Sedimentology of the Catalina Dome and Taxonomy of Mistaken Point Small Fronds

Home , Avalofractus, Avalon Peninsula, Drook Formation, Fermeuse Formation, Mistaken Point Formation, Pectinifrons

Sedimentology of the Catalina Dome and taxonomy of Mistaken Point small fronds

by:

Sara Joan Mason

A thesis submitted to the Department of Geological Sciences and Geological Engineering in conformity with the requirements for the degree of Master of Science

Queen’s University

Kingston, Ontario, Canada

October 2013

ABSTRACT

Research carried out in the Ediacaran of eastern Newfoundland focused on two projects: sedimentology of the Conception and St. John’s groups exposed on the Bonavista Peninsula; and taxonomic descriptions of the small, stemmed frondose fossils at Mistaken Point on the Avalon

Peninsula.

Sedimentological study of the upper Conception and lower St. John's groups at Catalina

Dome on Bonavista Peninsula extends our understanding of the Conception Basin, in which the oldest known complex, deep marine organisms lived, by a factor of two. Mudstone-rich turbidites dominate the succession, and a lack of wave-generated structures or other shallow- water indicators support the interpretation that the depositional environment was deep-marine.

The basal part of the succession contains seismoturbidites that show no evidence of horizontal translation, implying that deposition occurred on a flat basin plain. Strata higher in the succession exhibit horizontally slumped beds, implying a transition into slope deposition.

Turbidite ripple marks show a change in paleocurrent direction from eastward to southward in the Trepassey Formation, consistent with a change from convergent to strike-slip tectonics that occurred diachronously across the basin. Volcanic ash beds are more common in the Catalina

Dome succession than on the Avalon Peninsula, reflecting deposition closer to the volcanic source. These volcanic beds are associated with diverse fossil assemblages rich in rangeomorphs that locally persist into the Fermeuse Formation, in contrast with the Avalon Peninsula where the

Fermeuse Formation contains only simple discoid fossils. This taphonomic window lends support to the hypothesis that the form genus Aspidella represents the holdfasts of Ediacaran fronds.

Stemmed small frond fossils from Mistaken Point, Avalon Peninsula, have often been

informally referred to as “featherdusters”, but due to their small size and consequent poor

preservation, they have not until now been formally described. The small, stemmed fronds are

more diverse than previously realized, and include representatives of taxa described from elsewhere in Newfoundland, juveniles of other Mistaken Point fronds, and two new monospecific genera. This biodiversity suggests that the basal elevated tier that the small fronds occupied was competitive, with convergent evolution of frondose taxa showing distinct architecture and constructions, but broadly similar size and shape.

STATEMENT OF CO-AUTHORSHIP

While this thesis is substantially my own, original work, my supervisor and colleagues have contributed actively to its development. A modification of Chapter Two is published in the

Canadian Journal of Earth Sciences (Mason et al. 2013), and it is co-authored by Guy M.

Narbonne, Robert W. Dalrymple, and Sean J. O’Brien. Order of authorship is indicative of each individual’s relative intellectual involvement. We intend to submit Chapter Three to an international journal, and it is co-authored by Guy M. Narbonne.

TAXONOMIC STATEMENT

The systematic paleontology of the three new Ediacaran genera and species described in this thesis does not constitute the official description of these taxa in accordance with the

International Code of Zoological Nomenclature (third edition). The official descriptions of

Pricscipeniculus hofmanni and Broccoliformis alti will appear in a modified version of Chapter

Three that will be submitted to an international journal.

iii

ACKNOWLEDGMENTS

First and foremost I offer my appreciation and gratitude to Dr. Guy Narbonne for his

guidance, his support, and especially his patience, as well as for giving me this amazing,

unforgettable opportunity to study Ediacaran fossils in Newfoundland. I thank Dr. Marc

Laflamme for his words of wisdom and encouragement from my very first week as a grad

student onward, and now for an exciting opportunity to earn a PhD following this MSc. I thank

my co-authors for Chapter Two, Dr. Bob Dalrymple and Dr. Sean O'Brien, for helping me

grapple with sedimentology and stratigraphy, and get my first ever publication.

Calla Carbone was the best field assistant I ever could have hoped for, with her energy,

enthusiasm, fearlessness, careful driving, and productive conversations in the field. I also thank

Thomas Boag and Ryan Abrams for assistance with fossil and latex photography. I thank C. L.

Wilson of UBC for offering her Latin expertise toward naming two new fossil taxa.

This research and my time at Queen’s was generously funded by a scholarship from the

Natural Sciences and Engineering Research Council of Canada, an Ontario Graduate

Scholarship, Queen’s University scholarships and teaching assistantships, and support granted by the Natural Sciences and Engineering Research Council to my supervisor, Guy Narbonne.

Fieldwork in the Mistaken Point Ecological Reserve was carried out under scientific permits granted by the Parks and Natural Areas Division, Government of Newfoundland and

Labrador. I thank the people of Portugal Cove South, Trepassey, and Trinity Bay North,

Newfoundland, for their kindness and hospitality that made my stays in Newfoundland for field work an absolute pleasure.

I owe a great debt to the late Dr. Hans Hofmann, whom I never met, but whose research formed the backbone of much of this work.

I thank my parents for indulging a little girl's “I want to be a paleontologist when I grow up” phase, even when it never went away, and for always being there for me when I needed it most. I'd also like to thank Dr. Doug Haywick and Dr. Murlene Clark of the University of South

Alabama for introducing me to the world of geology and teaching me that there's more to

paleontology than dinosaurs, and that a lot of it, like the Ediacara biota, is actually even cooler!

My positive experiences in their classrooms as both a student and a teaching assistant convinced

me that a career in research and teaching is the right path for me.

I thank my friends in Kingston and around the world for making the last several years

unforgettable, with special thanks to Carl Jackson, Laureline Sangalli, and Greg Burzynski for

being there for me and making sure I saw daylight every now and then. Carl Jackson in particular

deserves thanks for proofreading many drafts of many papers over the course of my time at

Queen’s.

I also thank everyone involved with Kingston's Impromptu Productions for the periodic

escape from my reality into the world of theatre. Whether I was helping out as a crew member or

simply watching Shakespeare in the Park, time with Impromptu was always a welcome (and

necessary) distraction.

I can’t give enough thanks my grandmother Joan Mason, my aunt and uncle Laurie and

David Giesecke, and my aunt's friends Jane and Ross Fargher, for giving me the most amazing

(and surprisingly educational) pre-thesis defense holiday imaginable in Adelaide and the Flinders

Ranges, South Australia. That was the best surprise ever for an aspiring Ediacaran paleontologist.

This thesis is dedicated to my partner Dominic Russo who has offered me years of

unconditional emotional support, and has patiently waited for me to figure out the next stage of

my life so we can finally move to the same city. We’re almost there!

