Only known from the type locality in BC. All studied specimens were reared from Choristoneura biennis -it is the only braconid species reliably reared from that lepidopteran (Fernández-Triana and Huber 2010).
Comments. Th e related species A. fumiferanae has a relatively wide range of hosts (Mason 1974;Fernández-Triana and Huber 2010), but has never been recorded parasitizing Choristoneura biennis. Th e diff erent host species and slight but consistent morphological and barcoding diff erences provide suffi cient evidence to consider A. huberi as a separate and distinct species.
Etymology. I dedicate this species to John Huber (CNC) as an appreciation for the many things I have learned from him during the last four years (his knowledge of Hymenoptera and kindness are both extraordinary); and also for all the shared chocolate! Diagnosis. Th is species is related to A. fumiferanae but it is diff erentiated by its slightly larger size; yellow tegula; less defi ned areola (mostly marked by a depression and with only apical carinae; contrasting with a complete areola, well defi ned by carinae in A. fumiferanae); medio tergite 2 (less transverse in A. jenniferae, thinner in A. fumiferanae); and meditergite 3 (A. jenniferae with some sculpture centrally in anterior margin basally, completely smooth in A. fumiferanae).
Mesosoma. Pronotum laterally with dorsal and ventral grooves well defi ned. Mesoscutum with relatively close punctures (distance between punctures about 0.5× its diameter). Mesoscutum 1.4× wider than long. Mesoscutum and scutellum uniformly covered by dense, silvered-coloured pilosity. Scutellum almost smooth, with very sparse and shallow punctures mostly on the margins. Scutellum length/width at base 1.0×. Scutellar suture well impressed, with 12 costulae, the central ones more spaced and deeply impressed than the lateral ones. Posterior band of scutellum polished. Scutellar lateral face with polished area semicircular slightly less than half the face height. Mesopleuron setose and with punctures on the anterior half; the posterior half glabrous and smooth except for a thin sulcus running from the posterior margin (at about half the length of that margin) towards the lower margin of mesopleuron (ending just before the punctures and setae of the anterior half ). Th in, crenulated sulcus separating meso and metapleura. Metapleuron mostly smooth and polished, with setae and punctures only dorsally and ventrally along margins; metapleuron with a very short, crenulate, longitudinal sulcus running from lower margin near metacoxa through spiracle. Metapleural carina with short lamella. Propodeum with areola defi ned mostly by a central impression than carinae -though the posterior carinae are visible; propodeum coarsely punctured in the anterior half, with transverse striation in the apical half, the only smooth area is centrally inside the areola.
Wings. Forewing vein R1a 1.0-1.1× as long as stigma length; length of R1a 6-7.0× as long as the distance between its end and the end of 3RSb. Vein r 1.0× (1.0-1.1×) the maximum width of stigma. Join of veins r and 2RS angulated and with a small knob at their junction; vein 2M 0.8× (0.7-0.9×) as long as vein (RS+M)b. Edge of vannal lobe of hindwing medially straight to convex and glabrous.
Colour: Maxillary and labial palps, tegula, two fi rst pairs of legs (except for coxae), and basal half of metafemur yellow; head, meso and metasoma dark-brown or black; wing base and all coxae brown; metafemur, apical half of metatibia and metatarsus yellowish-brown to orange-brown. Most of veins very light brown, almost hyaline; stigma light brown.
Male. As female except for longer fl agellomere, antenna longer than body length, darker hind legs (with metafemur dark brown), and less transverse medio tergite 2 (which is almost quadrate and with striation arranged in a concentric way).
Distribution and biology. Th e species is widely distributed in eastern Canada, where it has been reared from Choristoneura rosaceana.
Comments. Th is species and the previous one (A. huberi) illustrate well the need for a review of what Fernández-Triana and Huber (2010) called "the A. fumiferanae species-complex". It is becoming obvious that many species are hidden under that name, and a comprehensive approach combining detailed morphology, biology (especially verifi ed host records) and molecular data will be required to unravel the rest of the species within that complex.
Etymology. I dedicate this species to Jennifer Read (CNC) to thank her for the many hours she spent taking photos for several projects we worked upon together; and as recognition for her superb photographic and editing skills. Diagnosis. Th is species looks similar to Apanteles cockerelli Muesebeck, 1921; and it will run to that species in the available keys (e.g. Muesebeck 1921). It diff ers in the stigma colour in the fore wing (pale with only brown borders in A. masmithi, completely brown in A. cockerelli); the shape of vannal lobe in the hind wing (straight and with short setae medially in A. masmithi, concave and glabrous in A. cockerelli), and the relative length of the metatibial spurs (about the same length in A. masmithi, the inner spur longer than the outer one in A. cockerelli). Th e two species also have diff erent geographic distribution, diff erent host species, and diff er in 14 base pairs within the barcoding region (more details below under the sections Molecular data, Distribution and biology and Comments).
Mesosoma. Pronotum very smooth and polished, laterally with dorsal and ventral grooves thin but deep and well defi ned. Mesoscutum mostly smooth, with shallow punctures (distance between punctures about its diameter), punctures a little closer and deeper in the posterior margin. Mesoscutum 1.2× (1.1-1.2×) wider than long. Mesoscutum and scutellum covered by sparse, silvered-coloured pilosity. Scutellum almost smooth, with very sparse (distance between punctures twice its diameter) and shallow punctures concentrated mostly on the margins. Scutellum length/width at base 1.1×. Scutellar suture thin and shallow, with 16 (15-17) costulae. Posterior band of scutellum polished. Scutellar lateral face with the polished area triangular and about 4/5 the face height. Mesopleuron smooth and glabrous on most of its surface, with sparse setae and punctures (distance between punctures twice or more its diameter) only on the anterior and dorsal margins. Th in and shallow sulcus, with a few costulae, separating meso and metapleura. Metapleuron mostly smooth and polished, with setae and sparse punctures only dorsally and posteriorly along margins; metapleuron with a thin, longitudinal sulcus running from lower margin through spiracle. Metapleural carina with short lamella. Propodeum mostly smooth, with sparsely punctures in the anterior half and a few transverse striae in the apical half; propodeal areola absent but there is a short, postero-median longitudinal band of rugosity (consisting of several very short carinae radiating from nucha).
Wings. Forewing vein R1a 1.0× as long as stigma length; length of R1a 5.0× as long as the distance between its end and the end of 3RSb. Vein r 0.8× the maximum width of stigma. Join of veins r and 2RS angulated and with a small know marking the angulation (sometimes only slightly angulated and then know very small to absent); vein 2M 0.6× as long as vein (RS+M)b. Edge of vannal lobe of hindwing medially straight and with short setae that are slightly sparser than the rest of the lobe.
Colour: Mostly black to dark brown, except for: maxillary and labial palps (light brown to brown), wing base (light brown), profemur and part of most of all tibia and tarsi (light brown to yellow), meso and metatibial spurs (light yellow to witish). Wings hyaline, with most of veins transparent, except for C+Sc+R, R1, and occasionally r and 2RS which can be partially pigmented; stigma hyaline except for brownish borders.
Male. Similar to females but slightly smaller in size and with longer antennal segments (especially the apical ones). Th e maxillary and labial palpi tend to be yellow, and the legs tend to be darker (mostly black with less yellow areas). Th e mediotergite 1 is fully smooth and polished, and narrows stronger toward apex (being thinner compared to that of females). Th e wing veins are paler, of milky coloration, including the stigma (which brown borders are very thin, almost disappearing in some specimens).
