Australian Assassins, Part I: A review of the Assassin Spiders (Araneae, Archaeidae) of mid-eastern Australia

Abstract The Assassin Spiders of the family Archaeidae are an ancient and iconic lineage of basal araneomorph spiders, characterised by a specialised araneophagic ecology and unique, ‘pelican-like’ cephalic morphology. Found throughout the rainforests, wet sclerophyll forests and mesic heathlands of south-western, south-eastern and north-eastern Australia, the genus Austrarchaea Forster & Platnick, 1984 includes a diverse assemblage of relictual, largely short-range endemic species. With recent dedicated field surveys and significant advances in our understanding of archaeid biology and ecology, numerous new species of assassin spiders have been discovered in the montane sub-tropical and warm-temperate closed forests of mid-eastern Australia, including several rare or enigmatic taxa and species of conservation concern. This fauna is revised and 17 new species are described from south-eastern Queensland and eastern New South Wales: Austrarchaea alani sp. n., Austrarchaea aleenae sp. n., Austrarchaea binfordae sp. n., Austrarchaea christopheri sp. n., Austrarchaea clyneae sp. n., Austrarchaea cunninghami sp. n., Austrarchaea dianneae sp. n., Austrarchaea harmsi sp. n., Austrarchaea helenae sp. n., Austrarchaea judyae sp. n., Austrarchaea mascordi sp. n., Austrarchaea mcguiganae sp. n., Austrarchaea milledgei sp. n., Austrarchaea monteithi sp. n., Austrarchaea platnickorum sp. n., Austrarchaea raveni sp. n. and Austrarchaea smithae sp. n. Adult specimens of the type species, Austrarchaea nodosa (Forster, 1956) are redescribed from the Lamington Plateau, south-eastern Queensland, and distinguished from the sympatric species Austrarchaea dianneae sp. n. A key to species and a molecular phylogenetic analysis of COI and COII mtDNA sequences complement the species-level taxonomy, with maps, habitat photos, natural history information and conservation assessments provided for all species.


Introduction
The 'assassin spiders' of the family Archaeidae are an ancient and iconic lineage of basal araneomorph spiders, characterised by a remarkable cephalic morphology and specialised araneophagic ecology. Archaeid spiders are obligate predators of other spiders, and all possess a grossly-elevated, 'pelican-like' cephalothorax and long chelicerae (Figs 1, 4A-C) which are used to hunt and capture their spider prey (Legendre 1961, Forster and Platnick 1984, Wood et al. 2007, Wood 2008. With extant species known only from Australia, southern Africa and Madagascar, the family was first described in Europe from Baltic amber fossil specimens, prior to the discovery of living representatives in the forests of Madagascar in the mid-19 th century (Cambridge 1881, Forster and Platnick 1984, Harvey 2002a, Wood et al. 2007). Other fossil assassin spiders -several congeneric with, and all remarkably similar to, extant taxa -have since been discovered in fossil strata of at least Mesozoic age, spectacularly illustrating the antiquity of the group (Penney 2003, Selden et al. 2008. Indeed, assassin spiders very similar to modern species were probably present throughout the Mesozoic; an observation further evidenced by recent higher-level phylogenetic research indicating the basal position of the Archaeidae relative to other araneomorph spider families (see Griswold et al. 2005, Rix and Harvey 2010.
Assassin spiders are iconic among arachnids due to the extraordinary history of their discovery, their remarkable appearance and antiquity, their limited distribution on the southern continents, their extreme endemism, and their highly specialised araneophagic biology (Forster and Platnick 1984, Harvey 2002a, Wood et al. 2007, Wood 2008. They are the emblem of Madagascar's rich spider fauna (Wood 2008) and have attracted a great deal of research interest in recent years as highly diverse and endemic faunas have been uncovered in Madagascar and southern Africa (see Platnick 1991a, Lotz 1996, 2006, Wood et al. 2007, Wood 2008. The Australian fauna is comparatively poorly-known relative to those from the Malagasy and African regions, despite the presence of dozens of species in south-western, south-eastern and north-eastern . The Recent archaeid fauna consists of 37 described species in three genera (Platnick 2011): Eriauchenius O.P. -Cambridge, 1881 andAfrarchaea Forster &Platnick, 1984 from the Malagasy and African regions; and Austrarchaea Forster & Platnick, 1984, endemic to mainland Australia (Figs 1-2). Only five species of Austrarchaea have previously been described from opposite corners of continental Australia: A. daviesae Forster & Platnick, 1984 from the Atherton Tableland, north-eastern Queensland; the type species A. nodosa (Forster, 1956) from the Lamington Plateau, south-eastern Queensland; A. hickmani (Butler, 1929) from Victoria; A. mainae Platnick, 1991b from the Albany region of south-western Western Australia (see also Main 1995, Harvey 2002a; and A. robinsi Harvey 2002a from the eastern Stirling Range National Park, south-western Western Australia. All five taxa were known only provisionally by their original taxonomic descriptions and subsequent collections, and recent research on Austrarchaea had not progressed beyond a simple recognition of the high levels of diversity and endemism present among Australian taxa (M. Rix, pers. obs.). In fact, the Australian archaeid fauna is far more diverse and widespread than expected even 10 years ago and, with recent advances in our understanding of archaeid biology and ecology, numerous new species and faunas have been discovered, including several species from regions previously assumed to be devoid of Archaeidae (e.g. southern South Australia and the south-eastern coast of Western Australia; see Fig. 2). In south-eastern Queensland and eastern New South Wales, the rainforests and montane wet sclerophyll forests along the Great Dividing Range provide habitats for at least 18 known species of Austrarchaea, most of which were undescribed, and all of which have relatively restricted, highly endemic distributions.
The current paper is thus a taxonomic revision of the species of Archaeidae known from 'mid-eastern' Australia, including those from south-eastern Queensland and eastern New South Wales, north of the Australian Alps (Fig. 2). The type species, A. nodosa, is redescribed from the Lamington Plateau, south-eastern Queensland, and an additional 17 new species are described from habitats between Kroombit Tops National Park in Queensland and the Badja State Forest in southern New South Wales. These 18 species were found to form a monophyletic clade in a molecular phylogenetic analysis (Fig. 3B), and the remaining Australian Archaeidae will be described in forthcoming monographic treatments.

