The rangeomorph fossil Charnia from the Ediacaran Shibantan biota in the Yangtze Gorges area, South China

Abstract

The terminal Ediacaran Shibantan biota (~550–543 Ma) from the Dengying Formation in the Yangtze Gorges area of South China represents one of the rare examples of carbonate-hosted Ediacara-type macrofossil assemblages. In addition to the numerically dominant taxa—the non-biomineralizing tubular fossil Wutubus and discoidal fossils Aspidella and Hiemalora, the Shibantan biota also bears a moderate diversity of frondose fossils, including Pteridinium, Rangea, Arborea, and Charnia. In this paper, we report two species of the rangeomorph genus Charnia, including the type species Charnia masoni Ford, 1958 emend. and Charnia gracilis new species, from the Shibantan biota. Most of the Shibantan Charnia specimens preserve only the petalodium, with a few bearing the holdfast and stem. Despite overall architectural similarities to other Charnia species, the Shibantan specimens of Charnia gracilis n. sp. are distinct in their relatively straight, slender, and more acutely angled first-order branches. They also show evidence that may support a two-stage growth model and a epibenthic sessile lifestyle. Charnia fossils described herein represent one of the youngest occurrences of this genus and extend its paleogeographic and stratigraphic distributions. Our discovery also highlights the notable diversity of the Shibantan biota, which contains examples of a wide range of Ediacaran morphogroups.

UUID: http://zoobank.org/837216cd-4a4a-4e13-89e2-ee354ba48a4c

Introduction

The late Ediacaran Period witnessed the rise and fall of the Ediacaran macrobiota (~574–539 Ma; Linnemann et al., 2019; Matthews et al., 2021), a group of macro-organisms that are complex and soft-bodied. Although some of the Ediacara-type macro-organisms are very likely metazoans (Fedonkin and Waggoner, 1997; Bobrovskiy et al., 2018; Chen et al., 2019), most of them remain phylogenetically unresolved. The Ediacaran macrobiota consists of three assemblages that may broadly represent three stages of evolution (Waggoner, 2003), although environmental conditions may have also played a role in their spatio-temporal distribution (Gehling and Droser, 2013; Grazhdankin, 2014). These stages are exemplified by the Avalon (~574–560 Ma), White Sea (~560–550 Ma), and Nama (~550–539 Ma) assemblages (Waggoner, 2003; Boag et al., 2016). Overall, the three assemblages have distinct taxonomic groupings and representative taxa, although a small number of taxa, for example, the frondose fossil Arborea, are known to occur in all three assemblages (Xiao and Laflamme, 2009; Droser et al., 2017). Like Arborea, the rangeomorph fossil Charnia is also previously known from the Avalon (Hofmann et al., 2008; Narbonne et al., 2014; Liu et al., 2015; Wilby et al., 2015), White Sea (Martin et al., 2000; Gehling and Droser, 2013), and Nama assemblages (Grazhdankin et al., 2008), therefore representing one of the longest-ranging genera of Ediacara-type macrofossils in terms of stratigraphic distribution.

Charnia was first described from the Ediacaran Charnian Supergroup in England (Ford, 1958). It is perhaps one of the most studied taxa of the Ediacara-type macro-organisms (Laflamme et al., 2007; Antcliffe and Brasier, 2008; Dunn et al., 2018, 2019, 2021; Butterfield, 2022). Different phylogenetic affinities for Charnia have been proposed since its discovery. It was originally described as an alga (Ford, 1958) but later classified into the family Charniidae (Glaessner, 1979), which may belong to the class Rangeomorpha of a proposed extinct phylum, the Petalonamae (Pflug, 1970, 1972; Jenkins, 1985). Charnia and other rangeomorphs have been variously interpreted as pennatulacean cnidarians (Glaessner, 1959. 1984; Glaessner and Wade, 1966; Gehling, 1991), lichens (Retallack, 1994), fungi (Peterson et al., 2003), members of the extinct kingdom Vendobionta (Seilacher, 1992), or members within the total-group Metazoa (e.g., Xiao and Laflamme, 2009; Dunn et al., 2018; Butterfield, 2022). Although Charnia and modern sea pens both possess a similar leaf-like shape, ontogenetic analysis reveals that Charnia and modern sea pens may have opposite growth polarities (Antcliffe and Brasier, 2007, 2008). Recently, detailed morphological descriptions and functional analyses, coupled with cladistic investigations (Dececchi et al., 2017, 2018), have led to the phylogenetic interpretations of rangeomorphs as stem-group metazoans (Xiao and Laflamme, 2009; Budd and Jensen, 2017; Darroch et al., 2018; Dunn et al., 2018), stem-group eumetazoans (Dunn et al., 2018, 2021; Hoyal Cuthill and Han, 2018; Butterfield, 2022; Runnegar, 2022), or stem-group cnidarians (Dunn et al., 2018; Butterfield, 2022). The feeding strategy of rangeomorphs also remains debated. Several studies have suggested that rangeomorphs, as well as erniettomorphs and dickinsoniomorphs, may have been capable of osmotrophy due to their high surface area to volume (SA/V) ratios (Laflamme et al., 2009; Sperling and Vinther, 2010; Ghisalberti et al., 2014). Liu et al. (2015) argued that osmotrophy in rangeomorphs may have been limited by the availability and recalcitrant nature of dissolved organic carbon. In addition, Butterfield (2022) contended that the large size, hence elevated Reynolds and Péclet numbers, of rangeomorphs are not conducive for osmotrophy. Alternative feeding strategies of rangeomorphs include suspension feeding (Butterfield, 2022) and intracellular symbiosis (Dufour and McIlroy, 2016; McIlroy et al., 2021), but these hypotheses have not been critically evaluated on the basis of detailed morphological observation and theoretical modeling.

The occurrence of Charnia in the Shibantan biota was first reported by S. Xiao et al. (2020), but a detailed description and morphometric assessment were not presented. Here we provide a systematic description of Charnia fossils, including the type species C. masoni Ford, 1958 emend. and C. gracilis n. sp., from this Lagerstätte. The fossils are preserved in thin-bedded bituminous limestone, which was deposited in subtidal environment (Q. Xiao et al., 2020), from the terminal Ediacaran Shibantan Member of the Dengying Formation in the Yangtze Gorges area of South China. The Shibantan Member is geochronologically constrained to between ~550 Ma and ~543 Ma (Huang et al., 2020; Yang et al., 2021). Thus, these Shibantan specimens not only extend the paleogeographic and paleoenvironmental distributions of Charnia, but also represent one of the youngest fossil records of this genus. They help us to obtain a more comprehensive picture of the rise and fall of the Ediacara-type macro-organisms.