Variation. Th ere is some variation in size among the diff erent localities (it is shown in the description) and also the maxillary and labial palpi range from dark brown to yellow.
Molecular data. Barcodes of 6 specimens of A. masmithi and 3 of the related species A. cockerelli were compared. Because all specimens but one were collected between 1951 and 1969 it was only possible to obtain mini-barcodes (144 bp). Th e only recent specimen (a paratype of A. masmithi, collected in 2008) rendered a full barcode (657 bp) which fully matched the other specimens with mini-barcodes. Th e molecular results confi rmed that they are indeed diff erent species, with at least 14 (9.7%) of base pairs divergence (Fig. 22). Interestingly, specimens of A. cockerelli within the US (from CA, MO and TX) seem to comprise more than one species -but that is beyond the geographical scope of the present work, thus they will be dealt with in a diff erent paper.
Distribution and biology. Th e species is widely distributed in Eastern Canada, where it has been recorded parasitizing Limnaecia phragmitella (Gelechidae) on Typha spp. heads (cattail grass). Some paratypes from Nova Scotia had written on their labels that the host could also be Dycimotomia julianalis (Pyralidae), also on cattail; however, this record needs to be confi rmed. Th is is the fi rst Microgastrinae (and Braconidae) species recorded as parasitoid of L. phragmitella.
Comments. Th e specimens of A. masmithi were identifi ed by W. Mason as a different but related species to A. cockerelli. Th e latter species has been recorded from US in the following ten states: CA, IA, ID, MI, MO, NE, NM, OR, SD, TX (Yu et al. 2005). Th e diff erent host species (Isophrictis sp. (Gelechidae) for A. cockerelli), 14 (9.7 %) of base pairs divergence within the barcoding region, and slight but consistent morphological diff erences, provide suffi cient evidence to consider A. masmithi as a distinct species.
Etymology. I dedicate this species, which DNA barcoding helped to recognize, to M. Alex Smith (University of Guelph) as an appreciation for the many parasitoid wasps he has helped to barcode, study and publish about; and also for sharing with me his superb knowledge on molecular approaches. Diagnosis. Th is species looks similar to A. stagmatophorae Gahan, 1919, and it will run to that species in the available keys (e.g. Muesebeck, 1921), but they differ in several characteristics. In A. roughleyi the vannal lobe of hindwing is medially straight and glabrous (slightly convex to slightly straight but with uniform setae in A. stagmatophorae), the ovipositor sheaths are longer (1.7× compared to 1.2×), the metafemur is thinner (3.5× as long as wide compared to 3.2×), and the propodeum is more sculptured (in A. stagmatophorae the propodeum is mostly smooth, with very shallow and small punctures). Th e two species have a very separate distribution (BC in western Nearctic for A. roughleyi; Maryland, in eastern Nearctic for A. stagmatophorae). Th e known host are also from diff erent families: Choristoneura sp., Tortricidae, for A. roughleyi; Periploca gleditschiaeella (Chambers, 1876), Cosmopterigidae, for A. stagmatophorae (more details below in the section Distribution and biology).
Mesosoma. Pronotum laterally with dorsal and ventral grooves thin, but well defi ned and deep. Mesoscutum with very shallow and sparse punctures (distance between punctures 1.5-2.0× its diameter). Mesoscutum 1.4× wider than long. Mesoscutum uniformly covered by silvered-coloured pilosity; scutellum almost glabrous, with just a few setae on margins. Scutellum almost smooth, with very sparse, small and shallow punctures mostly on the center. Scutellum length/width at base 1.1×. Scutellar suture very thin and shallow, with about 20 small and not well defi ned costulae. Posterior band of scutellum polished. Scutellar lateral face with polished area semicircular about 0.6× the face height. Mesopleuron setose and with sparse punctures only on the anterior margin; the rest glabrous, smooth and polished. Th in, crenulated sulcus separating meso and metapleura. Metapleuron mostly smooth and polished, with setae and punctures only dorsally and ventrally along posterior margin; metapleuron with a thin sulcus running from lower margin near metacoxa through spiracle. Metapleural carina with a very short lamella. Propodeum mostly punctured, with a few striae posterolaterally; propodeal areola absent, but there is a central smooth area (contrasting with rest of the propodeum sculpture) and also there is a short, postero-median longitudinal band of rugosity (consisting of several short carinae radiating from nucha).
Legs. Metatibial inner spur 1.1× the length of outer spur, and 0.5× the length of metatarsomere 1. Metafemur 3.5× as long as wide.
Wings. Forewing vein R1a 1.3× as long as stigma length; length of R1a 5.7× as long as the distance between its end and the end of 3RSb. Vein r 1.0× the maximum width of stigma. Join of veins r and 2RS slightly angulated; vein 2M 1.1× as long as vein (RS+M)b. Edge of vannal lobe of hindwing medially straight and glabrous.
Colour: Mostly black to dark brown, except for: maxillary and labial palps (yellow); tegula and wing base (light brown); fi rst two pairs of legs (yellow except for coxae which are partially light brown); hind legs (mostly yellow-brown, with metacoxa brown and dorsal brown marks on metafemur, metatibia and metatarsi). Wings hyaline, with most of veins brown, including stigma.
Male. Unknown. Distribution and biology. Th e host information (Choristoneura fumiferana) was recorded originally in 1965, i.e., before Freeman (1967) split the genus Choristoneura and changed the species boundaries. Th e actual host is either C. occidentalis or C. pinus, but there is no way to determine which.
Comments. Th e specimen bears a label by W. Mason, dated 1978, stating that it may actually be a new species related to A. stagmatophorae. Comparison with two paratypes of the later species (housed in the CNC) confi rms that the two species are distinct. In spite of the fact there is only one known specimen, the species is described to provide a name because of its potential economic importance (Fernández-Triana and Huber 2010).
Etymology. I dedicate this species to the late Rob Rougley (University of Manitoba) who passed away when this paper was starting. We all miss you dear friend and colleague, but I am sure you should be chasing heavenly Ditiscidae beetles right now!  ON, Mal. Trap, 19.viii-11. ix.1973. Diagnosis. Th us far this is the only Nearctic species of Apanteles with a significantly short antenna (half the body length); vein 2M very short, almost obliterating with vein 2RS; and antenna with yellow scape/pedicel and brown fl agellomere. Th e combination of those characters makes A. samarshalli one of the most distinctive and recognizable species within the genus.
Mesosoma. Pronotum laterally with dorsal and ventral grooves well defi ned. Mesoscutum with coarse, close punctures (distance between punctures less than half its diameter). Mesoscutum 1.2× (1.1-1.2×) wider than long. Mesoscutum and scutellum covered by uniform, large, silvered-coloured pilosity. Scutellum almost smooth, with very shallow and sparse punctures in the margins. Scutellum length/width at base 0.8×. Scutellar suture width 1/6 scutellum length, with 12-14 costulae. Posterior band of scutellum polished. Scutellar lateral face with the polished area triangular and about 4/5 the face height. Mesopleuron with close punctures and setae on the anterior half, smooth and glabrous on the posterior half. Th in and shallow sulcus, with fi ne costulae, separating meso and metapleura. Metapleuron mostly punctured and with setae, smooth, polished and glabrous only around the spiracle; metapleuron with a longitudinal sulcus running from ventral to dorsal margin of metapleuron through spiracle. Metapleural carina lamellate. Propodeum sculpture reticulate, postero-laterally with longitudinal striation; propodeal areola absent but there is a short, postero-median longitudinal band of rugosity (consisting of several short carinae radiating from nucha). Metasoma. Mediotergite 1 evenly and slightly narrowing toward apex, with a wide and deep basal depression; basal width/apical width 1.4×; length/apical width 1.5; mediotergite 1 mostly sculptured (except for smooth basal depression and central knob on the posterior margin), with longitudinal striation on its apical 2/3. Mediotergite 2 smooth and polished, transverse, and wider centrally; basal width/apical width 0.8×; length/apical width 0.3×. Mediotergite 3 2.0-2.5× the length of mediotergite 2. Mediotergite 3 and following unsculptured, polished and covered by sparse setae on the posterior margins. Hypopygium striate, with acute tip slightly protruding beyond apical tergites. Ovipositor sheaths fully setose, short, 0.6× as long as metatibia length.
