A remarkable new species of Himalusa Pace from Thailand (Coleoptera, Staphylinidae, Aleocharinae): phytophagous aleocharine beetle with potential for bio-control of skunkvine-related weeds in the United States

A new aleocharine species, Himalusa thailandensis Pace, Klimaszewski & Center sp. n., found feeding on sewer vine, Paederia pilifera Hook. f., in Th ailand, is described and illustrated. Th is is the fi rst record of the genus Himalusa for Th ailand. Himalusa was previously known from the Himalayan region in Nepal where it was originally described. New data on bionomics and distribution are provided, including a discussion on the potential use of this species in the biological control of weeds in the United States. A short diagnosis, description, colour habitus images of adult and larva, and black and white genital images are provided. A key to the world species of Himalusa is also given. A new tribe Himalusini Klimaszewski, Pace & Center, is erected to accommodate Himalusa species. ZooKeys 35: 1–12 (2010) doi: 10.3897/zookeys.35.329 www.pensoftonline.net/zookeys Copyright Her Majesty the Queen in the Right of Canada. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. RESEARCH ARTICLE Launched to accelerate biodiversity research A peer-reviewed open-access journal


Introduction
Himalusa Pace (2006) was previously known from a single species, H. annapurnensis Pace, described from the Annapurna region of the Himalayas, Nepal. Recently, an additional undescribed species of the same genus was discovered in Th ailand. Th is constitutes the fi rst generic records of Himalusa in Th ailand. Adults of the new species were observed feeding on Paederia pilifera Hook. f. (Rubiales: Rubiaceae), by gnawing little holes in plant leaves and scraping the plant tissue using their mandibles. Th is may be the fi rst confi rmed record of Staphylinidae/Aleocharinae species feeding on live plant tissues. Most aleocharine species are predators of other small arthropods. Th e purpose of this paper is to describe a new species of Himalusa from Th ailand, and to enhance further studies of this beetle, which has the potential to be a biological control agent of Paederia foetida L. and P. cruddasiana Prain, invasive environmental weeds in the southern United States.

Material examined
Sixteen adults of Himalusa representing a new species were found in Th ailand. All specimens were dissected. Th e genital structures were dehydrated in absolute alcohol and mounted in Canada balsam on celluloid microslides and pinned with the specimens from which they originated. Th e photographs of the body images (Figs 1,2,17) and the genital structures were taken using an image processing system (Nikon SMZ 1500 stereoscopic microscope; Nikon digital camera DXM 1200F; Nikon View 5 COOLPIX NSA, Version 5.1.2, and Adobe Photoshop software). Th e colour images of entire beetles  were generated using an image processing system consisting of a Wild M420 stereomicroscope (Leica), SPOT™ RT™ Slider camera (Diagnostic Instruments) and Adobe Photoshop software.
Comments. Th e specimens from Prachuap Kin Kan Province were excluded from the paratype series of Himalusa thailandensis on the grounds of smaller body (length 1.9-2.0 mm) and slight diff erences in the median lobe of aedeagus and spermatheca (Figs. 20,21,25). We consider these diff erences as infraspecifi c variations. Should further molecular studies prove the opposite, the status of this form should be reconsidered.

Collection and habitat data
Adults were captured in July (by type material) and April (by non-type material) feeding on skunkvine -Paederia pilifera Hook. f. (Rubiales: Rubiaceae). Closely related species are invasive weeds in the southern United States.
Th e adult beetles feed on the upper leaf surfaces of P. pilifera, parallel to a leaf vein, most often the mid-vein, where they excavate oblong holes completely through the leaf (Figs. 26,27). Th ese holes conform to the size of their bodies. Th e adults often nestle within them and, when disturbed, exit through to the underside of the leaf. Th ey also scrape the lateral laminar surfaces causing an irregular skeletonized pattern. Th e leaf tissues adjacent to the feeding sites blacken and necrosis follows. Th is produces signifi cant levels of foliar damage. Th e larvae burrow and feed within the leaf petioles. As they grow, the petioles swell and become almost gall-like. At some point, the petiole splits and the larva drops out. Th e images show a swollen petiole that contains a larva (Fig. 28), and the larva that emerged from it (Fig. 29). Pupation probably occurs in the soil or under leaf litter. Other aspects of their biology are unknown and merit further investigation.

Introduction
Th e Cryptophaginae is the nominate subfamily of the Cryptophagidae (the silken fungus beetles). Most species feed on fungal hyphae, spores, and conidia while others are saprophagous. Th ey are typically found in decaying organic materials that are colonized by fungi. Some species of Cryptophagus are found on a variety of stored products. Members of the genus Antherophagus are phoretic on bumble bees (Bombus spp.) whereas species in the genus Telmatophilus are associated with the fl owers of various aquatic plants (Leschen and Skelley 2002). Leschen and Skelley (2002) included 53 species in nine genera in the North American fauna. In Canada,  listed 42 species, although only 7 of these were recorded from Atlantic Canada (the provinces of New Brunswick, Nova Scotia, Prince Edward Island, and Newfoundland and Labrador). Th e present study contributes further records of species of this family from the region, provides keys to identifi cation of species in the genera Antherophagus, Cryptophagus, Henotiderus, and Telmatophilus in the region, and discusses the composition, zoogeography, and bionomics of the group in Atlantic Canada.

Methods and conventions
Specimens of Cryptophaginae originating from Atlantic Canada in a variety of collections were examined and identifi ed. Th ese collections yielded 766 specimens; 195 from Nova Scotia, 21 from New Brunswick, 49 from Prince Edward Island, and 501 from Newfoundland and Labrador. Abbreviations (largely following  of collections that contributed specimens or data to this study are: 1. Length Th ere are variations in character states (such as the position of the lateral tooth on the pronotum), and in some species the adeagi exhibit few distinctive features. Some character states, such as the shape of the lateral margins, are based on very subtle diff erences.
Where quantitative measurements are used to help distinguish species, it is likely that some specimens will fall outside the normal range. As a result there are some species that "… are found to be so nearly indistinguishable that their correct identifi cation on external characters becomes more of an art than a science." (Woodroff e and Coombs 1961: 183).
Th e keys, detailed descriptions, and illustrations of external anatomy and genitalia pro-

Caenoscelis.
Th e six species of Caenoscelis Th omson found in northeastern North America can be identifi ed with the key provided in : 1010-1011.

Results
Eighteen species

Antherophagus convexulus LeConte, 1863
Antherophagus convexulus was reported from both New Brunswick and Nova Scotia by . Th ere are a number of records from both the mainland of Nova Scotia and Cape Breton Island ( Fig. 1), however, we have not been able to locate voucher specimens or published records of this species from New Brunswick. We provisionally retain this species for the New Brunswick fauna (subject to confi rmation) since it is also known from the neighbouring jurisdictions of Québec and Maine (Procter 1946;Laplante et al. 1991), and is to be expected in the province. Antherophagus species are phoretic on bumblebees (Bombus spp.) and are often found in their nests or at fl owers where their larvae apparently feed on organic detritus Leschen and Skelley 2002).

Antherophagus ochraceus Melsheimer, 1844
Antherophagus ochraceus was reported from New Brunswick, Nova Scotia, Prince Edward Island, and insular Newfoundland by  and it is widely distributed in the region (Fig. 1). Antherophagus species are associated with bumblebees (Bombus spp.) (see A. convexulus). Blatchley (1910)   Cryptophagus acutangulus is newly recorded in Nova Scotia, on both the mainland of the province and Cape Breton Island. It was reported previously from insular Newfoundland by   (Fig. 4). It is probably a Holarctic species, generally distributed in North America (Woodroff e and Coombs 1961). Woodroff e and Coombs (1961) noted that in Europe this species occurs on stored produce and it is possible that its North American distribution may have been infl uenced by commercial activities. In the Palaearctic region it is found throughout Europe, in North Africa, across the Middle East and Caucasus, through Central Asia, the Russian Far East, south to Japan, North Korea, and China. It is also found in sub-Saharan Africa and the Neotropical Region ). It occurs both on stored products and in wild habitats (Woodroff e and Coombs 1961). A very distinctive species on account of the strongly produced lateral pronotal callosities.  Lake, August 30, 1977, D. Larson (5, MUN). Cryptophagus diffi cilis is newly recorded in insular Newfoundland, Labrador and in Atlantic Canada (Figs. 3,6). In Canada, previous records of the species are from British Columbia, Alberta, the Yukon, and Québec; in the United States it has been found from Idaho to New Jersey, and south to Georgia (Woodroff e and Coombs 1961). In British Columbia it has been collected from fungus (Woodroff e and Coombs 1961). It is a fl ightless species, and in most of the specimens examined from this region, the wings were vestigial (although one specimen had reduced wings equal to approximately half the length of the abdomen). Cryptophagus fallax is newly recorded in New Brunswick (Fig. 2). Previous North American records are from Maine (Woodroff e and Coombs 1961), Sable Island, Nova Scotia (Howden 1970), and Prince Edward Island (CAIPR 1973). Th e above record from Halifax also establishes the presence of this species on the mainland of Nova Scotia. It is an adventive species in North America and is found in the Palaearctic region throughout Europe, east through the Caucuses to Kazakhstan and Uzbekistan, and south to Egypt ). All the specimens collected in this region have been found in buildings on stored products except for two collected on Sable Island in an Ipswich Sparrow (Passerculus sandwichensis princeps Maynard) nest (Howden 1970). Th e narrow quadrate pronotum (compared to the enlarged, convex elytra), and the large pronotal callosities with their face turned upwards, make this a distinctive species amongst the North American fauna (Woodroff e and Coombs 1961).

