Ediacaran paleobiology and biostratigraphy of the Nama Group, Namibia, with emphasis on the erniettomorphs, tubular and trace fossils, and a new sponge, Arimasia germsi n. gen. n. sp.

Abstract

Ediacaran fossils, obtained in stratigraphic context in 1993, 1995, and 1996, with the assistance of A. Seilacher, IGCP project 320 scientists, and the Geological Survey of Namibia, are described for the first time. Most are from the Kliphoek and Buchholzbrunn members of the Dabis Formation and the Huns and Spitskop members of the Urusis Formation, Witputs subbasin, but a significant number, including Pteridinium, are from the Kliphoek Member, Zaris Formation, and the Neiderhagen Member, Nudaus Formation, north of the Osis arch, which separates the two subbasins. We extend the stratigraphic ranges and geographic distributions of several important taxa, including Archaeichnium, Ernietta, Pteridinium, and Swartpuntia, provide reassessments of the paleobiology of these and other organisms, and describe a new sponge—possibly an unmineralized archaeocyath—Arimasia germsi n. gen. n. sp. We also describe and illustrate various ichnofossils (including the oldest known traces from the Nama Group), narrow down the first appearance of Treptichnus in the Nama succession, and reinforce the idea that there was a prolific infauna of micrometazoans during the latest Ediacaran by naming and describing previously reported microburrows found on the surfaces of gutter casts as Ariichnus vagus n. igen. n. isp.

UUID: http://zoobank.org/8c267425-135a-4b0a-98b6-cf726515cbf2

Non-technical Summary

This work describes and illustrates Ediacaran (latest Precambrian) body and trace fossils collected in Namibia with the assistance of the Geological Survey of Namibia during 1993–1996. All of the fossils are impressions left in sandstones by the remains or activities of soft-bodied animals that have no obvious living counterparts. The challenge has been to understand the morphology of these organisms, describe their anatomy, and find places for them in the tree of life. The focus is on three erniettomorphs, Ernietta, Pteridinium, and Swartpuntia; a problematical organism named Archaeichnium that may be related to sea anemones; a new simple unmineralized sponge (Arimasia); and tubular fossils and trace fossils, all attributable to worms. We show how these fossils fit into the well-established stratigraphic context of the Nama sedimentary basin and briefly comment on their importance for the evolution of early animal life.

Introduction

The superbly exposed latest Ediacaran to earliest Cambrian succession in southern Namibia has produced some of the most outstanding body and trace fossils of soft-bodied Precambrian animals since first explored early last century (Gürich, 1933; Richter, 1955; Haughton, 1960; Pflug, 1970a, b, 1972; Germs, 1972a, b, 1973; Glaessner, 1978, 1979a; Crimes and Germs, 1982; Hahn and Pflug, 1985a, 1988; Narbonne et al., 1997; Jensen et al., 2000; Grazhdankin and Seilacher, 2005; Jensen and Runnegar, 2005; Wilson et al., 2012; Vickers-Rich et al., 2013; Elliott et al., 2016; Ivantsov et al., 2016, 2019; Buatois et al., 2018; Darroch et al., 2021; Turk et al., 2022; Bowyer et al., 2023). We continue that tradition by describing and interpreting new material from the southern (Witputs) and northern (Zaris) subbasins that house the Nama succession and extend the known distribution of Ediacaran organisms both stratigraphically and geographically. We first describe the stratigraphic context for our samples (Figs. 1–5), then deal with their systematics (Figs. 6–26) and conclude with a discussion of the implications of our findings.

Figure 1. Nama Group outcrop (shaded area) and locality map of southern Namibia. Cities, towns, and major geological features are shown in larger lettering; farms and other local features are smaller.

This study began with a generous invitation from Dolf Seilacher for JGG and BR to participate in a field discussion of Seilacher's vendobiont hypothesis at the now-famous Amphitheatre site on Aar farm and subsequently at H.D. Pflug's home in Lich, as chronicled by Mark McMenamin in The Garden of Ediacara (McMenamin, 1998). During that trip, we also tried to re-collect Germs and Richter's localities at Arimas, Chamis, Kliphoek, Kuibis, and Vrede, with limited but encouraging success. Subsequent work in 1995 and 1996, with the assistance and support of the Geological Survey of Namibia, produced much of the material described here. The work was and is aimed at documenting the Ediacaran biodiversity of Namibia. We illustrate, describe, and discuss the taxa studied but do not deal with some other well-known Namibian forms, such as Rangea Gürich, 1930 (Gürich, 1930a), which are well treated elsewhere. A few uncommon taxa are illustrated but not described.

Geological setting

Lithostratigraphy and geochronology

The largely undeformed stratigraphy of the Nama Basin in southern Namibia was well described by Germs (1972a, 1974, 1983) and then put into a sequence stratigraphic framework by Saylor et al. (1995, 1998 2005), Smith (1999), and Saylor (2003). Briefly, the Nama Basin is divided into three subbasins by topographic highs that were present during sedimentation. Ediacaran fossils are found only in the two western subbasins, Zaris and Witputs, which are separated by the Osis arch (Fig. 1). Upper Nama Group sediments belonging to the Cambrian Fish River and largely Ediacaran Schwarzrand subgroups extend across the Osis arch; the older Ediacaran Kuibis Subgroup is thicker, more carbonate-rich, and apparently more complete in the north (Fig. 2). The only distinctive Kuibis unit that crosses the arch is the Kliphoek Member of the Dabis Formation, which projects as a tongue into the southern edge of the Zaris subbasin (Fig. 2; Germs, 1983, fig. 3; Germs and Gresse, 1991, fig. 3). Fortunately, the limestone overlying this tongue, which is clearly the Mooifontein Member of the Zaris Formation, preserves the older rising limb of a pronounced positive carbon isotope excursion (OMKYK, Figs. 2, 3; OME of Bowyer et al., 2022) that is well characterized from thick, carbonate-rich sections in the Zebra River area (Figs. 1–3; Grotzinger et al., 1995; Saylor et al., 1998; Smith, 1999; Wood et al., 2015) and is older than a prominent volcanic ash bed with a U–Pb age of 547.36 ± 0.23 Ma (Grotzinger et al., 1995; Bowring et al., 2007; Schmitz et al., 2020). As the peak of the Omkyk excursion can be followed southward in the Mooifontein Member (Fig. 3), its presence above fossiliferous horizons of the underlying Buchholzbrunn and Kilphoek members in the Witputs subbasin provides a minimum age of about 548 Ma for those assemblages (Fig. 2; Saylor et al., 1998).

Figure 2. Lithostratigraphy of the Nama Group and stratigraphic ranges of taxa in this work. The lower parts of the successions of the Zaris and Witputs subbasins differ across the Osis ridge, and the ~2 million-year hiatus (vertical lines) proposed here for the Witputs subbasin is novel. Rock units shown are mainly members of the Kuibis (K) and Schwarzrand (S) subgroups rather than the formations as these are the most commonly used and mappable lithological subdivisions. The recommended nomenclature for the first-order sequence stratigraphy is developed from Saylor et al. (1995, 2005), Smith (1999), and Saylor (2003), and the terms apply to the sequences above each labeled boundary. The stratigraphic ranges shown are limited to genera and larger taxonomic groups to provide a clear overview of the distribution of key elements of the biota; specific details are provided in the systematic paleontology section. “OMKYK” shows the stratigraphic position of the Omkyk positive carbon isotope excursion (Bowyer et al., 2022) in both subbasins (Fig. 3), and the gray triangles within the Feldschuhhorn Member represent pinnacle reefs.

Figure 3. Measured stratigraphic sections from a north–south transect across the Osis ridge at farms Aar, Mooifontein, and Mamba (Fig. 1); rock types are siliciclastics (red/dark gray), carbonates (brick-wall patterns), and siltstones (green/light gray). The canyon section (Aar) incudes almost all of the Dabis Formation from granitic basement to the top of the Mooifontein limestone member; the Amphitheatre section, which transects the famous Pteridinium locality on Aar (UCLA 7307), is the source of the carbon isotope values shown by triangles (Supplemental dataset 1); the thinner and unfossiliferous Mooifontein section, closer to the Osis ridge, provided the carbon isotope values shown as gray filled circles (Supplemental dataset 1). The Mamba section, on the north side of the Osis ridge, has a siliciclastic tongue of Kliphoek Member between the Mara and Mooifontein limestone members (Fig. 2; Germs, 1983, fig. 3).); carbon isotope values from this section (black filled circles, Supplemental dataset 1) and those from the far thicker section along the Zebra River (gray triangles; Saylor et al., 1998) are correlated by normalizing the thicknesses between the basal unconformities and a distinctive stromatolitic marker bed, visible in both sections. Fossiliferous horizons sampled in this study are shown by their UCLA numbers. The single- and double-headed arrows are current directions, and the rose diagram illustrates the orientation of 10 transported specimens of Pteridinium simplex measured at UCLA 7307 (all measurements corrected for –19° magnetic declination). Lithologic symbols: brick wall patterns = limestones, dark (black) and light (gray); rhomboidal brick patterns = dolomites; red = sandstones and arenites; recessive units, gray or green = mainly siltstones.

A volcanic ash bed near Nooitgedacht (Fig. 1) at the “basal contact” of the siliciclastic Nudaus Formation, the oldest unit of the Schwarzrand Subgroup, has a U–Pb age of 545.27 ± 0.11 Ma (Nelson et al., 2022), thus implying an approximately 2 million-year hiatus between the Kuibis and Schwarzrand subgroups in the Witputs subbasin, where the Nudaus lies directly upon the Mooifontein (Fig. 2); carbonate and low-energy siliciclastic sedimentation seems to have continued through this hiatus in the north.

Another volcanic ash bed ~10 km southwest of Nooitgedacht in the Nasep Member of the heterogeneous Urusis Formation, which completes the Ediacaran section of the Schwarzrand Subgroup in the Witputs subbasin, has a U–Pb age of 542.65 ± 0.15 Ma (Fig. 2; Nelson et al., 2022). This constrains the fossiliferous intervals of the Schwarzrand Subgroup to between about 543 and 539 Ma in the Witputs subbasin, but the older Nudaus Formation is fossiliferous north of the Osis arch (UCLA 7320, Fig. 2; Darroch et al., 2016) and perhaps at Gründoorn, about 60 km from Karasburg (Figs. 1, 2; Haughton, 1960; Glaessner, 1978). The Nama section at Charliesput, ~95 km east of Karasburg, is almost entirely siliciclastic (Germs, 1972a, fig. 22; 1974, fig. 4). Glaessner (1978) followed Haughton (1960) in attributing the two slabs of sandstone that preserve the type specimens of Archaeichnium haughtoni Glaessner, 1963 to the Kuibis Formation equivalent, the Nababis Formation, but the stratigraphic range of Archaeichnium elsewhere suggests a younger, Schwarzrand equivalent provenance (Fig. 2).

The best-dated part of the Nama succession spans the Ediacaran–Cambrian boundary (Grotzinger et al., 1995; Linnemann et al., 2019; Nelson et al., 2022), and the relevant U–Pb ages for this work are summarized in Figures 2 and 5. The Precambrian–Cambrian boundary has traditionally been placed at a profound erosional surface at the base of the Nomtsas Formation (e.g., Germs, 1972a; Grotzinger et al., 1995; Wilson et al., 2012), but recently it has been suggested that the boundary should be lowered into the underlying Spitskop Member of the Urusis Formation (Fig. 2) because trace fossils of the Treptichnus pedum (Seilacher, 1955) type have been found within the Spitskop Member (Linnemann et al., 2019; Bowyer et al., 2022) and treptichnids lower down (Jensen et al., 2000; Darroch et al., 2021). The significance of these observations remains a matter for debate; here we follow the traditional view on the grounds that bone fide examples of Treptichnus pedum are found abundantly in the Nomtsas Formation (Fig. 25.3, 25.4; Wilson et al., 2012) whereas those present in the Spitskop Member (Fig. 25.1, 25.2; Germs, 1972b, pl. 2, fig. 1; Jensen et al., 2000; Darroch et al., 2021; Turk et al., 2022) are treptichnids but not Treptichnus pedum, a pattern seen elsewhere (Jensen, 2003), including Nevada (Tarhan et al., 2020). The persistence of characteristically Ediacaran fossils such as Pteridinium and Swartpuntia to near the top of the Spitskop Member is mirrored by the presence of Ernietta and other Ediacaran taxa immediately beneath the well-characterized Precambrian–Cambrian boundary in Nevada and Sonora, Mexico (Smith et al., 2016, 2022; Hodgin et al., 2021; Nelson et al., 2023). Thus, we regard all of the fossils discussed in this work, with the exception of T. pedum from the Nomtsas Formation, to be Ediacaran, not Cambrian, in age.

Biostratigraphy

The regional distribution of the Ediacara fauna and associated calcareous fossils and trace fossils was first documented by Germs (1972a–c, 1973). He showed that Ediacaran fossils (Rangea, Pteridinium, Ernietta) are common in the triangular area bounded by Aus (16.25°E, 26.67°S), Helmeringhausen (16.82°E, 25.88°S), and Goageb (17.22°E, 26.75°S) at the siliciclastic to carbonate transition from the Kliphoek to Mooifontein members, since separated out as the Buchholzbrunn Member (Germs and Gresse, 1991; Germs, 1995)—which we use here (Figs. 2, 3)—or as the Aar Member (Hall et al., 2013). Germs also found one specimen of Rangea higher in the succession, just above the Mooifontein limestone in the Neiderhagen sandstone Member, Nudaus Formation, at Chamis (17.00°E, 26.05°S), but as the fossil was not in situ, its stratigraphic position may be questionable. A rather different assemblage, interbedded with carbonates, was found at a single site in the base of the Huns limestone, Urusis Formation, Schwarzrand Subgroup at Arimas (17.00°E, 27.70°S). Thus, there seemed to be two principal assemblages of Ediacaran organisms, an older one characterized by Pteridinium and Ernietta and a younger one that lacked both genera but yielded a variety of tubular and trace fossils, plus the new genus Nasepia (Germs, 1972a, c, 1973). This situation languished until Grotzinger et al. (1995) showed that Pteridinium and Swartpuntia (Narbonne et al., 1997) extended almost up to the disconformity that separates the Cambrian Nomtas Formation from older Schwarzrand units. There has been one recent report of an “indeterminate erniettomorph” in the Nomtsas Formation in South Africa (Nelson et al., 2022), but both the fossil and its stratigraphic level need further assessment.

In the carbonates, Germs (1972a, b) reported Cloudina from the Mara, Mooifontein, and Huns limestone members of the Wiputs subbasin, but all of the described material was from a bioherm, the Driedoornvlagte reef (Grotzinger et al., 2000, 2005; Adams et al., 2004; Wood et al., 2015), in the Zaris subbasin (Fig. 1). All occurrences of Cloudina in both subbasins were subsequently summarized by Grant (1990) and Yang et al. (2022). Namacalathus is found with Cloudina in the Driedoornvlagte reef (Germs, 1972c, pl. 1, fig. 4; Grotzinger et al., 2005; Penny et al., 2014; Wood et al., 2015) and in the Omkyk Member in the Zebra River section on Donker Gange (UCLA 7319; Grotzinger et al., 2000), but apparently without Cloudina in the pinnacle reefs of the Feldschuhhorn Member at Swartkloofberg, although both are present near ash 4 in the Dundas section on Swartpunt (Fig. 5; Wood et al., 2015, fig. 14).

Trace fossils have been important components of Nama Group biostratigraphy since the pioneering studies of Germs (1972a, c) and Crimes and Germs (1982). Their taxonomy has been reviewed and revised by Darroch et al. (2021), leading to the elimination of characteristically Phanerozoic genera such as Zoophycos and Diplocraterion. What was left are putative cnidarian resting or dwelling traces (Conichnus, Bergauria) and possible narrow, horizontal burrows (Helminthopsis) in the Kuibis Subgroup and more diverse ichnofossil assemblages in the Schwarzrand Subgroup (Darroch et al., 2021, fig. 18b). The only sizable, Cambrian-like traces are Streptichnus nabonnei Jensen and Runnegar, 2005 from the uppermost Spitskop Member (UCLA 7375, Fig. 5) and Parapsammichnites pretzeliformis Buatois et al., 2018 from lower in the Spitskop in the Fish River area; the rest are narrow, subhorizontal burrows or levée-lined trenches (Archaeonassa, Gordia, Helminthoidichnites, Helminthopsis) that are similar to co-occurring tubular body fossils and, when inadequately preserved, may be confused with them. Although Streptichnus is “Cambrian-like” and has been invoked to lower the eon boundary into the Spitskop Member (Linnemann et al., 2019), its only other known occurrence is in the Ediacaran of China (Xiao et al., 2021; Mitchell et al., 2022).

The taxon of greatest interest, first found by Germs (1972a, b), is an ichnospecies that Jensen et al. (2000) referred to Treptichnus, but not T. pedum (Fig. 25.1, 25.2; Darroch et al., 2021, fig. 13). Our work reinforces this picture (Fig. 2), but we present clear examples of horizontal burrows (Gordia sp.) in the Kiphoek Member (Fig. 24.11, 24.13, 24.14), provide evidence for bioturbation down to depths of several centimeters in the Nasep and Huns members, and show that Archaeichnium haughtoni (Figs. 20, 21) is a body fossil rather than a trace fossil (Glaessner, 1963; Turk et al., 2022). We also suggest that structures (Fig. 12) that Darroch et al. (2021) called “guitar strings” are the tool and mold marks of current-transported erniettomorphs—probably Pteridinium—rather than sponge wall fragments.

Chemostratigraphy

Carbonate hand samples, spaced 1 m apart where possible, were collected by MRS and BR from limestone sections measured by them and/or JGG from the Mooifontein Member on Aar, Mamba, Mooifontein, and Twyfel farms; from the Omkyk Member at Swartmodder on Omkyk; from the Urikos, Neiderhagen, and Vingerbreek members on Saurus; and from the Huns Member on Arimas and Swartkloofberg; but samples from only some of those sections were processed because of funding constraints. All isotopic data are tabulated in Supplemental dataset 1 and plotted against stratigraphic heights in Figure 2. Our results confirm previous and subsequent studies (Kaufman et al., 1991; Saylor et al., 1998; Smith, 1999; Wood et al., 2015). Overall, the record is monotonous apart from the Omkyk excursion, which stands out from background but is not well expressed elsewhere except, perhaps, South China (Bowyer et al., 2022). Whether or not the negative values from the Mara and other lower Kuibis members, which Bowyer et al. (2022) term the “basal Nama excursion” (BANE), represent a post-Shuram, pre-basal Cambrian excursion (BACE) negative event of intercratonic significance is uncertain, given the paucity of other occurrences (Chai et al, 2021; Yang et al., 2021). Conversely, the position of the BACE, which is missing from the Nama succession, remains unresolved. Previously, it was thought to have been eliminated by the basal Nomtsas disconformity, but recent U–Pb ages suggest other possibilities (Linnemann et al., 2019; Hodgin et al., 2021; Bowyer et al., 2022; Topper et al., 2022; Nelson et al., 2023). In summary, the positive Omkyk excursion and the lower Kuibis negative intervals are the only features of the carbon isotope record of the Nama Group that are potentially useful for more than regional correlation.

Stratigraphic information

See Figures 2–5, Appendix, and the discussion under Materials and methods.

Locality information

We used the UCLA locality numbering system (e.g., UCLA 7307) for sites, occasionally at the same place in stratigraphic order, and numbered each piece of rock collected accordingly. Important fossils were then given decimal numbers corresponding to localities (e.g., 7307.1, 7307.2, etc.). Parts of an individual fossil were given the same decimal number, and different fossils on the same slab are identified by letters (e.g., 7307.3A, 7307.3B, etc.). These UCLA numbers for individual fossils were replaced by National Earth Sciences Museum numbers (GSN F) after the collection was repatriated to Namibia, but the UCLA locality numbers (e.g., UCLA 7307) still pertain. Locality details are described in the Appendix. Because this work was carried out before GPS became widely available, particularly in the Southern Hemisphere, the geographic positions of localities were plotted on photocopies of the relevant 1:50,000 scale topographic maps of South West Africa issued by the Surveyor General, Department of Justice, Government of Namibia. These map records were used recently to find the exact locations of the sites on Google Maps and to obtain their decimal longitudes and latitudes (Appendix); Google Maps uses the WGS84 standard.

Names of higher taxa

There is an emerging consensus that the majority of Ediacaran soft-bodied organisms are metazoans, but their phylum and class level assignments remain uncertain. To provide a framework for discussion, we assign taxa to some extinct higher taxa (e.g., class Archaeocyatha) and indicate under remarks and discussion where such plesions probably join the tree of life (e.g., stem Demospongiae).

Materials and methods

Most of the material used for this study was obtained during fieldwork carried out in Namibia in 1993, 1995, and 1996 with the assistance and cooperation of A. Seilacher, University of Tübingen (August 1993) and the participants in an International Union of Geological Sciences–International Geoscience Programme-funded field workshop on the Terminal Proterozoic System (May 1995) and the support and advice of the Geological Survey of Namibia (May 1995 and August–September 1996). In addition, we had access to the Pflug collection, housed in Germany before it was returned to Namibia (August 1993), to the Richter collection (Richter, 1955) in the Senckenberg Museum of Natural History, Frankfurt (July 1993), to the Haughton types and some of the Germs collection (Haughton, 1960; Germs, 1972a, 1973) in the Iziko South African Museum, Cape Town (August 1993), and to a large number of plaster casts of specimens of Pteridinium held by the State Museum of South West Africa, Windhoek. Those casts were made at UCLA in 1966 by LouElla Rankin Saul, then a Museum Scientist and curator of the paleontological collections (Groves and Squires, 2023), from material that was borrowed and returned about that time by Preston Cloud (Cloud and Nelson, 1966).