Legs. Metatibial inner spur 1.3× the length of outer spur, and 0.5× the length of metatarsomere 1. Metafemur 2.7× as long as wide.
Wings. Forewing vein R1a 1.3× (1.2-1.5×) as long as stigma length; length of R1a 4.0× (4.0-5.0×) as long as the distance between its end and the end of 3RSb. Vein r about the same length than maximum width of stigma. Join of veins r and 2RS evenly curved, not angulated; vein 2M very short, almost obliterating with 2RS, length of 2M 0.3× as long as vein (RS+M)b. Edge of vannal lobe of hindwing medially strongly concave and glabrous.
Male. Unknown. Molecular data. From all specimens studied, only the holotype rendered a partial sequence (390 bp, approximately 60% of the barcoding region). Th e specimen matches almost perfectly (99.96%) a Costa Rican species named as Apanteles Rodriguez151 (Smith et al. 2008).
Distribution and biology. Th e species has been found from the southwestern part of ON (Rondeau Provincial Park, 42°N) to about 25°N in FL (Everglades National Park and Florida Keys). None is know of its host, but most of the specimens have been collected in hammock forests.
Comments. Despite the two widely separate areas of distribution (ON and FL), I have not been able to fi nd any diff erence between the Canadian and US specimens; thus they are considered as conspecifi c here. As for the relation with Apanteles Rod-riguez151, I have not been able to examine specimens of the latter. If proven conspecifi c, it would be even more puzzling to explain the distribution of the species. All of those areas share in common the presence of oaks, but the data available is not enough as to draw any solid conclusion at present.
Etymology. I dedicate this species to a great friend and entomologist, Steve A. Marshall (University of Guelph). I hope you have many more collecting and photography trips in the near future! Mesosoma. Pronotum with ventral groove present, dorsal one almost obliterated. Mesoscutum with shallow, sparse punctures (distance between punctures about the same as its diameter); punctures almost disappearing in the notauli and posterior area of mesoscutum. Notauli not impressed, visible only because of the contrast of diff erent coloration and smoother area than most of the mesoscutum. Mesoscutum 1.3× wider than long. Mesoscutum and scutellum uniformly covered by dense, silvered-coloured pilosity. Scutellum almost completely smooth. Scutellum length/width at base 1.2×. Scutellar suture shallow and thin with 8-9 costulae some of them confl uent. Posterior band of scutellum polished. Scutellar lateral face with polished area about 1/3 the face height. Mesopleura smooth and glabrous, except for a few punctures and setae on the margins; sternaulus marked by a shallow impression with transverse striae. Crenulated sulcus separating meso and metapleura. Metapleura smooth in basal half, apical half punctuated and with setae; metapleura with a crenulated, longitudinal sulcus running from lower margin near the metacoxa through the spiracle. Metapleural carina with lamella. Propodeum weakly punctuate, almost smooth; propodeal areola absent but there is a short, postero-median longitudinal band of rugosity (consisting of several short carinae radiating from nucha).
Metasoma. Mediotergite 1 parallel sided for over 3/4 of its length, then slightly narrowing towards apex where it is rounded at posterior end; basal width/apical width 1.8×; length/apical width 3.6×; mediotergite 1 essentially unscultured except postero-laterally near apical margin; with broad excavation medially over anterior half. Mediotergite 2 subtriangular but with lateral margins weakly defi ned; basal width/apical width 0.3×; length/apical width 0.5×; essentially smooth, with fi ne, longitudinal grooves sublaterally, delimiting a central, raised region that is more or less rectangular in shape. Mediotergite 3 1.2× longer than mediotergite 2. Mediotergite 3 and following unsculptured, polished and with sparse setae. Hypopygium evenly sclerotized, truncated and slightly longer than apical tergites. Ovipositor sheaths very short (visible part 1/10 the length of metatibia), the tip blunt and with very sparse, tiny setae (those setae much shorter than hypopygium pilosity). Legs. Metatibial inner spur 1.5× the length of outer spur, and 0.7× the length of metatarsomere 1. Metafemur 3.6× as long as wide. Protarsus with Protapanteles-like spine. Tarsal claws basally with a large lobe that extends more than half the claw length.
Wings. Fore wing vein R1a as long as stigma length; length of R1a about 5.0× as long as the distance between its end and the end of 3RSb. Vein r 0.8× the maximum width of stigma. Vein r meeting 2RS in a distinct angle marked by a knob. Vein 2M about the same length that vein (RS+M)b. Hindwing with margin of vannal lobe medially straight and without setae in the fl at area.
Distribution and biology. Th is species represents the northernmost record of the genus. Nothing is known of its biology.
Comments. Based on morphology only, the limits between D. carolinae and D. solanae seem weak; however, morphological similarities are common within this genus. For example: D. solanae shares a number of characteristics with Neotropical species (see Grinter et al. 2009). I consider the Canadian specimen as a new and distinct species based on the major diff erences within the localities and habitats (California's inner Coast Range and Oregon Cascade Mountains for D. solanae; King Mountain in Gatineau Park, Quebec, for D. carolinae). In the CNC there are several specimens representing at least another undescribed species from the Nearctic (southern and eastern US); the study of those specimen will clarify in time the limits of the North America species.
Mesosoma. Pronotum XX. Mesoscutum uniformly sculptured by dense and well impressed punctures (distance between punctures about half their diameter). Mesoscutum 1.5× wider than long. Mesoscutum and scutellum uniformly covered by dense, silvered-coloured pilosity. Scutellum similarly sculptured than mesoscutum, though punctures slightly shallower and sparser. Scutellum length/width at base 1.2×. Scutellar suture thin and shallow, with 8-9 costulae. Posterior band of scutellum polished. Scutellar lateral face with polished area about 1/2 the face height. Except for a few punctures on the upper anterior margin, mesopleuron smooth and glabrous, setae over all of mesopleuron margins. Crenulated sulcus separating meso and metapleura. Metapleuron smooth in basal half, apical half punctate and with setae, metapleuron with a crenulated, longitudinal sulcus running from lower margin near metacoxa through spiracle. Metapleural carina with short lamella. Propodeum with median carina well defi ned and raised over its entire length; propodeum fully sculptured with transverse carinae, some radiating from the median carina.
Metasoma. Mediotergite 1 parallel sided for the basal 0.6× of its length, then narrowing towards apex, basal width/apical width 1.3× (1.2-1.3×); length/apical width 3.1×; mediotergite 1 with deep medial groove over its basal half, fully sculptured with longitudinal to transverse striae (except for a very small basal area surrounding the beginning of the groove and a small, polished apical knob). Mediotergite 2 transverse, subtriangular to trapezoidal in shape; basal width/apical width 0.4×; length/ apical width 0.4×; fi ne, longitudinal striae covering most of the surface (sometimes apical third smooth). Mediotergite 3 more than twice the length of mediotergite 2. Mediotergite 3 and following unsculptured, polished and uniformly covered by sparse setae. Hypopygium striate, with acute tip protruding beyond apical tergites. Ovipositor sheaths fully setose, 1.0-1.2× as long as metatibia length.
