New species of Australian arid zone chelonine wasps from the genera Phanerotoma and Ascogaster (Hymenoptera: Braconidae) informed by the ‘Bush Blitz’ surveys of national reserves

Abstract Here we focus on the poorly studied braconid wasp subfamily Cheloninae for the arid zone of the Australian continent, using material, in part, resulting from comprehensive surveys of three arid zone reserves. The Bush Blitz programme is a multi-institutional project with the aim of documenting the diversity of the flora and fauna in Australia’s National Reserve System, with describing new species being a key focus of the programme. In total, 11 species from the genera Ascogaster and Phanerotoma are treated, with species’ delimitation based on both molecular and morphological data. Two species are redescribed (Phanerotoma behriae Zettel, 1988a and P. decticauda Zettel, 1988a) and nine species are described as new (Ascogaster brevivena sp. nov., A. ferruginegaster sp. nov., A. prolixogaster sp. nov., A. rubriscapa sp. nov., Phanerotoma bonbonensis sp. nov., P. bushblitz sp. nov., P. lutea sp. nov., P. nigriscapulata sp. nov. and P. witchelinaensis sp. nov.). Keys to the arid zone species of these two genera are provided, along with a species richness estimation of Australian chelonine wasps.


Introduction
Although about 67,000 insects have been described for Australia (Australian Faunal Directory 2013), more than 45% of the continent has never been comprehensively surveyed for terrestrial invertebrates (Chapman 2009). To ameliorate this shortcoming, a biological survey programme was initiated in 2010 to document the biodiversity of the national reserve system, with an emphasis on describing new species of plants and animals utilising the expertise of systematists from museums, herbaria, universities and other research institutions (e.g. Lambkin and Bartlett 2011;Namyatova et al. 2011;Baehr and Whyte 2012). This initiative, titled the Bush Blitz programme, is a partnership between the Australian Government, BHP Billiton and Earthwatch Australia, and is coordinated by the Australian Biological Resources Study (Bush Blitz 2013a, 2013b. As part of the Bush Blitz programme, the braconid wasp subfamily Cheloninae was surveyed for three arid zone reserves in South Australia: Bon Bon Station, Hiltaba Station and Witchelina Station ( Figure 1). Cheloninae is a moderately large subfamily of braconid wasps with more than 1375 described species in 17 genera worldwide (Yu et al. 2005;. The group can easily be distinguished from other braconid subfamilies on characters associated with the metasomal carapace, the postpectal carina and fore wing venation (Shaw 1983(Shaw , 1997van Achterberg 1990;Zettel 1990c). Members of this subfamily are solitary egg-larval koinobiont parasitoids of Lepidoptera, where oviposition occurs into a host egg but development of the parasitoid is delayed until the host has emerged as a larva (Shaw and Huddleston 1991;Shaw 1997;LaSalle 2003).
The taxonomy of chelonine wasps is generally poorly known for Australia. The first species, Phanerotoma australiensis, was described by Ashmead (1900), followed soon after by Szépligeti (1900) who described species of Phanerotoma, Phanerotomella and Chelonus. In the following years, several authors described additional new species (Cameron 1911;Turner 1917;Girault 1924;Baker 1926). After a hiatus of more than 50 years, further species were described by Zettel (1988aZettel ( , 1989c and Huddleston and Walker (1994). Most recently, (Kittel and Austin 2014; ) have described two new genera, Austroascogaster and Phanaustrotoma, comprising six new species; presented a key to genera and a synopsis of the 65 species known for the continent; and described 18 new species of Phanerotomella. Three interesting aspects can be identified in this overview of the continent's chelonine fauna: first, most previous species descriptions, with a few exceptions, are based on very few specimens; second, little material has been added to Australian collections since about 1980; and third, the majority of described species have come from mesic coastal regions, particularly areas around Sydney and Perth, while very few taxa have been described from arid central Australia. As a first step to documenting the largely unknown chelonine fauna of the continent, particularly of the arid interior, we used material collected on the Bush Blitz surveys of the three arid Australian reserves above (Figure 1), supplemented by material from major collections, to describe four new species of Ascogaster and five new species of Phanerotoma, to redescribe two widely distributed Phanerotoma species and to present keys to facilitate identification of the species known from the central arid region. In so doing we took an integrative approach to delineate species, employing both fixed phenotypic differences and phylogenetic analysis of the cytochrome c oxidase subunit 1 (COI) barcoding region utilising generalised mixed Yule-coalescent (GMYC) and Poisson tree processes (PTP) analyses to estimate species boundaries. We also employed a Chao species richness estimator to calculate the likely diversity of the Australian fauna, and compare this to the number of morphospecies so far recognised.