Cryptophagus histricus Casey, 1900
Cryptophagus histricus was reported by Woodroff e and Coombs (1961) from a bumblebees' nest in Tallik Nagvak Fjord in northern Labrador (Fig. 6). Th is is the only record of this Nearctic species in the region. It is otherwise recorded from scattered states in   (Fig. 2). Nearctic specimens of this species were previously known under the name of Cryptophagus confertus (synonymized by Johnson et al. 2007. It is a Holarctic species found across Canada and the northern United States, in Europe in the Alps and Scandinavia, across Siberia to the Russian Far East, and south   Lohse (1967) describes it as a boreo-alpine species. Some specimens of this species can be separated from Cryptophagus bidentatus Mäklin on the basis of external morphology only with great diffi culty. Th e adeagi of both species are indistinguishable, however, the shape of the parameres of the two species are quite distinct (Woodroff e and Coombs 1961). Cryptophagus laticollis is newly recorded in Nova Scotia and eastern North America (Fig. 2). It is an adventive Palaearctic species, previously recorded in North America from the west coast in British Columbia, Alberta, California, Oregon, and Washington (Woodroff e and Coombs 1961;. It is a cosmopolitan species found throughout the Palaearctic in Europe, North Africa, and Asia, and in Australia, sub-Saharan Africa, and the Orient ). It is found on stored products and in vegetable refuse (Woodroff e and Coombs 1961). Th e small conical eyes of C. laticollis distinguish this species from the similar Cryptophagus setulosus (Woodroff e and Coombs 1961). Cryptophagus mainensis is newly recorded in insular Newfoundland, Nova Scotia, and Canada as a whole (Fig. 3). Th is is a Nearctic species previously known only from Maine, New Hampshire, New York, and Ohio (Woodroff e and Coombs 1961;. In Nova Scotia specimens were collected in red spruce (Picea rubens Sarg., Pinaceae) and mixed eastern hemlock (Tsuga canadensis (L.) Carr., Pinaceae) and red spruce forests varying in age from old growth (120+ years) to 45 years. Th is is the fi rst habitat information about this species. It is a very distinctive species on account of its small size (1.5-1.7 mm.) and very large pronotal callosities which subtend 1/3 of the lateral margin of the pronotum. Cryptophagus scanicus was reported in North America by Woodroff e and Coombs (1961) from the fi ve specimens collected in St. John's in 1949 by W.J. Brown. Additional records are given indicating that the species is established and occurs more widely in southern Newfoundland (Fig. 3). It is an adventive Palaearctic species found throughout Europe, in parts of North Africa, across Siberia to the Russian Far East, and in Turkey and Iran in the Middle East. It is also found in sub-Saharan Africa ). In Europe, it is found in various habitats including stored produce (Woodroff e and Coombs 1961). Cryptophagus setulosus is newly recorded on insular Newfoundland (Fig. 4). It was previously recorded by Woodroff e and Coombs (1961) from one specimen collected in Labrador. Additional records above indicate that it is abundant and widespread in southern Labrador (Fig. 6). Th ere are scattered records of this species from a number of Canadian and American provinces and states. Woodroff e and Coombs (1961) refer to it as a possibly Holarctic species, while  lists it as Holarctic. Woodroff e and Coombs (1961) noted that in Europe this species occurs on stored produce and it is possible that its North American distribution may have been infl uenced by commercial activities. In the Palaearctic region it is found throughout Europe, in Algeria and Morocco in North Africa, and in Turkey and Central Asia north to Eastern Siberia . In Europe it is most frequently found in nests of bees and wasps, but also in fungi and on stored produce (Woodroff e and Coombs 1961). Cryptophagus tuberculosus is newly recorded on insular Newfoundland, in Nova Scotia, and in Atlantic Canada (Fig. 3). Th is Holarctic species has previously recorded in North America from Alaska, Alberta, Arizona, British Columbia, California, Colorado, Idaho, New Mexico, Oregon, Utah, Washington, and Wyoming in western North America (Woodroff e and Coombs 1961), and in Maine and New Hampshire in eastern North America . In the Palaearctic region it is found in Scandinavia and northern Russia, east to the Russian Far East, Mongolia, and Inner Mongolia . Woodroff e and Coombs (1961) reported it from coniferous trees, nests, and stored produce (especially dried fruit). In Nova Scotia it has been found on dried corn and in a coniferous forest. Th is is a highly variable, polytypic species which can be diffi cult to separate from Cryptophagus subfumatus Kraatz. Th e shape of the parameres, however, are quite distinctive (Woodroff e and Coombs 1961).

Cryptophagus tuberculosus
Th e occurrence of C. tuberculosus in eastern North America is noteworthy. Woodroff e and Coombs (1961) drew attention to the wide Holarctic distribution of this species and its apparent absence in eastern North America. Th ey also noted three "forms" of this species in western North America and suggested that the species was rather easily isolated by geographical or ecological barriers. Th ey further proposed that the species may have dispersed to North America at a time when a land bridge existed across the Bering Strait. Th erefore, the occurrence of C. tuberculosus in eastern North America may suggest that this species has a typical Holarctic distribution. Henoticus serratus is newly recorded in insular Newfoundland and Nova Scotia (Fig. 1).  reported the species from Labrador, although it is not clear where this record originates.  reported the species from Alaska and British Columbia east to Québec, and  reported records from a variety of eastern and western states in the USA. In the Palearctic region this Holarctic species is found throughout Europe (except for the Iberian peninsula, the southern Balkans and the Mediterranean islands) southeast to Georgia, and east across Siberia to the Russian Far East ). All the specimens collected in Atlantic Canada are from wild habitats such as red spruce and red oak (Quercus rubra L., Fagaceae) forests. Blatchley (1910) reported it from dry fungi about the stumps of oaks.  noted that adults in the genus Henoticus are usually found in leaf litter, fungi, under the bark of dead or dying trees, and on leaves of trees or shrubs.

Henotiderus centromaculatus Reitter, 1877
Henotiderus centromaculatus was recorded from Nova Scotia by   (as H. obesulus). Th ere are scattered records from the central part of mainland Nova Scotia ( Fig. 2). Th is Nearctic species has been recorded from Alaska and across Canada from the Northwest Territories east to Nova Scotia, and in New Hampshire and New York . It is also found in the Eastern Palaearctic in China, Japan, and the Russian Far East ). Species of Henotiderus are found in leaf litter or associated with polypore fungi such as Polyporus, Pleurotus, and Fomes in forested habitats . All the specimens collected in Nova Scotia were found in red spruce forests, varying in age from young to old growth, except for one specimen collected in a mixed red spruce-eastern hemlock forest (unpublished data). Pteryngium crenatum is newly recorded in Nova Scotia and Atlantic Canada (Fig. 2). Th is species has previously been recorded in North America from British Columbia, Indiana, New Hampshire, Oregon, Québec, and Washington . It is an adventive Palaearctic species found throughout Europe ). In Nova Scotia, all the specimens were collected in coniferous forests, most on bracket fungi (Polyporaceae). Telmatophilus americanus is newly recorded in insular Newfoundland, New Brunswick and Nova Scotia (Fig. 1). It has previously been recorded in Canada from British Columbia to Québec, south to Colorado, Indiana, and New York in the United States . It is found on the fl owers of Sparganium spp. (Sparganiaceae) Leshen and Skelley 2002). Blatchley (1910)  Telmatophilus typhae was reported in North America from New Brunswick, Nova Scotia, and Prince Edward Island by Hoebeke and Wheeler (2000) (Fig. 5). Th e specimens collected by Hoebeke and Wheeler (2000) were from 1995 and 1997. It is an adventive Palaearctic species known in North America only from these three provinces. Th e above records establish the presence of this species in Nova Scotia, and hence North America, from as early as 1986. In the Palaearctic region it is found throughout Europe and in the Russian Far East, Japan, North Korea, and China. In North America it is found on the male fl owers of cattails, Typha latifolia L. and Typha angustifolia L. (Typhaceae). Caenoscelis basalis is newly recorded in Newfoundland, Nova Scotia, and Atlantic Canada (Fig. 4). Previous records of this Holarctic species are from Maine, New Hampshire, New York, and Québec . Specimens in Nova Scotia were collected in red spruce forests, and in Newfoundland it was found in a balsam fi r (Abies balsamaea (L.) Mill, Pinaceae) forest. Members of this genus are found in leaf litter, organic matter, nests of small mammals, and beneath the bark of logs ).