Stratigraphic sections were measured using a Jacob staff, and stratigraphic thicknesses were checked, where possible, with a Thommen Altitronic Traveller altimeter on the assumption that the dips are negligible in the sections measured. The accuracy but not precision of the altimeter was checked at the trigonometric station at the top of Dundas Hill on Swartpunt farm on 23 August 1996, when the altimeter recorded an elevation of 1,124 m versus the surveyed height of 1,169 m. There are, however, some discrepancies between our measurements and those of others who have studied the same sections. To illustrate these discrepancies, we tabulated the measured heights of these and other features, such as bed boundaries and distinctive rock types identified by us, Saylor (1996), and Turk et al. (2022) in the Arimas section (Fig. 4).

Figure 4. Measured stratigraphic section, Arimas; rock types are siliciclastics (red/dark gray), carbonates (brick-wall patterns), and siltstones (green/light gray). This section was also measured by Saylor (1996, section 20, p. 296–304) and Turk et al. (2022, fig. 4), so normalized estimates of common distinctive horizons (KAT, Turk et al., 2022; BZS, Saylor, 1996; and ALT, altimeter) are listed for comparison with JGG's Jacob staff measurements. Fossiliferous levels sampled in this study are shown by their UCLA numbers. The S3B/S3A sequence boundary at the top of the second limestone is a marine flooding surface (MFS) that is correlated with prominent karst horizons at Witputs and Swartkloofberg (Saylor, 1996, p. 269, 339; Saylor, 2003). Lithologic symbols: brick wall patterns = limestones, dark (black) and light (gray); rhomboidal brick patterns = dolomites; red = sandstones and arenites; recessive units, gray or green = mainly siltstones.

In the case of the Dundas section on Swartpunt farm (Figs. 1, 5), the fossiliferous interval includes a deformed slump or fault block that varies in thickness along strike and is, at least, slightly allochthonous, although its internal stratigraphy is thought to be intact (Saylor, 1996; Saylor and Grotzinger, 1996; Narbonne et al., 1997; Darroch et al., 2015; Linnemann et al., 2019). Differences between our measurements and those of other authors may therefore be partly attributable to the fact that we measured a thinner section of the slumped sequence. However, we also place the upper four of five dated volcanic ash beds at lower elevations in the profile than did Saylor (1996) and Linnemann et al. (2019), despite the fact that our thickness measurements agree with those of the other authors overall. As correct superpositional order is more important than absolute thicknesses, except for relocating sampled horizons, these discrepancies in measured thickness are not considered significant.

Figure 5. Measured stratigraphic sections, Dundas, Swartpunt, and Swartkloofberg farms; rock types are siliciclastics (red/dark gray), carbonates (brick-wall patterns), siltstones (green/light gray), and volcanic ash beds (red/gray with white v pattern); the cones represent pinnacle reefs that grew upward from the S4A/S3B sequence surface and were buried by the Feldschuhhorn Member (Saylor and Grotzinger, 1996; Grotzinger et al., 2000). The uppermost 50 m of this section (left) has also been measured by Saylor (1996, section 14, p. 291–293) and Linnemann et al. (2019, fig. 3), so estimates of the elevations of significant horizons, normalized in each case from the surveyed summit of Dundas (1,169 m; 2716B Rekvlakte 1:50 000 topographic map, 1979), are also shown (UL, Linnemann et al., 2019; BZS, Saylor, 1996) for comparison with JGG's Jacob staff measurements. Most of the right column is taken from Saylor (1996, section 14, p. 288–293) and Saylor and Grotzinger (1996, fig. 4C), again shown as if measured downward from the summit of Dundas. Fossiliferous horizons sampled in this study are listed by their UCLA numbers; “Fossil Bed A” and “Fossil Bed B,” named by Narbonne et al. (1997), are shown at our measured stratigraphic positions; the U–Pb ages are from Schmitz (2012) and Linnemann et al. (2019), but the plotted levels of ashes 2–5 are from our own observations. Pteridinium and Swartpuntia have also been reported from near the top of our unexposed interval by Saylor (1996) and Darroch et al. (2015). Lithologic symbols: brick wall patterns = limestones, dark (black) and light (gray); rhomboidal brick patterns = dolomites; red = sandstones and arenites; recessive units, gray or green = mainly siltstones; wavy pattern = stromatolitic horizon; v pattern = ash beds.

Preparation of the fossils has been minimal. Field photographs and images of specimens taken during the 1990s were made with a Minolta X700 35 mm FSLR camera equipped with Minolta MC Macro Rokkor-QF 50 mm lens using Kodak Ektachrome Professional film. Color slides and negatives were digitized using an Epson Perfection V700 Photo scanner. Digital images, taken more recently, were made with a Nikon D3100 DSLR camera equipped with Nikon AF-S Micro Nikkor 40 mm lens. Preparation of the figures was carried out with Adobe Photoshop, Adobe Illustrator, Aldus Super3D, and Synergy KaleidaGraph.

Carbonate hand samples for isotopic analysis were collected at 1 m intervals, where possible. Carbonate powders were obtained, so far as was practical, from micritic sections of sawn and smoothed slabs, avoiding fractures and secondary cements. No attempt was made to vet the samples for diagenesis using chemical or optical methods on the grounds that diagenetically altered samples are relatively easy to distinguish using oxygen isotope measurements in densely sampled sections. All isotope measurements were made at Harvard University (Mooifontein section) or the University of California, Santa Cruz, using standard methods (e.g., Kaufman et al., 1991; Zachos et al., 1997).

Repositories and institutional abbreviations

Types, figured, and other specimens examined in this study are (or were) deposited in the following institutions: National Earth Sciences Museum, Ministry of Mines and Energy (GSN), Windhoek, Namibia; State Museum of South West Africa (SMSWA), Windhoek, Namibia; Iziko South African Museum (ISAM), Cape Town, South Africa; Senckenberg Museum of Natural History (SMNH), Frankfurt, Germany; Yale Peabody Museum (YPM), New Haven, Connecticut, USA; North Carolina State Museum of Natural Sciences (NCSM), Raleigh, North Carolina, USA; Department of Geology, University of North Carolina (UNC), Chapel Hill, North Carolina, USA; Los Angeles County Museum of Natural History (LACMNH), Los Angeles, California, USA; Department of Earth, Planetary, and Space Sciences, University of California, Los Angeles (UCLA), Los Angeles, California, USA. Some specimens remain in the field, as noted in the figure explanations.

Systematic paleontology

Kingdom Animalia Linnaeus, 1758
Phylum Porifera Grant, 1836
Class Archaeocyatha? Bornemann, 1884
Order Monocyathida? Okulitch, 1935
Genus Arimasia new genus

Type species

Arimasia germsi n. gen. n. sp. from the Huns Member of the Urusis Formation, Schwarzrand Subgroup, Arimas farm, Namibia.

Diagnosis

As for the type species by montypy.

Etymology

Named for Arimas farm, the type locality.

Remarks

Nothing similar to Arimasia has been described from the Neoproterozoic or, so far as we are aware, from the Phanerozoic. It seems to be a one-walled, solitary, and sessile animal, preserved as a composite mold of both inner and outer surfaces, perhaps resembling an unmineralized version of a monocyathid archaeocyath or a vauxiid sponge.

Arimasia germsi new species
Figure 6

Holotype

GSN F 1960H from the Huns Member of the Urusis Formation, Schwarzrand Subgroup, UCLA 7376, Arimas farm, Namibia.

Diagnosis

Centimeter-scale, porous, rugose, horn-shaped skeletons that appear to have been unmineralized or, perhaps, demineralized.

Description

The holotype (Fig. 6.1, 6.4) is a narrow, conical object, 2 cm long, that has a sealed rounded base and about eight irregular, co-marginal rugae in the lower two-thirds of the structure; the open end of the skeleton was apparently circular but is now flattened by compaction that extends downward toward the rugose part of the cone; the cone surface is evenly granular, giving the impression of a fine mesh, which cannot be fully resolved because of the finite grain size of the matrix; a paratype (Fig. 6.2, 6.3) displays the mesh more clearly; the cells are 200–300 μm apart and appear to be packed like honeycomb; other specimens are more regularly rugose (Fig. 6.6, 6.7) and eight of 10 individuals on one small slab seem to be preferentially oriented, suggesting that the cones may have been tethered to the substrate (Fig. 6.5).

Figure 6. Arimasia germsi n. gen. n. sp., Huns Member, Urusis Formation, UCLA 7326, Arimas farm. (1, 4) Holotype, GSN F 1960H, showing rugose form, the apparently porous nature of the body wall, and other individuals (GSN F 1960A, GSN F 1960B, GSN F 1960C) on the same surface. (2, 3) GSN F 1960C, also showing the porous body wall. (5) A single surface with at least 10 specimens of A. germsi, eight of which (white numerals) are opening upward in this view, and the other two (yellow numerals) are facing downward, GSN F 1954. (6) External mold of one of the largest specimens, GSN F 1958. (7) Three, possibly current-aligned, specimens, GSN F 1955. (1, 2) scale bars = 5 mm; (3) horizontal scale bar = 2 mm; inclined scale bar = 1,000 μm; (4, 6) scale bars = 1 cm; (5, 7) scale bars = 2 cm.

Etymology

Named for Gerard J.B. Germs, in celebration of the fiftieth anniversary of the publication of his groundbreaking, University of Cape Town, Ph.D. dissertation on the stratigraphy and paleontology of the lower Nama Group (Germs, 1972a).

Materials

Eight specimens (GSN F 1953–1960), each with one or several specimens from UCLA 7376.

Remarks

Antcliffe et al. (2014) reviewed all of the then published reports of the oldest fossil sponges and recommended caution in making such claims. They proposed two selection criteria that should always be met and summarized a passing grade as: “The characters claimed for are useful for detecting sponges in the fossil record and have been reliably shown to be present in the particular candidate fossil.” In their opinion, the oldest fossil sponges are siliceous hexactinellid spicules from the Fortunian of Iran (also China; Chang et al., 2017) and Archaeocyatha from the Tommotian (Cambrian Stage 2) of Siberia, a conclusion that has not been effectively challenged by subsequent reports of sponge-like fossils of Ediacaran or earlier ages (e.g., Turner, 2021). Antcliffe et al. (2014, p. 999) also concluded that “the ancestral archaeocyathan sponges must occur in the [Fortunian] Purella antiqua Zone and would have consisted of small, simple rounded cups, each provided with a single, weakly calcified wall, perforated by simple pores.”

Arimasia germsi appears to pass these two selection criteria by having the sponge characters described in advance by Antcliffe et al. (2014) if our interpretation of the granular texture of the fossil is correct. The cell size of the wall mesh, 200–300 μm (Fig. 6.3), is comparable to the average interpore distance in single-walled Archaeocyatha, such as Archaeolynthus contractus Hill, 1965 (~330 μm; Hill, 1965, pl. 1, fig. 1), but is larger than the diameter of the pores, which in double-walled Archaeocyatha is commonly ~100 μm or less (Gravestock, 1984; Antcliffe et al., 2019). Thus, Arimasia may be viewed as an unmineralized, apparently single-walled archaeocyath, and perhaps also as a stem group demosponge, if that is where the Archaeocyatha fit into the Porifera (Antcliffe et al., 2014). There are also possible similarities to the unmineralized vauxiid sponges, which first appear in the Cambrian Stage 3 Chenjiang biota (Wei et al., 2021) and are regarded by some as being on the pathway to the keratose demosponges, now thought to be the monophyletic or paraphyletic sister group of the spiculate Heteroscleromorpha (Erpenbeck et al., 2012; Wörheide et al., 2012; Plese et al., 2021).

Although there is widespread agreement that the Archaeocyatha are hypercalcified aspiculate sponges (Rowland, 2001; Debrenne et al., 2012; Antcliffe et al., 2014) rather than some kind of calcified alga (Kazmierczak and Kremer, 2022), their position within the poriferan total group remains so uncertain as to be almost ignored (e.g., Botting and Muir, 2018). Arimasia may provide a way forward in that it is demonstrably older than any known spiculate sponge, was apparently unmineralized, and is similar in body form to the organic-walled vauxiids (Rigby, 1980, 1986; Botting et al., 2013; Luo et al., 2020; Wei et al., 2021), which Luo et al. (2021) have suggested might be demineralized archaeocyaths. Alternatively, Arimasia, Vauxia, and the archaeocyaths may all have been aspiculate stem group sponges, and therefore the vauxiids are not demineralized archaeocyaths (Luo et al., 2021) but, instead, were unmineralized members of the lineage leading to the aspiculate demosponges. This hypothesis would require the acquisition of siliceous spicules independently in the Hexactinellida and the Demospongiae, a proposal that has been vigorously rejected by nearly all sponge paleobiologists (e.g., Botting and Muir, 2018) but has recently received some molecular support (Aguilar-Camacho et al., 2019).

Class Erniettomorpha Pflug, 1972
Family Pteridinidae Richter, 1955
Genus Pteridinium Gürich, 1933

Type species

Pteridinium simplex Gürich, 1933 from the Kliphoek Member of the Dabis Formation, Kuibis Subgroup, Aus district, Namibia, by monotypy.

Other species

?Paradoxides carolinaensis St. Jean, 1973, from the Floyd Church Formation, Albemarle Group, Stanly County, North Carolina, USA; Pteridinium nenoxa Keller in Keller et al., 1974; Inkrylovia lata Fedonkin in Palij et al., 1979.

Diagnosis

Frondose organisms, up to at least 0.4 m long, that are made of three equal-sized organic-walled vanes that are set lengthwise about an axis that extends from the curved proximal end to the acute distal growing tip; each vane is constructed from sealed tubular modules that meet alternatively or oppositely at the axis, depending on their order about it, and are concave toward the distal end of the frond; margins of the vanes are defined by smooth, thickened edges against which the modules terminate without narrowing.

Occurrence

Kliphoek, Buchholzbrunn, and Mooifontein members of the Kuibis Subgroup and Nudaus, Nasep, Huns, and Spitskop members of the Schwarzrand Subgroup, Nama Group, Namibia (Fig. 2; Appendix); basal Ediacara Member, Rawnsley Quartzite, South Australia (Glaessner and Wade, 1966; Gehling and Droser, 2013); Floyd Church Formation, Albemarle Group, North Carolina, USA (St. Jean, 1973; Gibson et al., 1984; McMenamin and Weaver, 2002); Syuzma/Verkhovks member, Ust-Pinega Formation, Onega Peninsula, Russia (>552.85 ± 0.77 Ma; Keller et al., 1974; Fedonkin, 1981; Ivantsov and Grazhdankin, 1997; Grazhdankin, 2004; Ivantsov et al., 2019); Shibantan Member, Dengying Formation, China (<550.1 ± 0.06 Ma; Chen et al., 2014; Xiao et al., 2021; Yang et al., 2021); doubtfully, Ukraine (Fedonkin, 1983), Canada (Narbonne and Aitken, 1990), and Iran (Vaziri et al., 2021).

Remarks

It is not clear why Pflug's (1972) class Erniettomorpha has been widely adopted in preference to his Pteridinomorpha, which has page precedence, but we follow that practice for the probable clade (Dececchi et al., 2017) that includes Pteridinium Gürich, 1933, Ernietta Pflug, 1966, Phyllozoon Jenkins and Gehling, 1978, Swartpuntia Narbonne, Saylor, and Grotzinger, 1997, and perhaps Ventogyrus Ivantsov and Grazhdankin, 1997, and Miettia Hofmann and Mountjoy, 2010. Inkrylova lata, the type species of Inkrylova Fedonkin in Palij et al., 1979, appears to be a junior synonym of Pteridinium nenoxa Keller in Keller et al., 1974. The relationships of these and other species of Pteridinium are discussed under the description of P. simplex. Pteridium Gürich, 1930 (Gürich, 1930b) was a nomen nudum replaced, perhaps unnecessarily, by Pteridinium Gürich, 1933; Onegia Sokolov, 1976 is a nomen nudum applied to Keller's species nenoxa by Sokolov (1976) and Grazhdankin (2004). As discussed under the genus Ernietta Pflug, 1966, the holotype of the type species, E. plateauensis Pflug, 1966, appears to be a specimen of Pteridinium simplex, so technically Ernietta becomes a subjective junior synonym of Pteridinium. However, we propose that an application be made to the International Commission on Zoological Nomenclature (ICZN) to replace the holotype of E. plateauensis with a neotype, the holotype of Erniograndis sandalix Pflug, 1972, to retain current usage of this well-established name.

Pteridinium simplex Gürich, 1933
Figures 7–9, 10.1–10.4, 11.1–11.3, 11.6–11.8, 19.1, 19.2, 19.7

1930b

Pteridium simplex Gürich, p. 637, nomen nudum.

1933

Pteridinium simplex Gürich, p. 144, fig. 4a–c.

1955

Pteridinium simplex; Richter, p. 246, pls. 1–6, figs. 1–10, pl. 7, fig. 11.

1963

Pteridinium simplex; Glaessner, p. 8, pl. 1, figs. 1–4, pl. 2, fig. 1.

1966

non Pteridinium cf. P. simplex; Glaessner and Wade, p. 616, pl. 101, figs. 1–3.

1966

Ernietta plateauensis Pflug, p. 22, pl. 1, figs. 1–7.

1966

Pteridinium simplex; Cloud and Nelson, fig. 1A, C.

1972a

Pteridinium simplex; Germs, p. 173, pl. 21, figs. 1, 2.

1972

Ernietta plateauensis; Pflug, p. 163, pl. 34, figs. 1–4, 6.

1977

non Pteridinium simplex; Keller and Fedonkin, p. 926, pl. 2, fig. 4.

2002

Pteridinium simplex; Grazhdankin and Seilacher, fig. 1, pl. 1, figs. 1–3.

2014a

Pteridinium simplex; Meyer et al., figs. 2–6.

2022

Pteridinium simplex; Runnegar, p. 1103, fig. 9A.

2022

Pteridinium simplex; Darroch et al., figs. 2–5.

2022

non Pteridinium simplex; Darroch et al., fig. 7.

Neotype

Gürich's (1933) specimens were lost during World War II so Richter (1955) nominated a neotype, a sandstone cast of part of a frond with two visible vanes (SMNH XXX 660f), probably from the Kliphoek Member, Kuibis Formation, on Plateau or Aar farm, Aus district, Namibia (Richter, 1955, pl. 1, fig. 1a, b; Darroch et al., 2022, fig. 1a, b).

Diagnosis

A “three-vaned, ribbon-like frond” (Ivantsov et al., 2016, p. 540) in which the modules are less well expressed in the outer halves of the vanes.

Description

Elongate, frondose organisms formed of three equal-sized, undivided vanes that radiate from a common axis and may exceed 0.4 m in length without signs of expansion or tapering in width; vanes are composed of curved to straight tubular modules that are commonly, but not always, more topographically expressed near the axis than the periphery; modules maintain a similar-sized cross section across the vane and terminate abruptly at the distal margins; vanes terminate axially in closed, polyhedral ends that either alternate with those of other modules in a zig-zag fashion or are directly opposed, depending on position around the axis (Fig. 8.8–8.10; Runnegar, 2022, figs. 9, 10); in two specimens with visibly narrowing vanes, the curvature of the modules is convex in the direction of narrowing and the angle of narrowing is 10° or less (Fig. 11.1, 11.2; Richter, 1955, p. 249, pl. 6, fig. 7; Runnegar, 2022, fig. 9a); whether the tapering of the vanes is unidirectional or bidirectional is unknown; vane margins are commonly obscure but when well preserved are delineated by a narrow differentiated edge (Figs. 9.2–9.4, 10.3) that was stiff enough to imprint other individuals (Fig. 7.3–7.5); vane curvature generally coaxial but inconsistent; two of the vanes are frequently opposite each other at the axis and either lie parallel to bedding or curve quasi-symmetrically to partially embrace the third vane (Fig. 11.6–11.8), thus producing W-shaped cross sections (Fig. 7.5).

Figure 7. Pteridinium simplex Gürich, 1933, Aarhauser sub-member, Kliphoek Member, Dabis Formation, UCLA 7307, Aar farm. (1) A 25 cm thick bed of micaceous quartz sandstone overlain by dislodged joint blocks of the bed, many of which are fossiliferous. (2) Overturned piece of the same bed showing current scoured base. (3) Reassembled pieces of one joint block that contains two subparallel specimens of P. simplex, GSN F 1855A and GSN F 1855B, each composed of three vanes, A1, A2, A3, etc., lying with their axes parallel to bedding. (4) Two parts of the same block, viewed perpendicular to bedding, with the upper edges of vanes A1 and A3 indicated by white and black arrows, respectively. (5) Foreshortened oblique view of the reassembled block showing cross sections of vanes A1 and A3 on the sawn surface. (6) Lateral view of vane B3 with its lower edge parallel to bedding. (7) End piece viewed from the top to show the relative positions of vanes A3, B2, and B3. (8) Sawn edge of the end piece in (7) showing the cross-sectional curvature of vanes A1, A3, and B1. (1) Camera lens cover = 60 mm; (2) brush = 25 cm; (3, 4) scale bar = 5 cm; (5) scale bar = 3 cm but variable scales due to foreshortening; (6–8) scale bar = 3 cm.

Materials

Seven specimens (GSN F 1853–1859) from UCLA 7307; ~20 plaster casts of SMSWA specimens; ~20 specimens, mostly from Plateau and Aar farms in the SMNH collection (Richter, 1955); the Pflug (1970a) collection, plus numerous examples observed in the field at Aar farm and in the “museum” at Plateau farm, including the excavation and casting of the Seilacher slab and other specimens in 1993 (Fig. 10.1, 10.2; Crimes and Fedonkin, 1996; Seilacher, 1997, 2007; McMenamin, 1998; Grazhdankin and Seilacher, 2002; Ivantsov et al., 2019).