TABLE OF CONTENTS

Abstract………………………………………………………………………………………...... i

Statement of Co-Authorship.………………………………………………………….…...….. iii

Taxonomic Statement ………………………………………………………………..…...…….iii

Acknowledgments.……………….…………………………………………………..………….iv

Table of Contents…………………………………………………………………...……...... ….vi

List of Figures.…..…………………………………………………………………………….....ix

List of Tables…………….……………………………………………………………………..…x

Chapter One: General Introduction……………….……………………………………….…….1

1.1 Importance of the Ediacaran Period...... 2

1.2 The Ediacaran of the Avalon Peninsula……………………………………………………….5

1.2.1 Sedimentology of the Ediacaran of the Avalon Peninsula…………………………..7

1.2.2 Paleontology of the Mistaken Point Biota……….………………………………….9

1.3 Ediacaran succession on the Bonavista Peninsula.………………………………....…....…..12

1.4 Purpose of Thesis…………………………………………………………………………….16

Chapter Two: Paleoenvironmental analysis of Ediacaran strata in the Catalina Dome, Bonavista

Peninsula, Newfoundland. ……………………………………………………………………....17

2.1 Abstract………………………………………………………………………………………18

2.2 Introduction……………………………………………………………………………….….19

2.3 Methodology…………………………………………………………………………………24

2.4 Facies………………………………………………………………………………………...29

2.5 Facies associations…………………………………………………………………………...36

2.5.1 FA1……………………………………………………………………….36

2.5.2 FA2……………………………………………………………………….38

2.5.3 FA3……………………………………………………………………….39

2.5.4 FA4……………………………………………………………………….40

2.5.5 FA5……………………………………………………………………….40

2.5.6 FA6……………………………………………………………………….41

2.6 Ripple mark paleocurrents…………………………………………………………………...41

2.7 Frond fossil paleocurrents……………………………………………………………………45

2.8 Paleoenvironment and basin evolution………………………………………………………46

2.9 Paleobiological implications…………………………………………………………………50

2.10 Conclusions…………………………………………………………………………………51

Chapter Three: Two new small fronds from Mistaken Point, Newfoundland…………………54

3.1 Abstract………………………………………………………………………………………55

3.2 Introduction…………………………………………………………………………….…….55

3.3 Ediacaran frond morphology………………………………………………………………...59

3.4 Geological setting……………………………………………………………………………62

3.5 The Mistaken Point Assemblage……………………………………………………………..64

3.6 Methods……………………………………………………………………………….……...66

3.7 Systematic paleontology……………………………………………………………………..70

3.8 Other small fronds at Mistaken Point………………………………………………………..78

vii

3.9 Discussion……………………………………………………………………………………80

3.10 Conclusions…………………………………………………………………………………83

Chapter Four: General Conclusions…………………………………………………………....84

4.1 Objectives of Thesis………………………………………………………………………….85

4.2 Sedimentology of the Catalina Dome………………………………………………………..85

4.3 Small fronds of Mistaken Point……………………………………………………………...86

References……………………………………………………………………………………….88

viii

LIST OF FIGURES

Figure 1.1 Temporal and geographic distribution of select Ediacaran taxa……….……………...4

Figure 1.2 Map of Eastern Newfoundland showing important Ediacaran fossil sites……………6

Figure 1.3 Rangeomorph architecture and construction……...…………………………………11

Figure 1.4 Correlated Ediacaran stratigraphy of the Avalon and Bonavista peninsulas……...…14

Figure 1.5 Stratigraphic ranges of Ediacaran taxa extended by discovery in Catalina Dome...... 15

Figure 2.1 Location Map of the study area in the Catalina Dome………...... 20

Figure 2.2 Fossil photographs from the Catalina Dome……………………………………...…23

Figure 2.3 Stratigraphy of the Catalina Dome and the eastern Avalon Peninsula……………....25

Figure 2.4 Detailed stratigraphic logs of measured section of the Catalina Dome….…………..27

Figure 2.5 Representative examples of facies and facies associations……………………….....30

Figure 2.6 Composite stratigraphic section based on the measured section…………………….37

Figure 2.7 Rose diagrams showing distribution of paleocurrent directions…………………….43

Figure 2.8 Comparison of turbidite ripple mark paleocurrent directions……………………….44

Figure 3.1 Location map of the Mistaken Point area….…………………………………….…..57

Figure 3.2 Stratigraphy of the eastern Avalon Peninsula……………………………….……….58

Figure 3.3 Generalized small, stemmed frond……………..……………………………………60

Figure 3.4 Photographs of newly described small fronds……………………………….….…...71

Figure 3.5 Photographs of small fronds previously described from other areas…………..…….79

Figure 3.6 Morphometric chart comparing petalodia dimensions……………………….….…..82

LIST OF TABLES

Table 1.1 Proterozoic stratigraphy of the Avalon Peninsula……………………………………...8

Table 2.1 Facies……………………………………………………………………………….....31

Table 2.2 Facies associations…………………………………………………………………….32

Table 3.1 Measurements of specimens of new small frond taxa………………………………...69

CHAPTER ONE:

General Introduction

1.1 Importance of the Ediacaran Period

The Ediacaran Period (635-541 Ma) is the most recent official addition to the geologic

timescale (Knoll et al. 2004; 2006; Narbonne et al. 2012). It spans a crucial interval between the

globally-extensive glaciations of the informally named Cryogenian Period (850-635 Ma) – the

largest glaciation events in the history of the Earth (Hoffman et al. 1998; Hoffman and Schrag

2002; Hoffman and Li 2009) – and the beginning of the Phanerozoic Eon, when abundant shelly

fossils, pervasive bioturbation, and representatives of virtually all known animal phyla all begin to occur in abundance in the fossil record in a relatively short (10-20 million years) span of time in what's termed the “Cambrian Explosion” (Knoll and Carroll 1999; Erwin et al. 2011).

Designation of the Ediacaran Period reflected a 21st century solution of a 150-year-old problem popularly called “Darwin’s Dilemma”, the apparent absence of pre-Cambrian fossils that might have been precursors to the rich biological record of the Phanerozic – a problem that

Charles Darwin considered a major challenge to his theory of the origin of species by natural selection (Darwin 1859). When the long-expected terminal Neoproterozoic (Ediacaran) animal fossils were discovered in the 20th century (e.g. Gürich 1930; Sprigg 1947; 1949; Ford 1958) they were eagerly welcomed and classified as members of extant phyla, as they were believed to be the direct ancestors of modern organisms that Darwin predicted would be found. For example, many Ediacaran taxa were interpreted as cnidarian medusae or sea pens (e.g. Sprigg 1949;

Glaessner and Wade 1966; Anderson and Conway Morris 1982).

Seilacher (1989; 1992) challenged this dominant paradigm on the basis of problematic taphonomy and constructional morphology, and proposed that Ediacaran organisms belonged to

2 an extinct kingdom he termed the Vendobionta. While this view was not well-accepted, the criticisms of the previous classifications were difficult to ignore, and they catalyzed numerous other reinterpretations of Ediacaran affinities covering a diverse range of high-level clades, from bacterial colonies (Steiner and Reitner 2001) to fungi or fungi-grade organisms (Peterson et al.

2003), to terrestrial lichens (Retallack 1994, 2013a). Today, Ediacaran fossils are found all over the world (Narbonne 2005; Fedonkin et al. 2007), and while it is generally accepted that some taxa were probably stem-group metazoans related to extant phyla (Fedonkin et al. 2007; Xiao and Laflamme 2009; Sperling et al. 2011), many more taxa cannot convincingly be referred to any modern phyla. Figure 1.1 summarizes temporal and geographic distribution of the Ediacara biota as well as clades in which Ediacaran taxa have been classified.

Ediacaran paleontology as it exists today is relatively young field of study with many significant questions still without consensus, such as the affinities of the biota, their modes of life, the circumstances that led to their evolution after billions of years of microbial life, and the causes of their extinction. Ediacaran paleontology, sedimentology, geochemistry, and other fields of geology and biology are all essential to solving these problems, which have attracted great interest over the last decade. It is a time period of great interest both for understanding the origins of complex life on Earth, and for understanding the conditions necessary for the evolution of complex life in general that may be applied to the field of astrobiology.