Materials and methods

The Ediacaran succession in the Yangtze Gorges area of South China consists of the Doushantuo and Dengying formations (Fig. 1.1). The Dengying Formation consists of three lithostratigraphic members: the Hamajing, Shibantan, and Baimatuo members, in ascending order (Fig. 1.2). It is constrained by two CA-ID-TIMS zircon U–Pb ages of 551.1 ± 0.7 Ma (Condon et al., 2005) and 550.1 ± 0.6 Ma (Yang et al., 2021) from the uppermost Miaohe Member, which is regarded as the uppermost Doushantuo Formation (Xiao et al., 2017; Zhou et al., 2017; but see An et al., 2015), and a SIMS zircon U–Pb age of 543.4 ± 3.5 Ma from the Baimatuo Member (Huang et al., 2020).

Figure 1. (1) Geological map of the Wuhe section in the Yangtze Gorges area with location of the Wuhe quarry (marked with a red star) and the Huangling anticline in the Yangtze Gorges area. Inset map shows the location of the Huangling anticline (red star) and major tectonic terranes in China. (2) Generalized stratigraphic column of the Ediacaran succession in the Yangtze Gorges area, South China, showing stratigraphic distribution of fossils and U–Pb radiometric ages. Star marks the stratigraphic occurrence of Charnia in the Shibantan limestone. Modified from S. Xiao et al. (2020) and Wu et al. (2021). Geochronometric data sources: Condon et al. (2005) and Yang et al. (2021) for the lower Doushantuo Formation and Miaohe Member; Huang et al. (2020) and Zhang et al. (2022) for the Baimatuo Member; Okada et al. (2014) for the Cambrian Shuijingtuo Formation. Cry. = Cryogenian; Fm. = Formation; Mbr. = Member; Cam. = Cambrian.

The Hamajing and Baimatuo members are both characterized by dolostones with peritidal structures, such as tepees, karstification structures, and dissolution vugs (Duda et al., 2016; Y. Ding et al., 2019; S. Xiao et al., 2020). The Shibantan Member consists of 100–150 m dark gray, medium- to thin-bedded bituminous limestone with diagenetic chert nodules and bands (Chen et al., 2014; S. Xiao et al., 2020). Fine and crinkled laminae are dominant in the Shibantan Member, but hummocky cross-beds, rip-up clasts, intraclastic breccias, and graded beds are also common. Indeed, intraclastic breccias and graded beds in the Shibantan Member are interpreted as tempestites that were laid down in a subtidal environment between fair-weather and storm-wave bases (Q. Xiao et al., 2020). These tempestites may have contributed to the rapid burial of the soft-bodied Ediacara-type macrofossils (Q. Xiao et al., 2020). The wrinkled surfaces and dark crinkled micro-laminae in the Shibantan Member, interpreted as evidence for microbial mats (Chen et al., 2013; Meyer et al., 2014), may have played an important role in the preservation of the soft-bodied Ediacara-type macrofossils (Gehling, 1999; Callow and Brasier, 2009; Laflamme et al., 2011). These microbial mats, probably constructed mainly by cyanobacteria, have close associations with trace fossils from the Shibantan Member and may suggest that they provided oxygen oases as well as nutrients for the trace makers (Chen et al., 2013; Meyer et al., 2014; W. Ding et al., 2019; Xiao et al., 2019).

The Charnia fossils described here were collected from the Shibantan limestone at the Wuhe quarry (30.789°N, 111.051°E; Fig. 1.1), where a number of diverse fossils are preserved (S. Xiao et al., 2020). The Shibantan biota consists of algal fossils, Ediacara-type macrofossils, trace fossils, and some problematic fossils (S. Xiao et al., 2020). Recently, biomineralizing tubular fossils have also been discovered in the Baimatuo Member (Liang et al., 2020; Zhang et al., 2022) and also likely in the upper Shibantan Member (Chen et al., 2016), but these tubular fossils may be younger than the soft-bodied Ediacara-type macrofossils from Wuhe quarry. There are several frondose genera among the Ediacara-type macrofossils of the Shibantan biota (S. Xiao et al., 2020), including Arborea, Rangea, Pteridinium (Chen et al., 2014; Wang et al., 2020), and Charnia (herein). The Charnia specimens were collected from the basal Shibantan Member (Fig. 1.2) with known stratigraphic orientation (to determine whether fossils are preserved on the top or bottom bedding surface). Specimens were photographed with a Nikon D810 digital camera, and measurements of the specimens were carried out on these photos using ImageJ software. Specimens from Charnwood Forest in the United Kingdom (Wilby et al., 2015; Dunn et al., 2018, 2019, 2021), Newfoundland in Canada (Laflamme et al., 2007; Hofmann et al., 2008; Liu et al., 2013, 2015; Dunn et al., 2019), Sekwi Brook in Northwest Canada (Narbonne et al., 2014), the White Sea region and Olenek Uplift, Siberia in Russia (Sokolov and Fedonkin, 1984; Runnegar and Fedonkin, 1992; Grazhdankin and Bronnikov, 1997; Martin et al., 2000; Grazhdankin et al., 2008), Flinders Ranges, South Australia (Glaessner and Wade, 1966; Nedin and Jenkins, 1998; Gehling and Droser, 2013), and Oulongbuluke terrane in Northwest China (Pang et al., 2021) were also measured for comparison (Table 1). Measurements on Newfoundland specimens were conducted on retrodeformed photos from cited sources to account for tectonic deformation of the sediments.

Table 1. Biometric data for morphology comparison of Charnia specimens from the Shibantan biota and other localities.

*Incomplete petalodium.

Repository and institutional abbreviation

Specimens of Charnia illustrated in this study are reposited in Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences (NIGPAS), with NIGPAS museum catalog numbers (prefix NIGP-) provided for each specimen.

Systematic paleontology

Genus Charnia Ford, 1958, emended Dunn et al., 2019

Type species

Charnia masoni Ford, 1958

Charnia masoni Ford, 1958, emended
Figure 2.1, 2.2

1958

Charnia masoni Ford, p. 212, pl. 13, fig. 1.