Material and methods
All taxa were described and illustrated from specimens stored in 75% or 95% ethanol. Digital images were taken using a Leica MZ16A binocular microscope and a Leica DM2500 compound microscope, with auto-montage images captured using Leica DFC500 mounted cameras with Leica Application Suite Version 3.6.0 software. Male left pedipalps were dissected prior to imaging and bulbs were aligned for standardised comparison in the retrolateral and pro-distal positions illustrated; expanded pedipalps were illustrated in a retro-ventral position. Female genitalia were dissected and cleared in a 10% lactic acid plus 90% glycerol solution, prior to mounting on temporary glass slides. Illustrations were made on Utoplex tracing paper, using printed template automontage images.
Measurements. Measurements are in millimetres (rounded to the nearest hundredth of a millimetre) and were taken using an ocular graticule on a Leica M80 binocular microscope. Left legs were removed from specimens prior to taking measurements and imaging lateral body profiles. Lateral profile images were standardised for interspecific comparison by vertically aligning the centre of each left anterior median eye with the lower anterior margin of the carapace (above the labrum) (Fig. 6). Carapace height was measured in lateral view, from the margin of the pars thoracica above coxa II to the highest point of the pars cephalica (Fig. 6). Carapace length was measured from the lower anterior margin of the carapace (above the labrum) to the posterior margin of the pars thoracica (above the pedicel) (Fig. 6). 'Neck' width was measured in lateral view, at the narrowest point of the carapace, with total length, carapace width, abdomen length and abdomen width all measured in dorsal view.
Morphometrics. To quantify inter-specific variation in the shape of the cephalothorax and 'head', three morphometric ratios were derived from lateral profile images (see Figs 6-9). The carapace height to carapace length (CH/CL) ratio, used extensively by Wood et al. (2007) and Wood (2008), quantifies the relative dorsal elevation of the carapace, irrespective of gross body size (Fig. 6). The CH/CL ratio used here differs slightly to that described by Wood (2008), in that carapace height and length are measured directly from relative points on the carapace (Fig. 6), and not necessarily at right angles to each other (see measurement definitions, above), thus avoiding any variation caused by tilting of the 'neck' or the non-perpendicular alignment of specimens. For any given size class, mid-eastern Australian Austrarchaea have a CH/CL ratio of 2.00-2.44; taxa with a relatively taller, more greatly elevated pars cephalica have a CH/CL ratio > 2.20 (Fig. 6). The post-ocular ratio (P.O. ratio) (Figs 7-9) measures the length of the 'head' posterior to the AME, relative to the dorsal elevation of the pars cephalica above the level of the AME, and quantifies the significant inter-specific (and often sexually dimorphic) variation seen in the relative dorsal extension of the posterior 'head' region (e.g. see Figs 8D, 8G) . While most species of Austrarchaea from mideastern Australia possess a post-ocular ratio of 0.25-0.35 (Figs 7J,7N,8H), several taxa possess a strongly elevated dorsal pars cephalica, with a P.O. ratio > 0.37 (Figs 7C,8C,8E,9G). In contrast, the highest point of pars cephalica (HPC) to post-ocular length ratio  measures the position of the highest point of the dorsal pars cephalica, relative to the length of the 'head' posterior to the AME. It quantifies the equally significant variation observed in the position of the 'head' apex, from taxa with a more-or-less rounded or hemispherical 'head' in lateral view (HPC to post-ocular length ratio ~0.55-0.70) (Figs 7G,7J,7N) to taxa with a posteriorly extended, conical 'head' (HPC to post-ocular length ratio ~0.90) (Figs 7C, 8C-E).
Molecular and laboratory methods. For molecular analyses, specimens were preserved in 95% ethanol and stored at 4°C. Between two and seven legs of each individual were removed for DNA extractions and whole genomic DNA was extracted from leg tissue samples using the Qiagen DNeasy Blood and Tissue Kit protocol for animal tissues. Polymerase chain reaction (PCR) amplification of target gene regions was achieved using Invitrogen Platinum Taq Polymerase chemistry, in an Eppendorf Mastercycler ep gradient S thermal cycler. For each PCR reaction, 2 µl of extracted DNA, 0.25 µl of Platinum Taq (at 5 u/µl), 2 µl of MgCl 2 (at 50 mM), 2.5 µl of 10x PCR buffer, 5 µl of dNTPs (at 1 mM) and 5 µl of each primer (at 2 µM) were used in every 25 µl reaction. For most taxa, 1071 bp of the mitochondrial cytochrome c oxidase subunit I (COI) gene, along with 535-541 bp of the adjacent COII gene (~1609 bp in total), were amplified using the primers ArCO1 (newly-designed for this study) and C2-N-3661b (modified from Simon et al. 1994), or variants thereof (see Tables  1-2). For several taxa, additional internal primers were used to amplify the same region in two overlapping segments. The PCR protocol used was: 94°C for 1 min; 35x (94°C for 30 sec, 52.1°C for 30 sec, 72°C for 1 min); 72°C for 5 min. The presence of PCR products in PCR reactions was confirmed using standard agarose gel electrophoresis; if PCR products were detected, PCR reactions were then purified using the MoBio Ul-traClean PCR Clean-up Kit. Bi-directional sequencing of purified PCR products was performed by Macrogen Corporation (South Korea), using supplied PCR primers and additional internal sequencing primers (see Table 1). Sequence (.ab1) files for the coding and non-coding strands were assembled automatically as anti-parallel contigs, and then visualised using Sequencher 4.8 (Demonstration Version). Annotated sequences were saved as text files, and imported into ClustalX Version 1.83 (Thompson et al. 1997) for alignment using default parameters.
Conventions. Throughout this paper the term 'Border Ranges' is used to denote the mountainous geographic border region between south-eastern Queensland and northern New South Wales (see Fig. 28B For Material Examined sections, specimens of indeterminate identification (usually juveniles) are included for mapping purposes, and tentatively linked to named species according to their geographic proximity to type localities (or in the case of genotyped juvenile specimens, according to their molecular phylogenetic affinity); such specimens are highlighted in Figures 28-45, and individually listed for each species. Specimens not examined for the current revision, but currently housed at the California Academy of Sciences (due to ongoing research) are also listed separately (with identifications confirmed by H. Wood), along with unexamined juvenile specimens recently accessioned. Specimens sequenced for the molecular analysis are denoted by superscript codes, which correspond to specimen codes as shown in Figure 3B and Table 2. For species Diagnoses, molecular autapomorphies (e.g. see Harvey et al. 2008, Cook et al. 2010) are coded according their nucleotide number , as defined in Table 3 and Figure 3A.
Abbreviations used in the text are as follows:

ALE
Anterior lateral eye/s AME Anterior median eye/s CH/CL Carapace height (CH) to carapace length (CL) ratio F1/CL Femur I length (F1) to carapace length (CL) ratio HPC Highest point of pars cephalica HT 1-6 Abdominal hump-like tubercles 1-6 PME Posterior median eye/s TS 1-3 Tegular sclerites 1-3 Specimens described in this study are lodged at the following institutions:

Phylogenetic analysis
To complement and inform the morphological hypotheses presented for the specieslevel taxonomy (see below), and to provide molecular autapomorphies useful for distinguishing species of Austrarchaea from mid-eastern Australia, a molecular taxonomic approach was employed using mitochondrial DNA nucleotide sequences. A 1071 bp fragment of the cytochrome c oxidase subunit I (COI) gene, along with a 535-541 bp fragment of the adjacent COII gene (Fig. 3A, Table 3), were amplified in species of Austrarchaea (and outgroups) for analysis under a Bayesian framework. These data were generated and aligned as described in the Methods (above), and the resulting nexus file (see Appendix I) was analysed as highlighted (below). Taxa. Specimens of Archaeidae were collected throughout mid-eastern Australia in March-May 2010, for use in molecular analyses. At least three specimens from each major population were sequenced for COI and COII; for some populations,  Gray, 1987in Forster et al., 1987:  (Forster, 1956):   ATAGATAAAGTTCCTTTGTTTGTTTGGTCTGTATTAATTACAGCTATTTTATTAC-TATTATCTTTGCCTGTTTTAGCTGGGGCAATTACAATATTGTTAACAGATCGAAATTT-TAATACTTCTTTCTTTGATCCTGCGGGAGGTGGGGATCCTATTTTATTTCAACATT-TATTTTGATTTTTTGGTCACCCTGAAGTTTATATTTTAATTTTACCTGGTTTTGGTATT-GTTTCTCATGTTATTAGAGGATCAGTAGGTAAGCGTGAGCCTTTTGGTAGATTGGG-GATGATTTATGCTATAGTTGGAATTGGTGGGATAGGGTTTGTTGTATGAGCCCATCAT-ATATTTTCTGTTGGAATGGATGTGGATACTCGGGCGTATTTTACTGCTGCTACTATAAT-TATTGCAGTTCCCACTGGAATTAGGGTATTTAGATGGATAGCTACTTTATATGGGTCT-TATTTTAAATTGGAAGCTCCATTATTATGATGTGTGGGATTTGTGTTTTTATTTACTT-TAGGCGGGGTTACAGGAGTAGTTTTAGCTAATTCTTCTTTAGATATTGTTTTACATGA-TACTTACTATGTGGTTGCTCATTTTCATTATGTGTTAAGTATAGGAGCTGTATTTGC-TATTTTGGCTGGTATTACTTATTGATTTCCTTTGTTTTTTGGGGTAGTTCTGAATTCAA-GGAATTCTAATTTACAATTTTTTATTATATTTATTGGAGTGAATTTAACTTTTTTTCCT-CAACATTTTTTGGGGTTAAATGGTATACCACGTCGTTATTCTGATTATCCTGACGCTTT-TATTTACTGAAATATAGTTTCTTCTTTAGGGTCTTTATTATCTTTATTAGGAATTT-TATTTTTTATATTAATCATTTGAGATGGATTTATTTCAAAAAATTTAGGATTTTCGAAT-TATTATATATATTCTTCGTTGGAGTGAAATAATGGAGTTCCCCCATTAGATCATACATT-TAATCAGTTAGGACAATTGAATATTTAATTT  COII mtDNA (nucleotides 1072-1609)  TTGCCAACTTGAGGTTCATTGTATTTTCAAAATAGTTCTTCTTTTGTTATGGAGCAGT-TAATTTTTTTTCATGATTATACAATGGTAATTTTGATTATGATTATAGTTATTGTGGGG-TATTTATTAGTGAATTCTTGTTATGAAAATTATTATAATCACATGTTAAATGAGGGT-CAAGAGTTAGAGAGAATTTGGACTGTTCTTCCAGCTTTATTTTTGTTATTAATT-GCTTTTCCTTCTTTACAATTGTTATATTTAATAGAGGAAATAGAATTTCCTGAATTAAC-TATTAAAATTTTAGGTCATCAGTGATATTGATCTTATGAGTATAGAGATATAGGTTTA-GATTCGTTTGAGTCTTATATAATCAGAGGGGGGAGTGTACTTTTACGGCTTTTAGAG-GTTGATAATAATTTAGTGATCCCTTATAATTCTATTACTCGTATAATTATTTCTAGAAGA-GATGTTATTCATTCTTGAACTATTCCGTCTTTAGGTGTAAAAATAGATGCTATTCCAG-GTCGATTAAACCAAATTT fewer specimens were available. Most populations of Archaeidae previously known from mid-eastern Australia were successfully sampled and sequenced for the molecular analysis (see superscript DNA codes in the Material Examined sections, below), with numerous newly discovered populations also included. In total, sequences from 94 taxa were added to the final alignment (see Table 2), including 79 Austrarchaea from mid-eastern Australia, one archaeid specimen from north-eastern Queensland and eight Archaeidae from Victoria and Western Australia. A specimen of the Madagascan species Eriauchenius workmani O.P. -Cambridge, 1881 was also included, along with three other Palpimanoidea in the families Mecysmaucheniidae and Palpimanidae. The tree was rooted with the outgroups Hickmania troglodytes (Higgins & Petterd, 1883) (Austrochilidae) and Tarlina smithersi Gray, 1987 (Gradungulidae) (both in the super-family Austrochiloidea), shown to be sister or basal to the Palpimanoidea in previous analyses (see Griswold et al. 2005, Rix and Harvey 2010. Analysis. To infer phylogenetic relationships among sequenced specimens of Archaeidae from mid-eastern Australia, a combined, gene-partitioned Bayesian phylogenetic analysis was executed in MrBayes Version 3.1.2 Ronquist 2001, Ronquist andHuelsenbeck 2003). Prior to analysis, MrModeltest Version 3.7 (Posada and Crandall 1998) was used to choose the appropriate model of nucleotide substitution for each partition under an Akaike Information Criterion (AIC) framework; for the COI data, the GTR+I+G model was invoked with the following settings [Lset nst=6 rates=gamma]; for the COII data, the TVM+I+G model was invoked with the following settings [Lset nst=6 rates=gamma]. For each data partition, parameters were estimated independently ([Unlink tratio=(all) pinvar=(all) shape=(all) statefreq=(all) revmat=(all)]), rates were allowed to vary across partitions ([Prset applyto=(all) ratepr=variable]), and four Markov Chain Monte Carlo (MCMC) chains were run for 20 million generations, sampling every 1000 generations, with the final standard deviation of split frequencies < 0.01 and the first 2,000,000 sampled trees discarded as 'burnin' ([burnin=2000]). Burnin times and log likelikood trace files were visualised using Tracer Version 1.5 (Rambaut and Drummond 2009). Posterior probabilities were calculated and reported on a 50% majority-rule consensus tree of the post-burnin sample.
Results and discussion. The summary phylogenetic tree resulting from Bayesian analysis of the COI and COII data is presented in Figure 3B. The family Archaeidae and the genus Austrarchaea (as currently defined) were both monophyletic and strongly supported, with all mid-eastern Australian taxa similarly united in a monophyletic (although weakly supported) clade (highlighted green in Fig. 3B). Within this mid-eastern Australian lineage, evidence for at least 17 morphological species was supported by 17 equivalently-monophyletic and strongly supported molecular clades; inter-specific (i.e. sister-species) pairwise divergences for the combined (COI + COII) dataset ranged from 8-10%, with intra-specific divergences ranging from 0-6%. Three monophyletic clades from populations known only by juveniles (from the Kanangra-Boyd National Park, Willi Willi National Park and Badja State Forest) had sequence divergences in the range 8-9% (relative to sister-clades), suggesting that these populations may represent distinct species. Deeper species-group lineages were generally poorly supported by the COI and COII data, although A. monteithi sp. n. was clearly inferred as a basal taxon, sister to all other species from mid-eastern Australia (Fig. 3B).
The results of the molecular phylogenetic analysis highlight the utility of comparing molecular and morphological taxonomic techniques, and provide a first insight into the possible phylogenetic relationships among Australian Archaeidae. Despite their exaggerated morphology and specialised ecology, species of Austrarchaea are otherwise morphologically conservative haplogyne spiders, with only relatively subtle inter-specific somatic and genitalic differences between adults, and a diagnostic requirement in most species for adult male specimens. This morphological conservatism, combined with the general paucity of specimens in collections, the relative over-representation of juveniles in collections and in the field, along with the difficulties associated with collecting adult males, renders the identification of species of Austrarchaea difficult based on morphology alone. By sequencing juveniles and adults from across mid-eastern Australia, a much clearer picture of the distribution and limits of each species has been achieved; populations known only from juveniles and females could be confidently linked to type localities, and newly-collected juvenile specimens could for the first time be associated with conspecific adult specimens. In the case of collections made at Binna Burra (Lamington National Park) in April 2010, juvenile specimens of two sympatric species were successfully genotyped to determine their identification, and to test whether A. nodosa and A. dianneae sp. n. were truly sympatric on the Lamington Plateau (see Nomenclatural Remarks for A. nodosa, below).