Cambrian

Rangeomorphs Erniettomorphs Schwarzrand 545 Khatyspyt Nama Assemblage 550 Kuibis Discoidal Octoradial Bilateral Triradial etraradial T Pentaradial Ediacara Ernietta Cloudina 555 Zimmie Gory Arkarua White Sea Assemblage orgia endia Phyllozoon Y V Eoandromeda 560 Charnwood Rangea Spriggina entogyrus

Fermeuse Kimberella Conomedusites Pteridinium Dickinsonia V Swartpuntia ribrachidium Parvancorina Trepassey T Ediacara biota 565 Mistaken Point Eoandromeda Ernietta Assemblage Dickinsonia

570 Aspidella Palaeopascichnus Hiemalora Bradgatia Kimberella Cloudina Charnia Fractofusus valon A Briscal Charniodiscus Ediacaran Period Tribrachidium 575 Drook Doushantuo biota

Charniodiscus 580 Fractofusus Gaskiers Glaciation (582 Ma)

Doushantuo-Pertatataka acritarchs Doushantuo embryos

635 Lantian biota Marinoan Snowball Earth Glaciation (635 Ma)

Cryogenian

Figure 1.1 Temporal distribution and stratigraphic occurrence of representative taxa of the Ediacara biota. (From Xiao and Laflamme 2009; Narbonne et al. 2012)

4 1.2 The Ediacaran of Avalon Peninsula

The Avalon Peninsula of Newfoundland, Canada (Fig. 1.2), which forms part of the

Avalon Terrane (O'Brien et al. 1983; Myrow 1995), hosts the first described and the oldest

Ediacaran fossils known anywhere. The first Ediacaran fossil taxon described was Aspidella

terranovica, a centimetre-scale disc-shaped fossil described by Billings (1872) in downtown St.

John's in the northeast of the Avalon Peninsula. For many years it was dismissed as a

pseudofossil because it was believed that no macroscopic life existed before the Cambrian.

However, after the discovery of fossils such as Charnia and Dickinsonia in latest Neoproterozoic

of England (Ford 1958) and Australia (Sprigg 1947, 1949), and especially the discovery of clear

complex, macroscopic fossils at Mistaken Point on the Avalon Peninsula of Newfoundland,

Canada in the late 1960s (Anderson and Misra 1968), Aspidella is now well-accepted as an important Ediacaran body fossil with worldwide distribution (Gehling et al. 2000).

Mistaken Point in the southeastern part of the Avalon Peninsula (Fig. 1.2) is one of the most important Ediacaran fossil sites in the world due to its diversity, age, paleoenvironment, and preservational style (Narbonne and Gehling 2003; Narbonne et al. 2007). The area is home to the oldest-known probable animal fossils (Narbonne and Gehling 2003; Clapham et al. 2004;

Laflamme et al. 2007; Liu et al. 2012), including entire communities of deep-marine benthic organisms preserved in situ beneath beds of volcanic ash (Seilacher 1999; Clapham and

Narbonne 2002; Clapham et al. 2003; Darroch et al. 2013). The fossiliferous succession of the

Mistaken Point area directly overlies the Gaskiers Formation, which consists of glacial diamictite with a pronounced carbonate cap (Eyles and Eyles 1989; Myrow and Kaufman 1999) dated at

582 +/- 1 Ma (Narbonne et al. 2012). The Gaskiers Formation is associated with significant

B Eastern Newfoundland A 100 km

Catalina Newfoundland

Bonavista Peninsula

St. John’s

Spaniard’s Bay

Avalon Peninsula

50km Mistaken Point

Figure 1.2. Location map. (A) The island of Newfoundland with the Avalon Terrane represented in darker grey. (B) Eastern Newfoundland with important Ediacaran fossil sites on the Avalon and Bonavista peninsulas labeled with black circles.

6 changes in redox indicators such as iron speciation and sulfur isotopes that are interpreted as reflecting a shift from anoxic to oxic conditions in the water column (Canfield et al. 2007). The oldest-known large, architecturally complex fossils occur in the overlying Drook Formation

(Narbonne and Gehling 2003; Canfield et al. 2007) and are dated at 578.8 +/- 0.5 Ma (van

Kronendonk et al. 2008; Narbonne et al. 2012a), implying a causal relationship between glaciation, oxidation, and the origin of large multicellular life (Narbonne 2010, 2011).

1.2.1 Sedimentology of the Ediacaran of the Avalon Peninsula

The Ediacaran stratigraphy of the Avalon Peninsula (Table 1.1) was originally described and mapped by King (1988; 1990). The Conception Group consists of deep-water siliciclastic deposits that noncomformably overlie the mainly igneous rocks of the Harbour Main Group

(631-606 Ma; Krogh et al. 1988). The Conception Group comprises, from bottom to top, the non-fossiliferous, mudstone-dominated Mall Bay Formation; diamictite, and locally cap carbonate, of the Gaskiers Formation; the Drook Formation, which is lithologically similar to the

Mall Bay Formation in representing mainly muddy turbidites; the sandstone-rich turbidites of the

Briscal Formation (restricted to the eastern Avalon Peninsula); and the richly fossiliferous

Mistaken Point Formation. The overlying St. John's Group (Table 1.1) consists of the Trepassey

Formation, which is made up of mudstone-dominated turbidites with sparse Mistaken Point-style and slumped beds; the slumped, mudstone-dominated Fermeuse Formation, which contains only relatively simple fossils such as Aspidella; and the more sandstone-rich Renews Head Formation.

Broadly, the succession represents the infilling of a forearc basin (Wood et al. 2003); the depositional environment of the succession transitions from a flat basin plane to a slope and

Table 1.1. Proterozoic stratigraphy of the Avalon Peninsula. After King et al. 1988; Narbonne et al. 2005.

Group Formation Lithology (thickness in metres) (thickness in metres) Upper part: thick-bedded, buff weathering grey sandstone and CAPE BALLARD quartz granule conglomerate. Lower part: grey to purple shale (>260) and grey siltstone (50 m).

Thin- to medium-bedded grey and red sandstone. Red, wavy- FERRYLAND HEAD bedded sandstone and shale (High Rocks Member) locally at (250) SIGNAL HILL base. (>1500) GIBBETT HILL Thick-bedded, light grey sandstone, thin-bedded, dark grey (760) sandstone and siltstone, local calcareous sandstone ellipsoids.

CAPPAHAYDEN Laminated, fissile, light grey siltstone. (175)

Local erosional disconformity

RENEWS HEAD Thin, lenticular bedded, dark grey sandstone and minor shale. (300)

ST. JOHN’S FERMEUSE Grey to dark grey and black shale, thin lenses of buff weathering (~2000) (1400) sandstone and siltstone.

TREPASSEY Medium- to thin-bedded, graded, grey sandstone and shale. (250)

MISTAKEN POINT Medium-bedded, grey to pink sandstone, green, purple, and red (400) shale, minor tuff.

BRISCAL Thick-bedded grey sandstone, green to grey argillite, red (100-1200) sandstone and arkose, locally grey, thin-bedded siltstone and shale.

CONCEPTION DROOK White weathering, green, grey, and buff, and locally red to purple (>4000) (1500) argillaceous chert, siliceous siltstone, sandstone, silicified tuff, locally thick sandstone with shale, siltstone, and minor purple argillite.

GASKIERS Grey diamictite, intercalated rhythmites of mustone, siltstone, (250-300) and sandstone with dropstones and conglomerate.

MALL BAY Green siliceous siltstone and argillite, grey sandstone, black, (>800) green, and purple argillite and chert, tuffaceous sandstone, green siliceous tuff and agglomerate, white quartoze sandstone and minor limestone. Angular Unconformity HARBOUR MAIN Red, pink, and grey silicic tuff, agglomerate, pink to red rhyolite (>1500) and welded tuff; massive dark green to purplish basalt.

8 ultimately a delta as it shallows upward into the overlying Signal Hill Group (Williams and King

1979; King 1990; Narbonne et al. 2001).

Detailed sedimentological study of the part of the succession that contains complex

fossils has been done both in the Mistaken Point area in the southeastern part of the Avalon

Peninsula (Wood et al. 2003) and the Spaniard's Bay area in the northwestern part of the Avalon

Peninsula (Ichaso et al. 2007), both by MSc students at Queen's University. Both studies

concluded that the depositional environment was very deep marine, well below storm wave base,

based on the presence more than 1000 metres of stacked, fine-grained turbidites lacking any evidence of wave-generated structures or exposure. These depths imply an aphotic environment in which organisms could not have relied on photosynthesis as a mode of life. Another observation drawn from these studies was the transition from basin plane to slope deposition upward through the succession. Frondose fossils were predominantly oriented in a direction perpendicular to the down-slope direction as inferred from turbidite ripples, implying the presence of a weak contour-parallel current that could have served to aerate this deep-water basin

(Wood et al. 2003). Based on studies near Spaniard’s Bay in the northern part of the Avalon

Peninsula (Fig. 1.2), Ichaso et al. (2007) inferred a basin that was more topographically complex than previously realized, with an area they referred to as the Harbour Main High between the

Spaniard's Bay and Mistaken Point areas which caused ponding of turbidity currents and thick- bedded mudstone deposition in the northwestern Avalon Peninsula.

