by Jason A. Dunlop*1
Chasmataspidida (Fig 1) are rare, extinct arthropods known only from the early to mid Palaeozoic Era. They are probably closely related to either xiphosurans (horseshoe crabs; Fig 2) or eurypterids (sea scorpions; Fig 1); some chasmataspid fossils were originally misinterpreted as members of one of these two groups. They were first discovered in the 1950s, and were only recently recognized as a group distinct from horseshoe crabs. Chasmataspids are currently the oldest known examples of the Euchelicerata lineage —all Chelicerata except Pycnogonida (sea spiders) — and palaeontologists date the origins of these euchelicerates back to at least the late Cambrian Period.
Most chasmataspids are a few centimetres long. The body is divided into a prosoma at the front and an opisthosoma (or abdomen) at the back. The prosoma is covered from above by a carapace (a plate of exoskeleton), which is sometimes called the prosomal dorsal shield. Not all species are preserved in detail, but the more complete examples show a pair of median eyes (or ocelli) towards the middle of the carapace. There is a pair of larger lateral eyes either side of the median eyes. We can’t see the individual lenses, but eurypterids and horseshoe crabs have multi-faceted compound eyes, so we can assume that the lateral eyes of chasmataspids were similar. In chasmataspids of the genus Chasmataspis (family Chasmataspididae, Fig. 1,4), at least, the centre of the carapace seems to be raised, somewhat resembling the cardiac lobe of horseshoe crabs (a similar raised median structure of the carapace; Fig 2). In the genus Octoberaspis (family Diploaspididae) the lateral eyes appear to sit on a ridge. A similar structure in horseshoe crabs is known as the ophthalmic ridge. Similarities to horseshoe crabs are also apparent in the corners at the back of the carapace, which in well-preserved chasmataspids are drawn out into small (genal) spines.
The underside, or ventral surface, of these animals is not so well known, and no identified specimens reveal the chelicerae or other mouthpart structures. In Chasmataspis we have an example of a disarticulated leg that ends in a small claw, a bit like the leg of a modern horseshoe crab. By contrast, in diploaspids for which the legs are known, the limbs seem to have been fairly short, ending in a blunt point. In at least Diploaspis and Octoberaspis the sixth pair of legs have been transformed into small paddles, suggesting that these animals were able to swim. The genera Octoberaspis and Loganamaraspis reveal an unexpected feature: a small plate, shaped like an oval or heart, which sits over the back of the coxosternal region. This plate was originally thought to be unique to eurypterids, where it covers the chewing gnathobases at the bases of the legs. This is known as the metastoma, may thus have helped in some way with the feeding process.
The opisthosoma is one of the most interesting parts of the chasmataspids. It has a total of thirteen segments (Fig.3); the first is fairly small and inconspicuous. The second, third and fourth segments are wider, and together form a ‘preabdomen’. In Chasmataspis, the segments of the preabdomen are obviously fused together, and form a shield-like plate called a buckler. In diplospids, the three preabdominal segments are not so clearly fused. Behind the preabdomen comes a long ‘postabdomen’ of nine segments (segment numbers 5 to 13). These postabdominal segments may have been ring-shaped; in some species, they have small, spiny projections from the sides called epimera. The opisthosoma ends in a telson. In Chasmataspis it is quite long and spine-like, whereas in diploaspids it is smaller, shorter and in some genera flatter and plate-like in structure.
The underside of the opisthosoma is only known in a few species, but Octoberaspis and Loganamaraspis have a long projection here, usually known as the median abdominal (or genital) appendage. The feature is also seen in eurypterids, and probably played a role in mating. This chasmataspid genital appendage lies across a series of ventral plates (or opercula), which may well have carried the gills in life. These gill plates are not known in Chasmataspis, but some impressions left by a Chasmataspis-like animal sitting on the seabed, dating from the Cambrian Period, suggest that they had three large gill plates on the preabdomen and three more small (possibly gill-) plates, on the first three segments of the postabdomen.
The evolutionary position of the chasmataspids is still not entirely clear. Two families are currently recognised: Chasmataspididae (for Chasmataspis) and Diploaspididae (for all the rest). The two families are really quite different. Chasmataspis was originally thought to be an unusual horseshoe crab — and with good reason (Fig. 4). It has a horseshoe-shaped prosoma, a fused buckler plate rather like the fused abdomen (thoracetron) of the more modern horseshoe crabs, and at least one pair of its legs ends in a claw. By contrast, Diploaspididae are more like eurypterids. The prosoma is squarer, the abdominal segments are not really fused, and well-preserved examples reveal two features previously thought to occur only in eurypterids: the metastoma and genital appendage mentioned above. Some, but not all, diploaspids have paddles, which makes them look like small swimming eurypterids.
