David Legg

The interrelationships of the major extant arthropod groups are equivocal with various conflicting hypotheses being proposed. Recent large-scale molecular analyses have produced partly equivocal, sometimes novel and often surprising hypotheses of relationships. Some proposed relationships, such as a sister-taxon relationship between chelicerates and myriapods, seemingly lacked morphological support, thus prompting a reevaluation of character homologies and a search for potential synapomorphies. Despite this the novel molecular clades remain elusive in large-scale morphological phylogenies, which tend to favor “traditional” groupings. The causes of this phenomenon are poorly understood. Here we present a comprehensive phylogeny of 307 edysozoans (96 extant, 211 extinct) representing all major panarthropod groups and two cycloneuralian outgroups and utilizing data from morphology, development, behavior and gene-order. We recovered strong support for the inclusion of hexapods within a paraphyletic Crustacea as sister-taxon to remipedes and the Silurian crustaceomorph Tanazios. A remipede-hexapod relationship is consistent with recent molecular analyses but had not been previously recovered in a morphological phylogenetic analysis. This relationship is shown to be the consequence of the inclusion of fossil data within our analysis, reducing the effect of long-branch attraction between extant myriapods and hexapods resulting from convergent adaptations to terrestrialisation. Fossil evidence also indicates that the arthropods have an extensive and diverse stem-group which demonstrates the gradual acquisition of typical arthropod features. The diversity of this long stem serves to better polarize characters within Euarthropoda thus providing a better picture of basal arthropod features and reducing the effects of long-branch attraction amongst extant arthropod clades.

We present a comprehensive cladistic analysis, including 10 genera of aglaspidids sensu stricto, 6 aglaspidid-like arthropods and 42 Palaeozoic arthropod taxa. The results of this work provide a much clearer picture of the phylogenetic relationships among Lower Palaeozoic artiopodans.

The mid-Silurian Wenlock Series Herefordshire Lagerstätte preserves invertebrates in three-dimensions as calcitic void in-fills in carbonate nodules within a volcaniclastic horizon. A new arthropod from this fauna belongs to the ‘short great appendage’ (megacheiran) group. Phylogenetic analysis, based on the most detailed panarthropod character matrix yet assembled, recovers the new Herefordshire genus with other short great appendage genera in a clade that is the most derived of the euarthropod stem. It thus provides new insights into the nature of this stem and the morphology of immediately pre-crown group forms; its phylogenetic position, as well as that of other short great appendage genera, contrasts with the position of this group as determined in many other studies, in which they have been considered to be part of the chelicerate stem lineage; and the new genus represents the first identification of a short great appendage form in strata younger than mid-Cambrian in age.

A remarkable new horseshoe crab from the the mid-Silurian of Herefordshire (~425 Myr) preserves biramous limbs in the prosoma that can be homologized with those of living Limulus. The fossil shows that a true biramous limb evolved in chelicerates as well as in crustaceans, in contrast to the endopod with exite present in trilobites. The evidence of the new fossil supports models for the origin of the limb morphology of living horseshoe crabs through loss of gene expression.

Bivalved arthropods are a common constituent of many early Palaeozoic Konservat Lagerstätten; despite their ubiquity in such sites of exceptional preservation, however, their soft-parts are rarely well preserved. This has made determining their affinities difficult. Often considered relatives of crustaceans due to their carapace morphology, their limb morphology is at odds with this interpretation. Common crustacean features identified in these taxa, including two pairs of antenna and gnathobasic mandibles are actually absent. Recent discoveries from the Burgess Shale (Cambrian, Stage 5) and Chengjiang biota (Cambrian, Stage 3) have provided a wealth of new information regarding the anatomy of these arthropods and indicate they play a key role in understanding some of the fundamental questions in arthropod palaeobiology such as the acquisition of key “arthropod” characteristics in the euarthropod stem-lineage including biramy and head-limb specialisation. They also help to resolve issues regarding the segmental nature of early arthropod appendages. Specifically a number of these arthropods have been shown to possess both frontal antennae and “great-appendages” and calls into question the homology of the “greatappendage” with the appendages of extant arthropods. If homologous then the antenna of such taxa would be considered protocerebral and therefore unique amongst arthropods. The implications of this are discussed.

Whilst progress has been made in recovering a stable lower euarthropod stem group, basal stem groups systematics within the euarthropod crown remain problematic.  Current concepts include an expanded ‘Antennata’ encompassing the mostly Cambrian arthropods possessing a single antenna.  The stem group of the Chelicerata, conversely, is poorly populated, with no universally agreed members before the appearance of chasmataspid trace fossils towards the end of the Cambrian.  In short, the older consensus about the ‘Arachnomorpha’ whereby trilobites and their allies were considered to be related to chelicerates has thoroughly broken down.  A hypothesis in which the ‘great appendage’ arthropods were considered stem-group chelicerates based on their chelicera-like frontal appendages has in its stead gained considerable support.  However, this reconstruction pays insufficient attention to older ideas of stem-group chelicerates possessing an antenniform, not cheliceriform first appendage.  When the ‘xenopod’ taxa such as Sidneyia and Emeraldella are included in the analysis, an antennate first appendage is recovered as a chelicerate and thus euarthropod plesiomorphy.  This reconstruction has significant implications for both the composition of the stems of the pancrustaceans and euarthropods more generally, as well as raising the old problem of functional transitions within arthropod frontal appendages once more.

