A new arachnid order

In 1882, the German zoologist Ferdinand Karsch described a number of fossil arachnids from the coal measures of Neurode in Silesia (now Poland), including one he named Anthracomartus voelkelianus in honour of Herr Völkel, the foreman of the mine where it was discovered.^[9] This species was raised to a new, extinct, arachnid order which Karsch called Anthracomarti. The name is derived from ἄνθραξ (anthrax), the Greek word for coal. A number of other fossils which would eventually be placed in Trigonotarbida were discovered around this time. Hanns Bruno Geinitz described Kreischeria wiedei from the coal measures of Zwickau in Germany,^[10] although he interpreted it as a fossil pseudoscorpion. Johann Kušta described Anthracomartus krejcii^[11] from Rakovník in the Czech Republic, and published further descriptions in a number of subsequent papers.^[12][13][14] In 1884, Samuel Hubbard Scudder described Anthracomartus trilobitus from Fayetteville, Arkansas—the first trigonotarbid from North America.^[15]

Pantetrapulmonata

In the 1980s, Bill Shear and colleagues^[18] carried out an important study on well preserved Mid Devonian trigonotarbids from Gilboa, New York. They questioned whether it was appropriate to define a group of animals on a variable character state and carried out the first cladistic analysis of fossil and living arachnids. They showed that trigonotarbids are closely related to a group of arachnids which have gone under various names (Caulogastra, Arachnidea, etc.), but for which the name Tetrapulmonata has become most widespread.^[19] Members of the Tetrapulmonata include spiders (Araneae), whip spiders (Amblypygi), whip scorpions (Uropygi) and shorttailed whipscorpion (Schizomida) and, together with trigonotarbids, share characters like two pairs of book lungs and similar mouthparts with fangs operating rather like a pocket knife.^[20] In a 2007 study of arachnid relationships, the Shear et al. hypothesis was largely supported and a group Pantetrapulmonata was proposed which comprises Trigonotarbida + Tetrapulmonata.^[21] This has since been corroborated in more recent cladistic analyses.^{[22][23][24][25]}

Trigonotarbids and ricinuleids

In 1892, Ferdinand Karsch suggested that the rare and rather bizarre-looking ricinuleids (Ricinulei) were the last living descendants of the trigonotarbids.^[26] A similar hypothesis was reintroduced by Dunlop,^[16] who pointed out distinct similarities and possible sister group relationship between these arachnid groups. Both have opisthosomal tergites divided into median and lateral plates and both have a complicated coupling mechanism between the prosoma and the opisthosoma which 'locks' the two halves of the body together. Although cladistic analysis has tended to recover ricinuleids in their traditional position closely related to mites and ticks, further discoveries have revealed that the tip of the pedipalp ends in a small claw in both trigonotarbids and ricinuleids.^[17][27] If the hypothesis is true, ricinuleids, despite the lack of tetrapulmonate key characters (e.g. book lungs), may represent part of the pantetrapulmonate clade alongside trigonotarbids as well.^[16][17][20]

Internal relationships

The first cladistic analysis of the trigonotarbids was published in 2014.^[8] This recovered the families Anthracomartidae, Anthracosironidae, and Eophrynidae as monophyletic. In contrast Trigonotarbidae, Aphantomartidae, Palaeocharinidae, and Kreischeriidae were not. Two clades were consistently recovered with strong support—(Palaeocharinus (Archaeomartidae + Anthracomartidae)), and Lissomartus as sister group the 'eophrynid assemblage' (Aphantomartus (Alkenia (Pseudokreischeria (Kreischeria (Eophrynus + Pleophrynus))))).

Prosoma

The prosoma is covered by the carapace and always bears a pair of median eyes.^[1] In the probably basal families Palaeocharinidae, Anthracomartidae^[30]—and perhaps also Anthracosironidae—there is an additional pair of lateral eye tubercles which, at least in palaeocharinids,^[31] appear to have borne a series of individual lenses. In this sense palaeocharinids seem to be in the process of reducing a compound eye.^[32] Anterior margin of the carapace protrude into a projection referred to as clypeus.^[22]

The chelicerae are of the "pocket-knife" type consisting of a basal segment and a sharp, curving fang.^[1] The chelicerae are described as paleognathic: the fangs are held parallel to one another, like those of mesothele and mygalomorph spiders, but the chelicerae hang downwards like those of araneomorph spiders.^[33] There is no evidence in well-preserved fossils for the opening of a venom gland, thus trigonotarbids were probably not venomous. The chelicerae may have been slightly retractable into the prosoma. Well-preserved palaeocharinids show evidence for a small, slit-like mouth with an upper lip (a labrum or rostrum) and a lower lip (or labium).^[34] Inside the mouth there is some sort of filtering system formed from hairs or platelets which strongly suggests that trigonotarbids (like spiders and many other arachnids) could eat only preorally digested, liquified prey.^[34]

The pedipalps have the typical arachnid structure with a coxa, trochanter, femur, patella, tibia and tarsus. They are pediform, i.e. they look like small legs and were not highly modified.^[35] There is no evidence for a special sperm transfer device as in the modified palpal organ of male spiders. In at least the palaeocharinids and anthracomartids the tip of the pedipalp is modified into a small chela (claw) formed from the tarsal claw (or apotele) and a projection from the tarsus. As mentioned above, a very similar arrangement is seen at the end of the pedipalp in Ricinulei.^[17][27]

The walking legs again follow the typical arachnid plan with a coxa, trochanter, femur, patella, tibia, metatarsus and tarsus.^[1] The coxae surround a single sternum. In well preserved palaeocharinids there is a ring, or annulus, around the trochanter–femur joint which may be the remains of an earlier leg segment.^[35] The legs are largely unmodified, although in Anthracosironidae the forelegs are quite large and spiny,^[36] presumably to help catch prey. The legs end in three claws, two large ones and a smaller median claw.^[35]

Opisthosoma

The opisthosoma is largely suboval in outline with a flatten dorsal surface.^[35] It compose of 12 segments, with some of them had undergone degrees of fusion or reduction, hence the previous misinterpretation of around 8 to 11 segments.^[20] Tergite of the first segment partially covered by the posterior margin of preceding carapace, forming a complicated coupling mechanism known as 'locking ridge'.^[1][20] Tergites of segment 2 to 8 (segment 9 in some species) were all laterally divided into 3 (one median and two lateral) plates, with those of segment 2 and 3 fused to each other in most species.^[20] However, the corresponding tergites of the family Anthracomartidae are further subdivided into 5 plates.^[30] The last 3 segments are usually only visible from the ventral side,^[1] with the 2 final segments constricted into a tiny ring-like section known as pygidium.^[20]

Ventral side of opisthosomal segment 2 to 9 covered by series of lung-bearing opercula (2 and 3) and curved sternites (4 to 9).^[1][20] The first segment apparently lacking any ventral plates. Just like other lung-bearing arachnids (scorpion and tetrapulmonate), the book lungs of trigonotarbids formed by layers of trabecula-bearing lamellae, which is a feature adapted to a terrestrial, air-breathing lifestyle.^[37] A pair of ventral sacs located between the posterior operculum and following sternite had been observed in some species.^[38][4][20]