Is it possible that chasmataspids are not a natural group: that Chasmataspis is closely related to horseshoe crabs and diploaspids are closer to eurypterids. The problem with this is that all chasmataspids do share one very obvious feature: the way in which the opisthosoma is divided up. No other chelicerates have opisthosomas with a short first segment, three longer and wider segments and then nine narrower segments. That they share such an unusual body plan suggests that all chasmataspids have a single common ancestor.
It will probably take more discoveries before we have a final answer here. Part of the problem is that none of the chasmataspid fossils are really complete — they don’t show both the top and bottom of a whole animal. For example, we don’t know whether Chasmataspis had the eurypterid-like metastoma and genital appendage seen in the diploaspids. Some of the proposed similarities between chasmataspids and horseshoe crabs or eurypterids do not work quite so well when examined closely. Chasmataspis have fused abdominal segments, but the most primitive horseshoe crabs don’t. Diploaspids often have paddles, but the more primitive stylonurid eurypterids don’t. These problems must be taken into account when trying to place chasmataspids in the evolutionary tree.
Given that chasmataspids resemble both horseshoe crabs and eurypterids, we can deduce that they had a similar mode of life. They were probably predators, grabbing at prey with their chelicerae and legs. None of the fossils so far reveal toothed gnathobases at the base of the legs, but because all eurypterids and horseshoe crabs have such structures it would seem reasonable to conclude that they were there, and that chasmataspids used them to chew their food. Members of the Diplosapididae — at least those with paddles — perhaps actively swam after their prey. Species with a genital appendage must have used some sort of complex mating behaviour, but it is not clear what form this took. Most of the fossils seem to come from fairly near-shore environments, rather than deep water, but discussing any further aspects of their lifestyle would be speculation at this stage.
Eight species of chasmataspid have been described so far. The oldest potential record is from the late Cambrian of Texas. Rather than fossils of the whole animal, these are ‘resting traces’, left behind where the animal lay on, or perhaps burrowed into, the soft sediment. The traces show only the underside of the animal, but their outlines are so similar to the slightly younger fossils of Chasmataspis that the Cambrian fossils almost certainly belong to Chasmataspididae.
Chasmataspis laurencii comes from the mid-Ordovician of Tennessee (Fig. 1,4) and is the largest, and one of the best preserved, known species. A considerable number of specimens were recovered during the construction of a local dam. The famous Silurian locality of Lesmahagow in Scotland has produced Loganamaraspis dunlopi. This species is regarded as intermediate between Chasmataspis and the Devonian species; for example, it has a genital appendage but no swimming paddle.
The six remaining species are all from the Devonian Period. Diploaspis casteri and Diploaspis muelleri come from the Rhine region of Germany. Both are quite well preserved and include details of their legs and paddles. Also in good condition is Octoberaspis ushakovi, which has more exotic origins — it was found on October Revolution Island in the Russian Arctic. Another Russian species, this time from Siberia and currently named ‘Eurypterus’ stoermeri, is probably also a chasmataspid. There seem to be further species from Russia awaiting description. The last two specimens, Forfarella mitchelli (Fig. 5) and Achanarraspis reedi (Fig. 3), are both from Scotland but are not so well preserved. They were initially labelled as eurypterids, and are only recognizable as chasmataspids in outline through the characteristic division of the opisthosoma. Dating from the mid-Devonian, Achanarraspis is the youngest known chasmataspid.
Suggestions for further reading:
Dunlop, J. A. 2002. Arthropods from the Lower Devonian Severnaya Zemlya Formation of October Revolution Island (Russia). Geodiversitas 24, 349–379.
Dunlop, J. A., Anderson, L. I. & Braddy, S. J. 2004. A redescription of Chasmataspis laurencii Caster & Brooks, 1956 (Chelicerata: Chasmataspidida) from the Middle Ordovician of Tennessee, USA, with remarks on chasmataspid phylogeny. Transactions of the Royal Society of Edinburgh: Earth Sciences 94, 207–225 (doi:10.1017/S0263593300000626).
Poschmann, M., Anderson, L. I. & Dunlop, J. A. 2005. Chelicerate arthropods, including the oldest phalangiotarbid arachnid, from the early Devonian (Siegenian) of the Rhenish Massif, Germany. Journal of Paleontology 79, 110–124 (doi:10.1017/S0263593300000638).
1 Museum für Naturkunde, Leibniz Institute for Research on Evolution and Biodiversity at the Humboldt University Berlin, 10115 Berlin, Germany