Cambrian lobopodians have attracted much attention in that they are considered to have close affinities with the origin of arthropods. A new lobopodian species, Diania cactiformis Liu et al. 2011, was recently reported from the Chengjiang Lagerstätte. The authors suggested that this animal bears arthropod-like appendages and resolved D. cactiformis as sister-taxon to Arthropoda in their cladistic analysis. However, a reanalysis of the published dataset does not reproduce the phylogenetic placement of D. cactiformis. In our study, new observations on D. cactiformis are made using newly collected material, which rejects the existence of unequivocal evidence for arthropod-like articulated appendages. The criteria for defining arthropodized appendages are further discussed through comparative studies
among fossil lobopodians, anomalocaridids and arthropods. Our study also challenges the previous description of antero-posterior orientation, terminal claws and trunk tagmosis. A synthesis of recent datasets for lobopodians and stem-group arthropods is used to revisit the phylogeny of stem arthropods. Irrespective of how characters are weighted, D. cactiformis is resolved at a much more basal position to arthropods, and is actually one of the most basal Cambrian lobopodians, indicating that the morphology of this animal’s appendages does not fundamentally inform on the evolution of arthropodization.

With over 1,000,000 described species, arthropods are the dominant phylum on Earth today.  The interrelationships between their constituent clades (subphyla) remain poorly understood however, with conflicts arising between different molecular and morphological data sources.  The position of the pycnogonids (sea spiders) is particularly problematic, some analyses allying them to euchelicerates (horseshoe crabs and arachnids) and others placing them as sister-taxon to all other extant arthropods.  This problem is largely the result of an uncertain root position rather than topological conflicts per se; phylogenetic analyses incorporating stem-group euarthropod taxa have the potential to resolve these issues.  Cambrian lagerstätten contain many candidate stem-group arthropods, supplemented here by two new exceptionally-preserved taxa from the Cambrian of British Columbia: an elongate ‘great-appendage arthropod’ with multipodomerous trunk limbs; and a bivalved form lacking distinct tergal pleurae, a feature previously considered diagnostic of extant arthropods.  A comprehensive phylogenetic analysis was undertaken incorporating these and many other fossil and Recent arthropods; this resolved the stemlineage (including the new taxa) in some detail, documenting many steps in the sequential acquisition of euarthropod characters.  Pycnogonids resolved as sister-taxon to other euarthropods, implying that their chelicerate-like characters (e.g. chelate first appendage) are plesiomorphies rather than synapomorphies for euchelicerates.

Arthropods are a diverse group of animals including spiders, scorpions, centipedes, millipedes, insects and crustaceans. With over 1,000,000 described species they outnumber all other extant (living) phyla combined, and have been the dominant contributor to species diversity for the last 520 million years. They are also commonly used as model organisms for evolutionary studies. Despite all this, we understand very little about their actual relationships, either to each other or other animal groups. Nearly every possible arrangement of the arthropod subgroups has been proposed and each is largely dependent on the type of evidence used, i.e. morphological vs. molecular (DNA) data. Recently a ―total evidence‖ approach has been utilised which includes all available data, including information from extant and extinct species. Fossil data has been shown to be of particular importance in
phylogenetic studies as they tend to resolve major character conflicts created by extant taxa and provide a ―snapshot‖ of evolution. The aim of this talk is to provide an overview of recent works attempting to resolve arthropod interrelations and will demonstrate the importance of fossil data for
resolving this issue.

The marrellomorphs are a small group of Palaeozoic arthropods noted for their aberrant morphology, which has precluded resolution of their phylogenetic position. Marrellomorpha includes two clades, the marrellids and the acerostracans, allied by the shared possession of numerous (>25) trunk somites with appendages that reduce in size posteriorly. The marrellids are characterised by the possession of extensive anteriolateral head spines and typified by the Burgess Shale taxon Marrella; the acerostracans are characterised by the possession of a dorsal ―shield‖ that covers their entire body and appendages. Recent work on marrellomorph anatomy, including descriptions of unpublished material, has augmented our understanding of the morphology of the group, and enabled us to identify potential homologies with other arthropod taxa. A cladistic analysis was undertaken including all currently recognised marrellomorphs and a variety of extinct and extant arthropods. Preliminary results indicate that the marrellomorphs are monophyletic that includes: marrellids; acerostracans; and ―skaniids‖, a paraphyletic group of Cambrian arthropods that, like the acerostracans, possess a dorsal ―shield‖. This analysis also indicates that the marrellomorphs form part of a stem-mandibulate lineage that includes megacheirans, trilobites and various ―orsten‖ taxa.