Collecting locations, techniques and specimens
Three National Reserve System properties were surveyed as part of this project, all of which were previously large sheep-grazing properties (stations), now set aside to preserve their biodiversity. Witchelina Station was surveyed from late September to early October 2010, and is located in the northwestern part of the Flinders Ranges ( Figure 1). It was purchased by the Nature Foundation of South Australia and added to the National Reserve System in 2010. The property comprises 4200 km 2 of river red gum and coolibah (eucalypt) woodland lining the usually dry creek beds, and bluebush (Maireana spp.) shrubland. Bon Station was surveyed in October 2010. This property has been owned by Bush Heritage Australia since 2008, and is located 400 km west of Witchelina and 200 km south of Coober Pedy (Figure 1). The area is situated between the Great Victoria Desert and the large salt lakes of Eyre, Torrens and Gairdner. Its desert landscape is dotted with salt lakes, freshwater wetlands, shrubland, bluebush plains and arid zone woodland. Hiltaba Station, a 775-km 2 area purchased by the Nature Foundation of South Australia in 2012, was surveyed in November 2012. It is located north of the Gawler Ranges, and is approximately 260 km west of Port Augusta (Figure 1). It consists of rocky granite hills with open bluebush plains and mallee vegetation (Eucalyptus spp.).
A range of techniques were employed to collect material during the surveys, including yellow pan traps, light traps (LTs), Malaise traps (MTs) and sweep-netting of vegetation. All material was initially stored in 100% ethanol (for DNA extractions) and later pinned. All available chelonine specimens were also borrowed from the major Australian collections (approximately 5000 specimens in total) and, along with the material from Bush Blitz sites, were sorted to morphospecies. All relevant type material was examined and compared with the sorted morphospecies.

DNA sequencing
Genomic DNA was extracted from specimens preserved in 75-100% ethanol. The right hind leg was removed from each specimen and processed after the ethanol had evaporated. DNA extractions were performed using the Gentra Systems Puregene® DNA Purification Kit (Gentra Systems 2005): the leg was heated at 55°C in a 300 μL cell lysis solution with 1.5 μL Proteinase K solution. After 12-24 hrs, excessive proteins were removed by adding 100 μL Protein Perception solution. The DNA was washed in 300 μL isopropanol (to which 0.5 μL glycogen was added) and afterwards in 300 μL 70% ethanol. DNA was restored using 50 μL DNA hydration solution. Eppendorf thermal sequencers were employed to carry out the polymerase chain reaction (PCR) amplification. Each reaction of 25 μL comprised 14.4 μL nuclease-free water, 2.5 μL Taq Gold Buffer, 3 μL Magnesium chloride (MgCl) 2 , 2 μL 10 mM deoxynucleotide (dNTPs), 1 μL of each forward and reverse primer: forward CI-J-1718 5ʹ-GGAGGATTTGGAAATTGATTAGTTCC-3ʹ, reverse 1 CI-N-2191 5ʹ-CCCGGTAAAATTAAAATATAAACTTC-3ʹ (shorter) and reverse 2 CI-N-2329 5ʹ-ACTGTAAATATATGATGAGCTCA-3ʹ (longer; Simon et al. 1994), 0.1 μL AmpliTaq Gold® DNA Polymerase (Applied Biosystems Inc.), and 1 μL DNA. PCR settings started with a denaturation step of 9 min at 95°C, followed by 35 cycles of 30 sec at 94°C, 30 sec of 47°C , and an extension step of 1 min at 72°C. The final extension step was for 6 min at 72°C and 6 min at 24°C. PCR products were purified using the Ultraclean™ PCR Clean-up™ Kit (MoBio Biosystems Inc.) and sequenced by the Australian Genome Research Facility Ltd (AGRF). For specimens used and Genbank accession numbers see Table 1.