?1959

Charnia sp.; Glaessner, p. 1472, text-fig. 1b.

?1959

Rangea?; Glaessner in Glaessner and Daily, p. 387, pl. 46, fig. 2.

1961

Charnia sp.; Glaessner, p. 75, text-fig.

1962

Charnia sp.; Glaessner, p. 484, pl. 1, fig. 4 (non fig. 5).

1962

Charnia masoni; Ford, fig. 4 (non fig. 5).

1966

Rangea grandis; Glaessner and Wade, p. 616, pl. 100, fig. 5.

1973

Glaessnerina grandis; Germs, p. 5, fig. 1D.

1976

Charnia ex gr. masoni; Sokolov, p. 141, text-fig.

1977

Charnia ex gr. masoni; Sokolov, p. 441.

1978

Charnia sp.; Fedonkin, fig. 3 (9).

1979

Charnia masoni; Glaessner, p. A99, fig. 12 (3).

1981a

Charnia masoni; Fedonkin, p. 66, pl. 3, figs. 5, 6, pl. 29, fig. 1.

1981a

Zolotytsia biserialis; Fedonkin, p. 67, pl. 3, fig. 7.

1981b

Charnia masoni; Fedonkin, p. 100.

1981

Charnia masoni; Sokolov and Brekhovskikh, p. 3, text-fig.

1981

Glaessnerina grandis; Glaessner and Walter, fig. 6.11C.

1983a

Charnia masoni; Fedonkin, pl. 1, fig. 1.

1983b

Charnia masoni; Fedonkin, fig. 37.

1983

Charnia masoni; Sokolov and Fedonkin, p. 13, fig. 9.

1984

Charnia masoni; Sokolov, p. 6, text-fig.

1984

Charnia masoni; Sokolov and Fedonkin, fig. 3f.

1984

Charnia cf. C. masoni; Glaessner, fig. 2.21B.

1984

Charnia masoni; Glaessner, fig. 2.21A.

1984

Glaessnerina grandis; Glaessner, fig. 2.21C.

1985

Charnia masoni; Fedonkin, p. 99, pl. 12, fig. 4, pl. 13, figs. 2–4.

1985

Charnia cf. C. masoni; Jenkins, fig. 7C.

1985

Charnia masoni; Jenkins, fig. 7B.

1987

Charnia masoni; Fedonkin, pl. 15.

1987

Glaessnerina grandis; Preiss, fig. E.

1990

Charnia masoni; Fedonkin, p. 110, pl. 12, fig. 4, pl. 13, figs. 2–4.

1992

Charnia masoni; Fedonkin, figs. 28–30.

1992

Charnia masoni; Runnegar and Fedonkin, fig. 7.5.5A, 7.5.10A.

1994

Charnia masoni; Fedonkin, fig. 2A, B.

1995

Charnia grandis; Boynton and Ford, p. 168, fig. 1.

1996

Glaessnerina grandis; Jenkins, p. 35, fig. 4.1.

1997

Charnia masoni; Grazhdankin and Bronnikov, fig. 2a, d.

1999

Charnia grandis; Ford, p. 231, figs. 1, 3.

2000

Charnia; Martin et al., fig. 4A.

2004

Charnia; Grazhdankin, fig. 2.

2004

Charnia-like frond; Narbonne, fig. 3D.

2005

Charnia masoni; Narbonne et al., pl. 1L.

2005

Charnia; Grazhdankin et al., fig. 3d.

2007

Charnia masoni; Laflamme et al., p. 252, fig. 4A–J.

2007

Charnia cf. C. masoni; Fedonkin et al., fig. 276 (partim).

2007

Charnia cf. C. masoni; Fedonkin et al., figs. 304, 314 (partim).

2007

Charnia masoni; Fedonkin et al., p. 265, fig. 354.

2008

Charnia masoni; Hofmann et al., p. 16, fig. 13.1, 13.2.

2008

Charnia grandis?; Hofmann et al., p. 16, fig. 14.

2009

Charnia masoni; Bamforth and Narbonne, fig. 7.5.

2011

Charnia masoni; Wilby et al., figs. 2A, 3A.

2011

Charnia masoni; Grazhdankin, fig. 3a–d.

2012

Charnia masoni; Liu et al., figs. 4b, 5a.

2012

Charnia aff. C. masoni; Liu et al., fig. 4a.

2013

Charnia aff. C. masoni; Liu et al., fig. 1d.

2013

Charnia masoni; Liu et al., fig. 2a–d.

2013

Charnia sp.; Gehling and Droser, fig. 2Q.

2014

Charnia cf. C. masoni; Narbonne et al., p. 215, fig. 5.1–5.4.

2015

Charnia masoni; Noble et al., fig. 4A, G.

2015

Charnia masoni; Wilby et al., fig. 5.

2015

Charnia masoni; Liu et al., fig. 2D.

2016

Charnia masoni; Liu et al., fig. 3D.

2017

Charnia masoni; Antcliffe et al., fig. 4E.

2018

Charnia masoni; Dunn et al., figs. 1E, 3.

2018

Charnia masoni; Kenchington et al., figs. 1A, D, 7E (partim).

2019

Charnia masoni; Dunn et al., p. 16, figs. 1–3, 5–8, 10.

2020

Charnia masoni; Liu and Dunn, fig. 4d.

2021

Charnia masoni; Matthews et al., fig. 2B.

2021

Charnia masoni; McIlroy et al., fig. 1b.

2021

Charnia masoni; Pang et al., figs. 3A–C, S2A–E.

2021

Charnia masoni; Dunn et al., figs. 1 (A-2), S1–S2.

2022

Charnia masoni; Butterfield, figs. 1a, 2.

2022

Charnia masoni; McIlroy et al., fig. 10f.

Holotype

LEIUG 2328, from Bed B (Wilby et al., 2011), North Quarry, Charnwood Forest, UK.

Emended diagnosis

Charnia with ovate to parallel-sided petalodium consisting of sigmoidal first-order branches emanating alternately at an acute angle, typically >20°. First-order branches composed of series of near-rectangular second-order branches arranged acutely to almost perpendicularly to the first-order branches.