Taphonomy

The numerous specimens that have been observed, extracted, and studied from the Amphitheatre site (UCLA 7307) on Aar farm (Figs. 3, 7–9, 10.1–10.4, 11.1–11.3; Richter, 1955; Grazhdankin and Seilacher, 2002; Vickers-Rich, 2007; Meyer et al., 2014a, b, Darroch et al., 2022) are all from a set of quartz sandstone beds that have been named the Aarhauser sandstone submember by Hall et al. (2013) (Fig. 3). The bed that was the source of the Seilacher slab is about 0.4 m thick, has a deeply erosive base (Fig. 7.2), a horizontally bedded to low-angle cross-stratified upper part (Fig. 8.1, 8.2, 8.4), a middle zone with cylindrical specimens of Pteridinium (Fig. 10.4; Crimes and Fedonkin, 1996, pl. 2c, d), and a thicker, poorly laminated lower part that is richly fossiliferous. P. simplex is widespread in the laminated to upper cross-stratified part, typically as long, straight segments of two-vaned fronds seen in plan view on bed surfaces, as W-shaped intersections on east–west joints, and as upright, stretched single vanes on the faces of north–south joints (Figs. 8.1, 8.2, 9.2–9.4; Darroch et al., 2022, fig. 5). Almost invariably, the axes of vanes seen on joint faces lie parallel to bedding and are commonly at the bottom of the vanes, even when the whole organism is twisted through 180° about a horizontal axis (Figs. 8.1, 8.2, 8.4, 8.6, 8.7, 9.6, 11.6–11.8). There is no evidence that the upright middle vane, the “chaperone wall” of Grazhdankin and Seilacher (2002), routinely switches places with one of the other vanes during the 180° folding, as shown in the sketch (Fig. 8.5) from Grazhdankin and Seilacher (2002), as previously noted by Meyer et al. (2014a). That would require improbable twisting about two different rotational axes. The other common U-turn is a hairpin bend (Figs. 9.5, 9.7, 9.8, 19.1, 19.2; Richter, 1955, pl. 7, fig. 11; Vickers-Rich, 2007, fig. 108; Meyer et al., 2014a, figs. 3–6), in which the fold axis is vertical rather than horizontal. One such U-turn, found in situ in 1993 (Fig. 9.5, 9.7, 9.8), was at the upstream end of the hairpin-shaped fossil, as shown by the orientations and of 10 flat-lying specimens measured on the top surface of the same outcrop (Fig. 3). Thus, most if not all of the examples of P. simplex found in the upper, laminated to cross-laminated layers of the Aarhauser sandstone appear to have been transported by northward-flowing high-velocity currents, as suggested previously (Jenkins, 1985; Elliott et al., 2011; Darroch et al., 2022).

Figure 8. Folded specimens of Pteridinium simplex Gürich, 1933, Aarhauser sub-member, Kliphoek Member, Dabis Formation, UCLA 7307, Aar farm. (1) Field photograph of a dislodged but upright 20 cm thick joint block with a specimen of P. simplex folded about a horizontal axis (white dot with circle around it) and with longer upper part of the organism extending downstream. (2) Enlargement of (1) to show details of the limbs. (3) Field photograph of base of block excavated by the Seilacher team in 1993 (Seilacher, 1997) showing two similarly folded specimens of P. simplex, GSN F 758 and GSN F 576 (arrows indicate positions of horizontal fold axes), which became the basis for the “canoe” model for Pteridinium (Grazhdankin and Seilacher, 2002). (4) Field photograph of another dislodged joint block showing prominent horizontal lamination and one vane of a P. simplex that is tightly folded about a horizontal axis (white dot with circle around it). (5) Part of drawing (Grazhdankin and Seilacher, 2002, fig. 5C, republished with permission) used to explain both the canoe and vane substitution models for the growth of Pteridinium; note that vane substitution requires both twisting through 180° and folding about a horizontal axis. (6) Oblique view of small joint block that has been broken and sawn to reveal details of the kind of folds seen in (1, 2, 4), with the fold axis indicated by the arrows (GSN F 1857). (7) Same specimen as (6), lateral view. (8–10) Weathered fragment (GSN F 1856) that shows how the proximal ends of the three vanes, V1, V2, and V3, interlock; the modules of vanes V1 and V2 are opposite each other, whereas the modules of each alternate with those of V3 (9, 10). (1) Brush = 25 cm; (3, 4) camera lens cover = 60 mm; (6, 7) scale bar = 5 cm; (8–10) scale bar = 1 cm.

Figure 9. Field and other images of a U-shaped specimen of Pteridinium simplex Gürich, 1933, GSN F 1858, that had been exposed by excavation during or before 1993. (1) U-shaped end piece (5, 7, 8) in place and with the trailing vanes, V1 and V2, extending northward in the direction of downstream transport; an extracted piece with part of V1 (2–4) is in the foreground. (2–4) Field images of V1 that show the modules leaning downstream and the linear distal edge of the vane. (5) U-turn with axis (AX) at periphery. (6) Plaster cast of a specimen folded like a taco about a horizontal axis, SMSWA 45730.1 now GSN F 1878. (7, 8) Three parts of the U-bend showing the curvature of the axis and the positions and orientations of V1 and V2 on both sides of the turn (downstream is toward the bottom of the page). (1) Hammer = 33 cm; (3) coin = 23 mm; (4) scale bar = 2 cm; (5–8) scale bar = 3 cm.

A middle zone of closely packed tubular fossils is more problematical but is almost certainly a death association, perhaps created by close packing of enrolled, hairpin-shaped individuals (Fig. 10.4) rather than a population of sealed underground sausage-shaped organisms, as envisaged by Crimes and Fedonkin (1996). As the horizon is less accessible, it has not been well studied.

Figure 10. (1–4) Joint blocks of Aarhauser sandstone member, Aar farm (UCLA 7307), that had been split approximately in half parallel to bedding to reveal the lower sides of a large number of specimens of Pteridinium simplex Gürich, 1933 and then reassembled upside down for molding with silicone rubber by the Seilacher team (Seilacher, 1997). (1) Dolf Seilacher, second from left, with Mark McMenamin, Hans Luginsland, and Peter Seilacher viewing the “Seilacher slab” ready for molding, August 1993. (2) Richly fossiliferous two-thirds of the Seilacher slab (Seilacher, 1997, 2007, 2008), which inspired the “bathtub” or canoe models for Pteridinium living underground. (3) Seven aligned and four closely packed specimens of P. simplex seen in upper left corner of (2); in cross section, those in contact would resemble tubes. (4) Top of one of the joint blocks showing that the third vane (V3) may be underneath a pair of vanes (V1 and V2) exposed on the surface of the bed. (5) An in situ specimen of Pteridinium carolinaensis (St. Jean, 1973), Spitskop Member, UCLA 7373, Dundas Hill, Swartpunt farm. (2, 3, 5) Camera lens cover = 60 mm; (4) scale bar = 5 cm.

The lower layers are filled with P. simplex preserved in a somewhat different fashion, as can be seen from an extracted block in the Richter collection (Darroch et al., 2022, fig. 2). The lower part of this block, a similar specimen figured by Glaessner (1979b, fig. 11.1b), and the far more extensive Seilacher slab (Fig. 10.1–10.4; Seilacher, 1997, 2007; Grazhdankin and Seilacher, 2002) have doubly curved fronds that—to some—resemble inverted bathtubs or boats (Figs. 8.3, 9.2; Grazhdankin and Seilacher, 2002, text-fig. 1). These are the shapes that have given rise to the canoe model for Pteridinium (Pflug, 1970a; Buss and Seilacher, 1994; Ivantsov and Grazhdankin, 1997; Grazhdankin and Seilacher, 2002; Meyer et al., 2014b; Droser et al., 2017; Darroch et al., 2022) and to the hypothesis that Pteridinium lived on, in, or wholly within the sediment (Crimes and Fedonkin, 1996; Seilacher, 1997, 2007; Grazhdankin and Seilacher, 2002; Darroch et al., 2022). However, these lower layers also preserve specimens that are folded like tacos. In these cases, two oppositely directed vanes are bent through 180° about a horizontal axis, and for geometrical reasons, the third vane must follow one of the other two (Fig. 9.6). This cannot be a life orientation, so the fact that this postmortem topology is found among the canoes is evidence that all were transported before burial. As no one has described or illustrated convergence of the three vanes to form the “prow” or “stern” regions of any specimen of P. simplex, the canoe hypothesis is not supported by observation. Thus, we consider all of the material in the Aarhauser sandstone to have been transported by high-energy events. In this context, the experiments in computational fluid dynamics carried out by Darroch et al. (2022) may help understand the fact that in the laminated upper part of the Aarhauser sandstone, the horizontal laminae intersect the vertical or steeply inclined vanes with no trace of edge effects (Figs. 8.2, 8.4, 9.2–9.4; Crimes and Fedonkin, 1996; Elliott et al., 2011; Darroch et al., 2022). This would be possible if the sediment were moving by laminar rather than turbulent flow when transport is parallel to the vanes, as shown by the calculated streamlines (Darroch et al., 2022, fig. 10C). In this scenario, the only individuals to be trapped and buried were those that were concave enough to receive and retain sediment; presumably, all of the rest were blown away like discarded plastic shopping bags in the surf (see artwork by John D. Dawson in Monastersky and Mazzatenta, 1998). Additional support for frequent transport may come from widespread bed base rake and bump structures (Fig. 12), which we interpret as tool marks generated by Pteridinium or another erniettomorph.

Remarks

Pteridinium is an uncommon fossil except at Aar and Plateau farms. It is, therefore, difficult to find populations large enough to compare with P. simplex. The next most frequent occurrences are from localities on the Onega Peninsula, Russia, but there the fossils are fragmentary and frequently deformed (Keller et al., 1974; Fedonkin, 1981, 1985; Palij et al., 1983; Ivantsov and Grazhdankin, 1997; Grazhdankin, 2004). Fortunately, the second species of Pteridinium to be described, P. carolinaensis (St. Jean, 1973), seems to differ markedly from P. simplex, so we begin by examining the binary choice, simplex or carolinaensis? If all known specimens of Pteridinium can be comfortably referred to one of these two species, then this interim solution may serve until more quantitative information becomes available. If there are numerous intermediates that cannot be so allocated, then perhaps P. simplex should serve as the only known species of Pteridinium for the time being.

In a careful study of 18 specimens of carolinaensis and 25 specimens of simplex using modern morphometric methods, Meyer (2010) and Meyer and Xiao (2010) were unable to find any statistically significant differences between these two species on the basis of a landmark analysis designed to capture variability in size, vane shape, and module curvature. They therefore suggested that P. carolinaensis is a junior synonym of P. simplex. However, they could not incorporate rarely preserved features of the fossils, such as overall frond size and shape, in their analysis because these features are seen in too few specimens.

Perhaps the most striking feature of the seven specimens of P. carolinaenis that have been illustrated (St. Jean, 1973; Gibson et al., 1984; McMenamin and Weaver, 2002; Gibson and Teeter, 2011) is that six show terminations, even in specimens comparable in size (~20 cm) to many examples of P. simplex. Similar terminations are known from one specimen from the Spitskop Member (UCLA 7373) identified as P. carolinaensis by Narbonne et al. (1997) and from one of the ~50 then-known specimens of P. nenoxa Keller in Keller et al.,1974 (Fedonkin, 1981, pl. 5, fig. 2), but not from any of the numerous specimens of P. simplex. Whether this is a demographic difference is difficult to assess because all of the specimens in the Aarhauser sandstone at Aar could, conceivably, be members of a single long-lived cohort. Given the differences in frond size, module curvature, and module expression across the vanes, we continue to treat simplex and carolinaensis as separate species. P. nenoxa shares those characteristics with P. carolinaensis rather than P. simplex (Fedonkin, 1985), as do specimens of Pteridinium from Bed A (UCLA 7373) at Swartpunt farm (Figs. 10.5, 11.4, 11.5; Narbonne et al., 1997; Darroch et al., 2022, fig. 7), so we follow others in considering nenoxa to be a junior synonym of carolinaensis (Runnegar and Fedonkin, 1992; Narbonne et al., 1997; McMenamin and Weaver, 2002; Fedonkin et al., 2007). Inkrylovia lata Fedonkin in Palij et al., 1979 is probably a preservational variant of nenoxa resulting from expansion of the modules parallel to the axis as a result of soft sediment loading.

Figure 11. (1–3, 6–8) Pteridinium simplex Gürich, 1933, Aarhauser sub-member, Kliphoek Member, Dabis Formation, UCLA 7307, Aar farm. (4, 5, 9–13) Other occurrences of Pteridinium in southern Namibia. (1, 2) An unusual specimen of P. simplex that tapers proximally (right), Plateau farm collection, 1996, UCLA 7327.3 (plaster cast). (3) Severely weathered block of horizontally bedded sandstone with a fragment of one wide vane that preserves some of the distal edge, GSN 1859. (4) Vertically oriented vane of Pteridinium carolinaenis (St. Jean, 1973) for comparison with (3), GSN F 250, Spitskop Member, Urusis Formation, Dundas, Swartpunt farm, southern Namibia, photographed in Kingston, Canada, 1998. (5) Another specimen of P. carolinaensis from the same locality, GSN F 248, showing the distal edge of one vane well, photographed in Kingston, Canada, 1998. (6, 7) Field photographs, taken on Aar farm by Louis Mazzatenta in 1996, of a tightly folded and twisted specimen of P. simplex, with and without a removable piece, GSN F 1854 (8), that preserves vanes V2 and V3. (8) Another view of GSN F 1854 showing a configuration that has the topology implied by Grazhdankin and Seilacher's (2002) vane substitution hypothesis (Fig. 8.5). (9) Pteridinium cf. P. carolinaensis (St. Jean, 1973), plaster cast of SMSWA 45731 now GSN F 1905, “Uit Schwarzkalk” (Mooifontein Member, Zaris Formation), Kosis farm, near Helmeringhausen. (10, 12, 13) Pteridinium sp., three specimens, GSN F 1901, GSN F 1899, GSN F 1900, respectively, from UCLA 7320, Neiderhagen Member, Nundas Formation, Kyffhauser farm, that may be preservational variants of P. simplex. (11) Pteridinium sp., UCLA 7315, shale immediately below Mooinfontein Member, Buchholzbrunn Member, Dabis Formation, Namaland district, near Bethanien, showing axis with two vanes partly obscured by overfolding of another vane or individual, GSN F 1891. (1) Camera lens cover = 60 mm; (2) scale bar = 5 cm; (3, 10, 11) scale bars = 5 cm; (4) scale bar = 2 cm; (5) scale bar = 1 cm; (6–8, 12, 13) scale bars = 2 cm; (9) scale bar = 3 cm.

Pteridinium carolinaensis (St. Jean, 1973)
Figures 10.5, 11.4, 11.5, 11.9

1966

Pteridinium cf. P. simplex Gürich, 1933; Glaessner and Wade, p. 616, pl. 101, figs. 1–3.

1973

?Paradoxides carolinaensis (sic) St. Jean, p. 204, pl. 3, figs. A–D.

1974

Pteridinium nenoxa Keller in Keller et al., p. 133, figs. 1, 2, 4, 5.

1976

Onegia ?nenoxa Keller; Sokolov, p. 141.

1977

Pteridinium simplex; Keller and Fedonkin, p. 926, pl. 2, fig. 4.

1979

Inkrylovia lata Fedonkin in Palij et al., p. 70, pl. 56, figs. 1–4.

1981

Pteridinium nenoxa; Fedonkin, p. 66, pls. 5–7, pl. 29, fig. 2.

1981

Inkrylovia lata; Fedonkin, p. 68, pls. 8, 9.

1983

Inkrylovia lata; Palij et al., p. 81, pl. 56, figs. 1–4.

1983

Pteridinium nenoxa; Palij et al., p. 81, pl. 58, fig. 3.

1985

Pteridinium nenoxa; Fedonkin, p. 99, pl. 11, figs. 1–4.

1985

Inkrylovia lata; Fedonkin, p. 100, pl. 12, figs. 3–5.

1992

Pteridinium carolinaensis; Runnegar and Fedonkin, fig. 7.5.9E.

1997

Pteridinium carolinaensis; Narbonne et al., p. 956, fig. 5.1–5.4.

2002

Pteridinium carolinaensis; McMenamin and Weaver, figs. 2–6.

2004

Onegia; Grazhdankin, fig. 4.

2022

Pteridinium simplex; Darroch et al., fig. 7.

Holotype

Cast of distal end of frond (NCSM 4041; previously UNC 3062) from the Floyd Church Formation, Albemarle Group, Island Creek, Stanly County, North Carolina, USA (St. Jean, 1973, pl. 3A; Gibson et al., 1984, fig. 6; McMenamin and Weaver, 2002, fig. 2; Weaver and Ganis, 2013, fig. 4A), by original designation.

Diagnosis

A three-vaned, leaf-like frond in which the modules are well expressed across the whole width of the vanes.

Description

Frondose organisms apparently formed of three equal-sized, undivided vanes that radiate from a common axis; vane shape is approximately aerodynamic, with a pair of opposed vanes of larger specimens subtending an angle of ~20° at the distal end of the frond and rounded at the proximal end; vanes are composed of curved tubular modules that commonly taper in width across the vanes and terminate abruptly at the distal margins; vanes terminate axially in polygonal ends that alternate with those of other modules in a zig-zag fashion, so far as can be determined; when well preserved, vane margins are delineated by a narrow differentiated edge; vane curvature low to negligible, largely for taphonomic reasons; module curvature within vanes is convex toward the presumed proximal end of the frond (Fig. 10.5) and concave toward the opposite end (St. Jean, 1973; Gibson et al., 1984); consistency of vane curvature throughout suggests that the organism grew in only one direction.

Materials

Five reasonably complete specimens observed in the field at Dundas Hill, Swartpunt farm, three specimens described and illustrated by Narbonne et al. (1997), which were examined at Queen's University, and plaster casts of several specimens from White Sea localities kindly supplied by M.A. Fedonkin and the Museum of Paleontology, University of California, Berkeley.

Taphonomy

All specimens of P. carolinaensis from the peri-Gondwanan Carolina terrane are from float (Meyer, 2010) so that only the nature of the sediments that enclosed them is known. The environment of deposition is thought to have been shallow marine adjacent to a volcanic arc, but there is little to indicate whether the fossils were preserved in place or transported before burial. According to Ivantsov and Grazhdankin (1997), the White Sea specimens of P. nenoxa are preserved in the Nama manner, and one actually stands vertically in the sediment (Fedonkin, 1981, pl. 29, fig. 2, 1985, pl. 11, fig. 1, 1992, fig. 26), although it is unclear why. The fossils are mostly biconvex, ovoid when viewed from below, and are preserved in sandstone that filled broad, erosive-based channels, similar to those seen in the Buchholzbrunn Member.

Remarks

If the Peridinium from the Spitskop Member on Swartpunt farm is correctly identified as P. carolinaensis, then one specimen (Fig. 10.5) preserves the proximal end of the frond, as noted by Narbonne et al. (1997). The holotype, paratype, Rock Hole Creek, and Gleaning Mission Church specimens preserve the distal end of the fronds (Gibson et al., 1984, fig. 5; McMenamin and Weaver, 2002, fig. 4; Gibson and Teeter, 2011), and a small specimen from the Gleaning Mission Church has both (McMenamin and Weaver, 2002, fig. 5). These fossils show that the vanes of P. carolinaensis narrowed toward each end of the frond, but the curvature of the modules and overall shape of the frond remained unidirectional throughout growth. If P. simplex followed the same pattern of growth, then the specimen shown in Figure 11.1, 11.2 represents the proximal part of the frond, not its growing end.

Pteridinium sp.
Figure 11.10–11.13

Remarks

Three specimens from the Nudaus Formation on Kyffhauser farm, north of the Osis ridge (Fig. 11.10, 11.12, 11.13), and one from the Buchholzbrunn Member, Namaland, west of Bethanie (Fig. 11.11) are referred to Pteridinium but not easily to either P. simplex or P. carolinaensis because of the narrowness of their modules.

Pteridinium tool marks?
Figure 12.1–12.8, 12.10, 12.11?

Description

Marks on sandstone bed bases produced by a comb- or rake-shaped tool that had ~30 tines, each capable of producing rounded grooves, narrow channels, or pairs of closely spaced parallel scratches that are typically 2–5 mm apart. In one case, the paired grooves form a chevron-like pattern (Fig. 12.2).

Remarks

Three examples, two from the Arimas (Fig. 12.3, 12.5A) and one from Kyffhauser (Fig. 12.4), are clearly erniettomorph in character. Presumably, they are impressions of parts of bodies or vanes thrust against the underlying eroded surface by the channel-filling sand that accompanied the transported bodies. In other cases, the tools merely raked or lightly scraped the surface. We interpret the rake marks as being due to the axes of Pteridinium fronds, rather than to bodies of Ernietta, because of the need for a wide head to the rake and because these structures extend well beyond the known stratigraphic range of Ernietta (Fig. 2; Darroch et al., 2021). Comparable tool marks have been reported from the Ingletonian of Yorkshire (Rayner, 1957), the Ordovician of Ohio (Osgood, 1970), the Silurian of Scotland (Trewin, 1979), and the Ordovician of Estonia (Vinn and Toom, 2016). Some are attributed to rolling hard fossils, such as crinoids and corals, but Rayner (1957) and Trewin (1979) were able to substantiate rake marks produced by graptolite stipes and their thecae, which served as the tines. These rake marks—if correctly interpreted—prove the presence of Pteridinium in the absence of body fossils, show that premortem transport was ubiquitous, and imply that the axes of the fronds were as stiff as graptolite stipes. The properties of the bump marks rule out both a spicular sponge source and a non-biological origin for these structures (Darroch et al., 2021). Tool marks attributed to Pteridinium were described by Fedonkin (1976) and Fedonkin in Palij et al. (1983) as the trace fossil Suzmites volutatus Fedonkin, 1976. Whether this name should be applied to the Namibian structures remains a matter for future investigation.