Wings. Vein R1a 1.2-1.3× as long as stigma length. Length of R1a about 6.0× as long as the distance between its end and the end of 3RSb. Vein r X the maximum width of stigma. Vein r and 2RS evenly curved to very slightly arched, with no clear limits between the two veins. Vein 2M about twice as long as vein (RS+M)b. Edge of vannal lobe of hindwing medially straight to slightly convex and with uniform length setae shorter than those at base and apex of lobe.
Colour: Labrum, mandibles (except for black tips), scape and pedicel yellow; maxillary and labial palps light yellow; clypeus orange-brown; rest of antenna and head brown. Mesosoma, basal half of metacoxa and mediotergite 1 dark brown to black; mediotergite 2 completely, mediotergite 3 and following centrally, apical half of metacoxae dorsally, metatarsi and apex of metatibia, light brown; tegula, rest of legs, tergites 3 and following laterally, and all sterna, yellow to light yellow almost white; stigma and veins in forewing brown.
Male. Males have mediotergite 3 and following almost completely brown, clypeus, scape and pedicel darker, and metacoxa fully brown. Th e fl agellomeres are longer than those of females.
Variation. Some specimens have lighter body coloration. Molecular data. Eleven specimens rendered full barcodes, with four haplotypes showing up to 0.3% of variation (1-2 bp). Th ose specimens were compared with one unauthenticated specimen of Pseudapanteles dignus, the only Nearctic species with data available in GenBank. P. gouleti is very distinctive, with more than 18% of base pairs diff erent from the other species (Fig. 23).
Distribution and biology. All specimens have been collected in an area bounded by the St Lawrence and Ottawa rivers (44º-46º N and 74º-75º W) near Canada's capital. Th is is the northernmost known record of the genus Pseudapanteles. I studied 8 ♀ and 15 ♂ captured between mid July to mid September. One specimen was reared from the Maple Leafcutter, Paraclemensia acerifoliella (Fitch, 1856) (Incurvariidae). Th is is the third record of Braconidae parasitizing an incurvariid Lepidoptera; the other two being another Microgastrinae, Pholetesor ornigis (Weed, 1887), and a Braconinae, Bracon montowesei ; in all cases attacking the same incurvariid species (Marsh 1979;Yu et al. 2005;Whitfi eld 2006).
Etymology. I dedicate this species to Henri Goulet (CNC) in whose backyard (a biodiversity gem in Ottawa, fondly called by CNC researchers as "Goulet National Park") the holotype and several paratypes were collected. Henri wisely encouraged me to study the Microgastrinae and during four years has kindly given me access to his lab, collections and great expertise on many insect topics. Diagnosis. V. heberti is similar to V. pinicola Mason, 1981, and will run to that species in the recent key to the New World species (Whitfi eld et al., in press). V. pinicola is smaller (females: 1.6-1.9 mm, average=1.7 mm, N=8; males 1.7-2.4 mm, average=2.0 mm, N=5) than V. heberti (female: 2.2 mm; males: 2.0-2.4 mm, aver-age=2.2, N=8). Th e size (width and height) of the second submarginal cell in the fore wing (compared to the length of vein r, the width and the length of stigma) is smaller in V. pinicola -usually the values represent 0.6-0.8 of similar proportions for V. heberti. Th e males of V. pinicola have its veins mostly pigmented (as have the females of both species), contrasting with mostly unpigmented veins in males of V. heberti. Th e two species have diff erent geographical distributions: V. pinicola in west Canada/US (AB, BC, YT and ID) and V. heberti in Eastern Canada. Th e known hosts are diff erent: the Gelechids Coleotechnites milleri (Busck, 1914) and C. starki (Freeman, 1957) for V. pinicola; and the Gracillarid Caloptilia asplenifoliella (Darlington, 1949) for V. heberti. Th e two species also diff er in 12 base pairs of the barcode region (more details below under the sections Molecular data, Distribution and biology and Comments).
Mesosoma. Pronotum very smooth and polished, laterally with only the ventral groove well defi ned. Mesoscutum mostly smooth, with shallow but close punctures (distance between punctures 0.5-0.7 its diameter), punctures a sparser centrally along the posterior margin. Mesoscutum 1.2× (1.1-1.2×) wider than long. Mesoscutum and scutellum covered by sparse, silvered-coloured pilosity (sparser in the scutellum). Scutellum mostly smooth, with a few, shallow, very sparse punctures. Scutellum length/ width at base 1.0×. Scutellar suture width 1/7 scutellum length, with 16 costulae not very well defi ned. Posterior band of scutellum polished. Scutellar lateral face with the polished area semicircular, 0.3-0.4× the face height. Mesopleuron smooth and glabrous on most of its surface, with sparse setae and punctures (distance between punctures usually twice or more its diameter) only on the anterior, ventral and posterior margins. Deep sulcus, with costulae, separating meso and metapleura. Metapleuron setose and punctured along anterior and ventral margins; lower ¼ of metapleuron rugulose, and with a broad, crenulated sulcus running from lower margin through spiracle. Metapleural carina lamellate and with costulae. Propodeum mostly rugulose, especially on the apical third (which is concave and delimited from the rest of the propodeum Metasoma. Mediotergite 1 widened and rounded apically, with its widest part subapically; basal width/apical width 0.9×; length/apical width 1.5; mediotergite 1 rugulose, apical ¼ with longitudinal striation laterally and two pits at each side of a central, polished area (like a knob) that reaches the posterior margin of tergite. Mediotergite 2 trapezoidal in shape, centrally smooth and polished, laterally rugulose; basal width/ apical width 0.6×; length/apical width 0.6×. Mediotergite 3 twice the length of mediotergite 2. Mediotergite 3 and following unsculptured, polished and with few, sparse setae mostly along the posterior margin of tergites.
Legs. Metatibial inner spur 1.2× the length of outer spur, and 0.6× the length of metatarsomere 1. Metafemur 2.7× as long as wide.
Wings. Forewing vein R1a 0.7× as long as stigma length; length of R1a 2.7× as long as the distance between its end and the end of 3RSb. Vein r 0.6× (0.6-0.7×) the maximum width of stigma. Second submarginal cell height about the same length (or slightly smaller or larger) than vein r length; vein 2M 3.0× as long as vein (RS+M)b and 0.25-0.33× the stigma length. Edge of vannal lobe of hindwing covex and uniformly setose.
Colour: Mostly black to dark brown; pro-and meso-tibiae and tarsi yellowish brown, as it is apical 0.2× of metatibia, metatibial spurs, maxillary and labial palps. Wings hyaline, with most of veins transparent or whitish, except for C+Sc+R, R1, and 2M can be partially or totally pigmented; stigma brwon.
Molecular data. Full barcodes of 3 specimens of V. heberti and one specimen of the related species V. pinicola were obtained and compared (Fig. 24). Th e molecular data showed 12 (1.86 %) base pairs of diff erence between the two species.
Distribution and biology. Th e species is widely distributed in Eastern Canada (QC, NS, PE), where it has been realibly reared from Gracillaria asplenifoliella. In the CNC there is one specimen of V. heberti from BC with a label stating it was reared from Caloptilia invariabilis (Braun, 1927). Th is has to be a labelling mistake because C. invariabilis is only known from Eastern Canada (NS, ON, QC) and US, but has never been recorded from western Nearctic (De Prins and De Prins 2010).
Comments. When Mason (1981) described V. pinicola he mentioned some variations in the specimens from Eastern Canada compared to the West, but considered that as intraspecifi c variation. Th e consistent, though subtle, morphological and molecular diff erences; diff erent geographical distribution and hosts provide suffi cient evidence to consider them as distinct species. Because of the similarities between the two species, four former paratypes of V. pinicola (in the CNC) are here transferred as paratypes of V. heberti.
Etymology. I dedicate this species, recognized after DNA barcoding provided a fi rst clue, to Paul Hebert (University of Guelph), as an appreciation for his support; and also for allowing the gathering of thousand of Microgastrinae barcodes -which will hopefully contribute in a signifi cant way to the taxonomy of such a diffi cult and diverse group.