Occurrence

Shibantan Member, Dengying Formation, Yangtze Gorges area, South China (S. Xiao et al., 2020); Bradgate Formation, Charnian Subgroup, Charnwood Forest, UK (Wilby et al., 2011); Drook, Briscal, Mistaken Point, Trepassey, and Fermeuse formations, Newfoundland, Canada (Narbonne, 2004; Laflamme et al., 2007; Hofmann et al., 2008; Liu et al., 2012; Dunn et al., 2019; Matthews et al., 2021); Nadaleen Formation (previously known as “June Beds,” Moynihan et al., 2019), northwestern Canada (Narbonne et al., 2014); Khatyspyt Formation, Olenek Uplift, north-central Siberia, Russia (Grazhdankin et al., 2008); Verkhovka Formation, the White Sea region, Russia (Martin et al., 2000); Rawnsley Quartzite, Flinders Ranges, South Australia (Gehling and Droser, 2013).

Description

Only one specimen in our collections can be assigned to this species. The specimen is characterized by a uniterminal bifoliate frond comprising an incompletely preserved ovate petalodium connected to a discoidal holdfast at the basal end by a stem (Fig. 2.1). The incomplete petalodium is 113.5 mm wide, and its preserved length is 251.5 mm. Petalodium is composed of strongly constrained (sensu Narbonne et al., 2009) first-order branches (sensu Dunn et al., 2021) or primary branches, which emanate alternately on each side of the zig-zag central axis at an angle of 23.3–45.4° (mean = 34.0°). The longest first-order branch is 103.1 mm long and 11.8 mm wide. First-order branches are inclined toward the apex of the frond and are more or less sigmoidal in shape. First-order branches at the apical end of the petalodium are poorly preserved but seem to be shorter than those at the proximal end. First-order branches appear to be rotated and furled (sensu Brasier et al., 2012) or single-sided (sensu Narbonne et al., 2009). First-order branches are composed of about a dozen rectangular to near-rectangular second-order branches (sensu Dunn et al., 2021) or secondary branches. Second-order branches in the longest first-order branch of each specimen are 4.3–5.1 mm (mean = 4.7 mm, n = 5) wide and arranged perpendicularly to the first-order branch. Third-order branches (sensu Dunn et al., 2021) are not observed. The stem is twisted and 158.8 mm long (measured from the base of petalodium to the center of the holdfast). The holdfast is discoidal, 87.7 mm in diameter, with a few filiform-texture structures in the middle (Fig. 2.2).

Materials

One specimen (NIGP161628) from Shibantan Member, Dengying Formation at Wuhe quarry.

Remarks

Three species of the genus Charnia—Charnia grandis Glaessner and Wade, 1966, Charnia wardi Narbonne and Gehling, 2003, and Charnia antecedens Laflamme et al., 2007—have previously been erected in addition to its type species Charnia masoni Ford, 1958. However, C. grandis is considered a junior synonym of C. masoni (Wilby et al., 2011; Brasier et al., 2012), whereas C. wardi and C. antecedens were subsequently reassigned to Trepassia Narbonne et al., 2009 and Vinlandia Brasier et al., 2012, respectively. This Shibantan specimen is tentatively classified in the genus Charnia due to its constrained and alternately arranged first-order branches as well as its single-sided rangeomorph units. The divergence angle of first-order branches of this specimen, 34.0° on average, is similar to that of the C. masoni specimens from other localities and much larger than that of the other Shibantan Charnia specimens (Table 1). First-order branches of this specimen are more curved than those of the other Shibantan Charnia specimens (Table 1). Its ovate petalodium and sigmoidal first-order branches also share similarities with other C. masoni specimens elsewhere. Taking all these factors into account, we choose to assign this specimen to C. masoni.

This specimen bears a twisted stem and a relatively large holdfast (Fig. 2.1), which are rare in previously reported Charnia specimens. The stem, together with the proximal end of the petalodium, seems to have been affected by water currents. The holdfast bears filiform textures in the middle (Fig. 2.2), somewhat different from Hiemalora-like holdfasts, which have radially arranged tentacle-like structures around the rim of the central disc (e.g., Chen et al., 2014, fig. 4; Shao et al., 2019, fig. 2). The filiform textures could represent drag structures generated by uprooting of the holdfast (Tarhan et al., 2010), consistent with the twisting of the stem. However, there seem to be at least two sets of filiform textures that are perpendicular to each other, an observation not easily accounted for by uprooting. Alternatively, the filiform textures may be wrinkles resulting from the compression of an originally three-dimensional bulbous holdfast. They are also broadly similar to the radial and concentric bands and filamentous mesh present in an Ediacaria disc from the White Sea region, which was interpreted as possible “skeletal” structure by Luzhnaya and Ivantsov (2019).

Charnia gracilis new species
Figures 2.3–2.8, 3

?1972a

Rangea sibirica; Sokolov, p. 50.

?1972b

Rangea sibirica; Sokolov, pl. 1, fig. 3.

?1979

Glaessnerina sibirica; Glaessner, p. A99, fig. 12 (1).

?1984

Glaessnerina sibirica; Glaessner, fig. 2.21D.

?1998

Charnia masoni; Nedin and Jenkins, p. 315, fig. 1.

2007

Charnia sp.; Fedonkin et al., fig. 232 (partim).

2008

Charnia masoni; Grazhdankin et al., fig. 2A.

2014

Charnia masoni; Grazhdankin, fig. 2.3.

2020

Charnia sp.; S. Xiao et al., fig. 4f.

Holotype

NIGP161629, from Shibantan Member, Dengying Formation, Yangtze Gorges area, South China, illustrated in Fig. 2.4.

Figure 2. (1) Charnia masoni from the Shibantan limestone preserved in positive relief, bed sole view; black arrow points to twisted stem and white arrow points to discoidal holdfast; (2) magnified view of rectangle in (1), with black and white rectangles marking two different orientations of filiform structures on the holdfast; NIGP161628. (3, 4) Holotype of C. gracilis n. sp. from the Shibantan limestone preserved in positive relief, bed sole view: (3) magnified view of rectangle in (4), showing well-preserved second-order branches and possible third-order branches (arrow); note the straight first-order branches with small divergence angles in (4); NIGP161629. (5) Specimen of C. gracilis n. sp. with complete petalodium, preserved in positive relief, bed sole view; arrowhead points to apex of petalodium; NIGP161630. (6) Specimen of C. gracilis n. sp. preserved in positive relief, bed sole view; NIGP161631. (7) Incomplete specimen of C. gracilis n. sp. preserved in positive relief, bed sole view; arrow points to a possible and faintly preserved holdfast; NIGP161632. (8) Incomplete specimen of C. gracilis n. sp. preserved in positive relief, bed sole view; NIGP161633. (1) Scale bar = 5 cm; (2) scale bar = 2 cm; (3–8) scale bars = 1 cm.