The phylogenetic relationships inferred for Australian species of Archaeidae remain highly preliminary in the absence of additional genes and a greater taxon sample from southern and north-eastern Australia (M. Rix, unpublished data), however several key results are worthy of discussion. Firstly, the enigmatic A. monteithi sp. n., from the Gibraltar Range National Park (Fig. 19), was clearly inferred as a basal sister-species to all other Archaeidae from mid-eastern Australia, which together formed a monophyletic (although weakly supported) mid-eastern Australian clade (highlighted green in Fig. 3B) sister to an undescribed species from north-eastern Queensland. This result is congruent with morphology, in that the linear gradation seen in the number of dorsal hump-like tubercles on the abdomen (four in north-eastern Queensland taxa; five in A. monteithi sp. n.; six in all other mid-eastern Australian taxa; Figs 5E-G) matches the inferred gradation of clades in Figure 3B. Similarly, the observed gap in the distribution of archaeid species in central Queensland, roughly consistent with the 'St Lawrence Gap' (Webb and Tracey 1981) between Gladstone and Mackay (Fig. 2), seems to reflect a genuine phylogenetic barrier, rather than a collecting artefact. The other major gap in the distribution of Archaeidae in mesic eastern Australia, roughly consistent with the mountainous Australian alpine zone bordering New South Wales and Victoria (Fig. 2), seems to also reflect a second major phylogenetic barrier between a divergent clade of southern Australian taxa (highlighted blue in Fig. 3B) and all other Australian Archaeidae.
Clearly, applying molecular taxonomic methods to a morphological taxonomy is of great utility for species of Austrarchaea. For the current revision, molecular data are clearly linked to specimens and to museum registration numbers by using DNA taxon codes in Material Examined sections, each of which corresponds to an equivalent code in Table 2, and to branch terminals in Figure 3B. To fully integrate the molecular data with the morphological taxonomic hypotheses presented (below), species are also diagnosed (where possible) with unique molecular autapomorphies, in addition to standard morphological characters (see Conventions, above). This approach will facilitate the molecular identification of specimens in the future (as advocated by numerous authors, e.g. Cook et al. 2010), and assist in accurately genotyping juveniles for which genitalic and adult somatic characters are unavailable.  Platnick 1984, Wood 2008. The remarkable, elevated shape of the carapace (Figs 4A-C, 10A-B) and the very long chelicerae (Figs 4B, 4D) will also immediately separate this genus from all other Australian spiders.
Description. Small, haplogyne, araneomorph spiders; total length 2.5 to 5.0. Colouration: Body colouration cryptic and relatively uniform across species, usually with only subtle intraspecific variation in abdominal patterning; carapace, sternum and chelicerae tan brown to dark reddish-brown, interspersed with darker regions of granulate cuticle (Fig. 5), covered in highly reflective setae; legs tan-brown to darker reddish-brown, with pattern of darker annulations on distal segments; abdomen mottled with beige and variable hues of grey-brown (Figs 5E-G), with darker sclerites, scutes and sclerotic spots (Figs 5A-B); paler beige markings due to reflective, subcuticular guanine crystals (Fig. 5B); antero-lateral face of abdomen always with large, humeral patch of reflective guanine crystals (Figs 5A, 5E-G).
Legs and female pedipalp: Legs (longest to shortest) 1-4-2-3, covered with short plumose setae; spines absent; patella I long, greater than one-third length of femur I. Trichobothria present on tibiae and metatarsi of legs; tibiae I-IV each with two trichobothria; metatarsi I-IV each with single trichobothrium; bothrial bases with strongly ridged hood. Tarsi shorter than metatarsi, with capsulate tarsal organ and three claws; tarsi, metatarsi and distal tibiae of legs I-II usually with ventral and pro-ventral rows of moveable, spatulate setae. Female pedipalp with long, porrect trochanter and small tarsal claw; tibia with two dorsal trichobothria.
Abdomen: Abdomen arched anteriorly, rounded-subtriangular in lateral view, usually with four to six large hump-like tubercles on dorsal surface (Figs 5A, 5E-G); cuticle covered with short plumose setae and numerous sclerotic spots (Figs 5A-B). Epigastric region with sclerotised (setose) book lung covers and dorsal and ventral plates surrounding pedicel ( Fig. 5C) (plates fused in males); dorsal pedicel plate with transverse ridges; females with median genital plate and sclerotised lateral sigillae (Figs 5C-D); males with broad dorsal scute fused anteriorly to epigastric sclerites, with or without additional paired sclerites associated with hump-like tubercles (Fig. 5A). Six spinnerets, surrounded by thickened cuticle; ALS largest, PMS smallest; colulus absent. Posterior pair of divided tracheal spiracles situated anterior to spinnerets; males also with transverse row of epiandrous gland spigots situated closely anterior to epigastric furrow.
As noted by Wood (2008), the homology of the tegular sclerites among archaeid genera remains unclear, and this is especially true for Austrarchaea relative to Malagasy and African taxa. For the purposes of this revision, and for an easy comparison among species of Austrarchaea from mid-eastern Australia, the moveable tegular sclerites of the pedipalp are here numbered (1-3), relative to their pro-distal position within the unexpanded tegular cavity (e.g. see Figs 11F, 17F). Tegular sclerite 1 (TS 1) is a porrect, variably spiniform (Fig. 25F), rod-like ( Fig. 20E) or filiform ( Fig. 10F) process (breakable in some specimens; Fig. 21F) that originates near the prolateral base of the conductor, adjacent to the embedded base of the proximal embolic sclerite; during pedipalpal expansion this sclerite usually remains distally directed, positioned adjacent to the embolic haematodocha (Figs 26D, 27E). Tegular sclerite 2 (TS 2) is a distinctive, pointed, usually spur-like process, angled obliquely towards the conductor (Figs 11F, 25F), which is closely associated with the adjacent tegular sclerite 2a (TS 2a); in the unexpanded state, the sinuous, filiform TS 2a is usually obscured and 'locked' within a folded groove along the margin of TS 2 (see Forster and Platnick 1984, figs 60, 62). Tegular sclerite 3 (TS 3) is the most disto-dorsally positioned of the tegular sclerites, with a broader, more plate-like morphology relative to TS 1-2, usually visible as a distally pointed or rod-like projection beyond the retro-distal rim of the tegulum (Figs 14E-F, 17E-F, 20D-E).
Distribution. Assassin spiders occur in mesic habitats throughout south-eastern, south-western and north-eastern mainland Australia (Fig. 2), usually in montane rainforests (Figs 30C, 38C, 41C) and wet eucalypt forests (Figs 39C, 42C, 45C), but occasionally in temperate heathlands or lowland rainforests (Fig. 40C). In south-eastern Australia they occur on Kangaroo Island (South Australia) and along the Great Dividing Range, from Grampians National Park in south-western Victoria north to Kroombit Tops National Park in south-eastern Queensland. In south-western Western Australia they occur from the Leeuwin-Naturaliste National Park east to Cape Le Grand National Park, with outlying populations in the Porongurup and Stirling Range National Parks. In north-eastern Queensland archaeids occur along the Great Dividing Range, from Eungella National Park near Mackay north to the Mount Finnigan Uplands, near Cooktown. Although this distribution is markedly concordant with the distribution of closed and tall open forests in Australia's east and extreme south-west (see Specht 1981), assassin spiders appear to be notably absent from Tasmania, from the Australian Alps and from the 'St Lawrence Gap' (Webb and Tracey 1981) (Fig. 2), as evidenced by the lack of museum specimens and despite targeted searches by the senior author. Forster & Platnick, 1984, A. hickmani (Butler, 1929), A. mainae Platnick, 1991b, A. nodosa (Forster, 1956) and A. robinsi Harvey, 2002a -and the 17 new species from mid-eastern Australia: A. alani sp. n., A. aleenae sp. n., A. binfordae sp. n., A. christopheri sp. n., A. clyneae sp. n., A. cunninghami sp. n., A. dianneae sp. n., A. harmsi sp. n., A. helenae sp. n., A. judyae sp. n., A. mascordi sp. n., A. mcguiganae sp. n., A. milledgei sp. n., A. monteithi sp. n., A. platnickorum sp. n., A. raveni sp. n. and A. smithae sp. n.