Phylogenetic analysis of COI, species delimitation and concept
One hundred and twenty-one chelonine specimens were sequenced and three outgroups (Euphorinae, Ichneutinae, Miracinae, taken from Murphy et al. 2008) were used for the phylogenetic analyses. The sequences were aligned with the Clustal W ( Thompson et al. 1994) plug-in in GeneiousPro (Drummond et al. 2011). The COI sequences comprise of 489 bp and have open reading frames. Six of the 54 specimens of Ascogaster had a 3-bp indel, representing the five species 25, 28, 29, 34 and 35. This is rather uncommon, but not new among Hymenoptera (Schonfeld et al. 2011). Model selection was tested with JModeltest v. 0.1.1 based on the Bayesian information criterion (Posada 2008). The alignments were exported in the appropriate format for further analysis.
Bayesian phylogenetic analysis was performed using MrBayes v. 3.2.1 (Ronquist and Huelsenbeck 2003). Two independent analyses with four Monte Carlo Markov Chains (MCMC) each were run in parallel for 10 million generations under a GTR+I+G model. The first 25% were omitted as a burn-in. A consensus tree was created. Convergence was verified if the split frequencies fell below the 0.01 threshold.
The GMYC analysis is an established method to identify putative species based on molecular data. For these analyses the result of the phylogenetic analysis was used, although the optimal settings required the exclusion of outgroups (Astrin et al. 2012). The tree was first converted to an ultrametric tree using the package 'ape' (https://r-forge.r-project.org/projects/ ape/) in the R environment (R Development Core Team 2011). The GMYC analyses were then conducted using the 'splits' package (http://r-forge.r-project.org/projects/splits/). We tested both analyses, one allowing for only a single speciation event and the other allowing for multiple events.  n/a n/a EU106964* Ichneutinae indet. n/a n/a EU106967* Miracinae indet. n/a n/a EU106971* Notes: * Murphy et al. (2008). NSW, New South Wales; QLD, Queensland; VIC, Victoria; WA, Western Australia.
The PTP model was also used to establish a putative species estimate based on the molecular data (Zhang et al. 2013). This analysis requires a phylogenetic tree; however, it recommends using a randomised axelerated maximum likelihood (RAxML) tree obtained through RAxML 7.6.3 Blackbox on the CIPRES Science Gateway V 3.3 (Miller et al. 2010). The analysis was conducted using Python. We then compared four independent species estimates; recognition of morphospecies based on fixed phenotypic differences, the PTP estimate, GMYC with single a threshold and GMYC with multiple threshold estimates (Figures 2 and 3; see Pons et al. 2006;Fontaneto et al. 2007;Cook et al. 2010). Even though the PTP and GMYC analyses indicated the presence of cryptic species, we have refrained from describing them, and limited the study to only the description of the morphospecies (see Butcher et al. 2012), as the molecular data are based only on a single marker and the sequence variation is low in some cases. However, we indicate the presence of likely cryptic species in the comments section for the relevant species, and plan to use the results here as a basis for testing cryptic species boundaries with additional markers in the future (Pons et al. 2006;Fontaneto et al. 2007).

Species richness estimation
Species richness of Australian chelonines was calculated using a Chao species richness estimator implemented in the package 'SPECIES' (Wang 2011; http://cran.r-project.org/ web/packages/SPECIES/index.html) for the R environment (R Development Core Team 2011) for the morphospecies available, and based on frequencies of specimens per morphospecies (Wang and Lindsay 2005). These were calculated separately for each genus and summed, and also as a single estimate for Australian Cheloninae as a whole.