Diagnosis

A Charnia species characterized by a slender petalodium consisting of relatively long, thin, and straight first-order branches that have a parallel-sided blade-like shape. First-order branches emanate alternately from the central axis at an acute angle, typically ≤20°. First-order branches are composed of a series of rectangular or rhomboid second-order branches arranged acutely to perpendicularly to the first-order branches.

Occurrence

Shibantan Member, Dengying Formation, Yangtze Gorges area, South China; Khatyspyt Formation, Olenek Uplift, north-central Siberia, Russia (Grazhdankin et al., 2008); Verkhovka Formation, the White Sea region, Russia (Fedonkin et al., 2007); possible occurrence in Rawnsley Quartzite, Flinders Ranges, South Australia (Nedin and Jenkins, 1998).

Description

Specimens are characterized by a centimeter- to decimeter-scale, uniterminal bifoliate frond comprising a nearly parallel-sided and spicate petalodium tapering gradually at the apical end (Figs. 2.3–2.8, 3). There are four specimens in our collection that bear a completely preserved petalodium, which is 72.4–172.3 mm long (mean = 102.5 mm, n = 4) and 10.7–20.5 mm wide (mean = 14.6 mm, n = 4). Incomplete petalodia of five other specimens can be measured for maximum width, which varies from 14.6 mm to 73.1 mm (mean = 37.0 mm, n = 5), and their preserved length varies from 70.0 mm to 555.7 mm (mean = 204.7 mm, n = 5). Two additional specimens are too incompletely or poorly preserved to allow reliable measurements of petalodium width and length; thus, they are not included in the measurement data. The petalodium is composed of about a dozen strongly constrained (sensu Narbonne et al., 2009) first-order branches (sensu Dunn et al., 2021). First-order branches emanate alternately on each side of the central axis at an angle of 9.1–21.7°, with the average divergence angle (11.6–17.7° for nine specimens) usually <20°. Opposing first-order branches are offset by half a branch width, forming a zig-zag central suture (e.g., Fig. 2.4). First-order branches are inclined toward the apex of the frond and are nearly straight, although they can be slightly curved at both proximal and distal ends, leading to a blade-like shape. Proximal first-order branches are usually more curved than distal ones. The longest first-order branch usually occurs near the middle of the petalodium, 24.3–164.8 mm long (mean = 52.8 mm, n = 9) and 3.3–8.5 mm wide (mean = 4.8 mm, n = 9). First-order branches are composed of about a dozen second-order branches (sensu Dunn et al., 2021) arranged parallel to one another. The widest second-order branches in the longest first-order branch are 1.2–4.7 mm (mean = 2.8 mm, n = 9) wide. Second-order branches vary in shape and orientation in specimens of different sizes. In smaller specimens (e.g., Fig. 2.3–2.8), second-order branches are generally near rectangular and arranged more or less perpendicularly to the first-order branch, whereas in larger specimens (e.g., Fig. 3.1, 3.2), second-order branches are rhomboidal and arranged more acutely to the first-order branch. The number of second-order branches in each first-order branch seems to remain more or less constant (Fig. 4.1), whereas the average size of second-order branches in each first-order branch increases by ~14 times as the length of the first-order branch increases ~eight times (Fig. 4.2). The shape of second-order branches can also vary from rectangular or rhomboidal in the central region to trigonal or trapezoidal in the proximal and distal regions of the first-order branch (e.g., Fig. 2.3). Rangeomorph units of both first-order branches and second-order branches are rotated and furled (sensu Brasier et al., 2012) or single-sided (sensu Narbonne et al., 2009). Third-order branches (sensu Dunn et al., 2021) are barely discernable in some second-order branches, characterized by obliquely arranged ridges (e.g., Figs. 2.3, 3.2). First-order and third-order branches are inclined toward the apex of the frond. A discoidal holdfast is faintly preserved in one specimen (Fig. 2.7), ~7.3 mm in diameter. A Charnia gracilis n. sp. specimen (Fig. 3.4) is preserved together with a Helminthoidichnites-like trace fossil on the same bedding surface (Fig. 3.3). A possible juvenile specimen of C. gracilis is in alignment with another poorly preserved, taxonomically unidentifiable frond (Fig. 3.5), indicating common orientation of tethered, erect epibenthic organisms by water currents.

Figure 3. Charnia gracilis new species from the Shibantan limestone. (1, 2) The longest specimen of C. gracilis n. sp. in the Shibantan limestone with an incomplete petalodium and well-preserved second-order branches, positive relief, bed sole view; (2) magnified view of the rectangle in (1), showing rhomboidal second-order branches and inclined third-order branches; NIGP161634. (3, 4) A juvenile specimen of C. gracilis n. sp. marked by rectangle in (3), positive relief, bed sole view; the specimen is preserved together with a Helminthoidichnites-like trace fossil, marked by arrow in (3); (4) magnified view of rectangle in (3); NIGP161635. (5) A possible juvenile specimen of C. gracilis n. sp. (bottom) and another poorly preserved unnamed frond (upper) preserved in positive relief, bed sole view. Arrows point to inferred direction of water current that felled and aligned the specimens. NIGP161636. (1) Scale bar = 5 cm; (2, 5) scale bars = 2 cm; (3, 4) scale bars = 1 cm.

Figure 4. Measurements of several specimens of Charnia gracilis new species with well-preserved second-order branches (NIGP161629; NIGP161630; NIGP161631; NIGP161634). (1) Cross-plot of first-order branch (1' branch) length versus number of constituent second-order branches (2' branches); (2) Cross-plot of first-order branch length versus average size of constituent second-order branches. The number of second-order branches in each first-order branch is evaluated by dividing the length of the first-order branch by the average width of constituent second-order branches that are measurable.

Etymology

From gracilis (Latin, slender), in reference to the slender shape of the petalodium as well as the first-order branches.

Materials

Eleven specimens in total, from Shibantan Member, Dengying Formation at Wuhe quarry.

