Trigonotarbus johnsoni, Pocock, 1911

Jones, Fiona M., Dunlop, Jason A., Friedman, Matt & Garwood, Russell J., 2014, Trigonotarbus johnsoni Pocock, 1911, revealed by X-ray computed tomography, with a cladistic analysis of the extinct trigonotarbid arachnids, Zoological Journal of the Linnean Society 172 (1), pp. 49-70 : 59-60

MORPHOLOGY OF TRIGONOTARBUS JOHNSONI

Carapace

The prosomal region of Tr. johnsoni is lacking the complex ornamentation seen in some taxa, and all aspects of the anatomy are easily seen in the hand specimen. Hence the features identified here differ little from previous interpretations. The original description by Pocock (1911: fig. 38) recognized an essentially triangular carapace with a rounded anterior tip (or clypeus). The carapace has a raised median band bearing a single pair of median eyes, with somewhat sunken lateral flanks. Pocock also figured the transverse groove towards the back of the carapace, although the depression behind the eyes was not shown and is more clearly expressed in some specimens than others. It is tentatively assumed to be a genuine feature and included in the present reconstruction ( Fig. 4 ; see also Dunlop, 1996a: text-fig. 4).

Appendages

Ventrally, Pocock (1911) correctly figured the leg coxae increasing in size posteriorly, but the presence of a sternum between the coxae and the small chelicerae was first recognized by Petrunkevitch (1949: fig. 111) and confirmed here. Neither author saw the lip-like endites on the first leg coxae revealed here by the tomographic reconstruction. The nearest equivalent are projections from the first coxae in Palaeocharinus and the anthracomartids; however, the detailed shape is rather different and it is unclear if these are homologous with the Tr. johnsoni structures. The pedipalps and legs are of a fairly simple construction (i.e. no spines or raptorial structures) and generally give the impression of a compact arachnid with fairly short, stubby appendages. Some ambiguities in the lengths of the leg articles compared with previous descriptions can be clarified here with the help of tomography, such as the fact that the patella is not much shorter than the tibia, whereas the metatarsus is noticeably shorter than the tarsus. A short metatarsus is also seen in palaeocharinids and anthracomartids and may be a plesiomorphic character within trigonotarbids. Tarsal claws could not be resolved here and were added to the reconstruction ( Fig. 4 ) based on comparisons with other trigonotarbid species.

Opisthosoma

Opisthosomal segmentation can be a contentious issue in trigonotarbids. Pocock (1911) did not give a segment count, but Petrunkevitch (1949) recognized only eight segments in Tr. johnsoni . Subsequent study of better preserved material (e.g. the Rhynie Chert trigonotarbids) suggests a ground pattern for trigonotarbids of 12 segments in total; whereby the first tergite is modified into a locking ridge that largely tucks under the back of the carapace, the first sternite may be absent, tergites 2 + 3 are often fused into a large diplotergite, and the last two segments (11–12) are ring-like and form a small pygidium (Garwood & Dunlop, 2010). However, a number of Carboniferous taxa seem to differ from this ground pattern, either by reducing the first tergite and/or having separate tergites 2 and 3; possibly a reversal. Trying to apply this to Tr. johnsoni we are forced to consider two competing interpretations of the relationship between dorsal and ventral segmentation.

The short anterior tergites in Tr. johnsoni imply that there is no diplotergite 2 + 3, and this is reflected in the current figures and idealized reconstruction ( Fig. 4 ). We do, however, recognize a thickening or ridge at the front of the anterior-most tergite, which we tentatively interpret as a highly reduced tergite 1 in which its ‘locking’ function beneath the carapace may have been lost. Counting back from our presumed tergite 2, we can recognize dorsal sclerites back to a posterior-most tergite nine ( Fig. 3A ). This is consistent with all other trigonotarbids in which nine tergites are universally visible dorsally (the first perhaps reduced or hidden). In this scenario ( Fig. 3A ), tergite nine of Tr. johnsoni is rather small and – like that of palaeocharinids – lacks a division into median and lateral plates. Ventrally, segments 11 and 12 should form the small round pygidium (see above) and our favoured scenario of segmentation thus necessitates a fusion of sternites 9 + 10 surrounding the pygidium in order for sternite 8 to match the margins of tergite 8 ( Fig. 4 ). This ventral fusion is potentially a unique character for Tr. johnsoni .

If an alternative scheme were to be chosen in which the first fully visible tergite is (despite its short length) a fused diplotergite 2 + 3, then tergite nine must be both laterally and longitudinally divided. A comparable condition is observed in Anthracomartidae ( Garwood & Dunlop, 2011) and some Eophrynidae . This would then no longer require a fused sternite 9 + 10, but would make the anterior ventral segmentation difficult to reconcile with either the hand specimens or the tomographic model. For this reason the former hypothesis of segmentation is preferred.

Figures

Figure 3. Tomographic reconstruction of Trigonotarbus johnsoni based on specimen NHMUK I. 15860. A, dorsal view. B, fourth right leg, podomeres labelled. C, ventral view. D, ventrolateral view of the coxo-sternal region. Abbreviations: 1–12, segment number; CE1−2, coxal endites 1–2; CH, chelicerae; CL, clypeus; FE, femur; L1−L4, legs 1–4; ME, median eye tubercle; MT, metatarsus; PA, patella; PP, pedipalp; PR, projection; TA, tarsus; TI, tibia; TR, trochanter. Scale bars: A, C = 2 mm, B = 1 mm, D = 0.5 mm.

Figure 4. An idealized reconstruction of Trigonotarbus johnsoni based on the computed tomography scan of NHMUK I. 15860 and additional hand specimens. Scale bar = 2 mm.