Terminology and imaging
Terminology follows Eady (1968), van Achterberg (1988) and Karlsson and Ronquist (2012). Images were taken using a Visionary Digital BK+ photo system with a K2 lens system attached to a 7D Canon digital camera. Final images were stacked from multiple images using Zerene Stacker software (version 1.04), and edited in Adobe Photoshop CS5 (extended version 12.0x64). Measurements were taken using a Zeiss stereomicroscope and Adobe Photoshop CS5 software. The distribution maps were produced with DIVA-GIS (Hijmans et al. 2004). The bioregions in Figure 1 are based on the Interim Biogeographical Regionalisation for Australia version 7 classification by the Australian government (28 November 2013).

Phylogenetic analysis and species delimitation
The phylogeny for Australian Ascogaster and Phanerotoma and the number of species predicted by the four estimation methods are shown in Figures 2 and 3, respectively. For Ascogaster there was strong concordance amongst the four methods and, overall, the molecular-based estimates for species boundaries largely match our morphospecies estimates. However, there are three exceptions. For species 36, here described as A. brevivena sp. nov., PTP and GMYC predict a single species, but tentatively we initially recognised it as two morphospecies. It turned out that the two morphospecies were represented by single but opposite sexes, so we made the pragmatic decision that they represent a single sexually dimorphic taxon. For species 38, PTP and GMYC (single) both predicted three species but, based on morphology and GMYC (multi), only one species could be identified, thus suggesting, albeit inconclusively, the presence of cryptic species. Lastly, for species 42 and 43, GMYC (single) suggested a single species, but the other three estimates including morphology suggested two separate species. In total, the data indicate 44 species of Ascogaster for which we have molecular data, four of which occur in the arid zone and are described here.
The results for Phanerotoma ( Figure 3) were similar to those for Ascogaster in that there was strong concordance between the molecular-based estimates and recognition of morphospecies. The major exception was for species 3, P. decticauda, which was estimated to represent seven species by PTP, GMYC (single) and GMYC (multi), thus suggesting the presence of cryptic species. The specimens showed some morphological variation, but none of the seven clades had any fixed differences. Further studies of male genitalia, host association and sequence data from additional markers will undoubtedly help confirm the validity of these likely cryptic taxa.
Minor differences were evident in species 1, species 14 and species 17/18. But in each case, two of the three molecular-based estimates matched the recognition of morphospecies. In total, the data indicate 18 species of Phanerotoma for which we have molecular data, seven of which occur in the arid zone and are described or redescribed here.

Species richness estimation
Examination of the approximately 5000 Australian chelonine specimens available led to the recognition of 195 morphospecies (Table 2). When the Chao species richness estimator was applied to the same set of specimens for each genus separately it generated a figure of 262 species in total, with a range of 225-377 species ( Table 2). When the data were pooled for all specimens (irrespective of genus), the Chao estimator predicted a total species richness of 278 species with a range of 253-370.
The Chao estimator requires an unbiased collecting effort of an area which is clearly not the case for most terrestrial invertebrates, as indicated for Australian chelonines where the collection intensity has been far greater in eastern Australia, compared with the relatively meagre collecting effort in the arid zone and the western part of the continent. Also, a large number of morphospecies are represented by a single specimen, suggesting that collecting intensity in general, as well as geographically, needs to be significantly improved. In this respect, programmes such as Bush Blitz can make a significant contribution and help ameliorate this situation. Although the accuracy of the Chao estimator is likely to be limited, it still provides a useful starting point and indicates that there are possibly 60-170 chelonine species yet to be discovered on the Australian continent.