1.2.2 Paleontology of the Mistaken Point Biota

The Mistaken Point assemblage is dominated by fossils of organisms belonging to an

extinct group called the Rangeomorpha (Figs. 1.1, 1.3). Rangeomorphs have a unique

architecture, with bodies constructed out of repeating modules (Fig. 1.3A), where each module

consists of a branching unit exhibiting at least three orders of self-similar (fractal) branching

(Narbonne, 2004). These modules were used to construct the wide array of species seen in the

Mistaken Point biota (Fig. 1.3B). The fractal structure could have increased the organisms' surface-area-to-volume ratio sufficiently that osmotrophic feeding on dissolved organic carbon was a viable mode of life (Laflamme et al. 2009). Rangeomorphs described from Mistaken Point include the bush-like Bradgatia (Flude and Narbonne 2008) originally described from

Charnwood Forest, England (Boynton and Ford 1995), the reclining spindle-shaped Fractofusus

(Gehling and Narbonne 2007), the comb-shaped Pectinifrons (Bamforth et al. 2008), the stemless frond Charnia (Laflamme et al. 2007; originally described from Charnwood Forest,

England by Ford 1958), and the stemmed frond Culmofrons (Laflamme et al. 2012). The frondose body plan is a particularly common one, also represented by the non-rangeomorph taxon Charniodiscus (Laflamme et al. 2004; also originally described from Charnwood Forest;

Ford, 1958). Because Charniodiscus lacks rangeomorph architecture, its similar shape to rangeomorph fronds such as Culmofrons is an example of convergent evolution. Another common type of fossil in the Mistaken Point assemblage are ivesheadiomorphs, which consist of large, irregular, lobate fossils whose origins are controversial (e.g. Boynton and Ford 1995;

Laflamme et al. 2011; Liu et al. 2011; Wilby et al. 2011).

The mode of preservation of fossils in the Mistaken Point area is also important because entire communities of Ediacaran organisms were preserved in situ beneath beds of volcanic ash

(Seilacher, 1999; Narbonne 2005). Preservation of these “census populations” permitted

Figure 1.3. (A) Rangeomorph architecture: elements formed by self-similar branching at several levels of magnification. From Narbonne et al. 2009. (B) Rangeomorph construction: modular construction of rangeomorph elements into distinct forms in different taxa: (1): Pectinifrons, (2): Bradgatia, (3) Fractofusus, and (4), Charnia. From Laflamme and Narbonne 2008a.

11 Clapham and Narbonne (2002), Clapham et al. (2003), and Darroch et al. (2013) to apply

ecological methods normally used only when analyzing modern living communities and to

compare these Ediacaran communities with Phanerozoic and modern marine communities.

1.3 Ediacaran succession on the Bonavista Peninsula

The Bonavista Peninsula of Newfoundland, northwest of the Avalon Peninsula, is also part of the Avalon Terrane (Fig. 1.2). When its geology was first mapped, almost the entire peninsula was assigned to the Musgravetown Group (Jenness 1963), a terminal Neoproterozoic succession characterized primarily by terrestrial deposits such as continental red beds and subaerial volcanics (O'Brien and King 2005; Normore 2010). In 2002, however, O'Brien and

King described marine and deltaic strata from the northeastern part of the peninsula that was found to be correlative with Ediacaran stratigraphy from the Avalon Peninsula: the lowermost

Signal Hill Group, the St. John's Group, and, exposed only in a small dome near the town of

Catalina termed the Catalina Dome, the uppermost Conception Group. This correlation was confirmed with the 2004 discovery of diverse, complex Ediacaran fossils similar to those known from Mistaken Point in the Catalina Dome (O'Brien and King 2004a; O'Brien et al. 2006).

Despite this correlation, there are some important stratigraphic differences between the

Ediacaran marine successions of the Avalon Peninsula and Bonavista Peninsula. On the

Bonavista Peninsula, the succession is thinner, and it lacks the Briscal Formation of the eastern

Avalon Peninsula. The Mistaken Point Formation at the top of the Conception Group and the

Trepassey Formation at the base of the St. John's Group on the Bonavista Peninsula are each

locally divided into two informal members (O'Brien and King 2005). This stratigraphy is

summarized in Figure 1.4. Perhaps most importantly from a paleontological perspective, the ash beds that occur throughout the Mistaken Point and Trepassey formations on both the Avalon and

Bonavista peninsulas, which are responsible for preservation of the diverse, complex Ediacara biota of Newfoundland, also persist upward into the Fermeuse Formation of the Catalina Dome.

The biota from the underlying strata persist into the Fermeuse Formation where these ash beds occur, which extends the stratigraphic range of the biota and provides an important taphonomic window into that formation.

Hofmann et al. (2008) described the paleontology of the Catalina Dome in detail, which included numerous Ediacaran taxa known from the Avalon Peninsula, as well as taxonomic description of several new taxa. The stratigraphic ranges of many known taxa, such as Charnia,

Charniodiscus, and Fractofusus, were extended upward in the succession on the basis of their discovery in younger strata in the Catalina Dome (Fig. 1.5). Newly described taxa included

Parviscopa and Primocandelabrum – two frond fossils only represented by small specimens.

As well as containing previously unknown species and extending the taxonomic range of known Ediacaran taxa, these findings effectively doubled the known size of the Conception

Basin in which the oldest known complex, deep marine organisms lived.

Catalina Dome, Eastern Avalon Peninsula Eastern Bonavista Peninunsula

Signal Hill Gp. (top not exposed) Signal Hill Gp. (>3000m) U-Pb (Ma) Renews Head Fm. Renews Head Fm. 100m 500m s Gp. Fermeuse Fm. Fermeuse Fm. s Gp. St. John’ St. John’ Trepassey Fm. Port Union Mb. 565 +/- 3 Mistaken Point Fm. Catalina Mb. repassey Fm. T Murphy’s Cove Mb. Briscal Fm. Goodland Point Mb Mistaken Point Fm. Shepherd Point Mb 579 +/- 1

Conception Gp. (base not exposed) Drook Fm.

Drook Fm. Conception Gp.

582 +/- 1 Gaskiers Fm. 584 +/- 1

Mall Bay Fm.

Harbour Main Gp. (base not exposed)

Figure 1.4. Correlated Ediacaran stratigraphy of the eastern Avalon Peninsula and Catalina Dome, Bonavista Peninsula.

14 Figure 1.5. Stratigraphic range of key taxa from the Ediacaran of eastern Newfoundland. Boxes represent stratigraphic ranges that were extended by the discovery of those taxa on the Bonavista Peninsula. From Hofmann et al. 2008.

15 1.4 Purpose of Thesis

This thesis consists of two projects, each of which contributes to a greater understanding

of the Ediacara biota of eastern Newfoundland. Chapter Two documents the previously unknown

sedimentology and basin evolution of Ediacaran strata on the Bonavista Peninsula to compare and contrast with previous studies of correlative strata on the Avalon Peninsula, to further constrain the development of the Conception Basin and the environment of the Mistaken Point biota. This study doubles the known strike length of the basin, and provides a more refined understanding of the complexity and evolution of the basin that was the environment of the world’s oldest-known large and complex eukaryotes.