Figure 12. Tool and impact marks presumably left by erniettomorphs on the bases of flat, undulating, and incised sandstone beds, Huns (2, 3, 5, 7, 8, 10, 11) and Feldschuhhorn (1, 6) members, Urusis Formation and Neiderhagen Member, Nudaus Formation (4) on Arimas (UCLA 7309, 7326), Swartkloofberg (UCLA 7323) and Kyffhauser (UCLA 7320) farms, plus a vane of Nasepia altae Germs, 1972 from Arimas farm (9). (1) Broad comb-like toolmark, field photograph, Swartkloofberg, 1996. (2) Unique bilaterally symmetrical chevron-shaped tool mark, GSN F 1933. (3) Field photograph of an impact cast attributable to Pteridinium. (4) Lower surface and cross section of a shovel-shaped gutter cast, found by D.E. Erwin in 1995, showing Pteridinium-like impact mark on one side (arrow and insert), GSN F 1948. (5) Lower surface of large slab, left in field, showing a Pteridinium-like impact cast (arrow A and insert) and obscure impressions of several co-aligned specimens of Archaeichnium (arrow B), field photograph, 1996. (6) Hand specimen from same site as (1) showing similar comb marks, GSN F 1936. (7) Base of thin sandstone with evenly spaced bifid comb marks, GSN F 1924. (8) Another thin sandstone base with several sets of comb marks, one of which resembles the evenly spaced, bifid scratches of (7), field photograph, 1996. (9) Probable vane of Nasepia altae Germs, 1972 from the type locality but preserved in sandstone rather than limestone conglomerate (Fig. 14.4), GSN F 1909. (10) Third example of evenly spaced bifid comb marks, GSN F 1926. (11) Deep gouge mark on base of sandstone bed, which may or may not have been produced by a biological agent, GSN F 1932. (1) Comb approximately 3 cm wide; (2) scale bar = 1 cm; (3) coin = 25 mm; (4, 6, 7, 9–11) scale bars = 2 cm; (5) camera lens cap = 60 mm; (8) coin = 24 mm.

Genus Swartpuntia Narbonne, Saylor, and Grotzinger, 1997

Type species

Swartpuntia germsi Narbonne, Saylor, and Grotzinger, 1997 from the Spitskop Member of the Urusis Formation, Schwarzrand Subgroup, Swartpunt farm, Witputs district, Namibia, by original designation and monotypy.

Other species

None.

Swartpuntia germsi Narbonne, Saylor, and Grotzinger, 1997
Figures 13, 14.1–14.3, 14.6–14.8

1972a

?Nasepia altae Germs, p. 176, pl. 22, figs. 1–8.

1973

?Nasepia altae; Germs, p. 8, fig. 2A–G.

1997

Swartpuntia germsi Narbonne, Saylor, and Grotzinger, p. 956, figs. 4, 6, 9, 10.

1998

?Swartpuntia-like frond, Jensen, Gehling, and Droser, p. 568, fig. 2b, c.

2000

?Swartpuntia cf. S. germsi; Hagadorn and Waggoner, p. 351, fig. 4.

2000

non cf. Swartpuntia sp., Hagadorn, Fedo, and Waggoner, p.735, fig. 3.1, 3.2.

2006

non ?Swartpuntia sp., Weaver, McMenamin, and Tacker, p. 130, figs. 8–10.

2013

non Nasepia sp., Gehling and Droser, fig. 2J.

2022

Swartpuntia germsi; Hoyal Cuthill, p. 1211, fig. 1a.

2022

non cf. Swartpuntia; Hoyal Cuthill, p. 1211, fig. 1b.

2022

Swartpuntia germsi; Nelson et al., fig. 6A, B, C?

2022

?non cf. Swartpuntia; Meinhold et al., fig. 4b.

Holotype

Incomplete frond displaying parts of three vanes, one of which appears to have both upper and lower surfaces preserved, GSN F 238-H, from fossil bed B, Spitskop Member, Urusis Formation, Schwarzrand Subgroup, Dundas Hill, Swartpunt farm, southern Namibia.

Description

Cardioid to heart-shaped frond (Fig. 13.1), formed from at least three equal-sized vanes (Fig. 13.3) that are attached to a voluminous, knobbly axial structure (Fig. 14.6) so that the 50–100 narrow, tubular modules forming the distal parts of the vanes meet the axial structure at an angle of about 45°; those closer to the proximal end meet it nearly perpendicularly or even obtusely, and there is no evidence for a stem or stalk (Fig. 14.1, 14.8); confusion about the number of vanes arises because some appear to have been filled with fine sediment (Fig. 13.3; Narbonne et al., 1997, figs. 6, 7); in other cases, upper and lower surfaces of the vanes may be superimposed by composite molding so that the spacing of the modules may be halved.

Figure 13. Swartpuntia germsi Narbonne, Saylor, and Grotzinger, 1997 from beds A, UCLA 7373 (2) and B, UCLA 7374 (1, 3, 4) of Narbonne et al. (1997), Spitskop Member, Urusis Formation, Dundas Hill, Swartpunt farm. (1) Paratype, GSN F 423, showing the cardioid shape of the vanes, no evidence of a stem, and preservation of the vane surfaces on at least three levels, photographed in Kingston, Canada, in 1998. (2) GSN F 1886, upper surface of bed and partly overlapped by a specimen of Pteridinium carolinaensis (St. Jean, 1973). (3) Topotype, GSN F 1887, showing preservation of three (V1–V3) or possibly four vanes if VI* is not just the other surface of vane V1. (4) Paratype, GSN F 245, part and counterpart, showing no sign of a stem but clear evidence for three vanes, as illustrated by Narbonne et al. (1997, fig. 9.2), photographed in Kingston, Canada, in 1998. (1) Scale bar = 3 cm and loonie = 26.5 mm; (2) scale bar = 5 cm; (3, 4) scale bars = 3 cm.

Figure 14. Swartpuntia germsi Narbonne, Saylor, and Grotzinger, 1997 from bed B, UCLA 7374 (1, 2, 6–8) Spitskop Member, Urusis Formation, Dundas Hill, Swartpunt farm, and UCLA 7376, top of Huns Member, Urusis Formation, Swartkloofberg farm (3), plus a paratype of Nasepia altae Germs, 1972 (4) and the “Arimas lycopod” (5), both from UCLA 7326, Huns Member, Urusis Formation, Arimas farm (5). (1, 2, 8) Three views of a three-dimensionally preserved specimen of S. germsi, GSN F 1888, which exposes the proximal parts of the frond folded through about 90° and displaying no evidence for a stem; arrows in (8) mark the edges of one vane; (2) is flipped horizontally to serve as a mirror image of (1). (3) GSN F 1890, stratigraphically oldest known specimen of Swartpuntia, found by M.L. Droser in 1996, preserved in silt-sized carbonate, with axis presumably embedded in the counterpart. (4) Paratype of Nasepia altae Germs, 1972 ISAM K1086, showing distal edge of one vane embedded in a limy matrix that includes rounded limestone clasts (arrow), photographed in Cape Town, South Africa in 1993. (5) The “Arimas lycopod” (enlarged in insert), GSN F 1910A, found by JGG in 1996, may be the decorticated axis judging from circumstantial evidence; the organization of its diagonal arrays of “leaf scars,” analogous to those seen in lycopods, resembles that of the axial nodes of Swartpuntia, which are arranged in a similar fashion (6, 7, arrows). (6, 7) Topotype GSN F 1889 and paratype GSN F 247 (after Narbonne et al., 1997, fig. 10, republished with permission), of S. germsi that have well-preserved axial nodes. Scale bars = 2 cm except black bar in insert of (5) = 1 cm.

Materials

Four specimens from UCLA 7374 (GSN F 1886–1889) and one from UCLA 7376 (GSN F 1890) on Swartpunt and Swartkloofberg farms, respectively.

Remarks

Swartpuntia was originally reconstructed as a vertically oriented, three-vaned frond supported by a stout cylindrical stem that was attached to an unseen holdfast in the sediment (Narbonne et al., 1997, fig. 11; Narbonne, 1998, fig. 1). Only the holotype was thought to show evidence for a distinct stem, but the feature interpreted as the stem (Narbonne et al., 1997, figs. 6, 7) may well be just an elevated section of the matrix. Hoyal Cuthill (2022, p. 1211) wrote: “It is notable . . . that the stalk originally described in Swartpuntia (Narbonne et al. 1997) is not clearly visible even in the classic Namibian material.” We attempted to investigate this problem by collecting an in situ specimen preserved in relatively unweathered carbonate. Preparation of the proximal end revealed that opposing vanes are folded through ~90° across the axis of the putative stem and that the edge of one of the vanes can be followed to the axis (Fig. 14.1, 14.2, 14.8). There is no sign of a continuation of a voluminous axial structure beyond the margins of the frond.

The nature of the axial structure is not well understood, but it appears to have a surface formed of similarly sized, equally spaced, rounded projections that perhaps are arranged like the scales of a pineapple or the Fibonacci spirals of a pinecone (Fig. 14.6, 14.7). JGG found a possibly comparable axis at UCLA 7326 (Arimas), which became known as the “Arimas lycopod” (Fig. 14.5) because of its similarity to the bark of Paleozoic lycopods such as Leptophloeum. It was found at the same level as another float specimen (Fig. 12.9) with a fragment of a vane of Nasepia altae Germs, 1972 (Germs, 1972a, 1973), the only other example recovered subsequently from the type locality. Thus, the Arimas lycopod may be the decorticated axial structure of Nasepia judging from their co-occurrence and previously recognized similarities between the vanes of Swartpuntia and Nasepia (Fig. 14; Grotzinger et al., 1995; Narbonne et al., 1997). However, the syntypes of Nasepia are preserved in a carbonate conglomerate (Fig. 14.4), whereas the Arimas lycopod and the new Nasepia vane are both in blocks of sandstone, one of which also contains a poorly preserved specimen of Archaeichnium (Fig. 21.5). The only other penecontemporaneous fossil worthy of comparison with the Arimas lycopod seems to be Gibbavasis kushkii Vaziri, Majidifard, and Laflamme, 2018 (Vaziri et al., 2018, 2021) from the Ediacaran of Iran, but the similarities, although striking, are almost certainly superficial.

Swartpuntia has been positively or tentatively identified from the earliest Cambrian of South Australia (Jensen et al., 1998), the latest Ediacaran of Nevada and California (Hagadorn and Waggoner, 2000; Hagadorn et al., 2000), the early Cambrian of California (Hagadorn et al., 2000), the Ediacaran of North Carolina (Weaver et al., 2006), and the Spitskop Member just over the international border in South Africa (Nelson et al., 2022). Although the South African specimens are only parts of single vanes, their morphology and geographic and temporal proximity give confidence to the identifications. The same is not true for other reports, which should be treated skeptically on a case-by-case basis. The one American specimen that shows more than a fragment of a corrugated surface is LACNMH 12793 (Hagadorn and Waggoner, 2000, fig. 4.1, 4.2), which has two characters that may support an assignment to Swartpuntia: fine, dihedral linear striations that may be impressions of modules and an apparently knobbly axial structure. However, neither character is particularly convincing when compared with material from the type locality (Figs. 13, 14). The remainder of the referred specimens, including a sizeable surface from the Cambrian Poleta Formation of California (Hagadorn et al., 2000, fig. 3.2) may be pieces of erniettomorphs, but in the Cambrian at least, there are many other possibilities (e.g., MacGabhann et al., 2019; but see Hoyal Cuthill, 2022). However, the Swartpuntia-like fossils discovered by Jensen et al. (1998) deserve further investigation; there is more than one specimen, which is an important first step, and they display bilateral symmetry, discrete margins, and subdivisions reminiscent of erniettomorphs. They also occur very close to the local base of the Cambrian, which may be equivalent in age to the Ediacaran–Cambrian boundary zone in both Namibia (Linnemann et al., 2019) and South Africa (Nelson et al., 2022).

Swartpuntia has been reported with Vendoconularia Ivantsov and Fedonkin, 2002, Ventogyrus Ivantsov and Grazhdankin, 1997, and Calyptrina Sokolov, 1965 from the Onega Peninsula of the White Sea area, northern Russia (Ivantsov and Fedonkin, 2002). According to Serezhnikova (2014), this association is approximately 550 million years old, which would make this the oldest record of the genus. However, until this material has been figured and described, affinity with the Nama occurrences cannot be evaluated. A frond-shaped fossil from northern Norway has been tentatively compared with Swartpuntia (Meinhold et al., 2022, fig. 4b). This too would represent an occurrence older than those of the Nama Group, but in view of the revised morphology of Swartpuntia presented here, this comparison now is unlikely.

In summary, Swartpuntia is a Pteridinium-like frond that probably had only three equal-sized vanes set about a knobbly, voluminous, axial structure, and it lacked any kind of stem or stalk. Its modules and vanes resemble those of the smaller frond, Nasepia altae, so the discovery of a lycopod-like fossil, similar to the axial structure of Swartpuntia, with Nasepia at its type locality provides circumstantial evidence for a close relationship between the two genera. Swartpuntia/Nasepia is known with certainty only from Namibia and nearby South Africa, but one fragmentary specimen from California may belong to Swartpuntia. All other identifications are based on specimens that are too fragmentary or too little studied to warrant confident assignment to the genus or even to the Erniettomorpha.

Family Erniettidae Pflug, 1972
Genus Ernietta Pflug, 1966

Type species

Ernietta plateauensis Pflug, 1966 from the Buchholzbrunn Member of the Dabis Formation, Kuibis Subgroup, Aar farm, Aus district, Namibia, by original designation and monotypy.

Other species

Numerous other generic and specific names, as well as two new orders, four families, and five subfamilies, were proposed by Pflug (1972) for material on Aar farm that is essentially topotypic. In preparing the Precambrian section of the Introduction volume of the Treatise on Invertebrate Paleontology, Glaessner (1979b) attempted to rationalize Pflug's excessive splitting by recognizing only two subfamilies and five genera: Ernietta, Erniofossa, Ernionorma (Erniettinae), plus Erniobeta and Erniograndis (Erniobetinae). Soon after, Richard Jenkins effectively overruled this assessment with the statement: “One of us (Jenkins) has examined Pflug's material and considers that all the specimens he refers to as the ‘Erniettomorpha’ belong to a single genus and species, Ernietta plateauensis Pflug” (Jenkins et al., 1981, p. 71). That interpretation has become the status quo. However, in his summary of the genus Ernietta, Glaessner (1979b, p. A101) wrote as follows: “Body compressed at base into U-shape; ribs strongly developed, separated by zig-zag median line; resembling a folded petaloid of Pteridinium.”

When Pflug (1966) first described E. plateauensis, he thought he had the dorsal carapace of a soft-shelled worm or isopod-like arthropod but noted that the zig-zag dorsal suture was more like a structure found in Pteridinium than any animal except, perhaps Dickinsonia. One distinctive feature was a triangular mark at the topographical pole of the holotype, which he designated segment z (Fig. 19.1). This, he thought, was matched by segment 0 on the opposite side of the axis, and the segments were numbered away from these structures on both sides of the body. If segment z is a real feature of the anatomy, it is not seen in any other known specimens of Ernietta. It is, however, seen occasionally in U-shaped specimens of P. simplex (Fig. 19.2; Pflug, 1972, pl. 34, fig. 1), and it appears to be a tear of the seam between two modules on one side of the organism. Thus, Glaessner's diagnosis of Ernietta was more perceptive than he realized because the holotype of E. plateauensis (Fig. 19.1) is probably the tip of a tightly folded, U-shaped specimen of P. simplex.

One argument against this interpretation is that the discovery site “C” is described as “slate between Kuibis quartzite and black limestone in the lower part of the Nama system” (Pflug, 1966, p. 22), which clearly places it within the Buchholzbrunn Member (Fig. 3). This is the level where Ernietta abounds and Pteridinium is rarely seen (Fig. 2; Elliott et al., 2016). However, together with Bob Brain, Mark McMenamin, and Friedrich Pflüger, an attempt was made to recollect Pflug's locality C in 1993. Mark McMenamin found the only fossil, a vertical vane of Pteridinium (Fig. 19.7; McMenamin, 1998), at about the same stratigraphic level and geographic position as the holotype. So far as we know, Pflug's site C has not been resampled since that time; it is ~1.5 km east of the eastern edge of the geological maps of Plateau and Aar farms in Hall et al. (2013) and Elliott et al. (2016).

If the holotype of E. plateauensis is a small specimen of P. simplex, as seems likely, then Ernietta becomes a junior subjective synonym of Pteridinium. That is an undesirable outcome, given the long history of the use of Ernietta for a well-understood generic concept (Elliott et al., 2016; Ivantsov et al., 2016). However, the holotype of E. plateauensis is already an unsatisfactory standard because, as Pflug (1972, p. 139) noted: “From the collection area around point C [Pflug, 1966, fig. 1b] come the [holotype and paratype] specimens of the genus Ernietta and the specimens numbered 393, 399 of Erniotaxis. Almost all other pieces, with the exception of an Erniobeta colony, were found in collection area E, F.” (~2 km south of the southern edge of the maps of Plateau and Aar and farms in Hall et al. [2013] and Elliott et al. [2016]). As discussed in the following, Erniotaxis is an unusual juvenile form of Ernietta, and a “colony of Erniobeta” could mean many things. Thus, it is very difficult to obtain a population of individuals of E. plateauensis on the basis of the topotypic principle that all similar specimens from a bed or set of beds at one place are likely to be conspecific. Without such a sample to assess intraspecific variability, application of the name plateauensis is difficult. For both of these reasons, we recommend that the holotype of another of Pflug's species, Erniograndis sandalix Pflug, 1972, be designated the neotype of E. plateauensis. This recommendation will need the approval of the ICZN before it can take effect; in the meantime, community input is invited. There are precedents for this type of action to preserve useful names.

If there are to be other valid species of Ernietta, then Namalia villiersiensis Germs, 1968 has priority over all of Pflug's species except plateauensis. Again, the holotype of N. villiersiensis (Germs, 1968, fig. 1, 1972a, pl. 23, fig. 1) is not ideal in terms of preservation and the availability of topotypic material, but even worse, it may be missing (it could not be found at the ISAM in 1993). The type locality, Buchholzbrunn, has yielded the juvenile specimens of Ernietta shown in Figure 16, but they are preserved in a very different fashion from the holotype of Namalia villiersiensis. A better comparison is with the sandstone cast of a fossil—similar to those commonly attributed to Namalia villiersiensis or Kuibisia glabra Hahn and Pflug, 1985 (Hahn and Pflug, 1985a)—from the Aarhauser sandstone at Aar (Fig. 19.8, 19.9). This specimen was uncovered by the Seilacher team during their excavation in 1993. Thus, Namalia villiersiensis may be the senior synonym of Kuibisia glabra, and both may or may not be conspecific with Ernietta plateauensis (Jenkins et al., 1981; Runnegar and Fedonkin, 1992; Vickers-Rich, 2007; Ivantsov et al., 2016; but see Grazhdankin and Seilacher, 2002). The apparent differences between N. villiersiensis/K. glabra and neotypic Ernietta plateauensis may be due to preservation in coarse sandstone instead of siltstone.

Returning to Glaessner's (1979b) revision of Pflug's taxa, do his generic categories help break up Ernietta into distinct morphotypes? His subfamily Erniobetinae comprised two genera, Erniobeta and Erniograndis. A swift survey of Pflug's material may be obtained from the reproductions of his 13 Palaeontographica plates by Vickers-Rich (2007). Excluding plate 34, which deals mainly with E. plateauensis, 10 of the plates are devoted to specimens that generally resemble the shapes shown in Figure 19.3–19.6. The remaining two plates, 38 and 39, illustrate the Erniobetinae—Erniobeta and Erniograndis—which are bulky, internal molds of large specimens such as the proposed neotype for plateauensis (Fig. 15.3; Pflug, 1972, pl. 38, 1, 2, 4; Vickers-Rich, 2007, fig. 124; Elliott et al., 2016, fig. 3). Thus, the Erniobetinae sensu Glaessner (1979b) may serve as a population concept for E. plateauensis if the proposed neotype is eventually adopted.

Figure 15. Ernietta plateauensis Pflug, 1966, Buchholzbrunn Member, Dabis Formation, Plateau farm (2, 4, 7) and approximately the same stratigraphic level, UCLA 7378, Twyfel farm (1, 5, 6, 8–10). (1) Classic “sock in a rock” preservation, found in place and photographed in the field, both specimens numbered GSN F 1876. (2) Colorized version of one of five sketches based on plaster cast, YPM 204 508, of specimen in the “museum” at Plateau farm after Seilacher et al. (2003, fig. 11), copyright 2003, the Palaeontological Society of Japan, republished with permission. (3) GSN F 389, holotype of Erniograndis sandalix Pflug, 1972, and proposed neotype for E. plateauensis, photographed in Lich, Germany, 1993, GSN F 389. (4) Duplicate of the plaster cast used as the model for (2), UCLA 7327.2, gifted by the Seilacher team, was used to count the 70+ modules (Fig. 18.6) after tracing the between-module seams with a soft pencil. (5, 6) Two excavated specimens, GSN F 1863 and GSN F 1864, that share rhomboidal distal cross sections (corners indicated by arrows in (5)), viewed from above. (7) Photograph taken in 1993 of the specimen used to make the casts used for (2) and (4). (8–10) Four other excavated specimens, GSN F 1865, GSN F 1866, GSN F 1867, GSN F 1868, respectively, that show the typically pointed shape of the toe and, in the smaller specimens, evidence for growth interruptions. (1) Scale bar = 5 cm; (2, 4, 7) scale bar = 3 cm; (3) scale bar = 3 cm; (5, 6, 8–10) scale bars = 2 cm.