Discussion
Eighteen species of Cryptophaginae are now known to occur in Atlantic Canada. Eight of these including Cryptophagus diffi cilis, C. jakowlewi, C. laticollis, C. mainensis, C.

Labrador
Although collecting eff ort for this group in the region is still far from adequate, some preliminary observations can be made. Cryptophagus diffi cilis and C. setulosus are abundant species in insular Newfoundland and Labrador (particularly in the latter region), and are completely absent from the Maritime Provinces of Canada (the provinces of New Brunswick, Nova Scotia, and Prince Edward Island). In general, species of Cryptophagus are rather seldom collected in the Maritimes, yet the above two species are abundant and frequently collected in Newfoundland and Labrador; apparently refl ective of diff erences between the forest micro-habitats of the Atlantic Maritime Ecozone on the one hand, and the Boreal Shield and Taiga Shield Ecozones characteristic of Newfoundland and Labrador on the other. Cryptophagus histricus has also been recorded in Labrador, although this study did not fi nd further specimens. Conversely native species such as Antherophagus convexulus, Cryptophagus jakowlewi, and Henotiderus centromaculatus, which have been recorded from a number of localities in Nova Scotia (and Prince Edward Island in the case of C. jakowlewi), have not been found in Newfoundland and Labrador.
Although islands in general are characterized by a diminished fauna, the number of native cryptophagines recorded from Prince Edward Island (2 species; A. ochraceus and C. jakowlewi) and Cape Breton Island (3 species; A. convexulus, A ochraceus, and T. americanus) is very low. Th is may in part be due to insuffi cient collecting eff ort in these areas.
Five of the 18 species (27%) of the fauna consists of adventive species. Th e dates of earliest detection of these species are shown in Table 2, although it is likely that these species have been present (and undetected) for substantially longer. Note that two of these fi ve species were fi rst detected in North America in Atlantic Canada. Records from this region indicate that except for Cryptophagus fallax, which has been found solely in association with stored products in synanthropic situations, all four other adventive species (C. laticollis, C. scanicus, Pteryngium crenatum, and Telmatophilus typhae) are all established in wild habitats.
Th e adventive species, Telmatophilus typhae, was fi rst recorded in North America by Hoebeke and Wheeler (2000) from specimens collected in 1995 and 1997. Th e present account includes a record from 1986, pushing back the date of fi rst detection of this species on the continent by nine years. Th e very extensive distribution of this species in New Brunswick, Nova Scotia, and Prince Edward Island would appear to indicate that it has been present for much longer, or that its rate of dispersal has been very rapid. Th e fi rst author has found it to be present everywhere that he has looked  (Hoebeke and Wheeler 2000), it almost certainly is present in eastern Maine and should be looked for in that state. Pteryngium crenatum, while reported to be an adventive species in North America, has been collected in a number of jurisdictions in Canada from British Columbia to Nova Scotia, south to Indiana and New Hampshire. Records in Nova Scotia are exclusively from wild habitats, in coniferous forests associated with Polyporaceae. Given this distribution and the non-synanthropic environments where it is found, the status of this species in North America should be further investigated to ascertain if it might be a native Holarctic species.
Four native Cryptophagus species, C. histricus, C. diffi cilis, C. jakowlewi, and C. mainensis, have been recorded only in wild habitats. Cryptophagus acutangulus, C. laticollis, C. scanicus, C. setulosus, and C. tuberculosus, including both Holarctic and adventive Palaearctic species, are known from both wild habitats as well as being associated with stored products. It appears that the species in this group have considerable ecological plasticity. Th e adventive, Palaearctic Cryptophagus fallax is known primarily from stored products. All other cryptophagines found in Atlantic Canada occur (at least in part) in wild habitats.
Generally cryptophagines are ecologically diverse. Species such as Cryptophagus diffi cilis, C. mainensis, Henoticus serratus, Henotiderus centromaculatus, Pteryngium crenatum, and Caenoscelis basalis are associated with decomposing wood and fungi, and are components of the region's saproxylic fauna. Antherophagus convexulus, A. ochraceus, Cryptophagus histricus, and C. setulosus are exclusively or frequently associated with bumblebee or other Hymenoptera nests where they apparently feed on organic detritus. Telmatophilus americanus and T. typhae feed on pollen of marsh plants.
Further research is clearly indicated in order to gain a better understanding of cryptophagines in Atlantic Canada. ster for making specimens, records, and information available. Particular thanks to Colin Johnson (formerly at the Manchester Museum) for assistance with the identifi cation of some Cryptophagus specimens. Two anonymous reviewers read an earlier version of the manuscript and contributed many useful suggestions. Th e fi rst author thanks David Christianson, Calum Ewing, and Andrew Hebda at the Nova Scotia Museum for continuing support and encouragement. Th is work has been assisted by the Board of Governors of the Nova Scotia Museum. Leschen

Introduction
Th e genus Atomaria Stephens is the nominate genus of the Atomariinae, a subfamily of the family Cryptophagidae, the silken fungus beetles. Th ey are small (1.0-2.0 mm), somewhat shiny, oval-oblong, pubescent beetles, typically found in leaf litter, grass litter, decaying vegetation, and in habitats which promote mycelial fungal growth on which adults and larvae feed (Leschen and Skelley 2002).
Members of this genus are divided into two subgenera, Atomaria s. str. and Anchicera Th omson.  reported 12 species in the subgenus Anchicera and 10 in Atomaria s. str. in Canada, although none were recorded as occurring in Atlantic Canada. Poole and Gentili (1996) listed 30 species in Anchicera and 41 in Atomaria s. str. from North America (although at least eight of these names have been synonymized) while Leschen and Skelley (2002) reported that there were 10 species in Atomaria sensu lato in North America (9 described; 1 undescribed). Th is confusion, and disparity in the numbers of species, is one indication that the genus is in need of extensive revision, the last comprehensive North American treatment being by .
In the Palaearctic region, Atomaria is much better known and understood. Recent publications by Lohse (1967) and Johnson (1992) have provided a solid taxonomic foundation for the genus, and  provided a complete catalogue of the genus in the Palaearctic region.
Th e paucity of knowledge about this genus in North America has impeded the identifi cation of specimens, and limited our understanding of the historical, zoogeographic, distributional, biological, and ecological dimensions of Atomaria. Th e present study is a contribution to the knowledge of this genus in North America.

Methods and conventions
Specimens of Atomaria originating from Atlantic Canada in a variety of collections were examined and identifi ed. Th ese collections yielded 725 specimens; 425 from Nova Scotia, 49 from New Brunswick, nine from Prince Edward Island, and 242 from Newfoundland and Labrador. Abbreviations (largely following Evenhuis 2009) of collections consulted and referred to in this study are:

Results
As a result of the present investigation, ten species of Atomaria have been recorded in Atlantic Canada; seven in New Brunswick, ten in Nova Scotia, one on Prince Edward Island, fi ve on insular Newfoundland, and four in Labrador ( Atomaria apicalis is newly recorded in insular Newfoundland, New Brunswick, Nova Scotia, and Atlantic Canada (Fig. 16). We have examined the holotype of Atomaria ovalis Casey 1900 deposited in the Smithsonian Institution and have found it to  Sjöberg (1947), Lohse (1967), Johnson ( , 1971) with additional material. Note: no illustration of the adeagus of A. distincta was available. be identical in all respects to A. apicalis Erichson. We therefore designate A. ovalis syn. n. as a junior synonym of A. apicalis.
In North America this species has previously been reported from Connecticut, Indiana, Iowa, Maine, Manitoba, Massachusetts, New Hampshire, New York, Québec (Blatchley 1910;Britton 1920;Leng 1920;Procter 1946;. It is widely distributed in the Palaearctic region being found throughout Europe and North Africa, east across Siberia to the Russian Far East, across the Middle East, to Central Asia, Mongolia, and Fujian in China ). Johnson (1993) reported that it was, "a grassland species which has been recorded especially around farms, gardens and parks but often occurs in other habitats. It has been collected in man-made heaps of refuse, especially cut vegetation, grass, compost and dung heaps, fl ood debris, and more rarely on carrion and rotting fungi." Description: A variable species; head, pronotum, and elytra piceous to castaneous or (more rarely) black; elytra sometimes paler in apical 2/5. Legs, prosternum, and abdominal sterna reddish-testaceous. Pronotum weakly constricted from middle to base; from middle to apex even more strongly. Rarely, in males the basal half of the pronotum is parallel-sided, or very weakly constricted basally. Punctures of the pronotum and elytra very variable; on the base of the pronotum, interspaces between punctures ~ 1.0 times the diameter of puctures. Punctation of elytra similar to that of pronotum, or fi ner and shallower, but always less dense. Distance between furrows of prosternal process approximately half the width of the prosternal process. Body strongly convex in cross-section (Fig. 2). Antennae: antennomere 1 swollen, slightly longer than 2; 3 as long as 2 but narrower; 4-8 short and bead-like; club distinct, antennomeres 9 and 10 slightly transverse ( Fig. 1.1). Body: width/length ratio, 0.41; length, 1.3-1.6 mm. Atomaria distincta is newly recorded in New Brunswick, Nova Scotia, and Canada (Fig. 14). Th is Nearctic species has previously been reported from the District of Columbia, Massachusetts, New York, and Québec (Leng 1920;. Description: Head piceous, fi nely sparsely punctate. Pronotum piceous, fi nely sparsely punctate; pronotum widest at or slightly ahead of midpoint, not appreciably fl attened across base; fi nely sparsely punctate; interspaces between punctures 1.5-2.0 times the diameter of punctures. Scutellum black. Basal 3/5 of elytra black, becoming abruptly yellow in apical 2/5, the yellow portion extending slightly anterad in the region of the suture. Punctation very fi ne and sparse; interspaces between punctures 2.0-3.0 times the diameter of punctures; elytral setae very fi ne and short, shorter than interspace intervals and appressed. Legs, prosternum, and abdominal sterna reddishtestacous; sterna darker basally and laterally (Fig. 3). Antennae: antenomere 1 short, almost quadrate, not longer than 2; 2 slightly shorter than 1 and slightly narrower; 3 as long as 2 and slightly more slender; 4, 5 and 7 slightly shorter than 3; 6 and 8 still shorter, each approximately 1/2 the length of 3, not bead-like; antennal club distinct, antennomeres 9 and 10 almost quadrate, not noticibly transverse ( Fig. 1.2). Body: width/length ratio, 0.43; length, 1.4-1.5 mm.  Lake, June 13, 1996, July 11, 1996, July 25, 1996, pitfall, August 8, 1996; Glide Lake, July-August, 1994, C. Bassler (1, MUN). Atomaria ephippiata is newly recorded in Labrador, New Brunswick, insular Newfoundland, and Nova Scotia (Figs 13, 17). Th is Nearctic species has previously been recorded in Alberta, British Columbia, Colorado, Connecticut, Idaho, Indiana, Iowa, Maine, Manitoba, Massachusetts, New Hampshire, New York, Ontario, Oregon, Québec, Rhode Island, Washington, and Wisconsin (Blatchley 1910;Britton 1920;Leng 1920;Procter 1946;Oatman et al. 1964;Schwert 1996;Jacobs et al. 2006;Sikes 2004). It has been reported on fungi and molds in litter and nests (LaBonte 1998). In Atlantic Canada it has been found in mixed forests, coniferous forests, along forest roads, in raised bogs, salt marshes, and pastures. Jacobs et al. (2006) and Hammond et al. (2001) both reported this species as a member of saproxylic beetle assemblages on Populus tremuloides Michx (Salicaceae) in Alberta.

Atomaria fuscata
Description: Dorsum and venter usually dark red-brown castaneous, pronotum occasionally darker than elytra. Legs reddish-testaceous. Pronotal punctures moderately fi ne and dense apically, becoming progressively coarser and denser basally; pronotum widest at middle, strongly narrowing basally and apically; base with a distinct transverse impression, the middle third of the basal margin with a discernable, fi ne marginal bead. Elytral punctures very fi ne and sparse; interspaces between punctures 3.0+ times the diameter of punctures. Setae the length of the interspaces, appressed (Fig. 5). Antennae: antennomere 1 swollen, longer and wider than 2; 3 as long as or longer than 2 but more slender; 4-8 small and somewhat bead-like; antennal club distinct; antennomeres 9 and 10 quadrate, not discernibly transverse ( Fig. 1.4). Body: width/length ratio, 0.47; length, 1.4-1.7 mm. Atomaria lederi is newly recorded in North America (Fig. 14). Th e reference to North America in  refers to some of the above records. In the Palaearctic region this species has been recorded from central and northern Russia, east across Siberia to the Russian Far East, and south to Mongolia, North Korea, and northern China (Liaoning) ). In Nova Scotia it has primarily Description: Dorsum and venter reddish-testaceous throughout, including legs. Head, pronotum, and elytra moderately coarsely, moderately densely punctate; interspaces 1.0-1.5 times the diameter of punctures; elytral setae the length of the interspaces, appressed. Pronotum widest at middle, narrowing slightly to base and more strongly to apex (Fig. 6). Antennae: antennomere 1 quadrate, swollen; 2 the same length and almost the same width as 1; 3 slightly shorter and narrower; 4-8 similar in size and bead-like; club distinct; antennomeres 9 and 10 quadrate, not transverse ( Fig. 1.5). Body: width/length ratio, 0.43; length, 1.2-1.4 mm. We have examined the holotype of Atomaria curtula Casey 1900 deposited in the Smithsonian Institution and have found it to be identical in all respects to A. lewisi Reitter. We therefore designate A. curtula syn. n. as a junior synonym of A. lewisi.

Atomaria lewisi
Atomaria lewisi is newly reported in New Brunswick, Nova Scotia, Canada, and North America (Fig. 14). Under the name A. curtula it was reported from Connecticut, Ontario Leng 1920;, and Maine (Procter 1946). In the Palaearctic region it is found throughout Europe, the Russian Far East, Central Asia, Mongolia, China, Japan, and North Korea, south to Bhutan and northern India ).