Genus Alphomelon Mason, 1981
Th is New World endemic genus was recently revised by Deans et al. (2003), but its distribution is mostly tropical and only one species is recorded from Canada. No additional species are expected from the region. Deans, 2003. ON, QC. Also recorded from ENA and the NEO (Mesoamerica).

Genus Apanteles Förster, 1862
Th is is one of the largest genera of Microgastrinae, with 35 species (32 of them described) recorded from the region. Th ere are hundreds of unidentifi ed specimens in the CNC and other collections, and the genus will have many more species when further studies can be carried out. Th ree species are left undescribed here until more studies of the Holarctic species allow establishing their identities with more accuracy. Th e only comprehensive key to the Nearctic species is in , a relatively useful although outdated work. Mason (1974) and Fernández-Triana and Huber (2010) provided keys to species parasitizing tortricids -but those papers are missing many other species attacking a diff erent spectrum of Lepidoptera. In his review of Apanteles sensu lato Papp (1976) and subsequent papers provided keys to the European species, and Chen and Song (2004) also provided a key to Chinese species. Th e genus badly needs a comprehensive review and, probably as importantly, a clarifi cation of its current limits, a controversial and unsolved problem (e.g. Mason 1981;Whitfi eld 1995Whitfi eld , 1997van Achterberg 2002b).
Even from a detailed morphological redescription of the species (Mason 1974) it could be inferred that the species belongs to Dolichogenidea. However, I have carefully examined the holotype and found that the vannal lobe is medially fl attened and with minute, sparse setae; a character that would put the species under the genus Apanteles. As stated before, our present understanding of those genera is far from complete; but pending more studies to clarify or improve the boundaries between them I prefer to keep the traditional treatment of the species as See more comments of its distribution under the species description above. Apanteles sodalis (Haliday, 1834). BC, NB, NL. Distributed in the HOL, introduced accidentally to Canada (Mason 1978). Apanteles starki Mason, 1960. AB, BC. Distributed in WNA and China. Apanteles victoriae Muesebeck, 1921. BC. Apanteles xanthostigma (Haliday, 1834). NL. Distributed in the PAL and with two references from Uganda (Yu et al. 2009). Introduced in Canada (Williamson 1963), though there is no published data about its establishment. Apanteles sp. 1 near nephoptericis. ON. Four specimens in CNC. Most likely it is a new species but, pending further study of the Holarctic fauna of Apanteles, it is not described in this paper. Apanteles sp. 2 near plesius. QC. A recent paper (Fernández-Triana et al. 2009b) recorded three specimens from Frelishburg, QC, as a diff erent but related species. Apanteles sp. 3 near pseudoglossae. QC. A recent paper (Fernández-Triana et al. 2009b) recorded one specimen from Frelishburg, QC, as A. pseudoglossae Muesebeck, 1921, which would represent a new record of the species for Canada. After checking the specimen and comparing it with other Nearctic material I now consider it a diff erent species. Th us, the known distribution of A. pseudoglossae at present does not extend to Canada -as stated by those authors-but it is restricted to IL, MD, MI and MN in the US.