Species distributions
In this study, we focussed on the arid zone species of Ascogaster and Phanerotoma. However, when we compared the material from the surveys with material from other habitats of Australia, we found that the majority of species treated here occur also in other regions. Some have a distribution into the tropics, such as P. lutea sp. nov., or into mesic areas, such as P. nigriscapulata sp. nov. (both Figure 16). Two species, P. behriae and P. decticauda, are distributed very broadly across the continent (Figures 17 and 18, respectively). Species with a more restricted distribution are A. brevivena sp. nov. and A. prolixogaster sp. nov. These two species can only be found south of 25°S ( Figure 15). Ascogaster rubriscapa, P. bonbonensis sp. nov. and P. bushblitz sp. nov. are the only species with an exclusively arid Australian distribution (Figures 15 and 16, respectively).  Shenefelt 1973: 814, Shaw 1983: 7, Huddleston 1984: 348, Tang and Marsh 1994. For diagnosis of Australian taxa, see .

Comments
Leptodrepana, described by Shaw (1983) from the New World, is accepted as a valid genus by some authors (e.g. Shaw 1997;Brajkovic et al. 2010), while others have treated it as a junior synonym of Ascogaster (van Achterberg 1990; Yu et al. 2005). Tang and Marsh (1994) followed van Achterberg's synonymy and treated the new species from China as Ascogaster, but also suggested that a comprehensive revision of the group was needed as some species showed characters intermediate between Ascogaster and Leptodrepana. Shaw (1997) discussed the difficult status of Leptodrepana, arguing for a separate genus since Ascogaster would otherwise be paraphyletic. However, no comprehensive analysis has yet been undertaken using a combined morphological and multigene approach to resolve this question. Here we treat all relevant species as Ascogaster, but point out that A. brevivena sp. nov. exhibits characters intermediate between these two genera, such as having equilateral ocelli as in Leptodrepana.  Head. Antenna with 22 antennomeres in females, 24 antennomere in males; ratio of length of third antennomere to fourth 1.1 in females, 1.25 in males; ratio of length of third, fourth, penultimate and terminal antennomere 3.5-3.8, 3.3-3.7, 0.9-1.2, and 1.3-1.6 in females, 3, 3, 1.3 and 1.7 in males times their width, respectively; ratio of length of eye in dorsal view to length of temple 4.3 in females, 4.2 in males; ocelli equilateral; imaginary line between anterior margins of posterior ocelli is not touching the anterior ocellus; ratio of width of face in anterior view to its height 1.9-2.1 in females, 1.8 in males; ratio of width of clypeus to its height 1.0-1.1; clypeus without teeth; ratio of length of malar space to base of mandible 1.6-1.7 in females, 1.4 in males; face and frons punctate; eyes with sparsely minute setae; ratio posterior ocelli:distance between the two posterior ocelli (POL):distance between anterior ocellus and posterior ocellus (LOL):distance between posterior ocellus and eye (OOL) 1.0:1.0:0.7-0.8:2.2-2.6.
Colour. Head brown, paler around eyes; anterior half of antenna light brown, posterior half dark brown; mesosoma black; wings with long brown hair given a smoky appearance of the wings, with a white band underneath the parastigma; legs white, with femur and apical end of tibia brown; anterior end of carapace white, extending dorsal into the posterior dark end.
Male. Head beige; antenna light brown; mesosoma black; legs as female but paler; wings infuscate; wing venation brown; metasoma anterior half white, posterior half black.

Diagnosis
Females differ from all other described Ascogaster in Australia by having a reduced SR-1 vein (not extending to the margin of the fore wing).

Biology
Unknown.

Etymology
The name 'brevivena' reflects the unusual short SR-1 vein in the fore wing of the females.
Metasoma. Shape of metasoma oval in dorsal view; ratio of width of metasoma to its length 0.5-0.6; carapace broadens to posterior end in lateral view; ratio keel to metasoma length 0.2-0.3; carapace fine rugose.
Colour. Head and metasoma black; antenna brown; legs and carapace red brown; wing venation, parastigma and pterostigma dark brown.