Remarks

The Shibantan specimens possess diagnostic features of the genus Charnia, including single-sided rangeomorph units, strongly constrained first-order branches, and a zig-zag midline. Biometric plots also show that these Shibantan specimens share similarities with Charnia specimens from other localities (Fig. 5; Table 1). For example, the petalodium length versus width (Fig. 5.1), as well as the width of the longest first-order branch versus the largest width of second-order branch in the longest first-order branch (Fig. 5.2), is similar between the Shibantan specimens and C. masoni from Charnwood Forest, UK (Wilby et al., 2015; Dunn et al., 2018, 2019, 2021), C. masoni from Newfoundland, Canada (Laflamme et al., 2007; Hofmann et al., 2008; Liu et al., 2013, 2015), Charnia cf. C. masoni from Sekwi Brook, NW Canada (Narbonne et al., 2014), C. masoni and C. gracilis n. sp. from the White Sea region, Russia (Sokolov and Fedonkin, 1984; Grazhdankin and Bronnikov, 1997; Martin et al., 2000), C. masoni from Flinders Ranges, South Australia (Germs, 1973; Gehling and Droser, 2013), C. masoni and C. gracilis from Olenek Uplift, Siberia (Runnegar and Fedonkin, 1992; Grazhdankin et al., 2008), and C. masoni from Oulongbuluke terrane, NW China (Pang et al., 2021). Thus, it is reasonable to assign the Shibantan specimens to the genus Charnia. However, there are several notable differences between the Shibantan specimens and C. masoni. Relative to C. masoni specimens from other localities, the Shibantan specimens have first-order branches that are slenderer, longer, and straighter, taper more gradually toward the distal end, have a higher length/width ratio (Fig. 5.3, 5.4), and present a blade-like rather than a sigmoidal shape (Fig. 5.5, 5.9). In addition, the first-order branches are straight and blade-like in the Shibantan specimens but sigmoidal in C. masoni. This difference can be quantified using a new morphologic descriptor defined as X = |(a − b)/(a + b)|, where “a” and “b” are the angles between the diagonal line and borderlines at the distal end of first-order branches (Fig. 5.9). It can be shown that the parameter X effectively distinguishes the sigmoidal (X ≥ 0.3 for C. masoni) or straight (0 ≤ X < 0.3 for C. gracilis) first-order branches (Fig. 5.5). Finally, the mean divergence angle of the first-order branches in the Shibantan specimens, ranging between 12° and 18°, is much lower than in C. masoni specimens elsewhere (Fig. 5.4, 5.5). Considering these morphological disparities as likely interspecific variations, we choose to erect a new species, and we use the mean divergence angle of the first-order branches to differentiate C. gracilis (≤20°) and C. masoni (>20°).

Grazhdankin et al. (2008) reported Charnia masoni from the Khatyspyt Formation in Siberia (Grazhdankin et al., 2008, fig. 2A). Although incompletely preserved, this Khatyspyt specimen has straight and thin first-order branches with an average divergence angle of 19.8° (Fig. 5.4, 5.5; Table 1). This Khatyspyt specimen shares more morphological similarities with C. gracilis specimens from the Shibantan Member than typical C. masoni material; therefore, it is more appropriate to reassign this specimen to C. gracilis. An incompletely preserved Charnia sp. specimen from the White Sea region (Fedonkin et al., 2007, fig. 232) may also be considered C. gracilis. This incomplete specimen possesses a near-parallel-sided petalodium and relatively straight first-order branches with an average divergence angle of 18.5° (Fig. 5.4; Table 1). An incomplete Charnia specimen from Siberia (Glaessner, 1979, fig. 12.1), which was assigned to Glaessnerina sibirica (Sokolov, 1973), is similar to the Shibantan C. gracilis specimens in its thin, straight, and parallel-sided first-order branches. Its inclined second-order branches are similar to those of the longest specimen in Shibantan limestone (Fig. 3.1). However, the poor preservation of the middle and left parts of its petalodium makes measurement of the divergence angle of the first-order branches difficult. Therefore, this specimen can be only provisionally placed in C. gracilis. A Charnia specimen reported by Nedin and Jenkins (1998) from the Rawnsley Quartzite, South Australia, also resembles the Shibantan C. gracilis specimens in its straight, slender, and acutely divergent (17.5–26.1°) first-order branches, but its mean divergence angle (21.7°; n = 5) is slightly larger than those of the latter. In general, the Charnia specimens from the White Sea and Nama assemblages can be related to or even reassigned to C. gracilis, on the basis of the morphology of first-order branches, which are more acutely diverged, less curved, and apically thinner and straighter than those from the Avalon assemblage (e.g., Dunn et al., 2019, fig. 1F). However, C. gracilis is also somewhat similar to some specimens from Newfoundland in their parallel-sided outlines of the petalodium (e.g., Laflamme et al., 2007, fig. 4f–h; Liu et al., 2015, fig. 2D; Dunn et al., 2019, fig. 8A), despite the fact that their first-order branches are different (Fig. 5; Table 1).

Dunn et al. (2019) studied Charnia masoni specimens from different localities and found that they are comparable in morphology but hard to accord with the morphological reconstructions of some other rangeomorphs such as Avalofractus (Narbonne et al., 2009) and Rangea (Vickers-Rich et al., 2013), which possess an internal central stalk (also refer to Dunn et al., 2021, fig. 5). Some researchers envisioned that an internal stalk may also be present in Charnia (Narbonne et al., 2009), but subsequent studies by other researchers found no evidence for a central stalk (Dunn et al., 2019; see also Dunn et al., 2021, fig. 5). The Shibantan C. gracilis specimens, which are similar in overall morphology to Charnia fossils reported elsewhere, show no sign of a central stalk.

Discussion

The differences between Charnia gracilis n. sp. and Charnia masoni lie mainly in their first-order branches, as revealed by biometric analysis of Charnia specimens worldwide (Fig. 5). The Shibantan specimens of C. gracilis show notable differences from Charnia specimens elsewhere, with a straight blade-like shape, lower divergence angles, and greater length/width ratios for their first-order branches (Fig. 5.4, 5.5). These differences are unlikely to be artifacts of tectonic or taphonomic deformation. Tectonic shearing can be ruled out because associated holdfast structures are perfectly circular in shape (Fig. 3.2). Although taphonomic deformation did occur in some Shibantan fossils (e.g., Dickinsonia, Wang et al., 2021; see also slightly C-shaped fronds in Fig. 3.5), the tightly constrained first-order branches (Narbonne et al., 2009) in C. gracilis left little room for their postmortem dislocation, rotation, or deformation. Postmortem compression or stretching of the petalodium, which would increase the length/width ratio of the petalodium, straighten the first-order branches, and decrease their divergence angles, also seemed unlikely to have happened in such a uniform fashion simultaneously.