Lohse (1967) called it an adventive species in Europe, and  referred to it as cosmopolitan. In Nova Scotia it has been found in coniferous and mixed forests, in boggy areas, on Craetegus sp. (Rosaceae), and on compost. Johnson (1993) reported that it was, "a grassland species which has been recorded especially around farms, gardens and parks but it often also occurs in other habitats. It has been collected in man-made heaps of refuse, especially cut vegetation, hay, grass, and compost." Description: Dorsum and venter brownish-yellow or reddish-yellow throughout, including legs. Head moderately densely, moderately fi nely punctate. Pronotum moderately densely and moderately coarsely punctate; interspaces 0.5 times the diameter of punctures; base with a distinct fl attened depression. Elytra somewhat more fi nely and sparsely punctate; interspaces 1.0-1.5 times the diameter of punctures; elytral margins strongly rounded; setae rather long and prominently erect giving a "bristling" appearance ( Fig. 7). Antennae: antennomere 1 long, swollen, and curved, apex nearly 1.5-2.0 times the width of base, 1.5 times longer than 2; 3 the same length as 2 but more slender; 4-8 almost equal in length and bead-like; club distinct; antennomeres 9 and 10 slightly transverse ( Fig. 1.6). Body: width/length ratio, 0.46; length, 1.3-1.6 mm.   September 30, 1972 (1, MMUE). Atomaria pusilla is newly recorded in New Brunswick and Nova Scotia (Fig. 15). In North America it has been recorded in Alaska, British Columbia, Connecticut, Idaho, Indiana, New York, Ohio, Oregon, Québec, and Washington (Blatchley 1910;Britton 1920;Notman 1920;Williams et al. 1995;. Blatchley (1910) referred to it as a "European" species and  listed it as "probably introduced" in North America. In the Palaearctic region it is found throughout Europe (including Turkey), across North Africa, and in central Asia in Iran, Afghanistan, and Kazhakhstan, northeast to Mongolia and the Russian Far East ). Blatchley (1910) reported it from the borders of a sphagnum marsh and in other damp vegetable debris. In New Brunswick it was found in a coniferous forest on a decaying gill fungus. Johnson (1993) reported that it was, "a grassland species which is found especially around farms and gardens and only rarely elsewhere, It has been collected in man-made heaps of refuse, cut vegetation, hay, grass, and compost heaps.
Description: Body (dorsally and ventrally) uniformly reddish-yellow. Antennae and legs uniformly yellow-colored. Head very fi nely and sparsely punctate. Pronotum widest at midpoint, constricted basally and apically. Pronotum moderately fi nely and closely punctate, interspaces 1.0 times the diameter of punctures; pronotal base with a complete fi ne marginal line running along the bottom of the basal impression. Elytral punctation similar to pronotum (Fig. 8). Antennae: antennomere 1 short, straight, not or slightly expanded at apex, slightly longer than wide, approximately as long as antennomere 2; 3 slightly shorter than 2; 4-8 short and bead-like; antennal club distinct; antennomeres 9 and 10 transverse ( Fig. 1.8). Th e smallest species of the genus; body width/length ratio, 0.40; length, 1.0-1.2 mm. Atomaria testacea is newly recorded in New Brunswick, Nova Scotia, and Canada (Fig. 15). In North America it has previously been recorded from southern California (Leng 1920) under the name Atomaria rufi cornis Marsham, 1802, synonymized by . LeConte (1869, 258) reported a species from South Carolina under the name of Atomaria testacea Zimmerman; however, as  makes clear, this refers to Cryptophagus ferrugineus Sahlberg. In the Palaearctic region it is found throughout Europe, in Algeria, Morocco, and Tunisia in North Africa, and in Asia in Turkey, Kazhakhstan, Mongolia, and western Siberia ). In Atlantic Canada it has been found in red spruce forests, boggy areas and marshes, in pastures, fi elds, meadows, and open areas, on beaches under wrack, in a ravine, and in compost. Johnson (1993) reported that it was, "mainly a grassland species which has been recorded around farms, gardens and parks, but it often occurs in other habitats. It has been collected in man-made heaps of refuse, especially cut vegetation, grass, hay, haystack bottoms and compost, more rarely in dung or dung heaps." Description: Head and prontotum piceous. Elytra piceous basally, gradually becoming paler in the apical half. Venter: pro-, meso-, and meta-sterna piceous; abdominal sterna and legs dark testaceous. Head fi nely, moderately sparsely, punctate. Pronotum and elytra moderately fi nely and densely punctate, interspaces 1.0-1.5 times the diameter of punctures, elytral punctures becoming fi ner apically; pronotum widest at midpoint, constricted towards base and apex; base with only a very shallow transverse impression (Fig. 9). Antennae: antennomere 1 swollen, slightly curved and expanded at apex; 2 smaller than 1 but also apically expanded; 3 almost as long as 2 and only slightly more slender; 4-8 short and somewhat bead-like; antennal club distinct; antennomeres 9 and 10 transverse ( Fig. 1.8). Body: width/length ratio, 0.46; length, 1.3-1.6 mm. Atomaria nigrirostris is newly recorded in Labrador, insular Newfoundland, Nova Scotia, and Canada (Fig. 16, 17). In North America, it has previously been recorded in Alaska (Leng 1920; under the name Atomaria fuscicollis Mannerheim, 1852, synonymized by . In the Palaearctic region it is found throughout Europe (including Turkey), in Algeria in North Africa, across Siberia to the Russian Far East, south to Mongolia, and in central Asia in Iran and Uzbekistan ). In Atlantic Canada it has been collected in eastern hemlock, red spruce, and mixed forests, along a stream, and in a gypsum sinkhole. Lohse (1967) reported it in Europe from along rivers, under leaves. Johnson (1993) reported that it was, "a woodland species, mostly associated with broadleaved trees, it occurs on and under dead wood on the ground, in moss, leaf litter, fl ood refuse and general ground litter. It also occurs under hedgerows." Description: Head, pronotum, and pro-, meso-, and meta-sterna piceous brown; elytra and abdominal sterna a slightly paler, fuscous brown; elytra becoming slightly paler in apical half. Head, pronotum, and elytra moderately densely, moderately strongly punctate; interspaces 1.0-1.5 times the diameter of punctures; apex of elytra slightly more fi nely punctate; elytral setae approximately the length of the interspaces, prominent and somewhat erect. Pronotum strongly constricted from midpoint to apex, parallel from midpoint to pronotal base; before base with a strong transverse impression, with several fi ne and sharp longitudinal ridges. Elytra scarcely, or not, wider than pronotum (Fig. 10). Antennae: antennomere 1 long and curved, twice as long as 2 and approximately twice as wide at apex than at base; 2 cylindrical; 3 as long as 2 but more slender; 4, 6, and 8 short and bead-like; 5 and 7 somewhat longer; club distinct; antennomeres 9 and 10 clearly transverse (Fig. 1.7). Body: width/length ratio, 0.31; length, 1.7-1.9 mm.  Atomaria wollastoni is newly recorded in Labrador, insular Newfoundland, and Nova Scotia (Fig. 13, 17). Dollin et al.'s (2008) records of A. ochracea are attributable to A. wollastoni. Leng (1920) reported it from "Canada." In Europe this species has a northern distribution having been recorded in central and northern Russia, Scandinavia, Estonia in the Baltic region, Great Britain, Ireland, the Netherlands, Germany, and France ). In Atlantic Canada it has been found in coniferous forests consisting of eastern hemlock, balsam fi r, red spruce, and black spruce (Picea mariana (Mill.) BSP., Pinaceae). In Europe, Schiegg (2000) classed it as a stenotopic saproxylic species characteristic of high dead wood connectivity, and noted that it is red-listed in Germany. It is also red-listed in Denmark (Stolze and Pihl 1997). In Norway, Hågvar (1999) found it associated with sporocarps of Fomitopsis pinicola (Fr.) Karst. (Polyporaceae). It would appear to be a characteristic saproxylic species in both Europe and North America.