Hofmann et al. (2008) described two new genera of small fronds from the Bonavista

Peninsula and Narbonne et al. (2009) described small fronds from Spaniard’s Bay in the northern

Avalon Peninsula. Small fronds had previously been mentioned informally from Mistaken Point

(Clapham and Narbonne 2002; Clapham et al. 2003) but no systematic study had been carried out on these forms. Chapter Three is an examination of the previously unstudied small fronds at

Mistaken Point, to compare the taxonomic framework and importance of small fronds on the

Bonavista Peninsula, Spaniard’s Bay, and Mistaken Point, and ultimately to determine their role in these early deep-water ecosystems at the dawn of complex multicellular life. Taxonomic description of the small fronds in this chapter essentially completes the taxonomy of the

Mistaken Point assemblage and permits full taxonomic comparisons with other Ediacaran assemblages worldwide.

CHAPTER TWO:

Paleoenvironmental analysis of Ediacaran strata in the

Catalina Dome, Bonavista Peninsula, Newfoundland

2.1 Abstract:

Ediacaran strata in the Conception and St. John's groups that are exposed in the Catalina Dome,

eastern Newfoundland, comprise a succession that is thinner but otherwise broadly similar to that

known from the well-studied outcrops near Mistaken Point in southern Avalon Peninsula and

Spaniard's Bay in northern Avalon. In all of these areas, strata consist of turbidites deposited in deep-water basin-plain and slope environments, but important differences help to constrain

interpretations of basin history and Ediacaran paleobiology of eastern Newfoundland. A turbidite

paleocurrent shift from easterly to southerly is consistent with the existing two-phase tectonic

model for basin evolution previously proposed for the Avalon Peninsula. In the Catalina Dome,

however, this shift occurred stratigraphically higher than at Mistaken Point but lower than at

Spaniards Bay in the Avalon Peninsula. Probable seismoturbidites are common in the lower part

of the Catalina succession, suggesting particularly active tectonism. Except in the very lowest 20

m of the succession (Drook and lowermost Mistaken Point formations) in which ash is absent,

volcanic ash beds are both more common and more volumetrically significant throughout the

succession than farther to the south and east, which suggests that deposition occurred closer to

the volcanic arc. Volcanic ash beds persist higher stratigraphically, occurring within the

Fermeuse Formation, which here contains diverse rangeomorph fossils in contrast with the low-

diversity assemblage of Ediacaran discs prevalent in the Fermeuse Formation of the Avalon

Peninsula. This distribution strongly reaffirms the importance of taphonomy in controlling the

composition of deep-water Ediacaran assemblages.

2.2 Introduction

The Avalon Terrane of Newfoundland (O'Brien et al. 1983; 1988; Myrow 1995) consists

primarily of Neoproterozoic to early Paleozoic strata, which cover much of the Bonavista,

Avalon, and Burin peninsulas (Fig. 2.1A). These strata were deposited on and adjacent to

Avalonia, a microcontinent that formed as a volcanic arc off the coast of Gondwana and was later accreted to eastern Laurentia and western Europe (Nance et al. 2002; Pisarevsky et al. 2011).

Ediacaran strata of the Conception, St. John's, and overlying Signal Hill groups are well exposed on the Avalon Peninsula (Williams and King 1979; King 1988), and comprise a siliciclastic and volcaniclastic succession that records the gradual infilling of a forearc basin

(Narbonne et al. 2001; Wood et al. 2003). The Conception Group consists of turbidite and glacial

strata, deposited in a deep, basin-axis or basin-plain setting that passes upwards into deep-slope

turbidites of the Mistaken Point Formation. This succession is overlain by upper slope deposits

of the St. John's Group, and finally by shallow marine, deltaic and terrestrial alluvial fan deposits

of the Signal Hill Group (Williams and King 1979; King 1990; Narbonne et al. 2001).

Paleocurrent measurements from both the Mistaken Point and Spaniard's Bay areas of the Avalon

Peninsula (Fig. 2.1A) record a stratigraphic transition from dominantly eastward transport to

southward paleoflow, with a simultaneous decrease in the prevalence of ash beds (Wood et al.

2003; Ichaso et al. 2007). These observations are consistent with a proposed two-phase tectonic

model based on structural and geochronological data (Murphy et al. 1999, Nance et al. 2002). In

this model, deposition originally occurred in a forearc basin setting, least (in present-day

coordinates) of the active, ash-generating arc, and west of a subduction zone. This phase was

followed by a diachronous transition to strike-slip tectonism. The basin plain and easterly

prograding lower slope would have eventually been overridden by a slope prograding from the

A Eastern Newfoundland FA3 B Catalina FA1 (study area) FA2 Bonavista Peninsula Shepherd Point Catalina St. John’s FA4 Spaniard’ s Bay F rinity Bay Avalon T Peninsula Goodland Point Burnt Point

50km Mistaken Point B FA2 A FA5 Catalina Harbour Murphy’s Cove FA3

Green Island FA4

C FA5 Port Union

D Back Cove FA6 E 500m N

LEGEND: Thick-bedded, muddy, coarsening FA1 Thinly laminated, muddy, Shepherd FA4 Point association upward, Catalina association Variegated, seismically disturbed, Thick-bedded, sandy, Port Union FA2 FA5 Goodland Point association association Thick-bedded, muddy, ash-rich, Murphy's Thin-bedded, muddu, slumped, FA3 Cove association FA6 Fermeuse association

A Measured section Fault

Figure 2.1. (A) Location map of eastern Newfoundland, including the Bonavista and Avalon peninsulas, showing positions of the study area (Catalina), Spaniard's Bay, Mistaken Point, and St. John's. (B)Map of the study area, Catalina Dome, showing distribution of facies associations and measured sections (after O'Brien and King 2005).

20 north due to this tectonic shift. Ichaso et al. (2007) additionally suggested that the basin was more topographically complex than previously realized, and that a topographic barrier, named the Harbour Main High, existed between the Spaniard's Bay and Mistaken Point areas. This high was responsible for the ponding of turbidity currents in basinal areas, resulted in unusually thick- bedded muddy turbidites.

The Conception and St. John's groups on the Avalon Peninsula are world-renowned for the Ediacaran fossils of the Mistaken Point assemblage (Billings 1872; Anderson and Misra

1968; Gehling et al. 2000; Narbonne 2005; Narbonne et al. 2007), which first appear coincident with a major oxidation event at the end of the Gaskiers glaciation 582 million years ago

(Canfield et al. 2007), and range throughout the remainder of the Conception and St. John's groups. These fossils represent some of the oldest architecturally complex macrofossils known anywhere (Narbonne and Gehling 2003; Narbonne 2011). Most are rangeomorphs, an extinct clade of possible stem-group animals (Narbonne 2004, 2005; Xiao and Laflamme 2009; Erwin et al. 2011), with some possible crown group sponges (Sperling et al. 2011). The Mistaken Point assemblage was the first deep-water occurrence of the Ediacaran biota reported anywhere (Misra

1971), an interpretation supported by detailed sedimentological analyses carried out across the

Avalon Peninsula (Wood et al. 2003; Ichaso et al. 2007). These studies concluded that the rocks in which the Newfoundland Ediacaran fossils are found were deposited at such great depths that photosynthesis would not have been possible, ruling out some hypotheses concerning the affinities of the Mistaken Point assemblage. Wood et al. (2003) and Ichaso et al. (2007) further demonstrated the prevalence of weak contour currents in the Mistaken Point Formation, indicated by the close alignment of benthic, frondose fossils in an orientation perpendicular to the turbidite transport direction, which was presumed to be downslope. These bottom currents

21 may have been an important mechanism for supplying nutrients to the biota, which may have fed osmotrophically on dissolved organic carbon (Laflamme et al. 2009). The deep-water Mistaken

Point assemblage of the Avalon Peninsula and roughly coeval fossils in central England (Brasier and Antcliffe 2009; Wilby et al. 2011) differ significantly in composition from the classic, younger, shallow-water assemblages described from Australia, northern Europe, and Namibia

(Grazhdankin 2004; Gehling et al. 2005; Narbonne 2005; Fedonkin et al. 2007), and these differences help to elucidate the evolutionary and environmental conditions in which early large and complex life diversified.