Genus Choeras Mason, 1981
Th ere are keys to some Palearctic species (e.g. Tobias 1986; van Achterberg 2002a; Chen and Song 2004;Kotenko 2007), although the genus has never been properly revised. Th ree species are known from the Nearctic region (Whitfi eld 1995) and all of them occur in Canada. Additionally, in the CNC there are many specimens from western Canada that most likely represent several species; but lacking comprehensive studies of the genus I am taking the conservative approach of keeping them as belonging to one species for now.

Genus Clarkinella Mason, 1981
Th ere are only two described species from this genus, one of them from Canada (Mason 1981). I do not expect more species from the region.
Clarkinella canadensis Mason, 1981. ON. Known previously only from holotype, two additional specimens were recently collected in Ottawa.

Genus Cotesia Cameron, 1891
Th is is probably the largest genus in the region. It is also one of the most cohesive taxa within Microgastrinae (Mason 1981), although sometimes specimens of Protapanteles (see below on remarks of that genus) could be confused with Cotesia. Th e keys in  and Papp (1990) will only work partially for identifying specimens, because a plethora of new taxa remain undescribed. Here 55 species are recorded for the region studied (51 of them described). At least four undescribed species are also mentioned but lack of a comprehensive study on the Holarctic fauna prevents me to formally describe them here. Additionally, thousands of specimens in collections remain unidentifi ed; some of them likely represent many additional species.
Cotesia acauda (Provancher, 1886). NS, ON, QC. Distributed in the ENA. Cotesia acronyctae (Riley, 1871). AB, ON, SK. Distributed in the NEA. Cotesia anisotae (Muesebeck, 1921  Cotesia tmetocerae (Muesebeck, 1921). NS. Cotesia xylina (Say, 1836). AB, MB, NS, ON, QC. Distributed in the NEA. Whether this species is valid or not has been questioned by Muesebeck (1921) and Papp (1986) who considered it a synonym of C. yakutatensis or C. tibialis (Curtis) respectively. Some of the Canadian specimens seem closely related to those of C. yakutatensis. However, when barcoded specimens from Canadian C. xylina, C. yakutatensis and related specimens are analyzed, 5 distinct clusters are obtained ( Fig.  25; Supplementary Appendix 2). Based on the keys of Muesebeck (1921), Papp (1986Papp ( , 1987Papp ( , 1989 and van Achterberg (2006), some of those specimens might be C. halli, C. melanoscela, C. eliniae, C. tetricus (the last two not yet recorded from the Nearctic), or just new species. Th e only way to solve this species-complex would be a study of the genus at Holarctic level, which is beyond this paper. Cotesia yakutatensis (Ashmead, 1902). AK, BC, MB, QC. Distributed in the NEA. See comments under C. xylina. Cotesia sp. 1. MB. Th is species is treated as Cotesia jft01 in a paper currently under review (Fernández-Triana et al. unpublished). Th e only available specimen, a male from Burnt Site, Churchill, MB, runs to C. nemoriae in Muesebeck (1921) and to C. subordinanius in Papp (1986), but it is neither of those species. Its very large metacoxae (half the length of the metasoma) are very distinctive and seem to support its status as a new species. Cotesia sp. 2. MB, NL, NU, PE, QC, YT. Th is species is treated as Cotesia jft09 in a paper currently under review (Fernández-Triana et al. unpublished). Additional specimens, mostly from northern localities, have been found later in several Canadian provinces and territories. It is related to Holarctic species with short antennae -e.g. C. arctica (Th ompson), C. astrarches (Marshall, 1889), and C. tenebrosa (Wesmael) -but differs from all of the species keyed by Papp (1976). Most likely it is a new species but, pending further study of the Holarctic fauna, it is not described in this paper. Cotesia sp. 3. AB, MB, ON, SK, YT. Th e specimens grouped here (in the CNC) comprise those near C. xylina and/or C. yakutatensis that are still unresolved but are clearly diff erent species ( Fig. 25; see also comments under C. xylina). I am taking the conservative approach of considering all those specimens as belonging to one species for now -though they most likely represent several. Cotesia sp 4. ON. Over 30 specimens reared from Plutella xylostella in Ottawa. It is none of the known species of Cotesia parasitizing Plutella, and most likely represents a new species. It is not described here because of the same reason than previous species.

Genus Deuterixys Mason, 1981
Th is is a small genus and the Nearctic species were revised by Whitfi eld (1985). No additional species are expected to be represented in the region.

Genus Diolcogaster Ashmead, 1900
At least 10 species (7 of them described) are recorded here; however, it is clear that the actual number of species is much higher. For example: based only on the specimens of Diolcogaster with barcode sequences currently available (135 specimens, 122 of them with more than 500 bp) there are more than 17 clearly delimited species in the region, even if a very conservative approach is taken ( Fig. 26; Supplementary Appendix 2). No further eff orts are made here to deal with those specimens because there is a pending taxonomic review at Nearctic level (Choi WY and Whitfi eld JB, unpublished) that should improve our present understanding of the genus.
Diolcogaster auripes (Provancher, 1886) NB, ON, QC. Distributed in the ENA and CNA. Diolcogaster bakeri   . BC, ON. Distributed in the NEA. Diolcogaster scotica (Marshall, 1885). BC, QC. Distributed in the HOL. Diolcogaster sp. 1. MB. Th is species is treated as Diolcogaster jft01 in a paper currently under review (Fernández-Triana et al. unpublished). It does not key to any described species, and the barcoding data suggests it is one of the most distinctive and unique species within the genus. Diolcogaster sp 2. AK. Th is is the fi rst record of the genus for Alaska. Th e specimen, collected on July, 1959 in Umiat (69°22'N, 152°09'W) and housed in the CNC is very distinctive and also represents the northernmost known record of the genus. Diolcogaster sp. 3. NT. One specimen, almost as far north as the previous species, was collected on July, 1971 at Kovaluk River (69°11'N, 131°00'W) and housed in the CNC. However, it represents a diff erent species.

