The Verdun Syndrome: simultaneous origin of protective armour and infaunal shelters at the Precambrian–Cambrian transition

All of the structurally identifiable latest Ediacaran and earliest Cambrian infaunal trace fossils represent shelters of animals feeding above the sediment surface. It is the case with the most complete and oldest radiometrically dated Precambrian–Cambrian transition strata along the Khorbusuonka River in northern Siberia, in the basal Cambrian succession at Meishucun in southern China, richest in small shelly fossils, as well as in the type succession of the Vendian in Podolia, Ukraine. The oldest traces of feeding within the mud are known from no earlier than the late Tommotian of Siberia, Mongolia, Sweden, and Poland. This suggests that the invention of hydraulic mechanisms of sediment penetration was enforced by predation, not by trophic needs. Various ways to protect the body by secretion of a mineral skeleton or building tubes by collected mineral grains were developed by other animals at the same time. Predation may thus appear to be the triggering mechanism for the ‘Cambrian explosion’. Subsequent increase in the depth of bioturbation resulted in a profound change of taphonomic conditions, artificially enhancing the effects of evolution. There are two primary sources of evidence on the Precambrian–Cambrian evolution of marine faunas: the ‘Ediacaran biota’ of imprints of softbodied organisms in sandstone facies and the ‘small shelly fossils’ assemblages of mostly secondarily phosphatized remains of mineral skeletons in limestone facies. Owing to recent progress in understanding the taphonomy of the Ediacaran fossils, significant anatomical data has rapidly emerged from the long-known fossil collections. Imprints of dorsal surfaces of decaying bodies, fixed by early diagenetic iron sulphide cementation of soft sand (‘Ediacaran death masks’; of Gehling (1999)); may offer information on the distribution and mechanical properties of the internal organs (Dzik 2003). Impressions of animal bodies passively or actively settling on the surface of a microbial mat (‘Shroud of Ediacara’; of Dzik (2003)) inform about the external morphology of Ediacaran animals. Among the minute phosphatic and secondarily phosphatized fossils, easy to extract from limestone samples with organic acids and frequently numerous, there are not only elements of the skeletal armour (scleritomes), but also phosphatized organic tissues or even whole embryos (e.g. Bengtson & Zhao 1997). These highly valuable sources of palaeontological information have their limitations, however. Early diagenetic phosphatization was a rare phenomenon, except for the Early Cambrian, and gradually disappeared during the early Palaeozoic, probably the result of increased bioturbation depth in pelagic sediments (Dzik 1994). Fossilization of entirely soft bodies, not easily escaping scavengers and bacterial decay, is even less likely. After their Ediacaran abundance, such fossils became extremely rare. In the Ediacaran White Sea locality Zimnie Gory in northern Russia, the Devonian Bundenbach slates of the Hunsrück Mountains in Germany and the Jurassic Solnhofen lithographic limestone of Bavaria, ‘death tracks’ occur, represented by short (frequently spiral) trace fossils with the carcass of the animal at the end, apparently thrown into a toxic environment. These were thus extreme environments, in the Ediacaran time supporting only bizarre, and probably chemoautotrophic organisms in place (Dzik 2003). Even if the stratigraphic ranges of the Ediacaran biota and small shelly fossils may partially overlap, there is a wide gap in our knowledge of Precambrian– Cambrian organisms in between these two taphonomic windows. The gap exists not only in their time distribution, but also in a range of less unusual environments. This gap can be filled partially with evidence offered by traces organisms left in their activities, where their bodies had no chance to fossilize. Much data on the latest Precambrian and earliest Cambrian trace fossils has accumulated, but the main difficulty in interpreting this data results From: VICKERS-RICH, P. & KOMAROWER, P. (eds) The Rise and Fall of the Ediacaran Biota. Geological Society, London, Special Publications, 286, 405–414. DOI: 10.1144/SP286.30 0305-8719/07/$15.00 # The Geological Society of London 2007. from their lack of precise dating. The situation has improved significantly with the acquisition of diverse and unusually informative trace fossil assemblages from the Mattaia Creek section at Khorbusuonka in northern Siberia (Dzik 2005). These occur primarily below the occurrence of the abundant Manykodes (1⁄4‘Phycodes’) pedum, the appearance of which defines the base of the Cambrian (although it probably also occurs somewhat below in the stratotype section of Newfoundland; Gehling et al. 2001), and