Until relatively recently, the geology of the Bonavista Peninsula had not been thoroughly studied or mapped. Almost the entire peninsula was assigned to the Musgravetown Group

(Jenness 1963), a terminal Neoproterozoic succession characterized by shallow marine and terrestrial strata and subaerial volcanics (O'Brien and King 2005; Normore 2010). Within the last decade, O'Brien and King (2002, 2004b, 2005) discovered that the stratigraphy in the northeastern end of the peninsula was more similar to the Conception and St. John's groups, and the base of the Signal Hill Group, all previously known only from the Avalon Peninsula, and roughly coeval with the Musgravetown Group. The discovery of distinctive Ediacaran body fossils on the Bonavista Peninsula (Fig. 2.2), including numerous taxa known from Mistaken

Point and other localities on the Avalon Peninsula (O'Brien and King 2004a; O'Brien et al. 2006;

Hofmann et al. 2008), added further support to this correlation. The Fermeuse Formation of the

St. John's Group is widely exposed across the eastern portion of the Bonavista Peninsula, as well as the overlying Renews Head Formation and Signal Hill Group. However, the Mistaken Point and Trepassey formations, and the underlying Drook Formation are only exposed in the Catalina

Dome, a small structural culmination several kilometres across near the town of Catalina (Fig.

5 cm

Figure 2.2. (A) Fossil-bearing surface with abundant Aspidella 3 m above the base of the Murphy's Cove Member, Mistaken Point Formation. (B) Segment of FS7 (see Fig. 2.4 for stratigraphic location), the Discovery Surface, Murphy's Cove Member, Mistaken Point Formation.

23 2.1B). O'Brien and King (2005) subdivided these formations into members unique to that area

(Fig. 2.3). The succession in the Catalina Dome is dominated by fine-grained turbidites and thin

interbeds of volcanic ash, beneath which Ediacara-type fossils of soft-bodied organisms are

preserved. O'Brien and King (2005) and Hofmann et al. (2008) observed that the complex,

Mistaken Point-style biota persists stratigraphically higher into the Bonavista Peninsula

succession than in correlative strata on the Avalon Peninsula.

The present study describes the detailed stratigraphy and sedimentology of the

fossiliferous Ediacaran succession in the Conception and St. John's groups in the Catalina Dome

area on the Bonavista Peninsula. The work reported here provides a third data set to complement

the studies in the Mistaken Point area on southern Avalon (Wood et al. 2003) and in the

Spaniard's Bay are on the northern Avalon (Ichaso et al. 2007), which helps to constrain further the paleogeography of the basin and to test previous correlations. When corrected for tectonic compression and transpression, the study area is north and west of the Spaniard's Bay site, separated from it by the same distance and in the same direction as Spaniard's Bay is from

Mistaken Point (about 100 km, Fig. 2.1A). The two-phase tectonic model predicts that the

Bonavista Peninsula study area would have been closer to both the volcanic arc to the west, and the strike-slip-related highland to the north (Calon 2001) which served as the sediment source for the overlying Signal Hill Group. This study provides an opportunity to evaluate the effects of this transition on Ediacaran sedimentation and paleontology in this region.

2.3 Methodology

Well-exposed, generally continuous coastal exposures were examined in the southern end of the Catalina Dome in the northeast of Newfoundland's Bonavista Peninsula. Strata were

Catalina Dome, Eastern Avalon Peninsula Eastern Bonavista Peninunsula

Conception Gp. (base not exposed) Drook Fm.

Drook Fm. Conception Gp.

582 +/- 1 Gaskiers Fm. 584 +/- 1

Mall Bay Fm.

Harbour Main Gp. (base not exposed)

Figure 2.3. Stratigraphy of the Catalina Dome, Bonavista Peninsula (after O'Brien and King 2005), correlated with equivalent strata on the Avalon Peninsula (after Williams and King 1979; Myrow 1995). Note the difference in scale. U–Pb dates from Benus (1988) and Narbonne et al. (2012).

25 measured at the decimetre-scale, and bed thickness, grain size, colour, sedimentary structures,

and fossil occurrences were recorded. Five sections were measured along the coastline south of

Catalina Harbour, resulting in a virtually continuous 375 m thick composite section, from the

upper Goodland Point Member (lower Mistaken Point Formation) through the lower part of the

Fermeuse Formation (Fig. 2.4). North of Catalina Harbour, an additional 56 m thick section was

measured, beginning in the Shepherd Point Member, slightly below the base of the Goodland

Point Member and ending at a thick covered interval that obscures the contact between the

Goodland Point and overlying Murphy's Cove members. The northern section can only be

roughly correlated with the longer southern section, but its measurement allows for a more

complete picture of the Mistaken Point Formation in this area.

Following previous studies of equivalent stratigraphy on the Avalon Peninsula (e.g., King

1990; Wood et al. 2003; Ichaso et al. 2007), the turbidites that dominate the succession are

described using the classic Bouma sequence model, using the designation T for turbidite with the

subscripts A–E for the Bouma sequence divisions (Bouma 1962). TF is used to designate

interturbidite pelagic or hemipelagic sedimentation (cf. Hesse 1975). Turbidites are also divided

into three thickness categories: turbidite beds under 10 cm thick are described as thin-bedded;

turbidites 10–30 cm thick are termed medium-bedded; and any turbidites exceeding 30 cm in

thickness are termed thick-bedded.

Paleocurrent orientations were measured from Bouma TC division current ripples where ripples are exposed in three dimensions, and also from oriented fossil fronds on bedding planes.

All orientation measurements were corrected for bedding tilt and tectonic deformation, following the methods used by Wood et al. (2003), with a correction to the formula for ripple orientations described by Ichaso et al. (2007).

Section B

140

130 FA5 A

120

110

100

FS9

FA4 70

FS8 60 LEGEND:

50 A A Aspidella Section A FSn Fossil surface Fault 50 40 Ripple mark A Convoluted bedding 30 FA3 40 Rip-up clasts A Undulating bedding

30 20 FS7 Orientations of frond fossils FS6 FS2 A Ripple orientation

20 10 Facies association boundary FS5

FS4 Correlation line

FS3 F Sand Silt C Sand M Sand 0 VF Sand 10 AFS1 Massive Sandstone Section F Volcanic ash F Sand Silt C Sand M Sand 0 VF Sand 50 Seismite (mixed sandstone FA2 and mudstone) 40 Thin-bedded turbidite

30 Medium-bedded turbidite A

Thick-bedded turbidite 20

Slump with muddy matrix 10 Coarse, sandy debrite with large FA1 clasts F Sand Silt C Sand M Sand 0 VF Sand Covered interval

Figure 2.4. Detailed stratigraphic logs of measured sections from the Goodland Point, Murphy's Cove, Burnt Point, and Back Cove areas.Sandstone size scale: C, coarse; M, medium; F, fine; VF, very fine. See map (Fig. 2.1) on page 20 for locations of sections.

27 Figure 2.4. (concluded).

Section E

FS12 LEGEND: A 120 A Aspidella

FSn Fossil surface 110 Fault A Ripple mark 100 Convoluted bedding A A Rip-up clasts

90 A Undulating bedding Orientations of frond fossils

80 Ripple orientation Facies association boundary Correlation line 70 Massive Sandstone

60 Volcanic ash A

50 Seismite (mixed sandstone and mudstone) Thin-bedded turbidite 40 FA6 Medium-bedded turbidite

30 A Thick-bedded turbidite Section D Section C

20 20 A Slump with muddy matrix 100

10 A 10 Coarse, sandy debrite with large A clasts FS10 90 FS 11 A A Covered interval F Sand Silt C Sand M Sand 0 VF Sand F Sand Silt C Sand M Sand 0 VF Sand

20 FA5 Section B 10 A

160 VF Sand F Sand Silt C Sand 0 M Sand

150 F Sand Silt C Sand M Sand VF Sand

28 The occurrence of three particularly distinctive and laterally extensive beds — a

deformed, undulating bed containing a 5-cm thick ash bed at the top of the Goodland Point

Member (Fig. 2.5E): a 4 m thick massive sandstone near the middle of the Port Union Member;

and a 2.5 m thick coarse sandstone with large, isolated clasts (Fig. 2.5F) 70 m above the base of

the Fermeuse Formation — allowed the five measured sections south of Catalina Harbour to be

correlated with a high degree of confidence.