Discussion
As a result of the present investigation, ten species of Atomaria have been recorded in Atlantic Canada. Of these, six including Atomaria distincta, A. fuscata, A. lederi, A. lewisi, A. testacea, and A. nigrirostris are newly recorded in Canada. Atomaria lederi and A. lewisi are newly recorded in North America.
Th e paucity of knowledge about many of these species in North America, makes determination of the zoogeographic status of some species uncertain (Table 1). Two species, Atomaria ephippiata and A. distincta, are clearly Nearctic, not having been recorded outside this region. Atomaria apicalis and A. pusilla, have previously been categorized as "probably" adventive Palaearctic species by . Th e designation of Atomaria apicalis, however, was prior to the synonymy with the Nearc-tic Atomaria ovalis (in the present study); consequently this designation needs to be re-evaluated since the wide distribution of A. apicalis is more indicative of a native Holarctic species. Atomaria fuscata has a wide distribution in North America on both Pacifi c and Atlantic coasts, and in the interior of the continent, also suggestive of a native Holarctic species. Th e northern boreal distributions of Atomaria lederi and A. wollastoni in the Palaearctic region could be indicative of species with a Holarctic distribution. However, there are very few records of these species in North America. Atomaria nigrirostris is widely distributed in the Palaearctic region, and records in North America (Alaska, Labrador, Newfoundland, and Nova Scotia) are consistent with a Holarctic distribution. Atomaria lewisi and A testacea are widely distributed in the Palaearctic region, but have only been reported from a limited number of locations in North America -possibly indicative of adventive species on this continent. Given the limited knowledge about this genus in North America, and the considerable taxonomic confusion that still exists, the zoogeographic status of all these species should be re-evaluated in the future when more information is available and the genus is better understood.
Within Atlantic Canada, few generalizations can be made with respect to the distribution of species as indicated in Figs 13-17. Atomaria apicalis, A. ephippiata, A. fuscata, A. nigrirostris, and A. wollastoni are represented by a substantial number of records and appear to be relatively widely distributed throughout Atlantic Canada (Figs 13, 16). All but A. apicalis are also found in southern Labrador (Fig. 17). Atomaria lederi, A. lewisi, and A. testacea are represented by a moderate number of records and generally distributed in Nova Scotia (although A. lewisi has not been found on Cape Breton Island), and the latter two species have additionally been collected in New Brunswick (Figs 14, 15). Atomaria distincta, and A. pusilla are represented by a small number of records and appear to be relatively restricted in their distribution (Figs 14, 15). However, collecting eff ort for this genus has been far from adequate in all areas of the region, with the possible exception of the mainland of Nova Scotia. Consequently generalizations about their distribution or abundance in the region must be provisional. Th e possibility of additional species occurring in the region has certainly not been excluded. Further collecting is clearly indicated in order to gain a better understanding of the members of this genus in Atlantic Canada.
It is evident that there are diff erences in habitat preferences amongst these species. Atomaria ephippiata, A. fuscata, A. lewisi, and A. testacea appear to be eurytopic species found in a wide variety of forested and open habitats; Atomaria lederi and A. wollastoni have been principally collected in coniferous forests, whereas Atomaria nigrirostris has been found in coniferous, deciduous, and mixed forests. All three appear to be primarily saproxylic species. Atomaria apicalis, A. distincta, and A. pusilla are all primarily found in grasslands and other open habitats. Atomaria fuscata, A. ephippiata, A. lewisi, and A. testacea have all been collected in compost heaps in this region, and A. distincta, A. fuscata, A. ephippiata, and A. testacea have all bean found in seashore, salt marsh, and beach drift environments. At least in Great Britain and Ireland, A. pusilla appears to be primarily a synanthropic species (Johnson 1993). Further bionomic investigations should be conducted to better understand the macro-and micro-habitat preferences of these species and the ecological roles that they play in their respective environments.

Introduction
Th omson (1867) described Mallodon hermaphroditum, noting that it was the only species known in the genus whose male lacks impunctate and shining facets on the pronotum, in contrast with the remainder of the surface. Th us, the pronotum is simi-lar to that of the female. Later,  transferred M. hermaphroditum to Stenodontes (Mallodon), described Stenodontes (Mallodon) popelairei, and noted that this species shows diff erent characters, than Th omson's species. Some characters as, for example, body slender, labrum tumid, and prothorax in male without evident sexual dimorphism, allow to allocate these species in a new genus.
Since Lameere (op.cit.), both species were simply included in catalogues, and were not fi gured until Fragoso and Monné (1995) studied the lectotype male of M. hermaphroditum, and Santos-Silva (2005) examined the lectotype male and the paralectotypes (two males and one female) of M. popelairei. Neither of the latter two authors questioned the generic assignment of these species.
During the study of another genus (Nothopleurus Lacordaire, 1869), which included a cursory analysis of several species of Mallodon, it became evident that both M. hermaphroditum and M. popelairei exhibited an exclusive set of characters that would permit them to be assigned to a new genus. Further evidence is presented to support the inclusion of this new genus in Macrotomini, which includes Mallodontini, sensu Monné and Hovore (2006) and Monné (2006).