Diagnosis
The colour of the carapace distinguishes this species from all those currently described species in Australia. In addition, A. ferruginegaster is one of only five species, along with A. prolixogaster, A. rubriscapa and two undescribed species (Ascogaster sp. 25 and Ascogaster sp. 28), that has a 3-bp indel in its CO1 sequence at position 271-273. Male. Unknown.

Etymology
The name refers to the red-brown carapace.

Distribution
Northern Territory, Queensland, and South Australia (Witchelina Station; Figure 15). Metasoma. Middle lobe of mesoscutum punctate; notauli weakly present; mesoscutellum shiny, fine punctate, convex; mesopleuron shiny, punctate; precoxal sulcus present; ratio of height of mesosoma to its length 1.7-1.8; hind coxa shiny, fine punctate; ratio of length of hind tibia to hind tarsus 0.9; ratio hind coxa, hind femur, hind tibia and hind Metasoma. Shape of metasoma long oval from dorsal view; ratio of width of metasoma to its length 0.4-0.5; carapace broadens to posterior end in lateral view and narrows down in dorsal view, with an extended tip for the ovipositor; ratio keel to metasoma length 0.03-0.04; carapace rugose.
Colour. Black; femur, tegula, tibia and tarsus dark brown to brown.

Diagnosis
The shape of the carapace with its elongated tip is unique among all Australian Ascogaster. In addition, A. prolixogaster is one of only five species, along with A. ferruginegaster, A. rubriscapa and two undescribed species (Ascogaster sp. 25 and Ascogaster sp. 28), that has a 3-bp indel in its CO1 sequence at position 271-273. Male. Unknown.

Biology
Unknown.

Etymology
The name reflects the elongated tip of the carapace.
Metasoma. Shape of metasoma long oval in dorsal view; ratio of width of metasoma to its length 0.48; carapace broadens strongly to posterior end in lateral view; ratio keel to metasoma length 0.11; carapace rugose.

Diagnosis
Superficially similar to A. prolixogaster, but A. rubriscapa lacks the elongated tip of the carapace. Overall, it can be distinguished from all other Australian Ascogaster species by the combination of its colour and size, the shape of the carapace and the number of antennomeres. In addition, A. rubriscapa is one of only five species, along with A. ferruginegaster, A. prolixogaster and two undescribed species (Ascogaster sp. 25 and Ascogaster sp. 28), that has a 3-bp indel in its CO1 sequence at position 271-273.

Biology
Unknown.

Etymology
The name refers to the red scape of the specimen.

Comments
Van Achterberg (1990) stated in his diagnosis of Phanerotoma the presence of three distinctive clypeal teeth, although the number of teeth varies between two and three (Zettel 1990c). However, for Australia only a minority of species have three clypeal teeth, and belong to a species group including P. behriae, P. lutea sp. nov., P. novaguineensis and P. pacifica which have a reduced r vein and a much longer 3-SR vein.
Metasoma. Shape of metasoma oval in dorsal view; ratio of width of metasoma to its length 0.6; carapace flat in lateral view; ratio of keel to length of metasoma 0.2-0.3; first Colour. Head and mesosoma orange; antenna beige, gradually darker to tip; legs beige with posterior half of tibia orange; hind tibia black-white-black striped; wing venation brown, pterostigma and parastigma brown, white between them; wing hyaline; first and second metasomal tergite white, third orange.

Biology
The type for this species was reared from Etiella behrii Zeller (Pyralidae). Some specimens have been reared from the 'Currajong Bag shelter moth' (Dichocrocis clytusalis Walker), but due to the poor condition of the specimens it is difficult to determine whether they actually belong to a separate species or not. One specimen was reared from an unknown host on Dundas Mahogany (Eucalyptus brockwayi C.A. Gardner) Myrtaceae, and additional specimens have been collected from true mulga (Acacia aneura F.Muell. ex Benth.), Gundabluie (Acacia victoriae Benth.) Fabaceae, toothbrush plant (Grevillea sp.), silver needlewood (Hakea leucoptera R. Br.), Snottygobbles (Persoonia bowgada P.H. Weston) Proteaceae, tea tree (Leptospermum sp.) Myrtaceae and American pepper (Schinus molle L.) Anacardiaceae, and in a wheat crop.