2.4 Facies

The measured section is dominated by mudstone and is largely monotonous, although thick, anomalously coarse beds exist in several intervals. Over 90% of the succession consists of turbidite beds of variable thickness, from centimetre to metre scale, which are tabular on an outcrop scale. Thin (typically millimetre-scale) beds of volcanic ash, typically decimetres apart, punctuate nearly the entire succession. Soft-sediment deformation due to both probable seismic disturbance (in the lower part of the succession) and gravitational slumping (in the upper part of the succession; see further discussion later in the text) occurs sporadically. Mudstone-rich divisions are typically darker colours, from green to blue or brown, whereas sandstone beds are lighter, commonly beige or light gray. Based on grain size, sedimentary structures, and

composition, nine facies have been identified within the measured units (Table 2.1), and are

organized into six distinct facies associations (Table 2.2). Most of these facies and facies associations also occur in the Mistaken Point (Wood et al. 2003) and Spaniard's Bay (Ichaso et al. 2007) areas, and these papers contain full descriptions and references for these facies. Facies descriptions that are unique to the Catalina Dome are described in greater detail in this chapter.

A B F7 F1/F4

F1/F2 F5

F1/F4

C F3 F2

Figure 2.5. Representative examples of facies and facies associations. (A) Slump bed (F7) overlying thin-bedded, mudstone-rich turbidites (F1/F2) representative of FA6. (B) Thick-bedded turbidites and interturbidites (F1/F4) with a 12 cm thick bed of volcanic ash (F5) representative of Fa3. Measuring stick in 10 cm increments. (C) Thin-bedded turbidites (F1) with ripple marks (F2.) Ten-centimetre scale bar. (D) Contact between Bouma TE-division and TB-division, mixed mudstone-rich and sandstone-rich turbidites, representative of FA5. Scale bar 5 cm. (E) Seismoslump (F6) with vertical folding, representative of FA2. Measuring stick in 10 cm increments. (F) Large mudstone rip-up clasts in coarse, sandstone matrix (F8). Measuring stick in 10 cm increments.

30 Table 2.1. Descriptions and interpretations of facies in the Catalina Dome, Bonavista Peninsula

Facies Bed thickness Sandstone/ Name Grain size Structures Colour Process and origin FA (cm) mudstone ratio F1 Graded mudstone 1 – 30 Silt – Vf sand Tabular bedding, normal grading, 10/90 – 30/70 Green to blue, Slow, weak, turbidity FA1, FA2, without ripple cross grading upward sharp and non-erosive basal gray, rarely currents, ponding FA3, FA4, lamination to clay mud contact, parallel lamination reddish resulting in thicker FA6 beds (Bouma TDE) F2 Graded mudstone with 2 – 50 Vf – F sand Tabular bedding, normal grading, 20/80 – 40/60 Green to blue, Moderate to weak, low FA2, FA3, ripple cross lamination grading upward sharp and non-erosive basal brown, light concentration turbidity FA4, FA5, to clay mud contact, parallel lamination, ripple gray or tan currents FA6 cross lamination, convolute

bedding (Bouma T(B)CDE) F3 Graded sandstone to 10 – 100 M sand to Vf Tabular bedding, normal grading, 20/80 – 90/10 Light gray to Medium to strong FA2, FA3, mudstone sand ripple marks, scour surfaces, tan turbidity currents FA4, FA5 muddy rip-up clasts concretions

(Bouma TAB(C)DE) 31 F4 Laminated mudstone .05 – 1 Clay – silt Thinly laminated, tabular bedding 0/100 Dark green, Hemipelagic deposition FA1, FA2, (Bouma T blue, brown between turbidite flows FA3, FA4, F) FA5, FA6 F5 Crystal tuff <1 – 90 -- Tabular bedding, heavily cleaved -- White, cream, Volcanic ash-fall FA2, FA3, and recrystallized, obscuring pink, purple events FA4, FA5, original structures FA6 F6 Mixed sandstone and 10 – 170 Clay – F sand Patches of lighter sand and darker 50/50 – 90/10 Green-blue Soft sediment FA2 mudstone mud, mm- to cm-scale, vaguely and tan deformation, probably horizontal but chaotically mixed. the result of seismic activity mixing unconsolidated turbidite beds F7 Contorted mudstone >60 Clay – silt Folded and discontinuous internal 0/100 – 20/80 Brown Gravitational slumping FA5, FA6 lamination of stratified turbidite beds F8 Coarse sandstone >200 C sand matrix, Variable, decimetre-scale rip-up 90/10 Light gray to High energy debris flow FA6 with large clasts muddy clasts up clasts of laminated mud, and tan matrix, to meters in transported cobbles, in a matrix of variable clasts length coarse sand.

F9 Massive sandstone 400 M sand Mostly structureless, indistinct 100/0 Light gray to Very strong turbidity FA5 horizontal parting tan current or grain flow Table 2.2. Description and interpreted origin of facies associations in the Catalina Dome, Bonavista Peninsula

Facies Process and origins Correlative formation and Association Name Thickness Description outcropping area Thinly laminated, Thinly laminated mudstone (F4, minor F1) with no ash. Dominantly hemipelagic Shepherd Point Member, Drook FA1 mudstone-rich, 4.9m deposition; very distal Formation; between Goodland Shepherd Point turbidity currents Point and Shepherd Point association

Variegated, Varied and laterally discontinuous association, mainly thin- and Weak to moderate Goodland Point Member, lower FA2 disturbed, Goodland ~70m medium-bedded turbidites (F1 is the most common facies, with turbidity currents, Mistaken Point Formation; at Point association some F2, F3, minor F4), as well as beds showing varying hemipelagic fallout, soft Goodland Point and Murphy's degrees of soft sediment deformation (F6). Thin ash beds (F5), sediment deformation Cove, north and south of more abundant and thicker upward, beneath which fossils are likely due to seismic Catalina Harbour

sometimes preserved. Fossil orientations inconsistent. TC ripple activity paleocurrent directions variable but generally to the east.

Thick-bedded, Medium- to thick-bedded turbidites and mudstones with little Hemipelagic deposition, Murphy's Cove Member, upper 36m FA3 mudstone and ash- sand (F1 and F4, rare F2), thick ash beds up to a metre (F5). TE and weak, low- Mistaken Point, Murphy's Cove rich, Murphy's Cove concentration turbidity and Burnt Point, south of 32 and TF divisions make up the largest volume of this association. association Fossils preserved in T layers beneath ash beds, fossil currents. High level of Catalina Harbour F volcanic activity, ash fall orientations inconsistent. T ripples very rare, only one C events measurable, to the east

Thick-bedded, Thick- and some medium-bedded turbidites: F1, F2, F3, F4 all Hemipelagic deposition Catalina Member, lower FA4 mudstone-rich, 50m present. F5 ash beds are no less common as in FA3, but are and weak to moderate Trepassey Formation, Burnt coarsening upward, generally thinner. Overall this association coarsens upward. turbidity currents. Point

Catalina association Paleocurrent from TC ripples to the east-northeast.

Thick-bedded, Medium- to thick-bedded sandy turbidites (mainly F3, F2, minor Moderate to very strong Port Union Member, upper FA5 sandstone-rich, Port 90m F1), one muddy slump near the base (F7). Single 4 m bed of turbidity currents, one Trepassey Formation, from

Union association F9, minor amounts of F5. TC division ripple paleocurrent shifts occurrence of short Burnt Point area to Back Cove from east-trending to southeast approximately 50 metres from distance slump movement the base of the Port Union Member.