Arthropod tagmosis is the differentiation of segments along the anterior/posterior axis, producing appendages specialized for sensation, feeding, and locomotion. Characters relating to segment and limb number and morphology are important in constructing phylogenies that include fossil taxa. Chelicerates are a diverse group, including the extant spiders, mites, harvestmen, scorpions, xiphosurans, and the extinct trigonotarbids, eurypterids, chasmataspids, and synziphosurans. The extant pygnogonids (sea spiders) may also belong to this group. The chelicerate body is divided into two major tagmata: prosoma and opisthosoma. There is variation, however, within this bauplan (e.g. number of segments in a limb, fusion of dorsal tergites). Developmentally, patterns of tagmosis are associated with changes in the expression of Hox genes, segment polarity genes, and limb gap genes. The distribution of extant taxa for which Hox expression patterns are known is sparse. Therefore, we studied morphological traits for which the developmental basis is understood in extant chelicerates. We constructed a new phylogeny for chelicerates from over 400 morphological characters. Representative extant taxa with sequence or developmental gene expression data were included, as well as fossils with unique patterns of tagmosis or other morphological traits. Using this morphological topology, ancestral character states were inferred. Results under different models of character evolution (parsimony, likelihood, Bayesian) are compared.

Arthropod tagmosis is the differentiation of segments along the anterior/posterior axis, producing appendages specialized for sensation, feeding, and locomotion. Characters relating to segment and limb number and morphology are important in constructing phylogenies that include fossil taxa. Chelicerates are a diverse group, including the extant spiders, mites, harvestmen, scorpions, xiphosurans, and the extinct trigonotarbids, eurypterids, chasmataspids, and synziphosurans. The extant pygnogonids (sea spiders) may also belong to this group. The chelicerate body is divided into two major tagmata: prosoma and opisthosoma. There is variation, however, within this bauplan (e.g., number of segments in a limb, fusion of dorsal tergites). Developmentally, patterns of tagmosis are associated with changes in the expression of Hox genes. The distribution of extant taxa for which Hox expression patterns are known is sparse. Therefore, we studied
morphological traits for which the developmental basis is understood in extant chelicerates. We constructed a new phylogeny for chelicerates from over 400 morphological characters. Representative extant taxa with sequence or developmental gene expression data were included, as well as fossils with unique patterns of tagmosis or other morphological traits. Using this morphological topology, ancestral character states were inferred. Results under different models of character evolution (parsimony, likelihood, Bayesian) are compared.

With over 1,000,000 currently recognised species, arthropods are the most speciose phylum on Earth. Four distinct major extant clades are recognised: Chelicerata (spiders, scorpions, ticks), Myriapoda (centipedes and millipedes), Hexapoda (insects), and Crustacea (crabs, ostracods, brine shrimp, etc). Although the monophyly of these clades is often well supported, their relationships to each other are hotly debated – with nearly every conceivable grouping proposed. In morphological analyses, the main problem is identifying homologous structures amongst the different clades. New evidence from molecular and fossil data is helping to resolve these problems and establish a consensus. The importance of these techniques for determining patterns of arthropod head segmentation is given as a case study and highlighted as an area for future research.

A restudy of the scorpion fauna of Glencartholm (Lower Carboniferous, Viséan), in Scotland, has revealed that many previously described species are invalid: most taxa represent ontogentic stages of a single taxon, Archaeoctonus glaber (Peach, 1882).  A new phylogeny of Palaeozoic scorpions indicates that the same sequences of changes that occur throughout A. glaber’s ontogeny also occur throughout Palaeozoic scorpion evolution, i.e. ontogeny recapitulates phylogeny.  In particular, abdominal plate morphology, an important taxobase in many previous classifications, varies considerably with ontogeny, and is therefore useful for determining the age of fossil scorpions.  Regression analysis indicates that many aspects of the anatomy of A. glaber grew isometrically.  Isometric scaling indicates that the largest specimen represents the ninth moult and may have grown to a length of over 510 mm.

Acanthoscorpio mucronatus Kjellesvig-Waering was one of three species of fossil scorpion described from the Lower Devonian (Pragian?), Beartooth Butte Formation (Wyoming). A restudy of the only known specimen has revealed this taxon to be a juvenile eurypterid. Previously unrecogonised features supporting this placement include a Dolichopterus-type swimming paddle and a type-A genital appendage. The overall size of the individual, limb proportions and eye placement indicate this specimen is a juvenile. Only two species of eurypterid have previously been described from the Beartooth Butte Formation: Strobilopterus princetonii (Ruedemann) and Jaekelopterus (?) howelli (Kjellesvig-Waering and Størmer). A. mucronatus is considered an early instar of S. princetonii based on swimming paddle morphology and allometric growth estimates. A. mucronatus is a junior synonym of S. princetonii and is thus suppressed under Article 23.1 of the ICZN.
A. mucronatus has previously been assigned to Mesoscorpionina, a clade of terrestrial scorpions. Its reassignment thus has implications for the understanding of scorpion systematics and terrestrialisation. A reanalysis of Jeram’s (1998) data matrix was undertaken with A. mucronatus removed. Greater levels of support were found and Mesoscorpionina became paraphyletic with regards to crown-group scorpions. Other problematic scorpion taxa are briefly discussed in this context.