Distribution
Previously only recorded from South Australia, New South Wales and Northern Territory (Zettel 1988a), but additional material shows this species to be distributed widely across mainland Australia with new records from Queensland, Victoria, and Western Australia. This species was also found at Bon Bon Station, Hiltaba Station and Witchelina Station (Figure 18). Metasoma. Shape of metasoma oval from dorsal view; ratio of width of metasoma to its length 0.5-0.6; carapace flat in lateral view; ratio of keel to length of metasoma 0.16-0.21; first and second tergite longitudinal strigose; third tergite rugose; posterior end of carapace deeply indented; ratio of the three metasomal tergites 1.0:0.9:1.4-1.5.
Colour. Head orange, face and temple somewhat darker; metasoma dark reddish brown; lateral lobes of mesoscutum orange; wing venation, parastigma, and pterostigma brown; carapace darker around the edges; third tergite completely dark.

Diagnosis
One of only four species of Phanerotoma that have an indented carapace. It differs from the other species by the ratio of metasoma to mesosoma being less or equal to 1 (P. bonboniensis: more than 1), ratio of length to width of antennomeres being less than 3.5 (P. filicornis more than 4), and its dark colour (P. australiensis is yellow brown with black teeth).

Etymology
Named after the type locality, Bon Bon Station.
Colour. Scape, pedicel and head orange; antenna brown; interocellar area dark; mesosoma reddish brown; fore and mid legs yellow; hind leg with brown tibia and femur; first two metasomal tergites beige; second tergite on the margin with a dark stripe; third tergite brown; wings golden infused; wing venation, parastigma and pterostigma brown.

Diagnosis
The wing venation 2-SR+M antefurcal makes this species unique among all Australian Phanerotoma.

Biology
Unknown.

Etymology
Species names refers to the Bush Blitz programme, during which this species was discovered.

CO1 sequence
Genbank accession numbers for this species are KJ438627 and KJ438628.
Colour. Head orange; antenna brown; interocellar area dark; mesosoma mostly reddish brown; pronotum, mesoscutum and prosternum orange; first tergite beige to light brown, gradually from second tergite darker to posterior end; third tergite completely dark reddish brown.

Diagnosis
Phanerotoma decticauda can be distinguished from the other species of Phanerotoma by the rounded carapace, the overall dark colour and the sculpturing. Molecular evidence suggests that this species may well be a species complex (Figure 3). However, the 358191); 1 ♀, Tonga 20 July 57 D.J.P. (QM); 1 ♂, East Claudie Riv,

Etymology
The name 'lutea' refers to the overall yellow appearance of the species.

Distribution
New South Wales, Queensland, and South Australia (including Bon Bon Station; Figure 16).

CO1 sequence
Genbank accession number for this species is KJ438623.
Metasoma. Shape of metasoma oval in dorsal view; ratio of width of metasoma to its length 0.6; carapace flat in lateral view; ratio of keel to length of metasoma 0.15-0.2; carapace longitudinal sculptured; both sutures straight; posterior end of carapace not indented; ratio of the three metasomal tergites 1.0:0.9-1.0:1.2.
Colour. Head orange with reddish brown face and temple; antenna brown; interocellar area dark; metasoma dark reddish brown; with middle lobe of mesoscutum dark orange; first tergite of carapace beige, second and third gradually darker, reddish brown; wing venation, parastigma and pterostigma brown.

Diagnosis
This species differs from all other species by not having an indented carapace on the posterior end, in its sculpturing, and by having a unique colour pattern where the scapula is black and the mesoscutum is light brown forming an 'M'.

Disclosure statement
No potential conflict of interest was reported by the authors.