Composition. Five described species -Austrarchaea daviesae
Remarks. At least three clades of Archaeidae can be recognised in Australia ( Fig. 3B; see also Wood et al. 2010): a mid-eastern Australian clade, distributed from southern New South Wales to south-eastern Queensland (including the enigmatic, basal species A. monteithi sp. n.); a north-eastern Queensland clade, endemic to tropical Queensland; and a southern Australian clade, known from Victoria, South Australia and south-western Western Australia. For the purposes of this revision, mid-eastern Australian species are diagnosed relative only to other related species from mid-eastern Australia (i.e. A. nodosa and its closest relatives; Fig. 3B), all of which possess five or six dorsal hump-like tubercles on the abdomen (Figs 5F-G) and have a carapace height to carapace length (CH/CL) ratio ≥ 2.00. Austrarchaea daviesae and related species from north-eastern Queensland have only two pairs of hump-like tubercles on the abdomen (Fig. 5E), and A. hickmani, A. robinsi and A. mainae from southern Australia have a carapace height to carapace length (CH/CL) ratio significantly less than 2.00 (M. Rix, pers. obs.).

Key to the species of Austrarchaea known from mid-eastern Australia (males required)
1 Abdomen with five dorsal hump-like tubercles ( Diagnosis. Austrarchaea nodosa can be distinguished from all other Archaeidae from mid-eastern Australia by the broad, flanged proximal portion of the embolic sclerite (Figs 10D-E; see also Forster and Platnick 1984, figs 61, 63) and the unique shape of the conductor (Figs 10D-E), which is thin, gently-tapered and slightly bent along its distal half. The presence of a shallow concave depression near the posterior margin of the 'head' (Fig. 7I) can also be used to distinguish females from most other species, including the sympatric A. dianneae sp. n.
Variation: Males (n=2) Distribution and habitat. Austrarchaea nodosa is known from rainforest habitats along the McPherson Range and 'scenic rim' of extreme south-eastern Queensland and north-eastern New South Wales, in the Lamington, Border Ranges and Mount Warning National Parks (Fig. 28). At Binna Burra (Lamington National Park) it has been found in sympatry with A. dianneae sp. n., in the only known example of twospecies sympatry among Australian archaeids (see Nomenclatural Remarks, below).
Conservation status. This species has a relatively widespread distribution in several National Parks protected under World Heritage legislation, and is not considered to be of conservation concern.
Nomenclatural remarks. The holotype specimen of A. nodosa, described by Forster (1956), is a juvenile (probably penultimate) female from the Tullawallal Nothofagus forest near Binna Burra, Lamington National Park. Although long assumed to have only a single species, the greater Binna Burra region is now the only locality in Australia known to have two species of Archaeidae living in close sympatry: numerous specimens of A. dianneae sp. n. were discovered near Binna Burra in April 2010, along the 'Ships Stern Circuit Track', along with one juvenile specimen of A. nodosa. Both A. dianneae sp. n. and A. nodosa are closely related (Fig. 3B) rainforest-dwelling taxa, rendering the identification of Forster's holotype specimen -and therefore the identification of the generic type species -questionable. To address this issue, and to determine which species was actually described by Forster (1956), two lines of evidence are discussed below.
'Tullawallal' -the type locality cited by Forster (1956) -is a well-known, highaltitude Nothofagus moorei cool-temperate rainforest, situated off Binna Burra's 'Border Track' at around 900 m elevation. The dominant rainforest surrounding Tullawallal is a closed, complex notophyllous vine forest (with isolated warm-temperate and cooltemperate elements), typical of higher elevations throughout the Lamington National Park and McPherson Range (Fig. 28C). In all of the higher-altitude and/or closed rainforests of the Lamington Plateau and Border Ranges National Park, only identifiable specimens of A. nodosa (as recognised above) have so far been collected. Furthermore, the two male specimens collected at or near Tullawallal (WAM T89592, CASENT 9018966) are also both A. nodosa as here recognised. In contrast, the three localities where A. dianneae sp. n. has been found (i.e. along the 'Ships Stern Circuit Track' near Binna Burra, Wojigumal Creek, and in the Tamborine National Park) are significantly lower in altitude than Tullawallal and the surrounding 'Border Track' region of Binna Burra (764 m, 570 m and 313 m, respectively), with more open 'mixed' rainforests and emergent eucalypts at the Binna Burra and Mount Tamborine localities (Fig. 29C).
Secondly, female specimens of both species possess a distinctive 'head' morphology; females of A. nodosa (as here recognised) are characterised by a shallow concave depression posterior to the highest point of the pars cephalica (HPC) (Fig. 7I), whereas females of A. dianneae sp. n. have no such depression and a significantly more pronounced posterior margin of the 'head' (Fig. 7H). The holotype juvenile specimen of A. nodosa has a clear concave depression posterior to the HPC, and 'head' proportions otherwise very similar to the female illustrated in Figure 7I. In contrast, the only known penultimate female specimen of A. dianneae sp. n., collected from near Binna Burra (WAM T112556), does not have a concave depression posterior to the HPC, and 'head' proportions otherwise similar to the allotype female A. dianneae sp. n. illustrated in Figure 7H.
Clearly, given the identification of specimens collected from the type locality and similar nearby habitats, and the morphology of the holotype juvenile specimen, we are as confident as possible in newly-diagnosing A. nodosa as the species described above, given an otherwise highly precarious nomenclatural situation. Paratypes: Allotype female, same data as holotype (QMB S90186); 2 males and 7 juveniles, same data as holotype (WAM T112557 DNA: Ar59-60-M/Ar59-61-J/Ar59-62-J ).