from above a radiometrically dated volcanic breccia (Bowring et al. 1993). All are traces of infaunal activities of animals penetrating the sediment but feeding above it. Their burrows were, thus, protective shelters. The first skeletal remains of a variety of organisms occur in the same rock unit (Khomentovsky & Karlova 1993), as well as in coeval strata elsewhere. Thus, it is likely that predation forced the latest Ediacaran animals to either seek shelter under the sediment surface or protect themselves with mineral skeletons (‘the Verdun Syndrome’; Dzik 2005). Such explanation of the sudden emergence of skeletonized animals at the beginning of the Cambrian was first proposed by Evans (1912). In the present paper this idea is developed further, with a more specific presentation of the probable anatomy of the earliest producers of infaunal trace fossils. Additional evidence on the succession of specific modes of infaunal penetration, as reflected by trace fossils from Podolia, Ukraine, and the Meishucun locality in Yunnan, China is also presented. The late Ediacaran infaunal trace makers The Olenek River region in polar Siberia is among the few sites in the world where the Precambrian– Cambrian transition is documented by radiometric dating (Bowring et al. 1993), a stable isotope record (Pelechaty et al. 1996; Kouchinsky et al. 2001), Ediacaran biotas (Fedonkin 1985), skeletal remains (Khomentovsky & Karlova 1993), and trace fossils (Fedonkin 1985). As in fossiliferous successions elsewhere, traces of infaunal activity are missing in strata with Ediacaran fossils in Siberia. Although Seilacher (1999) suggested that most Ediacaran trace fossils makers fed below the microbial mat, this was falsified by the discovery of intact mats in the Zimnie Gory section (Dzik 2003). Both body and trace fossils occur there on the upper surface of the mat. Only shallow surface burrows are known from deposits of such age elsewhere (Jensen 1995, 2003; Jensen et al. 2005). The oldest common traces of deep sediment penetration by worm-like animals have been recovered from the middle part of the Nemakit-Daldynian (Manykayan) Kessyusa Formation in exposures near the mouth of the Mattaia Creek in the Khorbusuonka River section of the Olenek region. Many types of trace fossils occur in the Kessyusa Formation. Virtually all bedding plane trace fossils assemblages there are monospecific, and it is difficult to separate results of evolutionary change from changes controlled by purely ecological and environmental factors. Different species seem to be restricted to specific kinds of sediments. Narrow U-shaped, shallow burrows are found in thin layers of dark mudstone intercalated with siltstone and sandstone. The mud was apparently firm, because the burrows remained open during life of their makers, preserving a cylindrical cross-section (Dzik 2005). Synsedimentary faults offer proof that the light-grey quartz mud, in which serial cylindrical burrows of Manykodes pedum were dug, was firm in its whole bed volume already before deposition of the overlying pure sand layer. Also, the small trilobate burrows of Podolodes were produced in a relatively firm, fine-grained mud. In a softer, but coarser, sediment trace makers tended to dig deeper and make burrows circular in cross-section (Dzik 2005). The horizontal galleries of Mattaia miettensis were excavated in green glauconitic sand. Continuous horizontal galleries in sand Probably the most bizarre Early Cambrian traces of infaunal activity are those produced by Psammichnites. As interpreted by Seilacher (1995) and McIlroy & Heys (1997; referred to as Plagiogmus), these were made by animals horizontally penetrating sandy sediment at a stable depth. Sediment stripes were left behind as a result of their movement with retrograde peristaltic wave. An organ exposed to the surface was cutting the sediment while the worm moved (Seilacher 1995; Jensen 1997) and collecting food together with the sediment from the surface. Continuous burrows of Mattaia miettensis are similar to those of Psammichnites. They occur within the glauconitic sandstone beds in the Mattaia section, as well as in the coeval strata exposed along the Olenek River in the same region of northern Siberia (Fedonkin 1985). Owing to a clear delimitation of particular rock layers by clay-rich laminae (possibly representing remnants of microbial mats), the cross-section of the rock shows the internal structure of the trail (Fig. 1a). The lack of deformation of layers below the trail indicates that it was not produced by action of any hydraulic pushing mechanism, but the sediment was simply abraded by the organism. The trace itself is filled with two bands of a homogenized sediment, laterally raised and depressed in the centre, where the bands are separated by a vertical fissure filled with clay (cf. Young 1972). The J. DZIK 406