Materials and methods
Specimens examined for this study are from the following institutions / private collections: Size from small to large (up to 2 cm). Integument brown to dark-brown. Male (Figs 1, 3). Body not notably depressed; ratio of head (without mandibles) + prothorax and elytra equals to approximately 1:2.5. Head (Figs 1, 3) proportionally large and wide; punctation of dorsal surface (Fig. 29) coarse, deep and confl uent. Eyes large; upper ocular lobe just narrower than lower ocular lobe; distance between upper ocular lobes (Fig. 29) more than 1.2 times the length of scape; distance between lower ocular lobes approximately 1.4 times the length of scape. Ocular carina (Fig. 29) low, narrow and short, distinct only at vertical area of upper ocular lobe. Clypeus wide. La- brum vertical, but with the basal portion (Figs 6, 10) wide, coplanar or almost coplanar with clypeus, tumid and distinctly visible in dorsal view. Hypostomal area coarsely and anastomosedly punctate. Galea surpassing base of second palpomere of maxillary palps. Labial palps reaching approximately middle of palpomere III of maxillary palps. Mandibles (Figs 1, 3, 29), at most, as long as the head; dorsal carina ( Fig. 29) elevated on basal two-thirds; inner face densely pilose; lower, inner margin projected in plate evident sexual dimorphism; prosternal process (Figs 21-24) distinctly narrowed medially. In Mallodon body, in general, more robust (Fig. 5); mandibles of major male (Figs 30, 31) longer than the head; base of labrum (Fig. 7) not visible in dorsal view or when visible (Fig. 8), more distinct laterally, with its basal margin placed, in general, distinctly lower than the edge of clypeus (Fig. 9), and always strongly concave at middle region; lateral angles of prothorax of male (Figs 16, 18) placed in the same line or almost in the same line of posterior ones (sometimes, similar to the species of Allomallodon (Fig. 20)); pronotum, proepimera and prosternum of male (Figs 26,27) with sexual dimorphism very distinct; prosternal process (Figs 25-28) wide and not or, slightly narrowed medially (in female of some species, narrowed medially, but always moderately wide).
From the species of Physopleurus Lacordaire, 1869 with unarmed tibiae, Allomallodon diff ers, mainly, by the prosternal suture straight or substraight, and by the prosternum not tumid. In all species of Physopleurus the prosternal suture is distinctly curved, and the prosternum is tumid or strongly tumid.
Allomallodon diff ers from Mallodonhoplus Th omson, 1861 by the unarmed tibiae (distinctly spinose, mainly the protibiae, in Mallodonhoplus). It can be separated from the species of Stenodontes Audinet-Serville, 1832 by the shorter antennae that do not reach the middle of elytra. In males of Stenodontes the antennae surpass the middle of the elytra, and in females they reach the middle. It diff ers from Neomallodon Linsley, 1957 by the pronotum of male being distinctly and abundantly punctate laterally (slightly rugose in Neomallodon). From the species of Olethrius Th omson, 1861 it diff ers, mainly, by the scape not surpassing the posterior edge of eye (distinctly surpassing in Olethrius). Fragoso and Monné (1995) noted the following on the pronotum of A. hermaphroditum: "Besides the inappropriate epithet, the male pronotal disc shows a large, subcircular "tache luisante", with a cluster of points in the middle, as well as a few other points scattered at the periphery of the disc (latero-posteriorly more numerous); the sex-points are limited to lateral areas, close to pronotal bordes (including the "oreilletes")". Th ese statements, however, are not consistent with the lectotype and specimens examined by us. In fact, the epithet chosen by Th omson is quite appropriate, as the sexual dimorphism in the pronotum is minimal and, above all, there is no the sexual punctation as noted by Fragoso and Monné (op.cit.). Th e entire sculpture of pronotal disc is quite similar in both sexes of A. hermaphroditum (Figs 13, 14), as well as in A. popelairei (Figs 11,12), and very diff erent from what occurs in other species of Mallodon (Figs 15-20), in which the sexual dimorphism is strongly evident.

On the placement of Allomallodon gen. n.
Allomallodon belongs to the group that some authors (primarily European) exclude from Macrotomini, and include in Mallodontini. Although it may seem obvious that a genus carved out of Mallodon belongs to that group, such an assertions, in some cases, can be misleading. Prioninae is a small subfamily in comparison with other cerambycid subfamilies, but the relationship and placement of the genera and species, frequently, are complex. Th at, in our opinion, is the reason by which Mallodontini is not a well defi ned group, as suggested by some authors, for example Vitali (2008): "Th e systematics of the Macrotomini, already partially cleared by European authors (Th omson, Lameere, Quintin [sic], Villiers), still shows uncertainties due to the fact that the American authors (Monné and Hovore 2005;Silva-Santos [sic] and Martins 2005) still consider Macrotomini, Mallodontini and others only one tribe".
Th omson (1861) erected the group "Mallodonitae" (currently considered Mallodontini) and defi ning it by the following: head wide, not elongated behind eyes; antennae short, fi liform or submoniliform, with the antennomere III short, and barely longer or equal to IV; mandible subvertical or horizontal, robust, frequently elongated; lateral margins of prothorax crenulated or multi-spinose; elytron elongated, subdepressed; prosternum distinct; mesosternum fl at; legs cylindrical; protibiae unarmed (except in Mallodonhoplus); tarsi mediocre, with the tarsomere V usually not longer than I-IV together. Th omson (op.cit.)  According to Th omson (op.cit.), the characters of "Macrotomitis" that allow separation from "Mallodonitae" are: antennae longer, with antennomere III distinctly longer than IV; upper ocular lobes closer; mandibles not transversal; protibia often with spines and diff erent. Th e genera that were included in "Mallodonitae" make the group an amalgam of tribes, as currently accepted, be they by European or American authors: Macrotomini, Macrotomini/Mallodontini, Notophysini, Eurypodini, Platygnathini. Th us, the only merit of Th omson's (op.cit.) regarding "Mallodonitae" was to attribute a name to the group, without contributing substantively to the classifi cation of Prioninae, as noted by Vitali (2008).
Following his previous classifi cation, Th omson (1864) redefi ned "Mallodonitae" with the following changes sensu Th omson (1861): body wide, mainly in males; antennae just surpassing the middle of elytron (sometimes, shorter); scape thick, longer or as long as the antennomere III; pronotum in male with smooth facets, and scabrous or punctate in female; prosternal process slightly projected; mesosternal process laminiform; legs never with spines [mainly modifi cation]; tarsomere V as long as I-IV together; integument always brownish, shining. Th omson (1864) included in this group the genera recorded in Th omson (1861), except Mallodonhoplus (transferred to "Mac-rotomitae"), and added Opheltes Th omson, 1864. Again, this concept joins genera that currently are included in many tribes of Prioninae.
Th is concept of Mallodontini, based mainly on the absence of spines on the legs, is very unsatisfactory, and usually does not allow to reliably separat included genera. For example, in Physopleurus there are species with spines on tibiae as well as species with the tibiae unarmed. Many other characters are shared by these two species groups of Physopleurus (apex of antennal tubercles sub-horizontal and backward; prosternal suture absent or present and distinctly curved in males, etc), which does not allow the division in two genera, much less the inclusion of these species in diff erent tribes. It is important highlight that the species of Physopleurus show a general appearance similar to that of the species of Mallodon and often have facets on pronotum of males, mentioned by Th omson (1864), as well as all other characters pointed out by Th omson (1861,1864). In the description of Mallodon hermaphroditum, a species that does not have facets on the pronotum of males, demonstrates that Th omson was deliberately modifying the description of Mallodontini in each of his works, including and excluding genera. Th is exposes the fragility of the classifi cation of this group that, as previously seen by the concept of Th omson (1867), diff ers from Macrotomini only by the unarmed legs.
Lameere (1919) summarized his previous works ("Révision des prionides") and divided Macrotomini in several groups that currently are considered subtribes by some authors, and tribes by others: Archetypi; Basitoxi; Stenodontes; Cnemoplites; Macrotomae; Rhaphipodi; and Xixuthri. According to him, the subgroup "Stenodontes" is characterized: body, in general, large, more or less depressed; eyes not emarginated; antennal tubercle distinct and acute; scape elongated and longer than the antennomere III; mandibles with dorsal carina; ligulae small and weakly whole (almost undivided); prothorax wide, with the sides wide and a little bent down, lateral edge present and crenate, more parallel in male than in female; males with sexual punctation on prothorax; pronotum of males with callosities shining; legs robust, with or without spines; last urosternite of males more or less emarginated. As shown above, Lameere (op.cit.) modifi ed the concept of Macrotomini and Mallodontini established in the works of Th omson considerably. In particular, the main character used by Th omson (1867) to separate these two groups, was not accepted by Lameere (op.cit.), because the subgroup "Stenodontes", that included Mallodon, has species with legs that are spinose or not.
As in the divisions established in Th omson's works, the division proposed by Lameere maintains separate genera that share many characters (e.g. Xixuthrus Th omson, 1864 and Mecosarthron Buquet, 1840), and combined other genera with notably diff erent characters (e.g. some species of Nothopleurus, and Mallodon).
Despite showing inconsistencies in his descriptions of tribes, in which were included genera that contradict the characters that were noted (e.g. body more or less depressed in "Archetypi", in relation to Strongylaspis Th omson, 1861, whose species has body distinctly not depressed), the concept of "Stenodontes" proposed by Lameere is more consistent than that of Th omson, because it unites genera with generally similar shared characters (body form, length of scape in relation to the antennomere III, prothorax form, etc). Quentin and Villiers (1975) considered Macrotomini and Mallodontini as distinct and, in their key to the tribes that occur in Madagascar, separated these groups in the following manner: pronotum fl at, with sides explanate in Mallodontini, and convex and with the sides bent down in Macrotomini; presence of shining facets on pronotum of males, occupying almost the whole surface in Mallodontini, and without facets or with small facets in Macrotomini. Th at key, if applied to American genera, would separate into distinct tribes species in the same genus, including species that at that time were in Mallodon, as well as genera included in the subgroup "Stenodontes" by Lameere (1919). According to Quentin and Villiers (op.cit.), Mallodontini is characterized by: general form wide and depressed; eyes not or slightly emarginated, coarsely faceted, weakly protruding, not surpassing the antennal tubercles; antennae with eleven segments, fi liform, rarely surpassing the middle of elytra; scape, at least, as long as antennomere III; prothorax transverse, explanate laterally; pronotal disc, mainly in males, with facets or callosities fl at and shining; prothoracic episterna very wide; legs, in general, short and always unarmed. It is important to note that Quentin and Villiers (op.cit.) characterized the tribe not only for the single species present in Madagascar. Th is defi nition of Mallodontini distinctly modifi es, again, the limits of the group, resulting in the mandatory exclusion of genera included by Lameere (1919): Nothopleurus (part); Physopleurus; Mallodonhoplus; Olethrius; and Allomallodon gen. n. (originally included in Mallodon). Moreover, it may include species in genera that, by the defi nition of European authorities, could not belong to Mallodontini, as for example, Physopleurus rugosus (Gahan, 1894) and P. longiscapus Lameere, 1912, in which the proepisterna is not notably reduced and has all the other characters listed by Quentin and Villiers (op. cit.). Seemingly contradictory, Villiers (1980) allocated Nothopleurus in Mallodontini, without realizing that the type species of this genus, Nothopleurus ebeninus Lacordaire, 1869, does not fi t to his description of the tribe, primarily, by the pronotum of males is without impunctate and shining facets in contrast with the remainder of surface.
As shown above, the divisions proposed, in part by European authors, affi rmed the opinion of Vitali (op.cit.) (Mallodontini diff erent from Macrotomini), and did not "clear [up]" the classifi cation of Macrotomini. In fact it has made it more chaotic, mainly by contradicting the vastly diff erent concepts of Th omson (1861, 1864, 1867), Lameere (1919) and Quentin and Villiers (1975).
It is probable that Macrotomini is not a monophyletic group, but the divisions proposed up to now are inconsistent. Th us, we believe that the prudent action is to maintain Macrotomini as a single tribe, including and disregarding the subdivisions presented in Lameere (1919). Monné and Hovore (2006) listed the American genera of Macrotomini, following the concept of American authors.