In addition to the shape of first-order branches of Charnia gracilis specimens, it is unlikely that their lower divergence angles result from alignment by strong currents. Some Shibantan Charnia specimens do show evidence of alignment (Fig. 3.5), perhaps by water currents, although there is no sedimentary evidence for strong water currents in the fossil-bearing horizons (Chen et al., 2014; Duda et al., 2016). It is conceivable that the lower divergence angles of C. gracilis specimens may be a biostratinomic artifact related to deformation by strong currents, but this interpretation contradicts the observation that the Shibantan C. masoni specimen, which has a twisted stem and a wrinkled holdfast that may be caused by water currents (Fig. 2.1), has larger divergence angles than the Shibantan C. gracilis specimens (Figs. 2.4–2.8, 3.1). Some C. masoni specimens from Newfoundland (Dunn et al., 2019, figs. 7, 8) exhibit a slender frond with parallel-sided margins, similar to the Shibantan C. gracilis specimens, and they possess a long connecting region, which has been interpreted as an artifact caused by twisting upon felling (Dunn et al., 2019) that may also be affected by water currents. However, the divergence angles of these Newfoundland specimens are larger than 20°, and their first-order branches are sigmoidal in shape (Fig. 5.5), both of which are different from C. gracilis specimens.

The slender, straight, and blade-like shape, and the lower divergence angle, of first-order branches of Charnia gracilis are therefore considered a species-level taxonomic distinction. The more or less straight first-order branches of the Shibantan C. gracilis specimens lead to slightly jagged lateral margins and a sharp V-shaped apex of the petalodium (Fig. 2.4). The distal end of the first-order branches of Shibantan C. gracilis specimens is not distinctly curved (Fig. 2.4–2.8), in contrast to Charnia specimens elsewhere, indicating that the lateral margins of the petalodium in C. gracilis may have been somewhat unfurled and not morphologically influenced by water currents when alive. A petalodium with furled lateral margins would be better streamlined to reduce the drag of water currents, whereas one with unfurled lateral margins would fully expose the surface area of the frond to water currents, thus enhancing the feeding efficiency. However, considering that both C. gracilis and C. masoni are present in the Shibantan Member, and that there are relatively few specimens of either taxon, it is difficult to distinguish whether the Shibantan specimens are furled. It is also uncertain whether furling is a persistent and taxonomically informative character, an ecophenotypic behavior related to feeding strategies and water current intensity, or a biostratinomic feature.

Previous studies proposed that the growth of Charnia was achieved by the insertion of new first-order branches (e.g., Laflamme et al., 2007; Antcliffe and Brasier, 2008; Laflamme and Narbonne, 2008), but recent studies hypothesized that Charnia grew by the insertion of new first-order branches followed by their subsequent inflation (Wilby et al., 2015; Dunn et al., 2018). Two small frondose fossils (Fig. 3.4, 3.5) from the Shibantan limestone are recognized as juvenile specimens of C. gracilis. Although poorly preserved, it seems that these juvenile specimens contain fewer first-order branches (~12; Fig. 3.4, 3.5) than the largest specimen in our collection (>26; Fig. 3.1), supporting insertion of new first-order branches as a key growth mechanism. However, the number of second-order branches (~12–20) in each first-order branch does not seem to change much among specimens of different sizes (Fig. 4.1), whereas the shape of the second-order branches varies from near rectangular in the small- and medium-sized specimens (e.g., Fig. 2.3, 2.4), to axially rhomboidal in large-sized specimens (e.g., Fig. 3.1, 3.2); their size also increases greatly (Fig. 4.2). Considering that first-order branches increase in number and size during growth (Fig. 5.6, 5.7) whereas second-order branches increase mainly in size rather than number in every first-order branch (Fig. 4), it seems that the inflation of first-order branches was achieved mainly by inflation of second-order branches rather than insertion of new second-order branches. These observations support the hypothesis that growth of Charnia was accomplished by the insertion and subsequent inflation of first-order branches (Laflamme et al., 2007; Laflamme and Narbonne, 2008; Wilby et al., 2015; Dunn et al., 2018), which grew largely by inflation rather than insertion of second-order branches. Each first-order branch is basally initiated from the third to sixth second-order branches of the subtending first-order branch in the Shibantan specimens (e.g., Fig. 2.3), similar to those described by Dunn et al. (2021). The smaller first-order branches at the apical end of the petalodium (e.g., Fig. 3.4) suggest that new first-order branches were generated distally, in contrast to the long-considered analogs of sea pens (Antcliffe and Brasier, 2007). The presence of a stem in the larger C. masoni specimen (Fig. 2.1) and the absence of a stem in the juvenile specimens of C. gracilis implies that the stem may have been absent in younger stages but emerged later in the adult stage of the Charnia frond. Therefore, there may have been another generative zone at the proximal end of the Charnia frond where the stem was generated, in addition to the apical growth zone where new first-order branches were inserted (Dunn et al., 2018).

Figure 5. Biometric plots of Charnia from Shibantan limestone and other localities. (1) Cross-plot of petalodium length versus width. (2) Cross-plot of width of the longest first-order branch (1'branch) versus width of the widest second-order branch (2' branch) in the longest first-order branch. (3) Cross-plot of length versus width of the longest first-order branch. (4) Cross-plot of the length/width (L/W) ratio of the longest first-order branch versus mean divergence angle of first-order branches. Sketches at the lower right and upper left in (4) show first-order branches of two representative endmembers. (5) Cross-plot of mean divergence angle versus mean value of X = |(a − b)/(a + b)|; “a” and “b” represent the included angles between the diagonal line and borderlines at the distal end of first-order branches. (6) Cross-plot of petalodium length versus countable number of first-order branches. (7) Cross-plot of petalodium length versus length of the longest first-order branch. (8) Schematic diagram depicting how measurements for first- and second-order branches were made. (9) Schematic diagram depicting how measurements for X = |(a − b)/(a + b)| were made to distinguish the sigmoidal or straight shape of the first-order branches; note that in the lower left case, angle “b” < 0 and angle “a” > 0 (X > 1), whereas in the two other lower cases, both angle “b” and angle “a” are ≥0 (0 ≤ X ≤ 1). Localities are represented by symbols with different colors and shapes. Triangles with an attached arrow represent specimens with incomplete petalodium length in (1) and (7). Dots represent the Avalon assemblage, squares represent the White Sea assemblage, and triangles and crosses represent younger assemblages, possibly belonging to the Nama assemblage. Hollow symbols represent measurements of C. gracilis new species, whereas filled symbols represent measurements of other Charnia species.