Austrarchaea dianneae
Other Etymology. The specific epithet is a patronym in honour of the late Dianne Wojcieszek (1962Wojcieszek ( -2003, for her love of the Mount Tamborine Hinterland. Diagnosis. Austrarchaea dianneae can be distinguished from all other Archaeidae from mid-eastern Australia except A. cunninghami sp. n. by the shape of the conductor (Figs 11D-E), which is broad, foliate and curved laterally, with a triangular apex; and from A. cunninghami sp. n. by the longer, spiniform tegular sclerite 1 (TS 1) (Fig.  11F) and by the more conical, posteriorly elevated shape of the male 'head' (Fig. 8H).
This species can also be distinguished from other genotyped taxa from mid-eastern Australia (see Fig. 3B) by the following three unique nucleotide substitutions for COI (n = 6): T(303), G(798), A(1065).
Distribution and habitat. Austrarchaea dianneae is known only from subtropical rainforest habitats in the Tamborine and Lamington National Parks south of Brisbane, south-eastern Queensland (Fig. 29). At Binna Burra (Lamington National Park) it has been found in sympatry with A. nodosa, in the only known example of two-species sympatry among Australian archaeids (see Nomenclatural Remarks for A. nodosa, above).
Conservation status. This species is a short-range endemic taxon (Harvey 2002b), which although restricted in distribution, is abundant within the Tamborine National Park (M. Rix, pers. obs.) and is further protected within the World Heritage-listed Lamington National Park. It is not considered to be of conservation concern. Etymology. The specific epithet is a patronym in honour of British botanist and explorer Allan Cunningham (1791-1839), after whom the type locality of this species -Cunningham's Gap in the Main Range National Park -is named.

Austrarchaea cunninghami
Diagnosis. Austrarchaea cunninghami can be distinguished from all other Archaeidae from mid-eastern Australia except A. dianneae by the shape of the conductor (Figs 12D-E), which is broad, foliate and curved laterally, with a triangular apex; and from A. dianneae by the shorter, sharply-tapered tegular sclerite 1 (TS 1) (Fig. 12F) and by the more rounded, less conical shape of the male 'head' (Fig. 8G).
Distribution and habitat. Austrarchaea cunninghami is known only from rainforest habitats in the Main Range National Park of extreme south-eastern Queensland (Fig. 30).

Conservation status.
This species is a short-range endemic taxon (Harvey 2002b), which although restricted in distribution, is abundant within the World Heritagelisted Main Range National Park near Cunningham's Gap (M. Rix, pers. obs.). It is not considered to be of conservation concern. Etymology. The specific epithet is a patronym in honour of Australian naturalist, zoologist, conservationist, author, wildlife photographer and documentary film-maker Densey Clyne, for her landmark contributions to Australian natural history, and for having such a profound impact on the senior author during his formative childhood years.

Austrarchaea clyneae
Diagnosis. Austrarchaea clyneae can be distinguished from all other Archaeidae from mid-eastern Australia by the very long, spiniform tegular sclerite 1 (TS 1) (Fig.  13E) combined with the unique shape of the conductor (Figs 13C-D), which is thin, gently-curved laterally and pointed distally.
Female: Unknown. Distribution and habitat. Austrarchaea clyneae is known only from rainforest habitats in the Mount Clunie National Park of extreme north-eastern New South Wales (Fig. 31). A juvenile specimen from Tooloom National Park (near Urbenville) may also belong to this species based on proximity.
Conservation status. This species appears to be a short-range endemic taxon (Harvey 2002b), which although potentially restricted in distribution, seems wellprotected in at least one World Heritage-listed National Park. It is not considered to be of conservation concern. Etymology. The specific epithet is a patronym in honour of Dr Robert Raven, for his extraordinary contributions to arachnology, and for his ongoing efforts documenting the diverse spider fauna of south-eastern Queensland.

Austrarchaea raveni
Diagnosis. Austrarchaea raveni can be distinguished from all other Archaeidae from mid-eastern Australia by the very short, barely differentiated comb of accessory setae on the male chelicerae (Fig. 14C) combined with the unique shape of the conductor (Figs 14D-E), which is 'ear-shaped' with a large proximal lobe.
This species can also be distinguished from other genotyped taxa from mid-eastern Australia (see Fig. 3B) by the following three unique nucleotide substitutions for COI and COII (n = 6): G(9), G(843), T(1408).

Distribution and habitat.
Austrarchaea raveni is known only from rainforest habitats at Mount Glorious, Mount Nebo and Mount Mee, on the D'Aguilar Range north-west of Brisbane, south-eastern Queensland (Fig. 32).
Conservation status. This species is a short-range endemic taxon (Harvey 2002b), which although restricted in distribution, is relatively abundant within several National Parks and Forest Reserves (M. Rix, pers. obs.). It is not considered to be of conservation concern. Etymology. The specific epithet is a patronym in honour of Judy Rix, for her love of the Sunshine Coast hinterland, and for a lifetime of generosity and support to the senior author.

Austrarchaea judyae
Diagnosis. Austrarchaea judyae can be distinguished from all other Archaeidae from mid-eastern Australia by the small body size of males and females (Fig. 6) and by the unique shape of the conductor (Figs 15D-E), which is 'spade-shaped' and laterally incised.
This species cannot be distinguished from other genotyped taxa from mid-eastern Australia on the basis of unique nucleotide substitutions, but can be distinguished from all other genotyped taxa from south-eastern Queensland (see Fig. 3B) by the following three nucleotide substitutions for COI and COII (n = 6): G(1010), A(1413), T(1560).

Distribution and habitat.
Austrarchaea judyae is known from rainforest habitats on the Blackall and Conondale Ranges of south-eastern Queensland, in the Conondale National Park, Mapleton Forest Reserve and in the region surrounding Maleny/ Montville (Fig. 33). A juvenile specimen from Oakview State Forest (near Gympie) may also belong to this species based on proximity.