Pflug's plates 31 and 32 may best summarize the second morphotype, which Glaessner included in his subfamily Erniettinae; whether this morphotype may be specifically distinct from plateauensis is discussed in the following under the species description. Finally, plate 37, which shows specimens Pflug referred to Erniotaxis, is very different from all the others. Erniotaxis was one of five of Pflug's generic names that Glaessner (1979b, p. A102) dismissed as “unrecognizable.” Our discovery of this morphology on Twyfel farm, where it is associated with larger and more normal specimens (Fig. 17), allows us to show that Erniotaxis is a young growth stage that is allometrically different from larger individuals. The modified generic name “erniotaxid” may therefore serve as informal shorthand for this juvenile morphotype. Finally, we agree with Elliott et al. (2016) that Erniocarpus sermo Pflug, 1972 is not a specimen of Ernietta and suggest that while Erniocarpus orbiformis Pflug, 1972 may be, Erniocentrus centriformis Pflug, 1972 is certainly not.

Diagnosis

Sack-shaped, organic-walled bodies, oval or stadium shaped in cross section and U to V shaped in lateral profile, formed of tubular modules that meet in a zig-zag suture at the base, are attached to the outer wall, generated an inner wall by packing together during growth, and terminate distally in either stubby lobes or conical tips.

Occurrence

Kliphoek and Buchholzbrunn members, Dabis Formation, Kuibis Subgroup, Nama Group, Namibia (Fig. 2; Appendix); Lower Member, Wood Canyon Formation, Nevada, USA (Horodyski et al., 1994; Smith et al., 2017, 2022; Runnegar, 2022).

Ernietta plateauensis Pflug, 1966
Figures 15–17, 19.3–19.6, ?19.8, ?19.9

1966

non Ernietta plateauensis Pflug, p. 22, pl. 1, figs. 1–7.

1968

?Namalia villiersiensis Germs, figs. 1, 2.

1972a

Ernietta plateauensis; Germs, p. 174, pl. 21, figs. 4–9.

1972a

?Namalia villiersiensis; Germs, p. 177, pl. 23, figs. 1–7.

1972

non Ernietta plateauensis; Pflug, p. 163, pl. 34, figs. 4, 9.

1972

Erniodiscus rutilus Pflug, p. 158, pl. 27, figs. 3, 4.

1972

Erniodiscus clypeus Pflug, p. 158, pl. 27, fig. 1.

1972

Erniaster apertus Pflug, p. 159, pl. 28, figs. 1–3, 5–7.

1972

Erniaster patellus Pflug, p. 159, pl. 29, figs. 1, 4, 8.

1972

Erniofossa prognatha Pflug, p. 159, pl. 27, figs. 2, 6, 7.

1972

Ernionorma abyssoides Pflug, p. 160, pl. 29, figs. 6, 7, 10–12.

1972

Ernionorma peltis Pflug, p. 160, pl. 30, figs. 1, 7, pl. 29, figs. 2, 5.

1972

Ernionorma clausula Pflug, p. 160, pl. 31, figs. 2, 3.

1972

Ernionorma rector Pflug, p. 161, pl. 32, figs. 4, 6–9.

1972

Ernionorma corrector Pflug, p. 161, pl. 32, figs. 1–3, 5.

1972

Ernionorma tribunalis Pflug, p. 161, pl. 31, figs. 4–8.

1972

Erniobaria baroides Pflug, p. 162, pl. 31, figs. 11, 12, pl. 32, figs. 10, 11.

1972

Erniobaris gula Pflug, p. 162, pl. 33, figs. 1, 2, 4.

1972

Erniobaris epistula Pflug, p. 162, pl. 31, figs. 9, 10.

1972

Erniobaris parietalis Pflug, p. 162, pl. 33, figs. 3, 5, 6.

1972

Erniopelta scrupula Pflug, p. 163, pl. 33, figs. 7, 10.

1972

Ernietta aarensis Pflug, p. 163, pl. 34, figs. 5, 7, 8.

1972

Ernietta tsachanabis Pflug, p. 164, pl. 34, figs. 10–12.

1972

Erniocarpus carpoides Pflug, p. 164, pl. 35, figs. 5–9.

1972

Erniocoris orbiformis Pflug, p. 164, pl. 36, figs. 1–4.

1972

Erniotaxis segmentrix Pflug, p. 165, pl. 37, figs. 1–8, pl. 35, fig. 10.

1972

Erniograndis sandalix Pflug, p. 165, pl. 38, figs. 1, 2, 4, pl. 35, fig. 1.

1972

Erniograndis paraglossa Pflug, p. 166, pl. 38, fig. 3, pl. 39, figs. 7–9, 11.

1972

Erniobeta scapulosa Pflug, p. 166, pl. 39, fig. 6.

1972

Erniobeta forensis Pflug, p. 166, pl. 39, figs. 2–5, 10.

1981

Ernietta plateauensis; Jenkins, Plummer, and Moriarty, fig. 5A–E.

1985a

?Kuibisia glabra Hahn and Pflug, p. 5, pl. 2, 3.

2016

Ernietta plateauensis; Ivantsov et al., figs. 5, 6.

2016

Ernietta plateauensis; Elliott et al., p. 1019, figs. 3–5 (with additional synonymy).

2017

Ernietta; Smith et al., fig. 3.

2022

Ernietta plateauensis; Runnegar, p. 1104, fig. 3.

Holotype

Tip of U-shaped internal mold (GSN F 429; previously Pflug no. 227) from the Buchholzbrunn Member, Aar farm, Aus district, Namibia by original designation (Fig. 19.1; Pflug, 1966, pl. 1, figs. 1–3, 1972, pl. 34, fig. 4).

Proposed neotype

Nearly complete internal mold (GSN F 389; previously Pflug no. 192) from the Buchholzbrunn Member, Aar farm, Aus district, Namibia (Fig. 15.3; Pflug, 1972, pl. 38, figs. 1, 2, 4; Vickers-Rich, 2007, fig. 124; Elliott et al., 2016, fig. 3; Maloney et al., 2020, fig. 3A).

Description

Body sack-like, composed of as many as 70 tubular modules arranged side by side around the circumference and joined proximally (ventrally) in a zig-zag seam; body cross section is elliptical to stadium shaped; basal profile at right angles to the zig-zag seam is U shaped or rounded V shaped, and parallel to the seam it is U shaped; upper part of body usually truncated postmortally, with the upper parts frequently assuming the cross-sectional shape of a four-pointed star (Fig. 15.5) that is aligned with the symmetry axes; modules are approximately constant in width for any growth stage until they approach the distal (dorsal) margin, where they start to separate from each other and taper toward pointed ends (Smith et al., 2017, fig. 3d; J.G. Hall et al., 2020, fig. 1b; Runnegar, 2022, fig. 3a; possibly Narbonne, 2005, fig. 4b); modules are circular in cross section in the tapering tips, square to rectangular in cross section throughout much of the upper part of the body, and triangular to D shaped near the base, where they are in contact with each other only at the outer wall (Fig. 17.6); rare internal molds (Fig. 15.7) show that the sides of adjacent modules approached each other during growth and then coalesced about one-third of the way to the top of the body.

Materials

Numerous specimens (GSN F 1860–1877, 1880, 1881) from UCLA 7317 and UCLA 7378 on Buchholzbrunn and Twyfel farms and rare specimens from UCLA 7312, UCLA 7313, UCLA 7314, UCLA 7315, and UCLA 7381, all west of the road from Bethanie to Helmeringhausen (Fig. 1).

Ontogeny

Two sizeable blocks of sandstone, each part of a sand-cast gutter fill, were found on the floor of a small road metal quarry in the Buchholzbrunn Member on Buchholzbrunn by SJ and BR in 1995 (UCLA 7317; Fig. 16). The lower surfaces of these blocks preserve numerous small specimens of Ernietta that were transported with the sand and settled first, presumably because they behaved hydrodynamically like pebbles. The fossils are not well preserved, but they do display the modules well enough for them to be counted and compared with specimen size (Fig. 18.6). The smallest identifiable specimen, ~4 mm in diameter (E in Fig. 16.4, 16.5, 16.7), appears to have four modules (Fig. 16.7), although only three are visible in the gutter cast. An even smaller object to the upper left of it (Fig. 16.7) is preserved in the same fashion and may be an ~1 mm larval stage with only one module, such as the tiny White Sea specimens of Dickinsonia costata Sprigg, 1947 illustrated by Ivantsov and Zakrevskava (2022, pl. 1, figs. 1, 2). The subsequent growth of E. plateauensis is summarized in Figure 18.6, which is a plot of countable module number versus body size (length + width/2). The largest individual measured was a plaster cast of a specimen in the Plateau “museum” collection (Fig. 15.4, UCLA 7327.2, YPM 204 508; Seilacher et al., 2003, fig. 11, bottom row; Seilacher, 2007, fig. 1) that has ~70 modules. Thus, body size is a reasonable predictor of module number (Fig. 18.6), although Ivantsov et al. (2016) found a fairly constant number of modules (~26) in a cohort of similar-sized individuals preserved in a gutter cast. Presumably, modules are added at one or both ends of the zig-zag seam (Ivantsov et al., 2016), but it has not been possible to identify the most recently added modules in those areas because of inadequate preservation. At the base of the body, the modules terminate in separate, rounded ends (Fig. 15.7), so it is not obvious how new modules are generated; they may even be intercalated around the body during growth, as might be recorded by impressions of vertical seams on internal molds (Fig. 15.2, 15.4, 15.7; Jenkins et al., 1981, fig. 5B). However, these structures are better attributed to changes in module width and shape during growth, as discussed in the following.

Figure 16. Juvenile specimens of Ernietta plateauensis Pflug, 1966 preserved on the bases of two sizeable pieces of a gutter cast, found on the floor of a small road metal quarry, Buchholzbrunn Member, Dabis Formation, UCLA 7317, Buchholzbrunn farm, near Goageb. (1) Whole block, GSN F 1860. (2) Part of second block, GSN F 1861; arrows indicate directions of current flow. (3, 6) Enlargements of GSN F 1860 with individuals used for module counts (Fig. 18.6) indicated by letters. (4, 5, 7) Enlargements of parts of GSN F 1861 with smallest identifiable individual labeled E and three of its four modules indicated by L, M, and R (other arrows point to the ends of the modules of a larger individual); the bump above E in (7) may be the base of a tiny one-module postlarva (Fig. 18.6). (1, 2) Scale bar = 5 cm; (3, 4, 6) scale bars = 2 cm; (5) scale bar = 1 cm; (7) scale bar = 5 mm.

One of Pflug's (1972) taxa, Erniotaxis segmentrix, is a puzzling set of small objects that Glaessner (1979b) dismissed as “unrecognizable” and Elliott et al. (2016, p. 1024) thought were “part of the midline of a fragmentary Pteridinium fossil.” Our discovery of two similar small specimens on Twyfel farm (Fig. 17.1–17.4) validates Pflug's recognition of the importance of his find, and the two small collections reveal the same features of the internal anatomy of young examples of E. plateauensis (Fig. 16.1–16.5). The striking feature of both is the depth of the walls of the modules and their concave lateral surfaces. In these specimens, the inner edges of the modules are blade-like and close to the axis of the body (Fig. 17.2). The concavity of their lateral walls suggests that there may have been open space between the modules and that they were in contact only at the outer wall. This morphology is seen more clearly in a larger but still youthful specimen from the same site (Fig. 17.6), which is best understood as an inverted fragment of the lower part of the specimen shown in Figure 17.10 (in both examples, the modules taper distally). In these specimens, the modules were filled with carbonate following burial, so the whole structure is preserved in three dimensions and could in one case be separated from the internal mold (Fig. 17.6). The two halves of this specimen show clearly that, at an early growth stage, the modules were D shaped in cross section and in contact with each other only at the seams of the outer wall, best seen in the external mold (Fig. 17.6, left) and modeled in Figure 18.7, 18.8. Larger, more mature individuals have modules that are wholly in contact laterally, resulting in square to rectangular cross sections (Fig. 15.4; Pflug, 1972, pl. 38, fig. 3; Jenkins et al., 1981, fig. 6C; Elliott et al., 2016, fig. 4.2), except toward their growing terminations, where the modules separated and assumed a hydrostatic (circular) cross section (Runnegar, 2022, fig. 3a). There are also some contentious components of the growth of Ernietta: (1) whether there were two or more layers of modules in the body wall; (2) the nature of the growing terminations of the body; (3) the significance of waist-like constrictions that are seen in many specimens, including the proposed neotype; (4) whether sand was incorporated into the body during life. These matters are reviewed in the following under remarks.

Figure 17. Juvenile and small specimens of Ernietta plateauensis Pflug, 1966, Buchholzbrunn Member, Dabis Formation, UCLA 7378, Twyfel farm (1–4, 6, 7, 9, 10); UCLA 7308, Aar farm (5); and UCLA 7313, Klipdrif farm (8). (1–4) Unusual, globose specimens preserved in carbonate that resemble the Erniotaxis segmentrix morph (5) in having extraordinarily wide walls between adjacent modules (2) and highly curved outer walls, GSN F 1869 (1–3) and GSN F 1870 (4). (5) Holotype of Erniotaxis segmentrix Pflug, 1972, no. 396 now GSN F 449, photographed in Lich, Germany, in 1993. (6) Part and counterpart of a specimen with carbonate-filled modules that are convex in both outward and inward directions and are in lateral contact only at the outer surface (arrows); GSN F 1874. (7, 9) Excavated block shown in original orientation with two visible specimens of E. plateauensis, one of which is removable and is shown in inverted orientation in (9). (8) A deformed specimen found with other individuals in a small channel ~3 m below the first limestone of the Mooifontein Member, GSN F 1956. (10) Rare example of preservation of the outer surface of the organism as a result of carbonate-filled modules, GSN F 1872. Scale bars = 1 cm.

Figure 18. Growth of Ernietta. (1–5, 7, 8) Super3D models. (6) Scatter plot of size versus number of modules in Ernietta plateauensis Pflug, 1966 (filled circles); the holotype of E. plateauensis, thought to be a deformed specimen of Pteridinium simplex, is represented by the filled square. (1) Perspective view of two identical copies showing how the modules interdigitate along the proximal seam. (2–5) Orthographic views of the base of the model tilted about X by 30° (2) and 20° (3, 4) showing the progressive deconstruction of the model, which is based on rectangular modular cross sections found in mature individuals of Ernietta from Nevada. (7, 8) The basal part of the external layer of the model and three modules of the kinds seen in immature individuals from Namibia (Fig. 17.6), where the cross sections are D-shaped and end proximally in wedge-shaped terminations (arrow), reminiscent of the youthful modules of the erniotaxid morphotype (Fig. 17.2); it is assumed that the D-shaped modules merge distally into mature box-shaped ones.

Taphonomy

An evocative metaphor for the preservation of a mature individual of Ernietta is not a “rock in a sock” (Seilacher, 1992) but rather a “sock in a rock” (Knoll, 2003, p. 166; Fig. 15.1). Is this morphology the result of mass flow transport and burial or a life orientation? A recent consensus is the latter based on in situ specimens from localities west of the Aar homestead (Elliott et al., 2016) and sites on Weigkrup and Hansburg farms (Bouougri et al., 2011; Maloney et al., 2020), which are close to our localities UCLA 7378 and UCLA 7379 on Twyfel and Weigkrup (Fig. 1). Perhaps the best evidence for this interpretation is the fact that nearly all specimens are oriented with their zig-zag seams down and occur in clusters that are thought to have developed in depressions in the seafloor. However, some of these group occurrences appear to be secondarily transported (Ivantsov et al., 2016). At Twyfel, we also found that rare specimens in the same bed as the upright clusters are inverted, a configuration that is difficult to explain in an in situ community. Given the high probability that transported specimens partly filled with sand would aggregate seam downward in depressions, the life orientation of any of these clusters is questionable.

The carbonate infilling of the modules of the specimen shown in Figure 17.6 is unusual and previously unreported from Namibia. As the specimen is unique, no preparation of it was undertaken, so the identification of the fill, based on microscopic examination of fractured surfaces, is tentative. Another specimen found with it (Fig. 17.10) seems to be preserved in the same way and could be examined with computed tomography in the future. A possibly similar style of preservation has been reported by Ivantsov (2018) from the Ediacaran of Siberia.

Remarks

Jenkins et al. (1981, fig. 6) published a reconstruction of Ernietta based on Pflug's material, which Jenkins had examined in Giessen with Pflug's assistance. Two key observations, based on an unfigured syntype of Erniograndis sandalix (Pflug no. 182), were the presence, near the base, of a small piece of sediment that had filled the interior of a second outer palisade of modules and an “enigmatic ‘budding’ suture” that encircled the upper part of the internal mold and is also present in the holotype (Fig. 15.3; Pflug, 1972, pl. 38, figs. 1, 2, 4; Vickers-Rich, 2007, fig. 124; Elliott et al., 2016, fig. 3; Maloney et al., 2020, fig. 3A) and several other specimens from Aar (e.g., Fig. 15.2). Jenkins also sketched three cross sections of Ernietta on the basis of sawn specimens, including a paratype of E. sandalix (Pflug, 1972, pl. 39, fig. 1; Jenkins et al., 1981, fig. 5B), which is shown as displaying a voluminous inner layer of modules crossed by rare septa and a tiny, attached fragment of a second layer of modules. Pflug's (1972, pl. 39, fig. 1) figure of the paratype shows a thick layer of white sediment near the base of the organism, comparable to that illustrated by Ivantsov et al. (2016, fig. 6E, F) in a similar sawn section, and two or three linear features that could be the intersections of septa. However, their convexity is in the opposite direction to the septa shown in Jenkins's sketch, and the thickness of the adhering second layer of modules is negligible. Furthermore, the other two cross sections sketched by Jenkins each have only one layer of modules. Thus, evidence for a second layer of modules is limited to a few fragments adhering to the bases of syntypic specimens of E. sandalix (e.g., Elliott et al., 2016, fig. 4.2) and an image of the polished sawn surface of GSN F-1243 from the Teapot locality on Aar (Elliott et al., 2016, fig. 5.3), which show two layers of honeycomb-like cells. Given that the orientation of the polished surface with respect to the fossil is not clear, that the section may intersect part of the ventral zig-zag seam, and that some mineralized cracks may mimic mineralized organic walls, it also does not provide strong support for the dual- or multiple-wall hypothesis that forms the basis for the remarkably similar reconstructions of Jenkins et al. (1981, fig. 6A) and Ivantsov et al. (2016, fig. 7). Because there is no evidence for more than a single wall in the great majority of specimens of Ernietta (Fig. 17.6; Pflug, 1972; Hall et al., 2020; Runnegar, 2022) or Namalia (Grazhdankin and Seilacher, 2002), caution is recommended until a complete individual or population of individuals showing evidence for more than one palisade wall becomes available. We tentatively attribute those few specimens that have some evidence of more than one layer near their bases to abnormal development, regeneration after injury, or some unidentified taphonomic process.

This suggestion may also apply to Jenkins's encircling suture, which interrupts the modules so severely that they may be significantly narrower above it and not in register with the modules below it (Fig. 15.2, 15.3; Jenkins et al., 1981, fig. 5B), details not captured in the reconstructions (Jenkins et al., 1981, fig. 6A; Ivantsov et al., 2016, fig. 7). Perhaps an explanation for these sutures is that they also represent a response to traumatic injury, such as truncation of the top of the organism by storm surge, followed by subsequent regrowth. This may explain why the suture is lower down in specimens from Twyfel (Fig. 15.9) than in those from Aar (Fig. 15.3).

The nature of the growing ends of the modules is another feature of the two reconstructions that deserves reassessment. In Jenkins's reconstruction, the modules terminate in lappet-like edges, which border a pair of wide lips formed from three concentric palisades of modules (Jenkins et al., 1981, fig. 6A). Ivantsov et al. (2016, fig. 7) show two concentric rows of similar lappets that flare outward, away from the symmetry axis. These reconstructions contrast with the ones by Monastersky and Mazzatenta (1998), based on undescribed specimens of Ernietta from Nevada (Horodyski et al., 1994), which have the modules terminating in narrowly tapering cones with pointed ends (Runnegar, 2022, fig. 3a). The holotype of Kubisia glabra has lappet-like terminations (Ivantsov et al., 2016, fig. 8D), adding support for the Jenkins–Ivantsov reconstruction, but another specimen of Ernietta from Namibia seems to have distal terminations like the Nevada examples (Narbonne, 2005, fig. 4b; Smith et al., 2017, fig. 3d; Hall et al., 2020, fig. 1b; Runnegar, 2022, fig. 3a). Does this mean that Ernietta could withdraw and collapse its tentaculate terminations like anemones exposed at low tide, or that there are two distinct, co-existing morphotypes? These questions deserve further investigation.

As mentioned previously, Glaessner's (1979b) Erniettinae, best exemplified by the specimen shown in Figure 19.4, 19.5, may prove to be another distinct morphotype/species of Ernietta. If so, Pflug's Ernionorma abyssoides may be the name to use, which is why we re-illustrate an epoxy cast of the holotype (Fig. 19.3; GSN F 485; Pflug no. 280). This morphotype is commonly found as basal pieces formed of numerous closely spaced modules (Fig. 19.5, white dots) but may also have highly variable module widths (Fig. 19.6). A morphometric analysis of populations tied to the type localities for the holotypes of E. sandalix and E. abyssoides will be needed to answer this question.