Although Charnia is widely considered to have been an epibenthic organism, its posture on the seafloor has been debated. Some researchers consider it a reclining organism on the basis of the inference that most Charnia specimens preserve only one side of the frond, assuming that the two sides might be different (Grazhdankin, 2004; McIlroy et al., 2021). However, the twisted stem in a Shibantan C. masoni specimen (Fig. 2.1) implies the influence of water currents that rotated a standing frond, consistent with an erect living lifestyle (Laflamme et al., 2007; Narbonne et al., 2014; Wilby et al., 2015; Droser et al., 2017). The presence of a holdfast in some Charnia specimens (Fig. 2.1, 2.7) is also consistent with an erect lifestyle, although holdfasts are rarely preserved or observed. Some Charnia specimens from other localities preserve a small and bulbous holdfast (Laflamme et al., 2007; Wilby et al., 2015; Dunn et al., 2019), which has been taken as evidence that the holdfast was buried below the sediment–water interface (Burzynski and Narbonne, 2015). One Shibantan C. gracilis specimen preserves a faint holdfast (Fig. 2.7), and another C. masoni specimen bears a distinctly larger holdfast with filiform texture (Fig. 2.1, 2.2; Table 1). It is possible that the larger holdfast illustrated in Figure 2.1 represents an uprooted specimen that was pulled out of the sediment by water currents; this interpretation is also consistent with the twisted stem in this specimen, which may have been caused by rotation of the frond relative to the holdfast because of water currents. Overall, the evidence available seems to suggest that Charnia stood rather than lay on the seafloor.

The occurrence of Charnia in the Shibantan biota expands the paleogeographic distribution of this taxon and represents one of the youngest examples of this genus. The Shibantan Member preserves taxa that were thought to be characteristic of the Nama assemblage (e.g., Cloudina; S. Xiao et al., 2020) and White Sea assemblage (e.g., Dickinsonia; Wang et al., 2021). Previous researchers regarded the Shibantan biota as an example of the Nama assemblage (Boag et al., 2016; Muscente et al., 2019; Wu et al., 2021), an example of the White Sea assemblage (Laflamme et al., 2018), or a transition between these two assemblages (S. Xiao et al., 2020). Regardless, a recent radiometric date of 543.4 ± 3.5 Ma from the overlying Baimatuo Member (Huang et al., 2020) indicates that the Shibantan biota preserves one of the youngest occurrences of Charnia, roughly comparable in age to two other terminal Ediacaran occurrences, in the Khatyspyt Formation in Siberia (~553–544 Ma; Grazhdankin et al., 2008; Rogov et al., 2015) and in the Zhoujieshan Formation in Qaidam (~550–539 Ma; Pang et al., 2021). Meanwhile, the oldest occurrences of Charnia come from the Drook Formation in Newfoundland, Canada (~574–560 Ma; Narbonne and Gehling, 2003; Matthews et al., 2021) and the Bradgate Formation in Charnwood Forest, UK (~562–557 Ma; Ford, 1958; Noble et al., 2015); the genus is also present in the White Sea assemblage (Martin et al., 2000; Gehling and Droser, 2013). In terms of paleogeographic distribution, Charnia has been reported from almost all major Ediacara-type fossil localities ranging from low to high paleolatitudes (Fig. 6; see also Boddy et al., 2022). In terms of paleoenvironmental distribution, Charnia specimens have been reported from siliciclastic sediments in deep marine basins (Hofmann et al., 2008; Liu et al., 2015) to sandstones in shallow shelf environments, including lagoon/delta front and lower shoreface (or sheet-flow sands) (Gehling and Droser, 2013; McMahon et al., 2020) to carbonate shelves (Grazhdankin et al., 2008; this paper). Thus, Charnia is an Ediacara-type macrofossil genus with a remarkably long stratigraphic range and broad paleogeographic range (Fig. 6). This implies that the sessile Charnia must have had some sort of dispersal strategies, presumably through planktonic larvae (Darroch et al., 2013) or waterborne asexual propagules (Mitchell et al., 2015; Mitchell and Kenchington, 2018; see also Liu and Dunn, 2020). In addition, a juvenile specimen of Charnia is preserved together with a Helminthoidichnites-like trace fossil (Fig. 3.3), indicating that Charnia could survive in an environment occupied by bilaterian trace makers. The co-occurrence of Helminthoidichnites-like trace fossils and Ediacara-type body fossils has also been observed in other Ediacaran successions (e.g., Gehling and Droser, 2018). Such co-occurrences can provide key insights into the ecological interactions between trace-making animals and Ediacara-type soft-bodied macro-organisms and may help to test the biotic replacement hypothesis that bilaterian bioturbation may have led to the demise of sessile frondose taxa in the terminal Ediacaran (e.g., Seilacher, 1989, 1992; Laflamme et al., 2013; Darroch et al., 2015).

Figure 6. Paleogeographic distribution of Charnia (marked by yellow dots) worldwide. The paleogeographic map (~550 Ma) is based on Zhao et al. (2018) and Pang et al. (2021). C-Qil = Central Qilian; N-Qt = North Qiangtang; S-Qt-Ls = South Qiangtang and Lhasa; IC = Indochina.

Conclusion

A systematic description of Charnia masoni and Charnia gracilis n. sp. from the Shibantan biota (ca. 550–543 Ma) is presented in this paper. The Shibantan C. gracilis specimens show differences in the overall morphology, length/width ratio, and divergence angle of first-order branches from the type species, C. masoni. Nevertheless, morphological aspects of their first and second-order branches, their tightly constrained and one-sided first-order branches, and their zig-zag central suture indicate that they belong to the genus Charnia. The Shibantan Charnia specimens also present features that seem to support an insertion–inflation growth model and an erect sessile epibenthic lifestyle. Charnia masoni and C. gracilis from the Shibantan limestone represent one of the youngest occurrences of the genus Charnia and extend the paleogeographic, paleoenvironmental, and stratigraphic distributions of this genus. Charnia seems to be an evolutionarily resilient genus that persisted for ~30 Myr and witnessed the rise and fall of the Ediacara-type macro-organisms.