Austrarchaea harmsi
Etymology. The specific epithet is a patronym in honour of Danilo Harms, for his contributions to arachnology, and his invaluable assistance to the senior author during field work in south-eastern Australia.
Diagnosis. Austrarchaea harmsi can be distinguished from all other Archaeidae from mid-eastern Australia by the dense, pick-like tuft of accessory setae on the male chelicerae (Fig. 16C) and by the unique shape of the conductor (Figs 16D-E), which is 'shield-shaped' and twisted proximally.
Distribution and habitat. Austrarchaea harmsi is known only from araucarian rainforest habitats in the Bunya Mountains National Park of south-eastern Queensland (Fig. 34).
Conservation status. This species is a short-range endemic taxon (Harvey 2002b), which although restricted in distribution, is abundant within the Bunya Mountains National Park (M. Rix, pers. obs.). It is not considered to be of conservation concern. Etymology. The specific epithet is a patronym in honour of Aleena Wojcieszek, for her love of assassin spiders, and for her support of the senior author over many years.
Diagnosis. Austrarchaea aleenae can be distinguished from all other Archaeidae from mid-eastern Australia except A. alani sp. n. by the very large, porrect tegular sclerite 3 (TS 3) (Figs 17D-F); and from A. alani sp. n. by the dense tuft of accessory setae on the male chelicerae (Fig. 17C).
Variation: Males (n=3) Distribution and habitat. Austrarchaea aleenae is known only from rainforest habitats in the Kalpowar-Builyan region of south-eastern Queensland, in the Bulburin National Park and nearby Kalpowar State Forest (Fig. 35).

Austrarchaea alani
Etymology. The specific epithet is a patronym in honour of Alan Rix, for his great assistance in helping to collect this species, and for a lifetime of generosity and support to the senior author.
Diagnosis. Austrarchaea alani can be distinguished from all other Archaeidae from mid-eastern Australia except A. aleenae by the very large, porrect tegular sclerite 3 (TS 3) (Figs 18D-F); and from A. aleenae by the short comb of accessory setae on the male chelicerae (Fig. 18C).
This species can also be distinguished from other genotyped taxa from mid-eastern Australia (see Fig. 3B) by the following three unique nucleotide substitutions for COI and COII (n = 5): T(684), A(1218), C(1347).
Variation: Males (n=2) Distribution and habitat. Austrarchaea alani is known only from rainforest habitats in the Kroombit Tops National Park of south-eastern Queensland (Fig. 36).
Conservation status. This species appears to be a short-range endemic taxon (Harvey 2002b), which although potentially restricted in distribution, is abundant within the Kroombit Tops National Park (M. Rix, pers. obs.). It is not considered to be of conservation concern. Etymology. The specific epithet is a patronym in honour of Dr Geoff Monteith, for first discovering this species in the Gibraltar Range National Park.

The New South Wales fauna
Diagnosis. Austrarchaea monteithi can be distinguished from all other Archaeidae from mid-eastern Australia by the presence of only five dorsal hump-like tubercles on the abdomen (Fig. 5F).
Distribution and habitat. Austrarchaea monteithi is known only from subtropical rainforest habitats in the Gibraltar Range National Park of north-eastern New South Wales (Fig. 37).
Conservation status. This enigmatic species has an imperfectly known distribution, and although potentially restricted, appears to be relatively abundant within the World Heritage-listed Gibraltar Range National Park near Richardsons Creek (M. Rix, pers. obs.). It is not considered to be of conservation concern. Etymology. The specific epithet is a patronym in honour of Christopher Rix, for his close association with the Dorrigo region, and for his great achievements, both personal and professional.
Distribution and habitat. Austrarchaea christopheri is known from rainforest habitats throughout the Dorrigo and New England hinterland of north-eastern New South Wales (west and south-west of Coffs Harbour), in the Dorrigo, Cascades and New England National Parks (Fig. 38).
Conservation status. This species has a relatively widespread distribution in several National Parks protected under World Heritage legislation, and is not considered to be of conservation concern. Etymology. The specific epithet is a patronym in honour of Dr Norman Platnick and his wife Nancy. Dr Platnick's pioneering research into many different spider lineages -including Archaeidae -has inspired a generation of arachnologists.