Figure 19. Ernietta plateauensis Pflug, 1966, Buchholzbrunn Member, Dabis Formation, UCLA 7308, Aar farm (1, 3) and approximately the same stratigraphic level, UCLA 7379, Wegkruip farm, plus Pteridinium simplex Pflug and Namalia villiersiensis Germs, UCLA 7307, Aarhauser sub-member, Kliphoek Member, Dabis Formation, Aar farm (2, 7–9). (1) Plaster cast of holotype of Ernietta plateauensis Pflug, 1966, no. 227 now GSN F 429, probably a deformed and torn specimen of P. simplex (2) that should be replaced by a neotype such as the holotype of Erniotaxis segmentrix Pflug, 1972 (Fig. 15.3). (2) Plaster cast of a deformed and torn specimen of P. simplex, SMSWA 45370.2 now GSN F 1879, that shows a similar triangular lesion to the one in the holotype of Ernietta plateauensis, which Pflug (1972) termed an “apicostomatous aperture”; however, note that there are three vanes (V1–V3) preserved in this specimen. (3) Underneath view of epoxy cast of the holotype of Ernionorma abyssoides Pflug, 1972, no. 280 now GSN F 485, donated by H.D. Pflug, for comparison with specimens from Wegkruip farm (4–6); (4, 5) Underneath and lateral views of a weathered but otherwise well-preserved internal mold with the number of visible modules indicated by white dots, GSN F 1880. (6) Four similar-sized specimens to illustrate variations in module size and number, GSN F 1881, GSN F 1882, GSN F 1883, GSN F 1884, respectively. (7) Fragment of one vane of a specimen of P. simplex, embedded in a horizontally bedded sandstone, found by M.A.S. McMenamin at or near the type locality of E. plateauensis, field photograph, 1993, GSN F 2209 (McMenamin, 1998, p. 85, fig. 5.3). (8, 9) Two views of a specimen resembling the holotype of Namalia villiersiensis that was excavated by the Seilacher team at Aar farm, field photographs, 1993, GSN F 612 (Grazhdankin and Seilacher, 2002, text-fig. 9F–H). (1, 3–6) Scale bars = 1 cm; (2, 7–9) scale bars = 2 cm; (7) coin = 23 mm.

A fourth area of concern for understanding the biology and taphonomy of Ernietta is the long-standing questions as to whether they were epibenthic or endobenthic, whether they received sediment passively in their body cavities during life or actively incorporated sediment into their body tissues to enable them to remain upright if disturbed. There is no doubt that many are preserved with their modules filled with sediment, which is often coarser and better sorted than the matrix that surrounds them (Pflug, 1972; Ivantsov et al., 2016; Hall et al., 2020). This is particularly noticeable in Nevada, where the modules of corrugated bodies or their degraded bag-like remnants are full of clean quartz sand, quite unlike the deep-water, silty matrix in which they are interred (Hall et al., 2020). This suggests that only those bodies that were torn and filled with coarse grains during high-velocity transport were preserved. However, Ivantsov et al. (2016) opted for a division of function along the length of the modules, with active incorporation of sand in their basal parts, a fluid-filled hydrostatic function for their middle parts, and an aerobic/osmotic function for their distal parts. Evidence for sediment incorporation came from longitudinal thin and polished sections, which showed a wide zone of clean quartz sand between the outer and inner walls and sequential fill of less coarse sediment within the body cavity (Ivantsov et al., 2016, fig. 6E, F), similar to the sawn section illustrated by Pflug (1972, pl. 39, fig. 1; Jenkins et al., 1981, fig. 6B). However, the problem with these cross sections compared with internal molds of the E. sandalix type is that there is the paucity of partitions attributable to septa—even allowing for the small angles between sections and septa—and the distance between the inner and outer walls is proportionally large compared with the module depths recorded by internal molds. These discrepancies raise the possibility that, although the inner and outer walls were largely intact, the walls and septa had been sufficiently breached to allow coarse suspended grains to enter the wall cavity during transport. Thus, the Namibian and Nevadan specimens may have been preserved under similar conditions. The highly structured nature of the filling of the body cavity of a Namibian specimen (Ivantsov et al., 2016, fig. 6B, C) may provide a reason to doubt this interpretation, but it may also be explicable by waning storm surge sedimentation if the bodies had sufficient mechanical strength to remain open during burial.

Subkingdom Eumetazoa Bütschli, 1910
Phylum Cnidaria? Verrill, 1865
Family Mackenziidae? Conway Morris, 1993
Genus Archaeichnium Glaessner, 1963

Type species

Archaeichnium haughtoni Glaessner, 1963 from the Kuibis? or Schwarzrand Subgroup, Karasburg district, Namibia, by monotypy.

Other species

None. Two species of Archaeichnium erected on Paleozoic material, Archaeichnium kunmingensis Luo in Luo et al. (1994), from the lower Cambrian of Yunnan, China, and Archaeichnium(?) xizangensis Yang in Yang et al. (1983), from the Upper Carboniferous of Tibet, are trace fossils with longitudinal striations.

Diagnosis

Narrow conical organism with about 10–12 longitudinal ridges that may be the edges of, and internally septate, pleated body wall, possibly enclosed in similarly corrugated unmineralized epitheca.

Archaeichnium haughtoni Glaessner, 1963
Figures 20, 21

1960

?archaeocyathid Haughton, p. 38, pl. 3–5, pl. 3, fig. 1;.

1963

Archaeichnium haughtoni Glaessner, p.117, pl. 3, figs. 1, 2.

1978

Archaeichnium haughtoni; Glaessner, figs. 1, 2.

2000

non cf. Archaeichnium sp., Hagadorn and Waggoner, p. 351, figs. 3.5, 3.6.

2013

non Archaeichnium sp., Gehling and Droser, fig. 2M, N.

2016

microbially induced crack, Buatois and Mángano, fig. 2.7c.

2022

indeterminate trace fossils, Turk et al., p. 9, figs. 9.1, 9.2, 9.4, 9.5.

Holotype

One of several specimens probably preserved in convex hyporelief on a small slab of sandstone, ISAM K4812, ostensibly from the Nababis Formation, Kuibis Subgroup, on Gründoorn farm, near the Ham River, ~60 km east of Karasburg, southern Namibia (Fig. 20.1; Haughton, 1960, pl. 4, 5; Glaessner, 1963, pl. 2, fig. 1, 1978, fig. 1), but possibly from the younger Schwarzrand Subgroup in the same section (near 19.295356°E, 28.094153°S); by original designation and monotypy.

Figure 20. Archaeichnium haughtoni Glaessner, 1963, Nakop Member, Nababis Formation, Gründorn farm (57) (1) and Huns Member, Urusis Formation, UCLA 7309, Arimas farm (2–8). (1) Holotype of A. haughtoni, ISAM K4812, photographed in Cape Town, South Africa, in 1993. (2) Sandstone slab with two specimens, GSN F 1904A (3, 4, 6, 8) and GSN F 1904B (5), that reveal much of the anatomy of the form. (3, 4, 6, 8) Four views of GSN F 1904A taken with different lighting and equipment to show the nature of the body wall and its construction. (5) End piece showing likely origin of growth. (7) A co-occurring external mold that is longitudinally fluted and may represent a cast of the cuticle or tube of Archaeichnium. (1) Coin = 19 mm; (1, 2) scale bars = 2 cm; (3–5) scale bar = 1 cm; (6–8) scale bars = 5 mm.

Description

Longitudinally ridged, tubular fossils that taper gradually or rapidly to a closed end (Figs. 20.5, 21.1), which may have been fixed to the substrate (Fig. 21.1, 21.4); the greatest diameter is typically ~5 mm, and the tubes may exceed several cm in length; the number of longitudinal ridges is ~10–12 assuming that external molds, which have ~3.5 ridges (Fig. 21.6), represent about a third of the circumference; perimortem kinks (Fig. 21.2) and twists (Fig. 20.1) in the tubes suggest that they were unmineralized and flexible; a remarkable specimen (Fig. 20.3, 20.4, 20.6, 20.8) shows either an impression of the side of one ridge or the side view of a flange that either extended outward from the ridge crest or is an internal extension from the body wall; the structure has evenly spaced, radially oriented ridges that are about 0.5 mm apart and of similar length that may have provided structural support or had some other function (see Remarks); rare external molds suggest that, in life, the ridges were narrow and stiff and the intervening wall segments were concave so that the cross section resembled a concave or parabolic star with 10–12 points (Fig. 21.6).

Figure 21. Archaeichnium haughtoni Glaessner, 1963, Huns Member, Urusis Formation, UCLA 7309, Arimas farm (1), UCLA 7325, Holoog River (2), and Neiderhagen Member, Nudaus Formation, Kyffhauser farm (3–6). (1) Four longitudinally striated individuals with pointed terminations (arrows), presumed to be the origins of growth, on the base of a 3 cm thick sandstone bed with “old elephant skin” texture, GSN F 1906. (2) Two specimens from the Holoog River, one of which is severely kinked (insert), GSN F 1962 and GSN F 1975, respectively. (3) Superb bed base, GSN F 1939, found by D.E. Erwin in 1995, with at least eight tethered and current-oriented individuals, six facing right and two facing left, with the three best-preserved ones indicated by arrows and shown in (4). (4) Three panels enlarged from (3) to show left-facing (top, GSN F 1939A) and right-facing individuals (middle, GSN F 1939B, bottom GSN F 1939C). (5) External mold, GSN F 1949. (6) An external mold, photographed in the field and then discarded, figured as a pseudofossil by Buatois and Mángano (2016, fig. 2.7c) that clearly shows the pleated nature of the body wall; image kindly provided by Luis Buatois, rotated through –90° so that it appears in positive rather than negative relief. (1) Scale bar = 2 cm; (2, 2 insert, 3, 5, 6) scale bars = 1 cm; (4) scale bar = 5 mm.

Materials

Six pieces of sandstone from Arimas (UCLA 7309), two from the Holoog River (UCLA 7325), and one from Kyffhauser (UCLA 7320), most having more than one specimen of Archaeichnium, plus an image of a specimen (Fig. 21.6) figured by Buatois and Mángano (2016, fig. 2.7c), generously provided by Luis Buatois.

Remarks

The holotype and other specimens on the same surface are tubular fossils that are longitudinally ridged and may or may not taper to pointed ends. Haughton (1960) thought the pointed ends were closed and possibly attached to the seafloor; Glaessner (1963, 1978) rejected Haughton's comparisons to archaeocyathids and thought that the tapering of the tubes was caused by their trajectory out of the bedding plane. We are confident that the tubes tapered to closed ends because they overlay surfaces sealed by microbial mats before being buried by event sands and do not leave those surfaces (Fig. 21.1, 21.2).

Glaessner (1978) thought Archaeichnium was some kind of agglutinated sand worm tube, and Turk et al. (2021, 2022) have compared it to priapulid burrows, but it is clear from the specimen discovered by JGG at Arimas (Fig. 20.6) that Archaeichnium must be a body fossil, not a trace fossil. Presumably, the ladder-like feature preserved in this specimen is part of a tubular and perhaps pleated body wall. As it associated with one of the longitudinal ridges, it may be one of numerous similar structural elements, each comprising part of one of the ridges. If the external molds shown in Figure 20.7, 20.8 do represent casts of the exterior of Archaeichnium, on the basis of their co-occurrence, then all of the wall complexity would presumably lie inside this unmineralized epitheca. Thus, the ladder-like feature (Fig. 20.6) may have extended inward from the ridge crest in the form of a longitudinal septum. An associated external mold (Fig. 20.7) is another similar-sized, tubular object, but it also has ~8 longitudinal corrugations and regularly spaced commissural flanges. It may represent an external mold of a piece of the epitheca of Archaeichnium such that each longitudinal corrugation would house a projecting flange.

We also tentatively assign several specimens on the base of the slab from Kyffhauser (UCLA 7320; Fig. 21.3, 21.4) to Archaeichnium, although they may represent a completely different organism. However, they are conical, taper to a pointed and apparently attached end, are longitudinally ridged, and are comparable to Archaeichnium in size but not in length. Apart from their length/width ratios, they are similar to other specimens of Archaeichnium (e.g., Fig. 21.5). Whether the Kyffhauser specimens are assigned to Archaeichnium makes little difference to the biological interpretation of the fossil. In that context and starting from first principles, Archaeichnium appears to have had radial symmetry and a stiff but flexible body wall and probably lived attached to the substrate. Some of the Kyffhauser specimens somewhat resemble the Cambrian demosponge Takakkawia Walcott, 1920 (Rigby, 1986; Botting 2012), but the lengths of the longer ones and the flexibility of the walls effectively rule out a poriferan affinity. Perhaps a more plausible possibility is some connection with those “Precambrian macroorganisms” (Protechiuridae; Ivantsov and Fedonkin, 2002; Ivantsov et al., 2019) that possess unmineralized conical thecae. Although most are vastly different (e.g., Protechiurus edmondsi Glaessner, 1979a), there are intriguing similarities to some (e.g., Vendoconularia triradiata Ivantsov and Fedonkin, 2002) in the pleating of the walls, the hint of duodeciradial symmetry, and the possibility of longitudinal flanges. Ivantsov et al. (2019) pointed out some similarities of the protechurids to conulariids and anabaritids and suggested that all three groups might be basal scyphozoans, so a cnidarian affinity for Archaeichnium is one potentially viable possibility. The reconstruction of the putative Ediacaran anthozoan Auroralumina attenboroughii (Dunn et al., 2022) is also similar in some respects to Archaeichnium, most notably in its longitudinally pleated cup and stem. A third, and perhaps even better, cnidarian comparison is with the Cambrian Stage 4 “tubicolous enigma” Gangtoucunia aspera Luo and Hu in Luo et al., 1999, which is thought to be a sessile, tube-dwelling stem or early crown medusozoan (G. Zhang et al., 2022). Some specimens of Gangtoucunia aspera have 16–19 mesenteric septa that appear to extend along the length of the body. The tube is phosphatic and densely annulated like some Schwarzrand Subgroup tubes (Fig. 23), so it is possible that our speculative association of a radially fluted tube (Fig. 20.7) with the longitudinally ridged and possibly septate body of Archaeichnium may prove to be correct.

Another, and more likely, possibility is a relationship to the Burgess Shale and Chenjiang mackenziids Mackenzia Walcott, 1911 (Conway Morris, 1993) and Paramackenzia Zhao et al. 2021. Each has an elongate, sausage-shaped body with 10–20 longitudinal elements that are thought to be radial septa, either of the cnidarian type (Conway Morris, 1993) or the kind found between the modules of Ernietta (Zhao et al., 2021). However, in Paramackenzia, each septum houses shallowly inclined tubular structures that are ~1–2 mm long, ~1 mm apart, and ~0.25 mm in diameter, which are comparable in organization and dimensions to the ladder-like feature of Archaeichnium (Fig. 20.6). The tubular structures are thought to represent pore canals that were used to pump water into the body cavity of an Ernietta-like organism (Zhao et al., 2021), and some are filled with sediment, giving them the kind of topographic relief seen in the ladder-like feature of Archaeichnium. Although Zhao et al. (2021) advanced a strong case for Ernietta-like modular construction of Paramackenzia, the presence of an inner body wall is still debatable. Conway Morris's (1993, p. 610) description of the body wall of Mackenzia is closer to our concept of Archaeichnium (see Fig. 21.6): “It is conjectured that in life the circumference of the body was not simple but thrown into relatively deep folds and intervening ridges, the expression of which is now seen in the elevated lines and displaced margins. Further support for this comes from the distal end of some specimens which have a lobate appearance.” Thus, it is possible that even the mackenziids are total group cnidarians, although the inferred pore–canal system of Paramackenzia has no counterpart in the Cnidaria. Nevertheless, for all of these reasons, we tentatively refer Archaeichnium to the Cnidaria while acknowledging that new discoveries are needed to further explore that possibility.

Tubular fossils

Remarks

It is well known that the terminal part of the Ediacaran is replete with tubular fossils preserved in different ways: organic films, which may or may not be phosphatic or phosphatized; composite molds in siliciclastic sediments; calcareous skeletons, frequently originally aragonitic and recrystallized; and siliceous replicas of originally calcareous skeletons. Although the nature of the inhabitants of most of these tubes remains uncertain, the time immediately before the “Cambrian explosion” has become known as “Wormworld” (Schiffbauer et al., 2016; Darroch et al., 2018; Chai et al., 2021) or—less restrictively—as “tube world” (Budd and Jackson, 2015). The Nama biota is characteristic in that the dominant fossils through much of the succession are vendotaeniids (Cohen et al., 2009), mineralized tubes of Cloudina (Grant, 1990; Yang et al., 2022), composite molds of the bodies of Archaeichnium, and a variety of smooth or annulated tubes typically filled or cast in positive or negative hyporelief by sandy event beds (Figs. 20–23).

Order Sabelliditida Sokolov, 1965

Remarks

Similarities in the collar-in-collar construction of the tubes of Saarina hagadorni (Selly et al., 2020) and Cloudina hartmannae Germs, 1972 (Germs, 1972b) (Yang et al., 2020, 2022) have given rise to the term “cloudinomorph” as a group name. If these similarities are thought to be due to relatedness rather than convergent similarity, then the family and group names Saarinidae, Sabelliditidae, and Sabelliditida (Sokolov, 1965) should take precedence over Cloudinidae Hahn and Pflug, 1985 (Hahn and Pflug, 1985b) and cloudinomorph.

Siboglinid annelid worms abound in Russian Arctic waters because of the widespread availability of methane from seafloor clathrates and cold seeps (Karaseva et al., 2022). Discovery of this biodiversity (Ivanov, 1954, 1963) led to temporary acceptance of the phylum Pogonophora for the gutless siboglinids (Pleijel et al., 2009) and presumably to Sokolov's (1965, 1967) hypothesis that his Ediacaran genera Calyptrina, Paleolina, and Saarina—as well as Sabellidites Yanishevsky, 1926—are Precambrian examples of the phylum. A recent in-depth study of the tubes of Sabellidites has supported Sokolov's hypothesis (Moczydłowska et al., 2014), but the discovery by Schiffbauer et al. (2020) of a one-way gut in a cloudinomorph—which may be either Saarina or the related genus Costatubus (Selly et al., 2020)—would reinforce molecular evidence that the Siboglinidae are a significantly younger, highly derived clade of the Annelida (Hilário et al., 2011; Vrijenhoek, 2013; Georgieva et al., 2019, 2021; Capa and Hutchings, 2021). However, Eoalvinellodes annulatus (Little et al., 1999) is a pyritized annulated worm tube from a Silurian fossil hydrothermal vent site in Russia (Georgieva et al., 2019), which may imply that it had a chemosymbiotic lifestyle. Some other tube-dwelling polychaetes that inhabit vents and seeps obtain nutrients from bacterial symbionts in their respiratory crowns (Goffredi et al., 2020), a less-derived mode of chemosymbiosis that may have been in operation lower in the annelid tree. Thus, a non-siboglinid annelid affinity for the sabelliditid tubes remains a prime possibility and is compatible with the existence of a non-siboglinid, tube-dwelling polychaete (Dannychaeta) in a Cambrian Stage 3 fauna in China (Chen et al., 2020).

Landing et al. (2021) have also argued for extending the stratigraphic range of siboglinid and sabellid polychaetes into the Ediacaran on the basis of their reinterpretation of the early Cambrian stem gastropod Pelagiella exigua (Resser and Howell, 1938), which preserves two fan-shaped arrays of chitinous chaetae (Thomas et al., 2020). Their new sabellid genus Pseudopelagiella is based on P. exigua but is considered characteristic of species such as Pelagiella subangulata (Tate, 1892), which have triangular apertures (e.g., Mghazli et al., 2023) and an inner shell layer made from foliated aragonite (Runnegar in Bengtson et al., 1990, fig. 169B). The presence of an identical microstructure in Aldanella attleborensis (Shaler and Foerste, 1888; Qiang et al., 2023), which Landing et al. (2021) accept as a stem gastropod, and in the stem lineage bivalves Fordilla troyensis (Barrande, 1881) and Pojetaia runnegari (Jell, 1980; Runnegar and Pojeta, 1992; Vendrasco et al., 2011), makes the probability that species referred to “Pseudopelagiella” are annelids rather than mollusks vanishingly small.

Bobrovskiy et al. (2022) have recently redescribed one of Sokolov's sabelliditid species, Calyptrina striata Sokolov, 1967, and have extracted biomarker molecules from an organically preserved specimen of its tube. The proportion of cholestane, the diagenetically derived end product of cholesterol, was ~9% in Calyptrina, a little less than the background average for the Lyamtsa locality (~11%; Bobrovskiy et al., 2020) and thus very different from the much higher average amount (~50%) found in large specimens of Dickinsonia from the same site (Bobrovskiy et al., 2018). Furthermore, the ratio of two cholestane isomers (5ß/5α) was extraordinarily high (~4) in Dickinsonia but similar to that expected from diagenesis (~0.65) in Calyptrina. Conversely, other lighter and heavier steranes from Calyptrina have unusually low 5ß/5α ratios (~0.2) compared with Dickinsonia, where the average value (~0.7) is indistinguishable from the diagenetic expectation. Bobrovskiy et al. (2022) used these and other data to conclude that almost all of the steranes in Dickinsonia were derived from cholesterol in body tissue that had been decomposed by anaerobic bacteria rather than from dietary cholesterol in a one-way gut, the usual source of 5ß-cholestane in younger paleobiological, archaeological, and forensic contexts (Runnegar, 2022). Calyptrina, they suggested, had lost its tissue cholesterol by not being decomposed by anaerobes, and the low 5ß/5α ratios of its other steranes was due to the processing of dietary sterols derived from algal food sources by aerobic bacteria. This complex argument depends on many questionable assumptions, including a comparison with Kimberella (Glaessner and Wade, 1966), which they assumed to be a bilaterian with an alimentary canal. If that assumption is incorrect, then the case for a gut in Calyptrina, based on biomarkers, is even weaker. Given that 88% of the total steranes detected in Calyptrina come from green algae and that the clay underlying Calyptrina is similar to bulk rock extracts from the White Sea area (Bobrovskiy et al., 2020, table S1, 2022, table S1), perhaps the simplest explanation is that the part of the tube analyzed was not inhabited at the time of fossilization.