Introduction
South America is home to several large, metallic ichneumonids, mostly in the subfamily Cryptinae. Within the subfamily Pimplinae, metallic blue pimplines were known in the genus Pimpla, then Ian Gauld described a large, distinctive, metallic blue pim-pline from Costa Rica as a new genus and species, Umanella caerulea (Gauld, 1991). Since then, several more specimens of Umanella have been collected in Colombia and Ecuador but no other specimens are known from older collections. As Umanella are large and likely to attract attention there are probably not many undescribed species.
Here we evaluate the variation between specimens and describe those from Ecuador as a second species, Umanella giacometti Broad & Sääksjärviq sp. n.  Gauld (1991). Fore wing length is given from the hind edge of the tegula to the apex of the wing. Ovipositor length is measured from the base of the ovipositor, i.e. anterior to the end of the metasoma.

Taxonomy
Th e specimen base for this study is small as Umanella are rarely collected and apparently sparsely distributed. 'We have examined a total of 21 specimens, 12 from Costa Rica (10 females, two males) (BMNH, IAVH), four from Colombia (two females, two males) (BMNH, IAVH) and fi ve from Ecuador (two females, three males) (BMNH, USNM and ZMUT). Female specimens of U. caerulea predominate.
Although specimens from Ecuador are noticeably larger than those from Costa Rica and Colombia there are no discrete morphological diff erences between these populations. Umanella specimens lack most carinae, the integument being mostly unsculptured and metallic-looking. Th e only structural diff erences we could fi nd between females were (1) the shape of the sides of the swelling on the second tergite, and (2) overall size and the relative length of the ovipositor, as compared to fore wing length. In addition, there are distinct colour diff erences, concerning (3) the presence or absence of a violet tinge on the metasoma, (4) the colour pattern of the mid tibia and, (5) in the amount of white on the fore trochanter (see Table 1). Th e few known males off er some small diff erences between populations: the colour pattern of the mid femur and tegula diff ers between those from Ecuador and those from Costa Rica / Colombia. It seems that the single paratype of U. caerulea in BMNH is anomalously small and has a much broader fi rst tergite than the other males examined. Smaller specimens of both sexes are distinctly brown on the metasoma. Table 1 describes the characters that vary between specimens from Costa Rica, Colombia and Ecuador: we consider variation between Colombian and Costa Rican specimens to be minor, whereas there are distinct diff erences in several characters when comparing these specimens with those from Ecuador. Th erefore we are describing the Ecuadorean specimens as a new species.
Umanella Gauld, 1991Gauld (1991 provides an excellent description of the genus Umanella, which we are not repeating here, and provides a key to Costa Rican Pimplinae genera which serves to diagnose Umanella anywhere in South America. Of the South American Pimplinae, Umanella is the only genus with a long ovipositor, metallic blue colouration (Fig. 1) and lacking the epicnemial carina (Fig. 2). Some Pimpla are metallic blue but can be easily separated by the relatively short ovipositor (less than half as long as fore wing), generally stout body and presence of the epicnemial carina. Some Neotropical Dolichomitus resemble Umanella in body shape but are never metallic blue in colouration and lack the lateral denticles on the apex of the upper valve of the ovipositor. Gauld, 1991 Description. See Gauld (1991). Essentially similar to U. giacometti but diff ering in the characters listed in Table 1. Only characters that are useful in diff erentiating U. caerulea from U. giacometti are emphasised here, although complete descriptions of female and male colour patterns are given.
Colour: metallic blue, duller towards apex of metasoma, fading to dark brown. Metasoma with purple tinge, sometimes strongly so. Ground colour of metasoma brown. Female from Colombia (BMNH) with metasoma largely brown, but with some blue and purple gloss. Antennae black, except for white, sub-apical annulus on fi ve to seven fl agellomeres. Maxillary palps dark brown basally and apically, dull white on third and fourth segments.'Tegula centrally metallic blue, brown around edges. Wing venation, including stigma, dark brown. Wing membrane slightly infuscate basally, distinctly infuscate in apical quarter. Legs with coxae, trochanters and trochantelli shiny, metallic (dark) blue. Fore leg with large apical patch on trochanter (or entire front side) creamy white, extreme apex of trochantellus brown, conspicuous creamy streak along front edge of femur (from apical half to entire length); fore tibia with basal 0.6 brown (slightly paler sub-basally); fore tarsus brown fading to yellowish on third and fourth tarsomeres, fi fth tarsomere black. Mid femur with tiny basal patch to conspicuous basal streak creamy; mid tibia brown with dull yellowish mark sub-basally and fading to yellowish on apical quarter (black at extreme apex) (Fig. 4A, B) or almost entirely white (black very apically (Fig. 4C). Legs otherwise marked as in U. giacometti but pale markings are duller, usually more yellow.
Paratype male (BMNH). Whole insect: Fig. 6. Fore wing length 7.5 mm, body length (from antennal insertion to apex of genitalia) 7.8 mm. Submetapleural carina complete. First metasomal tergite 2.2 × as long as maximum width (Fig. 5A), second tergite 1.6 x as long as wide. Second and third tergites with strong diagonal, basal grooves cutting off raised, central section which is raised posteriorly. First and second tergites with strong setae laterally, fi rst tergite and basal half of second smooth dorsally, metasoma setose dorsally from second half of second tergite. Sclerotized part of fi rst tergite extending to 0.2 of distance between spiracle and hind edge. Some dorsal punctures on apical half of second tergite, following tergites regularly punctate and setose dorsally.
Colour: head and mesosoma metallic blue. Antennae black, white on 7 (22nd to 28th) fl agellomeres. Maxillary and labial palps white. Tegula brown. Wing venation, including stigma, dark brown. Wing membrane clear basally, slightly infuscate in apical quarter. Fore leg white on fore side, except apical tarsomere brown. Hind side of fore leg basally white, trochantellus and femur brown; fore tibia pale brown over basal quarter, apex narrowly dark brown; fore tarsus pale brown except apical tarsomere dark brown. Mid leg with coxa and trochanter white, trochantellus brown, fore side of femur brown on basal 0.15, remainder white, fore side of mid tibia vaguely infuscate brown basally and medially, dark brown at very apex, hind side of femur brown, hind side of tibia brown over basal 0.6, slightly paler patch sub-basally, mid tarsus dark brown on fi rst, second and fi fth tarsomeres, third and fourth white. Hind leg with metallic blue coxa, metallic darker blue on trochanter and trochantellus, turning to black on femur; apex of hind trochantellus and base of femur narrowly brown; hind tibia black with narrow annulus sub-basally, white on outer side, brown on inner side; hind tarsus dark brown on fi rst, basal 0.7 of second and apical 0.5 of fi fth tarsomeres, remainder white. Metasoma with fi rst tergite metallic blue, second and third tergites brown with dark brown apical rims, remaining tergites dark brown; fi rst sternite with sclerotized part metallic blue, sclerotized sections of other sternites brown, membranous parts of sternites white.
Variation. 'Th e single male in BMNH seems to be unusually small and stout, compared to male specimens from Colombia in IAVH and has deformed antennae (Fig. 7). One male in IAVH has the metasoma almost entirely metallic blue. Th ere is variation in the amount of white on the mid tibia of females (Fig. 4A, B, C), in the amount of metallic blue colouration on the metasoma and in the relative length of the ovipositor ( Table 1). Much of this variation we assume is size-related (smaller specimens seem to be more extensively white and brown), but one female specimen from Colombia in IAVH is particularly stout, with length to breadth ratios of the 1st and 2nd tergites of 3.4 and 2.1, respectively, and the ovipositor only 1.6 x the length of the fore wing (which is relatively large, at 14 mm). Th e mid tibia of this specimen is almost entirely white. Whether this specimen represents a third species or is just at the extreme end of variation within U. caerulea we are unable to say at present but it would be desirable to see more Umanella specimens from Colombia.
Biology. Nothing is known. Distribution. Known from from Costa Rica (see Gauld, 1991, Gauld et al., 1998 Description. Female. Whole insect: Fig. 1. 'Fore wing length 18.5 (holotype)-20 mm, body length (from antennal insertion to apex of metasoma) 31-34 mm, ovipositor length 41 (holotype)-45 mm, ratio of ovipositor to fore wing length 2.22-2.25. Occipital carina complete, joining hypostomal carina behind base of mandible, dipped mediodorsally with depression on vertex and deep groove on back of head. Epomia present for short distance across trough in pronotum. Notauli deep and long, converging but not meeting posteriorly, petering out at about half the length of mesoscutum. Epicnemial carina absent (Fig. 2). Mesopleurum with posterior suture weakly impressed dorsally, strong ventrally, smooth except for some weak crenulae, and deeply impressed furrow connnecting suture to episternal scrobe. Mesepisternal sulcus complete, strong and non-crentulate. Posterior transverse carina of mesosternum absent. Propodeum lacking all carinae except pleural and submetapleural carinae, which are complete, and stubs of median longitudinal carinae, present at posterior end of propodeum. Faint, narrow groove present on dorsal, ventral half of propodeum. Propodeal spiracle about twice as long as medially wide. Whole body elongate and integument entirely smooth and shining except mandible sparsely punctate basally, puncto-striate medially. Upper tooth of mandible slightly shorter than lower tooth. Hind coxa about 1.2 as long as dorsal face of propodeum. First tergite of metasoma 4.4 × as long as apically wide, second tergite 2.8 × as long as apically wide. First tergite with swollen, apical area, posterior end impressed behind it. Sclerotized part of fi rst sternite with low, raised bump just before level of spiracle; extends to half distance between spiracle and posterior end of tergite. Second tergite with deep, diagonal grooves cutting off anterior corners; narrow, drop-shaped median area defi ned, sides of raised area concave (Fig. 3C). Th ird and fourth tergites with deep, lateral grooves. Metasoma with strong setae laterally. All tarsal claws with acute basal lobe. Fore tibia with longitudinal patch of stronger, darker setae. Propodeum with coxal foramen narrowly separated from metasomal foramen by sclerotized bridge. Ovipositor with tip angled downwards. Lower valve of ovipositor apically slightly overlapping dorsal valve, with 13 visible teeth, regularly spaced and inclined. Dorsal valve with row of lateral denticles above teeth on each side.
Colour: metallic blue, duller towards apex of metasoma. Antennae black, except for white, sub-apical annulus on three and a half fl agellomeres (23rd to 26th) to fi ve (holotype, 23rd to 27th) fl agellomeres. Maxillary palps dark brown, a little paler centrally, ventrally. Tegula metallic blue. Wing venation, including stigma, dark brown. Wing membrane slightly infuscate basally (holotype) or distinctly brown, distinctly infuscate in apical quarter. Legs with coxae, trochanters and trochantelli shiny, metallic (dark) blue. Fore leg with apical edge of trochanter creamy white with white streak extending over basal half of anterior surface of fore femur, extreme apex of trochantellus and base of femur brown; fore tibia with basal 0.6 brown (slightly paler sub-basally); fore tarsus brown except apical half of third and fourth tarsomeres creamy, fi fth tarsomere black. Mid tibia black with small, dull creamy mark sub-basally and abruptly  white on apical quarter (black at extreme apex) (Fig. 4D); mid tarsus brown to black, except fourth and apical half of third tarsomere creamy. Hind tibia with white annulus sub-basally; hind tarsus black on fi rst, basal half of second and apical half of fi fth tarsomeres, remainder creamy. Paratype males. Whole insect: Figs 8, 9. Fore wing length 10-14 mm, body length (from antennal insertion to apex of genitalia) 15-21 mm. Submetapleural carina complete to about two thirds length of propodeum then abruptly weaker (barely traceable on smaller specimen) for remainder. First metasomal tergite 3.7 to 4.0 × as long as maximum width (Fig. 5B), second tergite 2.7 to 3.1 × as long as wide. Second and third tergites with strong diagonal, basal grooves cutting off raised, central section which is raised posteriorly (Fig. 10). First and second tergites with strong setae laterally but smooth dorsally. Sclerotized part of fi rst tergite extending to 0.5 of distance between spiracle and hind edge. Th ird tergite regularly punctate and setose dorsally. Fourth tergite onwards setose dorsally with inconspicuous punctures.
Colour: head and mesosoma metallic blue. Antennae dark brown to black, white on 3-11 (22nd to 32nd) fl agellomeres. Maxillary and labial palps cream coloured. Tegula white-transparent. Wing venation, including stigma, dark brown. Wing membrane clear basally, infuscate in apical quarter. Fore leg white on fore side, except apical tarsomere black or dark brown, other tarsomeres off white, basal half of fi rst tarsomere pale brown to white. Hind side of fore leg basally white, but coxa with slight brown hint, femur with broad ventral streak on apical 0.7, light or dark brown fading into metallic blue; fore tibia brown on basal half, apex narrowly dark brown. Mid leg with coxa and trochanter white, trochantellus dark brown to black, fore side of femur dark brown to black on basal 0.15, remainder white; mid tibia almost totally white or dark brown on basal half, except for slightly paler subbasal patch, narrowly black at apex; mid tarsus dark brown to black on fi rst and fi fth tarsomeres, second and third dark brown fading to off white, fourth white. Hind leg with metallic blue coxa, shiny black or dark brown trochanter, trochantellus and femur; inner apex of hind trochantellus off -white, base of femur narrowly whitish to brown; hind tibia black or brown with narrow white annulus sub-basally, but uniformly black or brown dorsally; hind tarsus dark brown to black on fi rst, basal Metasoma with fi rst to fi fth or sixth tergites metallic blue, fading to dark brown or near black, apical tergites dark brown to shiny black; fi rst sternite with sclerotized part metallic blue, sclerotized sections of other sternites shiny black to brown, membranous parts of sternites white. Variation. Th e male in USNM/ZMUT is smaller than the two males in BMNH and is more extensively white on the antennal fl agellomeres and mid tibia. Biology. No specimens of Umanella have been reared but the holotype and three paratypes (BMNH) were collected whilst fl ying around a standing, dead tree trunk. Th e hosts may be large coleopteran larvae. Th e ovipositor shape is rather similar to that of Dolichomitus species, which are parasitoids of beetle larvae in dead wood. Th ree specimens have been found at fairly high (1,100m) altitude whilst one male was collected by canopy fogging at fairly low (216 m) altitude. In this respect the new species resembles Umanella caerulea, which also inhabits tropical forests of various altitudes (Gauld et al., 1998 Etymology. We are pleased to give the name suggested by Mrs Jean Halperin, who won a competition to name this beautiful species in celebration of the opening of the Natural History Museum's Darwin Centre Two, where the holotype is housed. Th e    name refers to the wasp's resemblance to the slender, attenuated fi gures of the Swiss artist, Alberto Giacometti and is a noun in apposition.