Genus Distatrix Mason, 1981
Th is is a predominantly tropical genus, and a revision of the New World species was published recently (Grinter et al. 2009). I have found a new species reaching as north as QC, Canada, and representing the northernmost known record.

Genus Dolichogenidea Viereck, 1911
Th e keys dealing with Apanteles (mentioned above in the treatment of that genus) also cover the species of Dolichogenidea. Both genera are easily confused (Mason 1981;Whitfi eld et al. 2009), and the correct generic assignment of some of the species is often limited by subtle and subjective characters. A comprehensive study of those genera (plus a few others, see for example, van Achterberg 2002b) is badly needed to redefi ne its limits with accuracy. In the meantime, 19 species of Dolichogenidea (17 of them described) are recorded here for the region studied. Muesebeck, 1965. AB, BC, NB, NL, NS, MB, ON, PE, QC. Distributed in the NEA. Dolichogenidea breviventris (Ratzeburg, 1848). Previously recorded as D. mesoxantha (Ruschk a, 1971) by Whitfi eld (1995) in his list of Nearctic species. However, D. mesoxantha was synomized under D. breviventris by Papp (1978). NL. Distributed in the HOL. Dolichogenidea cacoeciae (Riley, 1881). ON, QC. Distributed in the NEA. Dolichogenidea californica (Muesebeck, 1921 (Muesebeck, 1933). SK. Distributed in the NEA and Cuba. Dolichogenidea lacteicolor (Viereck, 1911). NB, NS, QC. Distributed in the HOL and also some records in the Oriental region. Th is species was transferred to Apanteles by van Achterberg (2002b) but I am keeping it as Dolichogenidea based on its vannal lobe medially convex and uniformly covered by setae of the same length. Dolichogenidea laspeyresiae (Viereck, 1913). BC. Distributed in the WNA. Dolichogenidea longicauda Wesmael (1837). BC. Distributed in the HOL. Recently Fernández-Triana and Huber (2010) commented on the changing generic status of this species within the last years, and considered it as belonging to Dolichogenidea, a decision that is followed here. Dolichogenidea melanopa . BC, PE. Previously known from Connecticut. First record to Canada. Dolichogenidea paralechiae (Muesebeck, 1932). NB, ON, QC. Distributed in the NEA. Dolichogenidea phthorimaeae (Muesebeck, 1921 (Muesebeck, 1935). ON, QC. Distributed in the ENA. Dolichogenidea tischeriae (Viereck, 1912). QC. Distributed in the NEA. A recent paper (Fernández-Triana et al. 2009b) recorded the species for the fi rst time in Canada, but erroneously mentioned it as Apanteles tischeriae. Th e species was transferred to Dolichogenidea by Mason (1981

Genus Exix Mason, 1981
Th is genus has one Nearctic species (Mason 1981), known from only one specimen in Canada.

Genus Exoryza Mason, 1981
Like the previous genus, only one species is known from the Nearctic (Mason 1981), with no more species seen in collections (Whitfi eld 1995).

Genus Glyptapanteles Ashmead, 1904
This genus is considered one of the most diverse and dominant genera in tropical regions (e.g. Whitfield 1995;Whitfield et al. 2009), but is still very commonly seen and specious in Alaska/Canada, reaching even to the tip of Ellesmere Island (+82°N). Papers covering Apanteles (mentioned above under that genus) will deal with some species, but the limits of the genus are controversial (e.g. van Achterberg 2002b; and discussion under Protapanteles below) and there are no updated keys to the species. Many unidentified specimens remain in collections and the recorded species here (10, nine of them described) are just a fraction of the actual number.
Glyptapanteles alticola (Ashmead, 1902). AK, BC, NB. Distributed in the NEA. From the material housed in the CNC, Papp's (1983) statement that G. alticola is not diff erent from G. fulvipes (Haliday) seems valid. However, both species are kept here as valid until the type material can be studied. Glyptapanteles compressiventris (Muesebeck, 1921). MB, NT, NU, QC. Distributed in the HOL. Glyptapanteles fl avovariatus (Muesebeck, 1921 Papp (1988), an arrangement that has been accepted by subsequent authors (e.g. Yu et al. 2005;Kotenko 2007). Glyptapanteles sarrothripae (Weed, 1887). BC, NS, ON. Distributed in the ENA, here recorded for the fi rst time for WNA. Glyptapanteles websteri (Muesebeck, 1921). AB, NB, QC. Distributed in the ENA. Glyptapanteles sp. 1 near alticola. MB. Specimens from Manitoba form around half a dozen of distinct clusters based in barcoding data that might well represent different species related to G. alticola (Fernández-Triana et al. unpublished data). Th e barcoding diff erences are also supported by slight morphological diff erences (e.g. antennae colour, relative length of the last fl agellomeres, puncture density of head, seta density and length on the mesoscutum, scutellum punctures, wing base colour, propodeal carination, hind leg colour -especially tibia and tarsi-sculpture of mediotergites 1 and 2). However, without having molecular data from the type material and/or a comprehensive taxonomical review of the genus within the Holarctic, an unequivocal assignment of specimens is not possible at present. I am taking the conservative approach of considering all specimens as one species for now -though they probably represent many species.

Genus Hygroplitis Th omson, 1895
No more species are expected to be found within the Nearctic region (Whitfi eld 1995).

Genus Hypomicrogaster Ashmead, 1898
Some specimens in the CNC seem to be diff erent species than the two described species here recorded for the region. Th ere is currently a review of the genus underway for the New World fauna (Valerio A, pers. com.) and thus no further attempt is made here to deal with those unidentifi ed specimens.
Hypomicrogaster ecdytolophae . NS, ON, QC. Distributed in the NEA and the NEO.
Hypomicrogaster zonaria (Say, 1836). NB, NS, ON, QC. Distributed in the NEA Hypomicrogaster sp. ON. Th is is a new species that will be described elsewhere (Valerio A, pers. com.).