Family Saarinidae Sokolov, 1965
Calyptrina Sokolov, 1965

Type species

Calyptrina partita Sokolov, 1965 by original designation.

Other species

Calyptrina striata Sokolov, 1967 from the Syuzma beds of the Ust-Pinega Formation (>552.85 ± 0.77 Ma) in a borehole at Obozerskaya (40.31°E, 63.45°N), ~200 km south of Arkhangelsk, Russia (Sokolov, 1967; Stankovskiy et al., 1983; Xiao et al., 2002; Ivantsov et al., 2019; Bobrovskiy et al., 2022).

Coarsely and regularly annulated tubes, cf. Calyptrina striata Sokolov, 1967
Figure 22.2–22.6

Description

Sand-filled tubes, ~5 mm in diameter and up to 10+ cm long, that were probably originally circular in cross section, now slightly flattened, are in places ornamented with regularly spaced well-separated circumferential ridges, which presumably strengthened the tube wall.

Material

Eleven sandstone gutter casts, each containing several to many individual tubes, from Kyffhauser (UCLA 7320) and single specimens, possibly of the same form, on bed bases from Arimas (UCLA 7309) and the Holoog River (UCLA 7325).

Remarks

It is not known whether the corrugated parts of these tubes represent a different stage of growth from the smooth and seemingly thicker parts or are due to differences in preservation. However, there is little doubt that these corrugated tubes are neither body fossils nor trace fossils but rather the secreted dwelling structures of a worm-shaped animal. Rare specimens from the Holoog River (Fig. 22.5) and Arimas (Fig. 22.6) are tentatively included in this form taxon.

Figure 22. Coarsely and regularly annulated tubes, cf. Calyptrina striata Sokolov, 1967 (2–6), smooth tubes (1, 7), and two important specimens of Archaeichnium haughtoni Glaessner, 1963 (8, 9) from the Neiderhagen Member, Nudaus Formation, UCLA 7320, Kyffhauser farm (1–4), the Huns Member, Urusis Formation, UCLA 7325, Holoog River (5, 7), and UCLA 7309, Arimas farm (5, 8, 9). (1) Bed base with sandstone casts of numerous small, short, conical tubes plus one wider, coarsely annulated, kinked tube (arrow), GSN F 1941. (2) Base of gutter with sandstone cast of one coarsely annulated tube, GSN F 1944. (3) Top of tube-filled gutter cast, found by D.H. Erwin in 1995, with one annulated tube indicated by the arrow, GSN F 1943. (4) Top, end, and base of small section of a gutter cast with one enclosed coarsely annulated tube indicated by the arrow, GSN F 1945. (5) Cast of irregular annulated tube on bed base, Holoog River, GSN F 1973. (6) A somewhat similar structure, Arimas, GSN F 1934. (7) Small sandstone slab with casts, many presumably current-aligned smooth tubes, GSN F 1976. (8) Recognizable specimen of Archaeichnium haughtoni that is on the same surface as the “Arimas lycopod” (Fig. 14.5), thus demonstrating co-occurrence of these two taxa, GSN F 1910. (9) Quartz filling of Archaeichnium haughtoni that gives some information about its cross-sectional shape before burial and compaction, GSN F 1919. (1, 3, 4, 7) Scale bars = 2 cm; (2, 5, 6, 8, 9) scale bars = 1 cm.

A variety of sparsely annulated tubular fossils have been assigned to Calyptrina striata, which was based on a single compressed pyritized specimen (Sokolov, 1967; Bobrovskiy et al., 2022, fig. S4H, I). The redescription of the species from numerous White Sea examples preserved in different ways (Bobrovskiy et al., 2022) revealed that the apertural part of the tube had regularly spaced wall thickenings that were robust enough to leave deep grooves in external molds and are clearly visible in mineralized compressions (Bobrovskiy et al., 2022, fig. S4D, H, I).

Between the thickened annulations, the wall has fine longitudinal costae that would not be visible in our material because of the grain size of the sandstone gutter casts. Bobrovsky et al. (2022) also provided excellent evidence that the apertural end of the tube projected above the seafloor during life and that the longer, buried portion of the tube ran horizontally and changed character gradationally along its length to become finely and regularly annulated or finely and irregularly annulated, features we found in tubular fossils from other localities (Fig. 23.1–23.3, 23.6, 23.9). However, as there is no direct evidence for a biological connection between these variously ornamented tubes, we describe the finely annulated ones separately. Annulated structures from Ediacara identified as the meniscuate trace fossil Taenidium cf. T. serpentinum (Heer, 1877) by Jenkins (1995) are superficially similar to C. striata but are consistently short and banana shaped (Reid et al., 2017) and probably not circular in cross section.

Figure 23. Various annulated tubes, cf. Sinotubulites baimatuoensis Chen, Chen, and Qian, 1981 (1–3, 6, 9) and cf. Sekwitubulus annulatus Carbone et al. (4, 5, 7, 8), Feldschuhhorn Member, Urusis Formation, UCLA 7377, Swartkloofberg farm (1, 2), Urikos Member, Zaris Formation, UCLA 7383 and UCLA 7384C, Zaris farm and Zaris Pass (3, 5, 6), and Huns Member, Urusis Formation, UCLA 7309, Arimas farm (4, 7, 8) and UCLA 7325, Holoog River (9). (1, 2) Two views of a kinked tube on a presumed lower fine-grained carbonate bed surface, GSN F 1935. (3) Two finely annulated tubes on the lower surface of a carbonate slab that has OES texture, GSN F 1982. (4) Small piece of crisply annulated tube, GSN F 1918. (5) Another crisply annulated tube, GSN F 1984. (6) Bed base casts of finely annulated tubes (seen in positive relief in insert), GSN F 1966. (7, 8) Narrow annulated tube, seen in bed base context in (8), GSN F 1953. (9) Section of finely annulated tube, bed base, GSN F 1971. (1, 3, 5, 6 insert, 7, 9) Scale bars = 1 cm; (2, 4) scale bars = 5 mm; (6) scale bar = 2 cm; (8) scale bar = 5 cm.

Family uncertain
Genus Sinotubulites Chen, Chen, and Qian, 1981

Type species

Sinotubulites baimatuoensis Chen, Chen, and Qian, 1981, from the Shibantan Member of the Dengying Formation, Shibantan of Yichang City, China, by original designation.

Remarks

We tentatively refer some regularly annulated tubular fossils to this species, which is based on rather poorly preserved type material and has been identified from Ediacaran deposits in many parts of the world (Yang et al., 2022). As currently diagnosed, the species is a form taxon that is sufficiently broadly defined to accommodate the Namibian material for the time being. There are also some similarities to Wutubus annularis (Chen et al., 2014) as discussed in the following.

Finely annulated tubes, cf. Sinotubulites baimatuoensis Chen, Chen, and Qian, 1981
Figure 23.1–23.3, 23.6, 23.9

2021

Tubular and annulated body fossils, Darroch et al., fig. 4a, f.

Description

External molds preserved in calcareous siltstone and sandstone of a an originally cylindrical finely annulated tube, having either dispersed irregular narrow ridges (Fig. 23.1, 23.2) or tightly packed narrow annulations (Fig. 23.3, 23.6, 23.9); length of tube at least 140 mm, diameter 2–6 mm; longest known tube appears to taper from ~3.5 to ~2 mm; severely kinked specimen (Fig. 23.1) demonstrates that the tube wall was unmineralized and flexible.

Material

One large, kinked specimen from Swartkloofberg (UCLA 7377), three slabs with five specimens from Zaris (UCLA 7383), five slabs with one or more specimens from Zaris Pass (UCLA 7384C), and eight slabs from the Holoog River (UCLA 7325), only some of which may be tentatively included in this category.

Remarks

There are similarities to some specimens of Calyptrina striata (e.g., Bobrovskiy et al., 2022, figs S2E, S4C), but in the absence of examples with widely spaced coarse rugae, membership of that species is unlikely. Sinotubulites baimatuoensis has fewer distinctive features, being simply an irregularly to regularly annulated tube, but we tentatively refer these Namibian tubes to that form species. Wutubus annularis (Chen et al., 2014) is also similar, but some specimens taper rapidly to a closed apex that is presumed to be the site of attachment to the substrate. Our material tapers far more slowly (Fig. 23.6), and there is no evidence for a closed end. Annulatubus flexuosus (Carbone et al., 2015) is more regularly annulated and fits better with the Sekwitubus–Corumbella–Shaanxilithes morphotype according to the analysis by Dunn et al. (2022).

Regularly annulated tubes, cf. Sekwitubulus annulatus Carbone et al., 2015
Figure 23.4, 23.5, 23.7, 23.8

1972c

Taenidium sp., Germs, pl. 2, fig. 2.

2016

Shaanxilithes; Darroch et al., fig. 6d.

2021

Tubular and annulated body fossils, Darroch et al., fig. 4b, c, d.

2022

Tubular body fossils, Turk et al., fig. 11.2–11.4.

Description

Cylindrical tubes, 2–3 mm in diameter, ornamented with regularly spaced angular ridges that are not obviously collar shaped, are not tapered, and do not appear to have been mineralized.

Material

About half a dozen short pieces of tubular fossils found as external molds on the bases of slabs from Arimas (UCLA 7309), the Holoog River (UCLA 7325), and Zaris (UCLA 7383).

Remarks

There are many similarly ornamented tubular structures in the Ediacaran, and the morphology persists to the present, as exemplified by the living terebellid annelid Glyphanostomum pallescens (Georgieva et al., 2019). Sekwitubulus annulatus is a comparable, incompletely known regularly annulated tubular fossil from the Blueflower Formation, northwest Canada, described by Carbone et al. (2015), who compare it with previously described Ediacaran genera.

Other body fossils and body traces
Figure 24.1, 24.3, 24.4, 24.6–24.9

Remarks

We illustrate but do not describe specimens of Aspidella sp. and Beltanelliformis brunsae Menner in Keller et al., 1974 from Aar (UCLA 7307) and Palaeopascichnus sp. from the Namaland (UCLA 7315) for the sake of completeness. One locality has yielded three specimens of Pseudorhizostomites (Sprigg, 1949; Fig. 24.6), best interpreted as the removal trace of a frond (Tarhan et al., 2015). We also show two examples of scratch circles (Osgood, 1970; Jensen et al., 2018), one of which is on a bed base and has a conical plug at its center (Fig. 24.7) that resembles structures described (Darroch et al., 2021) as the conical burrows Conichnus (Männil, 1966) and Bergaueria (Prantl, 1946) and presumably was the entrance to the home of the producer. Other blister-like structures on bed bases (Fig. 24.5) are best interpreted as incipient syneresis cracks.

Figure 24. Miscellaneous body fossils and trace fossils. (1) Aspidella sp., Aarhauser sub-member, Kliphoek Member, Dabis Formation, UCLA 7307, Aar farm, GSN F 1894. (2) Aspidella terranovica Billings, 1872, Fermeuse Formation, St. John's Group, Ferryland, Avalon Peninsula, Newfoundland, UCLA 7335.1, for comparison with (1). (3) Beltanelliformis brunsae Menner in Keller et al., 1974, characteristic closely packed aggregate, top of Kliphoek Member, Dabis Formation, UCLA 7311, Kliphoek farm. (4) Palaeopascichnus sp., base of very thin sandstone, Buchholzbrunn Member, Dabis Formation, UCLA 7315, Namaland district, near Bethanien, GSN F 1892. (5) Lens-shaped blisters, possibly sandstone casts of syneresis cracks, base of bed, same locality as (4). (6, 9) radially grooved disks reminiscent of Pseudorhizostomites Sprigg, 1949, bed base and counterpart cast of bed base, same locality as (4), GSN F 1893 and GSN F 1895, respectively. (7) Sandstone cast of scratch circle and funnel-shaped hole made by rotating tethered object, Huns Member, Urusis Formation, UCLA 7309, Arimas farm, GSN F 1912. (8) Concentric scratch circles on ripple-marked bed top, same locality as (7), GSN F 1917. (10) Archaeonassa isp., positive relief, Urusis Formation, UCLA 7325, Holoog River, GSN F 1979. (11) Gordia isp. in positive hyporelief, thin sandstone bed, Urikos? Member, Zaris Formation, UCLA 7384A, Zaris Pass, GSN F 1970. (12) Helminthopsis isp., sinuous channel, either a trace or a body fossil, on a rippled bed top, same locality as (7), GSN F 1915. (13, 14) Gordia isp., two small slabs from the same bed with possibly the oldest known trace fossils from the Nama Group, Kliphoek Member, Dabis Formation, UCLA 7378, Twyfel farm, GSN F 1920 and GSN F 1921, respectively; the trace fossils occur with syneresis cracks, e.g., left of center in (13). (1–9, 11, 12) Scale bars = 2 cm; (10, 13, 14) scale bars = 1 cm.

Ichnofossils

Remarks

The Ediacaran ichnofossil record of Namibia has recently been reviewed by Darroch et al. (2021) and Turk et al. (2022). Our contribution is focused on bilaterian traces from the Kliphoek Member (Fig. 24.11–24.13) and evidence for a substantial infauna of small animals during Schwarzrand time (Figs 25.5, 26.1–26.8). We also discuss the evidence for the first occurrence of treptichnids in the Nama succession and briefly review ichnological arguments for and against treating the upper part of the Spitskop Member as earliest Cambrian.

Figure 25. Ediacaran and Cambrian trace fossils. (1) Treptichnus isp? and Helminthopsis isp., GSN F 1937, base of thin sandstone slab with continuous and intermittent traces preserved in convex hyporelief, both possibly made by the same organism, and comparable to the traces shown in (2), ~160 m above base of Huns Member, Urusis Formation, UCLA 7371, Arimas farm. (2) Treptichnus isp. and numerous microburrows, Ariichnus vagus n. igen. n. isp., all preserved in convex hyporelief, ISAM K4366, Huns Member, Urusis Formation, Arimas farm, found by G.J.B. Germs before 1972, photographed in Cape Town, South Africa, in 1993. (3, 4) Treptichnus pedum (Seilacher, 1955), lower bed surfaces, Nomtsas Formation, UCLA 7324, Sonntagsbrunn farm, GSN F 1951 and GSN F 1952, respectively. (5) Subhorizontal burrows excavated and filled by sediment filling a gutter as evidenced by breaks in the continuity of the borrows (arrow), GSN F 1923, found by A.J. Kaufman in 1995, Nasep Member, Urusis Formation, UCLA 7322, Swartkloofberg farm. (6) Gordia isp., looping traces on the top surface of a rippled slab, GSN F 1925, Huns Member, Urusis Formation, UCLA 7326, Arimas Farm. (1–4) Scale bars = 1 cm; (5, 6) scale bars = 2 cm.

Figure 26. Gutter casts and microburrows of Ariichnus vagus n. isp., Buchholzbrunn Member, Dabis Formation, UCLA 7314, Namaland district (3) and Huns Member, Urusis Formation, UCLA 7326, Arimas farm (1, 2, 4–8). (1, 2) GSN F 1911, sandstone cast of large gutter, viewed from side and bottom, with microburrow traces on the shallower parts of the cast. (3) Cross section of a sandstone-filled gutter cast, embedded in a thin sandstone event bed, and comparable to samples found as float elsewhere. (4) GSN F 1929, flat base of an event bed that cast erosional intersections with many microburrows. (5) Upper and lower surfaces of a channel cast topped by hummocky stratification (rectangle shows location of the holotype; arrow indicates ripple crest), GSN F 1931; lower surface enlarged in (6). (6) GSN F 1931, enlargement of lower surface of channel (5, 6) showing numerous casts of microburrows; insert is an enlargement of the holotype, which is on another part of the same surface (5). (7, 8) GSN F 1927, oblique and cross-sectional views of a well-formed channel that has cast microburrows above the level of the white arrows and below the level of black arrows, a stratigraphic interval of ~3 cm. (1, 2, 5, 7, 8) scale bars = 5 cm; (3) camera lens cap = 60 mm; (4, 6) scale bars = 1 cm; (insert in 6) scale bar = 1 mm.

Ichnogenus Archaeonassa Fenton and Fenton, 1937

Type ichnospecies

Archaeonassa fossulata Fenton and Fenton, 1937 from the Cambrian Series 3, Stage 5, Mt. Whyte Formation, Alberta, Canada (Fenton and Fenton, 1937; Yochelson and Fedonkin, 1997), by monotypy.

Archaeonassa isp.
Figure 24.10

1982

?Nereites sp., Crimes and Germs, p. 900, pl. 2, fig. 8.

2021

Archaeonassa Fenton and Fenton; Darroch et al., fig. 9 g.

2022

Archaeonassa; Turk et al., p. 6, fig. 7.1.

Remarks

A single specimen, GSN F 1979, from the base of the Huns Member at Holoog River (UCLA 7325) shows the characteristic U-shaped end of this groove and ridged trace; better examples were illustrated by Turk et al. (2022, fig. 7.1) from their Canyon Roadhouse site. Archaeonassa Fenton and Fenton, 1937 ranges from the late Ediacaran to the present (Jensen, 2003; Uchman and Martyshyn, 2020).

Ichnogenus Ariichnus new ichnogenus

Type ichnospecies

Ariichnus vagus n. igen. n. isp. from the Huns Limestone Member, Urusis Formation, Arimas farm, Southern Namibia.

Diagnosis

As for the type species by montypy.

Etymology

Contraction of Arimas (farm) and ichnos, Greek for footprint or track.

Ariichnus vagus new ichnospecies
Figures 25.2, 26.1, 26.2, 26.4–26.8

1972a

thin, straight, or curved thread-like trails, Germs, p. 208, pl. 26, fig. 5.

1972b

very thin, straight, or curved thread-like trails, Germs, p. 866, pl. 1, fig. 5.

2000

smaller ?trace fossils, Jensen et al., fig. 2.

2021

sub-millimeter scale burrows, Darroch et al., p. 15, fig. 9b, c.

2022

meiofaunal traces, Turk et al., p. 11, fig. 8.

Holotype

Burrow shown in insert of Figure 26.6, on slab GSN F 1931, from the Huns Limestone Member, Urusis Formation, Arimas farm, Southern Namibia.

Diagnosis

Narrow, subhorizontal, dichotomously branching burrows that follow irregular paths and occasionally cross over each other.

Description

Narrow, subhorizontal cylindrical burrows, ~0.3 mm in diameter, exhibiting occasional Y-shaped junctions that multiply the total number of terminations away from the burrow entrance; in the holotype, adjacent branches are irregular in the forward direction and cross each other without intersecting (see also Darroch et al., 2021, fig. 9c, right arrow; Turk et al., 2022, fig. 8.3); numerous circular cross sections presumably represent vertical segments that connect the subhorizontal tunnels into a complex three-dimensional network.

Etymology

Latin, vagus (roving, wandering), in reference to the paths of the primary branches of the burrows.

Material

Six sandstone gutter casts, GSN F 1924–1931, from the same locality, each with numerous examples of microburrows intersected and cast by the sandstone channel fills plus the specimen collected by Germs at Arimas.

Remarks

The three-dimensional geometry of these microburrow systems is difficult to reconstruct from the curved two-dimensional surfaces on which they are observed. It does, however, appear that they formed systems with true branching and are not merely the result of coincidental interference. The surface expression resembles planar sections of Chondrites burrow systems seen in thin and polished sections and core slices (Ekdale and Bromley, 1982; Bromley and Ekdale, 1984; Baucon et al., 2020). The diameter of the burrows is smaller than that of most ichnospecies of Chondrites but overlaps with Chondrites intricatus (Brongniart, 1828), in which the strings are smaller than 1 mm (Fu, 1991); Ekdale and Bromley (1982) and Uchman (1999) also recorded occurrences with a burrow diameter as narrow as 0.2–0.3 mm. The wandering pathways their primary branches seem to follow (Fig. 26.6; Darroch et al., 2021, fig. 9c; Turk et al., 2022, fig. 8.3) distinguish Ariichnus vagus from all previously described ichnospecies of Chondrites.

Reconstruction of the three-dimensional geometry is problematic because of the small size and style of preservation, but it appears to be less complex than is typical for Chondrites, so attribution of the new ichnospecies to Chondrites was not desirable. It should be noted that this would have been the first Ediacaran record of Chondrites, earlier reports of this age having been rejected (Jensen and Runnegar, 2005; L. Zhang et al., 2022). The ichnogenus is rare also in the Cambrian, without a single accepted Terreneuvian or Series 2 occurrence (Mángano and Buatois, 2014; but see Baucon et al., 2022 for possible exception). Chondrites from the Teltawongee Group of New South Wales, Australia (Webby, 1984), sometimes cited as Terreneuvian (e.g., L. Zhang et al., 2022), is in strata of poorly constrained age. Webby (1984) considered the occurrence to be no older than early or middle Cambrian, on the basis of the trace fossils, and they remain the main criteria for the maximum depositional age of the group, with minimum depositional ages obtained from cross-cutting volcanics dated at 505 and 515 Ma (Johnson et al., 2016).

An alternative ichnogeneric assignment of the Nama material could have been to Pilichnus, an ichnogenus that Uchman (1999) erected as part of the Chondrites group of branched structures. The Cretaceous type material has straight to winding strings 0.15–0.35 mm wide, with dichotomous branches. Subsequent reports have extended this ichnogenus to the Terreneuvian (e.g., Buatois and Mángano, 2012). The latest Ediacaran Pilichnus from the Tamengo Formation, Brasil (Adôrno, 2019) has larger dimensions and rare branching. A possible difference of the Nama material from Pilichnus is that the orientation of Pilichnus is mainly along a horizontal plane, although comparison of this ichnogenus with modern traces has been made with more vertically oriented traces (Hertweck et al., 2007).