Austrarchaea platnickorum
Diagnosis. Austrarchaea platnickorum can be distinguished from all other Archaeidae from mid-eastern Australia by the very long, spiniform tegular sclerite 1 (TS 1) (Fig. 21F) combined with the unique shape of the conductor (Figs 21D-E), which is thin and 'arrow-shaped', with a long triangular apex.
Distribution and habitat. Austrarchaea platnickorum is known only from rainforest and mesic closed forest habitats in the New England National Park of north-eastern New South Wales (Fig. 39).
Conservation status. This species has an imperfectly known distribution, and although potentially restricted, appears to be abundant within the World Heritage-listed New England National Park near Point Lookout (M. Rix, pers. obs.). It is not considered to be of conservation concern. Etymology. The specific epithet is a patronym in honour of Dr Greta Binford, for her pioneering research on spider venoms and for contributing to a highly successful basal clades tour.
Diagnosis. Austrarchaea binfordae can be distinguished from all other Archaeidae from mid-eastern Australia by the very long, spiniform tegular sclerite 1 (TS 1) (Fig.  22F) combined with the unique shape of the conductor (Figs 22D-E), which is thin and slightly curved laterally, with a ridged ventral margin.
Distribution and habitat. Austrarchaea binfordae is known only from lowland subtropical rainforest habitats in the Kerewong and Lorne State Forests, near Wauchope, New South Wales (Fig. 40). Two juvenile specimens from Mount Banda Banda (Willi Willi National Park) may also belong to this species, but possess divergent mtD-NA sequences indicative of possible speciation (Fig. 3B).
Conservation status. This species appears to be a rare short-range endemic taxon (Harvey 2002b), with populations in the Kerewong and Lorne State Forests potentially threatened by land-clearing, habitat degradation, fire and climate change. It is one of the few archaeids known to occur in lowland rainforest habitats in south-eastern Australia, and many of these habitats have been severely impacted by forestry activities. Female (WAM T112568): Total length 3.74; leg I femur 2.96; F1/CL ratio 2.24. Cephalothorax dark reddish-brown; legs tan-brown with darker annulations; abdomen mottled grey-brown and beige (Fig. 23A). Carapace tall (CH/CL ratio 2.02); 1.32 long, 2.67 high, 1.23 wide; 'neck' 0.65 wide; bearing two pairs of rudimentary horns; highest point of pars cephalica (HPC) near middle of 'head' (ratio of HPC to post-ocular length 0.59), carapace gently sloping posterior to HPC; 'head' not strongly elevated dorsally (post-ocular ratio 0.21) (Fig. 7L). Chelicerae without accessory setae on anterior face of paturon. Abdomen 2.10 long, 1.41 wide; with three pairs of dorsal hump-like tubercles (HT 1-6). Internal genitalia with dense cluster of ≤ 15 variably shaped spermathecae on either side of gonopore, clusters meeting near midline of genital plate (Fig. 23F); innermost (anterior) spermathecae longest, sausage-shaped, curved antero-laterally; outermost (posterior) spermathecae bulbous; other spermathecae variably pyriform, mostly straight, directed antero-laterally.
Distribution and habitat. Austrarchaea milledgei is known only from rainforest and mesic closed forest habitats on the Barrington Tops Plateau, in the Barrington Tops National Park and Barrington Tops State Forest, New South Wales (Fig. 41). A juvenile specimen from Chichester State Forest and a single female specimen from Gloucester Tops may also belong to this species (see Remarks, below).
Conservation status. This species is a short-range endemic taxon (Harvey 2002b), which although restricted in distribution, is abundant within the World Heritagelisted Barrington Tops National Park (M. Rix, pers. obs.). It is not considered to be of conservation concern.
Remarks. Specimens of Austrarchaea from the Barrington Tops Plateaux (i.e. Barrington Tops, Gloucester Tops and Chichester State Forest) seem to exhibit a larger than normal amount of morphological and molecular variation, as highlighted by the deep (~6%) COI and COII sequence divergences observed among specimens collected along the Barrington Tops Forest Road in April 2010 (Fig. 3A), and the incongruous CH/CL ratio of the single female (AMS KS102950) from Gloucester Tops relative to the only known female from Barrington Tops (WAM T112568) (see Fig. 6). It is possible that two cryptic species may occur in sympatry on the various mountains and plateaux that make up the Barrington Tops region, although the current paucity of specimens makes this difficult to ascertain. For the purposes of this revision, specimens from the northern Barrington Tops National Park (and State Forest) are recognised as conspecific with the holotype of A. milledgei, but further work is required to compare males from across the region, and to confirm the identification of the Gloucester Tops and Chichester State Forest populations.
Conservation status. This species is a short-range endemic taxon (Harvey 2002b), which although restricted in distribution, is abundant within the eastern Coolah Tops National Park (M. Rix, pers. obs.). It is not considered to be of conservation concern.
Distribution and habitat. Austrarchaea smithae is known only from wet eucalypt forest habitats in the Blue Mountains National Park west of Sydney, New South Wales (Fig. 43). Numerous juvenile specimens from the Kanangra Walls Plateau (Kanangra-Boyd National Park) may also belong to this species, but possess divergent mtDNA sequences indicative of possible speciation (Fig. 3B).
Conservation status. This species has an imperfectly known distribution, and although potentially restricted, appears to be relatively abundant within the World Heritage-listed Blue Mountains National Park near Mount Wilson (M. Rix, pers. obs.). It is not considered to be of conservation concern.
Other Etymology. The specific epithet is a patronym in honour of Helen Rix, for her love of the Illawarra Escarpment, and for her hospitality to the senior author during field work in eastern Australia.
Female: Unknown. Distribution and habitat. Austrarchaea helenae is known only from rainforest habitats in the Macquarie Pass National Park, on the Illawarra Escarpment of southeastern New South Wales (Fig. 44). A juvenile specimen from Barrengarry Mountain (Morton National Park) may also belong to this species based on proximity.
Conservation status. This species appears to be a rare short-range endemic taxon (Harvey 2002b), with populations on the Illawarra Escarpment potentially threatened by land-clearing, habitat fragmentation and fire. Much of the original rainforest of the Illawarra region has been cleared for agriculture and livestock, and only isolated fragments of forest remain. hump-like tubercles (HT 1-6). Internal genitalia with cluster of ≤ 12 variably shaped spermathecae on either side of gonopore, clusters meeting near midline of genital plate (Fig. 27F); innermost (anterior) spermathecae longest, sausage-shaped, curved anterolaterally; other spermathecae variably aciniform, straight, directed antero-laterally.
Distribution and habitat. Austrarchaea mcguiganae is known only from mesic closed forest habitats in the Monga National Park of southern New South Wales (Fig.  45). A female specimen from Deua National Park may also belong to this species based on proximity, and numerous juvenile specimens from the Badja State Forest possess divergent mtDNA sequences indicative of possible speciation (Fig. 3B).
Conservation status. This species appears to be a short-range endemic taxon (Harvey 2002b), which although potentially restricted in distribution, is abundant within the Monga National Park near Link Road (M. Rix, pers. obs.). It is not considered to be of conservation concern. port, discussion and collaboration on all things 'archaeid', and special thanks to Hannah for providing data on Australian specimens housed at the CAS, and for sending DNA extractions of Malagasy and African taxa for use in molecular analyses. Hannah Wood, Jeremy Miller and an anonymous reviewer provided helpful comments on an earlier draft of the manuscript. Figure 1. Habitus images of live Archaeidae from mid-eastern Australia: A-B, female Austrarchaea nodosa (Forster, 1956) from Binna Burra, Lamington National Park, Queensland; C-D, female A. mascordi sp. n. from Coolah Tops National Park, New South Wales; E-F, juvenile A. raveni sp. n. from Mount Glorious, Queensland. Images A-D by M. Rix; images E-F by Greg Anderson, used with permission.  . Molecular phylogenetic data analysed as part of this study. A, Schematic map of the mitochondrial cytochrome c oxidase subunit I-II (COI-COII) gene complex in Archaeidae and other basal Araneomorphae, showing (i) the position of primers used to amplify and sequence 1.6 kilobases of mtDNA, and (ii) the inferred stop and initiation codons for COI and COII, respectively. Note the centralised, overlapping position of the two internal sequencing primer sites (SeqF2a/SeqR1), and the TTG initiation codon for COII, present in all but one of the spider species sequenced for this study. B, Majority-rule consensus tree with re-estimated branch lengths, resulting from a combined, gene-partitioned Bayesian analysis of the COI-COII mtDNA data. Thickened branches represent clades with >95% posterior probability support, and individual support values are shown above other nodes.   Figure 6. Graphs depicting the relationship between carapace length (CL) and carapace height (CH) for species of Austrarchaea from mid-eastern Australia. Overall body size variation is quantified by the relative lengths of the carapace, whereas carapace shape variation is reflected by the CH/CL ratio; taxa with a very tall, greatly elevated pars cephalica have a CH/CL ratio > 2.20. Circles • denote New South Wales and southern Queensland species; and triangles ▲ denote Queensland species (from north of the Border Ranges). Note the relatively small body sizes of A. judyae sp. n., A. binfordae sp. n. and A. alani sp. n., and the relatively tall carapaces of most Queensland taxa. Note also the smaller body sizes and lower variance in carapace length among males relative to females.    (Forster, 1956) (QMB S75416). Asterisks (*) denote concave depressions.                     (Forster, 1956), distribution and habitat: A, topographic map showing the known distribution of Archaeidae in south-eastern Queensland and eastern New South Wales, with collection localities for A. nodosa highlighted in yellow (red highlighted localities denote juvenile specimens of tentative identification); B, satellite image showing detail of inset (A); C, subtropical rainforest near the type locality -Binna Burra, Lamington National Park, Queensland (April 2010). Image (C) by M. Rix.