Genus Iconella Mason, 1981
Th is genus has never been revised, though key to some Palearctic species can be found in Chen and Song (2004) and Kotenko (2007). Th ere is only one described species within the Nearctic (from US and the Neotropics) but none was previously recorded from Canada. I have found at least two species in Canada but, pending a study of the Holarctic fauna, the species are not described here.

Genus Illidops Mason, 1981
Th is genus has never been revised, though Kotenko (2007) provided a key to the Russian Far East. Th ere is only one described species within the Nearctic (from southern US) and a few undescribed have been mentioned from Arctic Canada and the Rocky Mountains (Mason 1981). Some of the Canadian species lack the posterior medioapically desclerotized tergites and/or the convergent eyes that characterize the genus. Th e majority of the specimens available were collected more than 30 years ago, and only one recent specimen rendered a full barcode, but for 18 older specimen partial barcodes (120-292 bp) were obtained. Both morphology studies and barcoding indicate that, even under the most conservative approach, there are at least four species in the studied region. Pending a larger study of the Holarctic fauna, the species are not described here.
Illidops sp. 1. MB. Th is species is treated as Illidops jft01 in a paper currently under review (Fernández-Triana et al. unpublished). Illidops sp. 2. MB. Th is species is treated as Illidops jft02 in a paper currently under review (Fernández-Triana et al. unpublished). Illidops sp. 3. MB. Th is species is treated as Illidops jft03 in a paper currently under review (Fernández-Triana et al. unpublished). Illidops sp. 4. AB, MB, NS, NT, ON, QC. Additional material from the provinces mentioned here is housed in the CNC. Th ey are diff erent from the previous three species, and probably represent more than one species, but for now are kept provisionally as one species.

Genus Lathrapanteles Williams, 1985
Th is genus was described and its species revised by Williams (1985) and no new additions are expected. Th e validity and relationships of this genus to other Microgastrinae might be questioned when future studies are made.

Genus Pholetesor Mason, 1981
Th e Nearctic species were revised by Whitfi eld (2006) and the genus is reasonably covered. However, Palearctic species need to be dealt with altogether with the Nearctic ones to avoid duplication of descriptions. For that reason, I am treating two of the 20 species recorded here as undescribed for now.
Pholetesor sp. 2. MB. Th is species is treated as Pholetesor jft02 in a paper currently under review (Fernández-Triana et al. unpublished). Two male specimens (Sample ID: 07PROBE-22417, 07PROBE-23399) diff er slightly from P. viminetorum regarding veins r and 2RS, length of metatibial spurs and shape of tergite 1 and 2. Th e barcode variation between these two species was 1.94%, and there are also two character states diff erences within the D2 region of the nuclear gene 28S. Th e combination of these three lines of evidence suggests that those males are a separate species from P. viminetorum. However, pending a study of the Holarctic fauna, the species is not described here.

Genus Protapanteles Ashmead, 1898
Altogether with Apanteles, the limits of this genus are one of the most controversial (e.g. Mason 1981;Whitfi eld 1995Whitfi eld , 1997van Achterberg 2002b;Yu et al. 2005). Mason (1981) provided some characters that supposedly defi ned the genera, but even within a geographical restricted area such as Canada, there are considerable variation (e.g. specimens with propodeum sculptured like a typical Cotesia, instead of smooth; mediotergite 1 strongly narrowing toward apex like typical Glyptapanteles; specimens looking like Sathon; etc). Th e North American workers have usually considered it to be a rather small genus, and have kept the other genera separated, but I am not sure if that is the best arrangement, or at least Glyptapanteles and Sathon should be part of an expanded Protapanteles genus -similar to the proposal of van Achterberg (2002b). Solving those problems is beyond the scope of this paper and for now I am following Whitfi eld's (1995) arrangement of the Nearctic species.
Protapanteles alaskensis (Ashmead, 1902). AK, BC, MB, NL. Distributed in the NEA. Protapanteles paleacritae (Riley, 1881). BC, MB, NL, NS, ON. Distributed in the NEA. Protapanteles phigaliae (Muesebeck, 1919). NB, ON. Distributed in the NEA. Protapanteles phlyctaeniae (Muesebeck, 1929). ON. Distributed in the ENA and CNA. Protapanteles sp. 1. AB, BC, SK, MB. A signifi cant number of specimens from western Canada is included here, most of them from reared material housed in the CNC and the Northern Forestry Centre, Edmonton. I am taking the conservative approach of considering all specimens as one species for now -though they likely represent several species

Genus Protomicroplitis Ashmead, 1898
Th is small genus had been reported within the Nearctic from central and eastern US, as north as NY (Yu et al. 2005). In the CNC there are several specimens of P. calliptera captured near Ottawa (Metcalfe and Stittsville) which represent the fi rst record of the genus for Canada and the northernmost distribution in North America. No more species are expected to be found in the region.

Genus Pseudapanteles Ashmead, 1898
Th is New World genus is mostly found in the tropics, with a few species reaching to the US. Here I record two species (one of them new) for Canada, expanding further north the known distribution of the genus.

Genus Rasivalva Mason, 1981
All the Nearctic species of this genus are dealt with in , and Whitfi eld (1995) stated that no clearly undescribed species had been seen in collections. I report here an additional undescribed Canadian species that does not fi t within the described ones.
Rasivalva perplexa . BC, NB, ON. Distributed in the NEA. Rasivalva rugosa . ON, QC. Distributed in the NEA. New record to Canada. Rasivalva stigmatica . AB, BC, QC. Distributed in the NEA. New record to Canada. Rasivalva sp. ON, NB. Two female specimens in the CNC that are diff erent to the previous species but that are not dealt with further until a study of the Holarctic fauna is done.

Genus Sathon Mason, 1981
Th e limits of this genus are controversial (see above comments under Protapanteles and also Whitfi eld et al. 2002Whitfi eld et al. , 2009). Williams (1985) provided a key to species.

Genus Venanides Mason, 1981
A small genus with four described species, one of them from the Nearctic (Mason 1981); with an additional couple of undescribed ones in southern US (Whitfi eld 1995). It is not likely that more species will be found in Canada/Alaska. (Mason, 1981). MB, ON. Distributed in the NEA and Brazil.

Genus Venanus Mason, 1981
Th e genus was recently revised by Whitfi eld et al. (in press), but from the seven recognized species within the New World, only one was recorded from the Nearctic. Th e barcoding data revealed a new species that had been overlooked by Mason (1981); it is clear now that the specimens from eastern and western Canada are diff erent (see more comments on the two species in the section describing the new species).