At the type locality, the microburrows are preserved on the sides and bases of sandstone gutter casts such as the one from the Buchholzbrunn Member seen in outcrop (Fig. 26.3). The Arimas gutter casts were float samples, but the level from which they came is well constrained (Fig. 4; Turk et al., 2022). The microburrows are confined to certain parts of the channel walls and are not found below or above the presumed bioturbated interval that is ~3 cm deep and lies 1–2 cm below the surface (Fig. 26.7, 26.8). This distribution eliminates all four of the hypotheses proposed by Turk et al. (2022) for their formation: opportunistic colonization of exposed sediment in gutters, selective preservation in gutters of a ubiquitous infauna, lithologic contrast between gutter-filling sandstone and guttered substrate, and reborrowing of the channel surface by small animals carried with the eroding fluid. However, they also said that “small bilaterian traces … might be more widespread in these intervals than is currently recognized,” as we suggest. As the tops of the channel sands are rippled (Fig. 26.5), the original height of the seafloor was probably lower than the tops of the channel-filling sand (Fig. 26.1, 26.3, 26.5, 26.8). Nevertheless, it is clear that the microburrowed interval began at least 1 cm below the sediment–water interface and continued downward for ~3 cm. This suggests that the tracemakers were exploiting a particular part of the redox gradient and, like Chondrites, may have been adapted to dysaerobic conditions (Bromley and Ekdale, 1984; Savrda and Bottjer, 1987).

Parry et al. (2017) described a dense occurrence of “meiofaunal ichnofossils” from the terminal Ediacaran of Brazil and identified them as Multina minima Uchman (2001), an Eocene species of Multina Orlowski in Orlowski and Zylinska, 1996 from the late Cambrian of the Holy Cross Mountains, Poland (Orlowski and Zylinska, 1996). However, both of these ichnospecies of Multina are meshes, not networks, and even M. minima has a much larger tunnel diameter than the Brazilian meiofaunal burrows. There is a better size comparison with the Namibian structures, but although the Brazilian traces are described as having “rare dichotomous branches” (Parry et al., 2017, p. 1456; Adôrno, 2019), the tomographic reconstructions show little evidence for branching. On the basis of the small diameter of their narrowest burrows, Parry et al. (2017) were able to exclude most bilaterian phyla as possible tracemakers and opted for a nematode-like worm that lacked the ability to move by peristalsis. We think that is unlikely for the Namibian microburrows because of the irregularity of the trajectories of the tunnels. Nematodes move by bending their bodies in a sinusoidal fashion and either make sinusoidal traces (Balinski and Sun, 2015) or generate “meioturbation” rather than well-defined burrows (Schieber and Wilson, 2021). Thus, it seems more likely that the Namibian microburrows were produced by small animals using hydrostatic processes. The significance of a sizeable bioturbated zone, well below the sediment–water interface and apparently decoupled from the widely assumed Ediacaran microbial mat communities, is explored under Discussion. Another kind of subterranean trace fossil community is indicated by Planolites-like burrows intersected by a gutter cast in the top of the Nasep Member on Swartkloofberg (UCLA 7322; Fig. 25.5). Thus, it seems that by the close of the Ediacaran, some bioturbation had moved well below the level of microbial mats. Although probably morphologically distinct, the Nama microburrows compare to, and predate, Fortunian material of Olenichnus irregularis Fedonkin in Sokolov and Iwanowskii (1985) from Siberia that Marusin and Kuper (2020) interpreted as complex three-dimensional endobenthic tunnel systems made by bilaterians.

Trace fossils are rarely preserved in pot and gutter casts and, if present, are thought to be post-depositional (Myrow, 1992; Jensen, 1997; Mángano et al., 2002). However, the Planolites-like burrows from the Nasep Member are interrupted by the gutter cast wall (arrow, Fig. 25.5), and the microburrows from Arimas were clearly exposed by the erosive action that created the gutters. By contrast, the body fossils, which are occasionally preserved in gutter casts or on the bases of channels (Figs. 12.3–12.5, 16, 21.3, 21.4, 22.4), are thought to have been transported by the eroding events and are, like the tool marks, synchronous with them.

Ichnogenus Gordia Emmons, 1844

Type ichnospecies

Gordia marina Emmons (1844) from an Ordovician “fine flagging stone” (calcareous turbidite) of the Giddings Brook slice, Taconic Allochthon (Landing, 2012), at “Mr. M‘Arthur's quarry,” Jackson, New York, by monotypy.

Gordia ispp.
Figures 24.11, 24.13, 24.14, 25.6

1972a

worm tracks, Germs, pl. 27, fig. 4.

1972b

bundled and individual cylindrical tubes, Germs, pl. 2, fig. 5.

2021

Gordia, Darroch et al., p. 8, fig. 7g.

2022

Gordia, Turk et al., p. 8, fig. 7.6

Material

Three small slabs from Twyfel, GSN F 1920–1922, one specimen from Arimas, GSN F 1925, and other examples observed in the field.

Remarks

Gordia is one of four Ediacaran ichnogenera/behaviors identified in simple horizonal trails (Buatois and Mángano, 2016) and is characterized by common self-crossings. The examples we illustrate are typical, but those from the Kliphoek Member on Twyfel (Fig. 24.11, 24.13, 24.14) are older than previous records from Namibia and extend this kind of behavior downward from the Schwarzrand Subgroup (Darroch et al., 2021, fig. 18b) into the Kuibis Subgroup. There has been some difference of opinion about the stratigraphic position of the fossiliferous intervals on Twyfel, Weigkrup, Hansburg, and Zuurberg farms (Bouougri et al., 2011; Maloney et al., 2020), but we agree with Maloney et al. (2020) in identifying these horizons as either upper Kliphoek Member or Bucholzbrunn Member rather than Kanies Member. Thus, these specimens of Gordia isp. and some associated finer trails are the oldest known trace fossils from Namibia. A looped trace from the Huns Member on Arimas (Fig. 25. 6) may also be referred to Gordia isp., but it is clearly different in detail from the Twyfel examples.

Ichnogenus Helminthopsis Heer, 1877

Type ichnospecies

Helminthopsis magna Heer, 1877 by subsequent designation of Ulrich (1904, p. 144) or Helminthopsis abeli Książkiewicz, 1977 (Han and Pickerill, 1995) or Helminthopsis hieroglyphica Wetzel and Bromley, 1996.

Remarks

Wetzel and Bromley (1996) proposed Helminthopsis hieroglyphica to retain Heer's generic name, which was based on material that should be referred to other ichnogenera, so H. hieroglyphica Wetzel and Bromley, 1996 has gained acceptance as the replacement type species (e.g., Šamánek et al., 2022).

Helminthopsis isp.
Figures 24.12, 25.1

Material

Only two possible examples of many similar structures are illustrated. The long continuous trace in Figure 25.1 is best characterized as the form genus Helminthopsis but may, in fact, have been generated by the producer of Treptichnus isp? (Fig. 25.1, arrows).

Remarks

According to Buatois and Mángano (2016, p. 41) “Helminthopsis displays a tendency to meander,” so we refer simple cylindrical traces with this property to Helminthopsis. However, as noted by many others, it may be difficult to distinguish such ichnofossils from tubular body fossils. The specimen shown in Figure 25.1 is fairly clearly a trace fossil and may, in fact, be a variety of Treptichnus, which occurs next to it on the same slab. The one shown in Figure 24.12 is more ambiguous, and there are many more poorly preserved traces like this in the Schwarzrand Subgroup that could be either trace or body fossils.

Ichnogenus Streptichnus Jensen and Runnegar, 2005

Type ichnospecies

Streptichnus narbonnei Jensen and Runnegar, 2005 from UCLA 7375, uppermost Spitskop Member, Urusis Formation, Swartpunt farm, southern Namibia, by original designation and monotypy.

Streptichnus narbonnei Jensen and Runnegar, 2005

2005

Streptichnus narbonnei Jensen and Runnegar, figs. 2, 3.

2019

Streptichnus narbonnei, Linnemann et al., fig. 4b.

Holotype

One of two adjoining slabs, GSN F 626 (Jensen and Runnegar, 2005, fig. 2b), from the Spitskop Member, Urusis Formation, UCLA 7375, 13 m below the summit of Dundas Hill, Swartpunt farm, southwestern Namibia.

Remarks

The complexity of the Streptichnus burrow system is comparable to that of Treptichnus pedum, so Linnemann et al. (2019) advocated lowering the Ediacaran–Cambrian boundary to just below the level of Streptichnus in the Dundas Hill section on Swartpunt (Fig. 5). However, the subsequent discovery of Streptichnus in the Shibantan Lagerstätte of South China (Xiao et al., 2021; Mitchell et al., 2022) confirms that it is associated with typical Ediacaran taxa.

Ichnogenus Treptichnus Miller, 1889

Type ichnospecies

Treptichnus bifurcus Miller, 1889 from the Carboniferous, Gzhelian, Mansfield Formation of Indiana (Maples and Archer, 1987), by original designation and monotypy.

Treptichnus pedum (Seilacher, 1955)
Figure 25.3, 25.4

1972a

Phycodes pedum; Germs, pl. 27, figs. 7, 8.

1972b

Phycodes pedum; Germs, pl. 2, figs. 7, 8.

1982

?Neonereites biserialis Seilacher, Crimes and Germs, p. 897, pl. 2, fig. 7.

1982

Phycodes cf. P. pedum; Crimes and Germs, p. 901, pl. 2, fig. 9.

2012

Treptichnus pedum; Wilson et al., figs.10–15.

Holotype

Specimen figured by Seilacher (1955, p. 387, fig. 4a) from the early Cambrian Neobulus shale (Khussak Formation), Salt Range, Pakistan (Buatois, 2018).

Material

Three specimens from the Nomtsas Formation on Sonntagsbrunn, GSN F 1950–1952.

Remarks

See Wilson et al. (2012) for a full description of this species and its occurrence in Namibia.

Treptichnus isp.
Figure 25.1, 25.2

1972a

Discontinuous trails with three ridges, Germs, p. 208, pl. 26, figs. 5, 7, pl. 27, fig. 1.

1972b

Trails with three parallel ridges, Germs, pl. 1, figs. 5, 7, pl. 2, fig. 1.

2000

Treptichnus; Jensen et al., fig. 2A–E.

2016

treptichnids; Buatois and Mángano, fig. 4b, c.

2020

treptichnids; Mángano and Buatois, fig. 2a.

2021

burrow similar to Torrowangea Webby; Darroch et al., fig. 9a.

2021

treptichnid-type traces; Darroch et al., fig. 13a–e.

Materials

A single slab from Arimas, ISAM K4366, described originally by Germs (1972a, b) and another similar-sized slab collected by us (Fig. 25.2) from Arimas (UCLA 7371).

Remarks

Germs's discovery slab was not in place, so the stratigraphic level from which it was derived is uncertain. Germs (1972b, fig. 2) showed the horizon as being immediately beneath the base of the Huns Limestone Member as he then defined it, which would probably place it above the level of UCLA 7326 (Fig. 4). Jensen et al. (2000) placed Germs's specimen lower down in the Arimas section on the basis of field observations of similar traces by JGG in 1993, and Turk et al. (2022, fig. 4) indicated a comparable level, just below their “gutter cast horizon” (= UCLA 7326). Buatois and Mángano (2016, fig. 4) suggested that treptichnids occur even lower down, well below the first prominent limestone bed, in the vicinity of UCLA 7309 (Fig. 4). The only sample in our collection we can confidently place in the stratigraphic section is the one shown in Figure 25.1, which was removed from outcrop by SJ in 1996. The similar slab collected by Germs could easily have moved downslope from that level. Thus, UCLA 7371 (Fig. 4) is the oldest certain occurrence of Treptichnus in Namibia.

We refer these traces to Treptichnus because of the great range of preservational variants found in some examples of the type species, including structures that closely resemble Germs's “discontinuous trails with three ridges” (Getty et al., 2016, fig. 4.1, 4.2). In any case, Treptichnus is a form taxon, which could have been produced by different kinds of animals in the Ediacaran and the Cambrian, so our use of the generic name does not necessarily imply biological continuity across the eon boundary.

Results

Stratigraphy

Recently published U–Pb ages for the Witputs subbasin (Linnemann et al., 2019; Nelson et al., 2022) have allowed us to postulate a hiatus of ~2 million years at the base of the Schwarzrand Subgroup south of the Osis arch (Fig. 1), but there is little biostratigraphic evidence for this break (Fig. 2). Our carbon isotope data from a section at Mamba, just north of the Osis arch, confirms the correlation of the peak of the OMKYK positive excursion, first identified by Grotzinger et al. (1995), with the top of the Mooifontein Member in the heart of the Witputs subbasin (Fig. 3). On the basis of these correlations, we propose a revision of the sequence stratigraphic terminology for the Nama Group (Fig. 2). We also constrain and extend the stratigraphic and geographic ranges of the key Ediacaran taxa: Archaeichnium, Ernietta, Pteridinium, Swartpuntia, and Treptichnus (Fig. 2). A conservative estimate for the first appearance of the genus Treptichnus is at the top of the Huns Limestone Member, higher than previously thought (Fig. 4), but the diagnostic basal Cambrian species of Treptichnus, T. pedum, has not been found below the Nomtsas Formation. Suggestions to lower the eon boundary to beneath Streptichnus narbonnei (Linnemann et al., 2019) are not supported by the recent discovery of Streptichnus with characteristic Ediacaran taxa in the Shibantan Lagerstätte of South China (Xiao et al., 2021: Mitchell et al., 2022).

Taphonomy and paleoecology

All of the evidence presented here indicates that few if any of the Ediacaran soft-bodied organisms were preserved in life orientation in their original habitats. That includes specimens from the Aarhauser sandstone at Aar excavated by Seilacher and his team that formed the basis of the canoe hypothesis for Pteridinium (Ivantsov and Grazhdankin, 1997; Seilacher, 1997; Grazhdankin and Seilacher, 2002) as well as the beds with abundant Ernietta on Aar, Twyfel, Wegkruip, Hansburg, and Zuurberg farms that have been used to propose and model a totally infaunal or partly buried lifestyle for Ernietta (Crimes and Fedonkin, 1996; Meyer et al., 2014a, b; Elliott et al., 2016; Ivantsov et al., 2016; Gibson et al., 2019; Hall et al., 2020; Maloney et al., 2020). All of the unmineralized tubular fossils we have studied appear to have been transported or perhaps toppled (Fig. 21.4) by water motion, as are most specimens of Cloudina. The only taxa that are unquestionably in situ are the trace fossils, Palaeopascichnus, and frond holdfasts such as Aspidella and Pseudorhizostomites.

Tool-marked bed bases are a prominent feature of the Nama Group succession. Comb and rake structures previously attributed to the transport of spicular sponge skeletons (Darroch et al., 2021) are shown to be bump and drag marks of erniettomorphs, most probably vanes of Pteridinium. Microscopic trace fossils on gutter casts are referred to new ichnogenus and ichnospecies Ariichnus vagus. These microburrows were produced by tiny animals that seem to have inhabited a 2–3 cm thick dysaerobic zone that began ~1 cm below the sediment–water interface.

Taxonomy

The holotype of the type species of Ernietta, E. plateauensis, appears to be a small, deformed specimen of Pteridinium simplex, and we recommend that it be replaced by a neotype, the holotype of Ernietta sandalix, to retain the use of Pflug's iconic generic name. The discovery of a problematicum—the Arimas lycopod—at the type locality of Nasepia, and its resemblance to the axis of Swartpuntia, raises the possibility that Swartpuntia is either a junior synonym or close relative of Nasepia. New evidence suggests that Swartpuntia lacked a stem and holdfast, which puts it closer in body plan to Pteridinium than previously thought.

Paleobiology

Arimasia germsi n. gen. n. sp. from the Huns Limestone Member is described as a simple sponge, which may have resembled an unmineralized, one-walled archaeocyath. We suggest that Arimasia, the Archaeocyatha, and the unmineralized vauxiid sponges may all have been aspiculate stem members of the Demospongiae. This hypothesis requires the independent origin of siliceous spicules in the Hexactinellida and the Demospongiae (Aguilar-Camacho et al., 2019).

Juvenile specimens of Ernietta from Buchholzbrunn (Fig. 16) show that growth proceeded from a stage with four or fewer tubular modules to an observed maximum of ~70 modules in the largest specimens (Fig. 18.6). The evidence for more than one layer of modules in the body wall is limited, so Ernietta is considered to be an epifaunal, bag-shaped organism formed of a single layer of tubular modules that were generated at the outer wall and coalesced in a proximal to distal direction during growth. New modules may have arisen at the ends of the zig-zag basal seam and/or by intercalation. Growth interruptions, which are obvious on many internal molds, are attributed to zones of damage and repair during life.

Archaeichnium haughtoni, previously thought to be an archaeocyath, an agglutinated worm tube, or a trace fossil, is shown to be a body fossil with a complicated, pleated body wall that resembles to some extent the polyp-like bodies of the Cambrian animals Mackenzia and Paramackenzia (Zhao et al., 2021). Consequently, Archaeichnium and the mackenziids are tentatively considered to be anemone-like cnidarians rather than Ernietta-like vendobionts.

Three different kinds of unmineralized, annulated tubes are illustrated and briefly described as possible examples of Calyptrina, Sinotubulus, and Sekwitubulus. Those identified as “cf. Calyptrina striata Sokolov” compare well to White Sea examples of that species illustrated and analyzed for biomarkers by Bobrovskiy et al. (2022). Although the biomarker argument for a one-way gut in Calyptrina is debatable, there is a developing consensus that at least some of these Ediacaran tubular structures were produced by annelid grade worms.

Discussion

Glaessner (1979b, p. A96) tentatively referred Pflug's “Petalonamae” to four families, Pteridiniidae, Rangeidae, Charniidae, and Erniettidae “until clear distinctions between observable and hypothetically postulated characters can be drawn.” The removal of the Rangeomorpha as an order of Octocorallia (Jenkins, 1985) or more plausibly as a plesion of the Eumetazoa (Dunn et al., 2022) leaves Pteridinium, Ernietta, Swartpuntia, and their candidate synonyms (Namalia, Nasepia, Inkrylovia, Kuibisia) as another potentially monophyletic clade, the Erniettomorpha (Pflug, 1972; Erwin et al., 2011). However, finding shared derived characters to support a monophyletic Erniettomorpha as distinct from other Ediacaran fronds has been challenging to impossible (Dececchi et al., 2017; Hoyal Cuthill and Han, 2018). If our interpretation of the anatomy of Swartpuntia germsi is correct, then it is far more similar to Pteridinium carolinaensis than previously suspected, and the question about the monophyly of the Erniettomorpha reduces to whether Pteridinium and Ernietta are closely related and whether Phyllozoon (Jenkins and Gehling, 1978; Gehling and Runnegar, 2022)—or any other taxon—is another member of this clade. The synapomorphies identified by Dececchi et al. (2017) for Ernietta, Pteridinium, and Swartpuntia—“undifferentiated tubular elements (modules) that are parallel to each other and all of the same width”—also apply to Phyllozoon but are insufficient to assess possible ingroup relationships. As these taxa show little if any sign of whole-body differentiation, their position outside the Rangeomorpha and/or Arboreomorpha seems secure. But are they similar to each other as a result of inheritance, convergence, or merely simplicity?

Unique features of some or all erniettomorphs include tubular modules that coalesce during early growth (Ernietta) and are in contact laterally via linear not planar seams (Ernietta, Pteridinium, Phyllozoon) as well as triradial symmetry about a unipolar growth axis (Pteridinium, Swartpuntia). The apparently bipolar growth of Ernietta may be an attribute of its topology, and it is conceivable that Ernietta inherited the unipolar patterning of most Ediacaran fronds (Runnegar, 2022). Thus, the bipolarity of Ernietta may be an adaptation to an epibenthic lifestyle, as it apparently was for Fractofusus within the Rangeomorpha (Gehling and Narbonne, 2007; Dececchi et al., 2017). If so, the case for a monophyletic Erniettomorpha remains viable but barely so. In the remaining part of this brief discussion, we address two interwoven topics, both ultimately dependent on taphonomy: paleoecology and extinction.

Pteridinium has been found only in distal mass-flow deposits in South Australia even though it must have persisted during the deposition of the classical Ediacara-style bedforms of the Ediacara Member (Glaessner and Wade, 1966; Wade, 1971; Gehling and Droser, 2013). So where did it live, given that many richly fossiliferous beds have been sequentially excavated at the Nilpena Ediacara National Park (Droser et al., 2019) without revealing a single specimen of Pteridinium? The same question could be asked about Ernietta in Namibia and Nevada (Smith et al., 2017; Hall et al., 2020), some of which are filled with clean quartz sand quite unlike the downslope silty matrix in which they are found. Perhaps the answer lies with Phyllozoon, which seems to have inhabited sites that were below storm wave base in South Australia (Gehling and Runnegar, 2022) or with unsuspected anchoring structures that are in plain sight, such as one of the many “triradialomorphs” (Hall et al., 2020). Conversely, where are the discoidal holdfasts and epibenthic recliners that are so characteristic of South Australian and Avalonian assemblages? The obvious answer is differential extinction between the White Sea and Nama assemblages (Darroch et al., 2015, 2018; Evans et al., 2022), but the almost complete absence of Ediacara-style bed surfaces in Namibia (UCLA 7315 being a notable exception) suggests that differential preservation may be just as important. If only the mass-flow deposits in South Australia were fossiliferous, then Ediacara and Nilpena would be “Nama” rather than “White Sea” sites (Gehling and Droser, 2013). Last but not least, increasingly sophisticated phylogenomic studies continue to require substantial Precambrian histories for the crown group clades such as the Cnidaria (McFadden et al., 2021) and the Ecdysozoa (Shi et al., 2022; for a contrary view, see Holmes and Budd, 2022). Although magnificently exposed, the Nama succession is sparsely fossiliferous and thus may represent only a small sample of late Ediacaran biodiversity.