Thin-bedded, Mainly thin- sometimes medium-bedded turbidites (F1 and F2, Slow and weak turbidity Fermeuse Formation, Back FA6 mudstone-rich, 178m rarely F3, minor F4) with metre-scale muddy slump beds (F7), currents, hemipelagic Cove area slumped, Fermeuse with one 2.5-metre coarse sandy bed with large clasts (F8). settling, short distance association Frequent thin ash beds (F5) preserve rangeomorph fossils with slump movement, one

inconsistent orientations. TC ripple marks indicate an overall occurrence of a high south/southeast paleo-flow, but within several metres of slump energy debris flow beds, paleocurrent measurements are more scattered and unreliable due to rotation. The turbidites in the succession can be divided into three facies. F1, graded mudstone

without ripple cross-lamination, is the most abundant facies (Figs. 2.5A, 2.5B). It consists of

Bouma TDE beds — parallel laminated silty mudstone overlain by structureless mudstone. F1

beds are thin- to thick-bedded, and fine upward. F2, graded mudstone with ripple cross-

lamination, is a slightly coarser facies, with T(B)CDE beds (Fig. 2.5C). While mudstone is

volumetrically most abundant, generally comprising 60%–80% of each bed, very fine to fine

grained sandstone at the base of beds shows current ripple cross-lamination. Beds of F3, graded

sandstone to mudstone (Fig. 2.5D), are typically medium- to thick-bedded, and contain a larger

proportion of sandstone (very fine to medium grained), with massive and (or) graded sandstone divisions overlain by parallel laminated sandstone, which grades into mud. Complete TABCDE and

TBCDE beds are common, with the TC division represented either by ripple cross-lamination or by

convolute bedding. Some beds lack clear TC divisions, or lack the upper parts of the Bouma

sequence, due to truncation by subsequent flows. Mudstone rip-up clasts are locally found at the

base of F3 beds that show erosional truncation of underlying F3 beds. F3 beds range from 0% to

60% mudstone.

Facies F4, laminated mudstone, consists of volumetrically small but paleontologically

significant interturbidite, or “TF” divisions, that overly any of the above three facies. Only in the

basal 5 m of the measured section, in the Shepherd Point Member, does thinly laminated F4

dominate the succession, and in this case it makes up 90% or more of the rock. This facies

represents the slow, pelagic and hemipelagic fallout of mud between turbidity current events.

Although not present in the lowest 20mof the succession, ash beds (F5) within F4 intervals sporadically preserve fossils of soft-bodied, benthic biota as epireliefs on bedding planes

composed of F4.

Beds or laminae of F5, crystal tuffs (Fig. 2.5B), are cream to light gray (or, less commonly, pink or purple) that persist throughout the measured section, except for the lowest 20 m. They typically occur as 1–5 mm thick laminae, but reach a thickness of slightly over a metre in rare cases. Recrystallization, cleavage, and the recessive nature of these ash beds obscure original grain size and sedimentary structures within this facies. The ash originated from a nearby volcanic arc, and is responsible for the preservation of fossil biota, in what Narbonne

(2005) termed Conception-style preservation. Because the recessive nature of these beds creates planes of weakness, the overlying layers tend to break off and erode away, creating flat benches in outcrop just beneath the ash layer.

Facies F6, mixed sandstone and mudstone, includes strata that reflect liquefaction and

(or) soft sediment deformation of originally horizontal turbidite beds (Fig. 2.5E). In some cases, bedding is complexly disrupted, with mixed lenses of darker mudstone and lighter sandstone (top of Fig. 2.5E), or even complete homogenization of the lithology. In other units, there are centimetre- or decimetre-scale, symmetrical undulations deforming originally horizontal beds

(middle of Fig. 2.5E). Some layers show both of these features: mixed and undulating bedding.

Beds of F6 range from subtle, decimetre-scale, symmetrical undulations in otherwise tabular turbidites to beds that show thorough mixing of sand and mud with only faint, residual horizontal stratification. Anomalously coarse or thick beds that could have caused rapid loading do not rest above any F6 occurrences, and load structures were not observed. There is no evidence of horizontal movement such as slump folds, slump scars, or asymmetric ball-and-pillow features

(cf. Myrow et al. 2002) in this facies, so gravity-driven slope failure is not indicated. There is no evidence of wave-generated structures in the entire succession, making a wave-generated origin for these features highly unlikely. The best explanation for the occurrence of this soft sediment

deformation in deep marine turbidites, given the lack of waves and the near-arc tectonic setting

this is likely to have been seismically active (Fuller et al. 2006), is seismic activity. The beds of

this facies can therefore be classified as probable seismites (cf. Seilacher 1969). Although seismicity can affect unconsolidated sediment in numerous ways, in fine, unlithified sediment

resting on a horizontal surface, it is known to cause non-directional undulations in bedding termed seismoslumps (Druschke et al. 2009), and homogenization of sediment called homogenites (Montenat et al. 2007). Because beds were originally turbidites prior to interpreted seismic disruption, they can collectively be referred to as probable seismoturbidites (cf. Mutti et al. 1984). The seismoslumps are primarily the result of hydroplastic deformation, and the homogenites the result of fluidization (Bezerra et al. 2001).

Facies F7, contorted mudstone (Fig. 2.3A), in contrast with F6, shows soft-sediment deformation strongly influenced by gravity-driven slope failure. Originally horizontal bedding and laminae are heavily contorted, commonly into asymmetric slump folds, forming thick

(decimetre- to metre-scale), contorted intervals. The folds are decimetres in amplitude, and beds of these folds are decimetres to over a metre thick. The asymmetric folds in these slumps imply lateral transport, and therefore deposition on a slope. Slumping might have been initiated by seismic activity, but simple failure without a specific trigger other than sediment loading is also possible. These features are remarkably similar to the slumps described by Dalrymple and

Narbonne (1996) from deepwater Ediacaran strata in northwestern Canada.

Facies F8, coarse sandstone with large clasts (Fig. 2.5F), is only found in a single 2.5 m thick bed in the Fermeuse Formation. The matrix in this bed contains the coarsest material found in the measured section (coarse to very coarse sand). The clasts, which are ungraded within the unit and centimetres to decimetres in diameter, consist mainly of mudstone rip-up clasts of

turbidites (some with lamination still visible) but also include transported extraformational

cobbles of both igneous and sedimentary origin. This unit was probably deposited as a debris

flow, and shows no evidence of bedding, lamination, or vertical grading of either the coarse

sandstone matrix or large clasts.

Facies F9, massive sandstone, occurs as a single 4 m thick bed within the Port Union

Member. This distinctive bed consists of structureless, fine- to medium-grained sandstone,

deposited either as a grain flow or as the TA division of a particularly large and high-energy

turbidity current. Deposition of massive, sandy turbidites such as this facies are products of rapid

deposition of sand from high-velocity flows (Vrolijk and Southard 1997) and have been linked to

extinction of turbulence in turbidity currents (Cantero et al. 2012).

Figure 2.4 shows the distribution of these facies throughout the measured section. The main turbidite facies, F1–F3, are not distinguished in this figure because the facies present alternate at scales of a metre or less; instead, variations in bed thickness and grain size within turbidites are represented. The finest-grained turbidites are F1, and the coarsest are F3. F4 and F5 occur throughout the succession, but generally in beds too thin to be shown in Fig. 2.4.

2.5 Facies associations

The nine facies described in the preceding section are grouped into six distinct facies

associations (Table 2.2; Figs. 2.4 – 2.6). Each of these associations is tens of metres thick, occurs

only once in the succession, and reflects member-scale variation in lithologic character of the

succession. They are therefore named for the members in which they occur.

2.5.1 FA1: thinly laminated, mudstone-rich, Shepherd Point association

Composite Fossil frond Ripple mark section: grain paleocurrent Ash bed thickness Turbidite thickness T /T Ratios Fossil occurrences LEGEND orientations ABC DE size and lithology directions A, Ch, H, Pr A A Thin-bedded turbidite

400 400 400 400 A 400 A Medium-bedded turbidite A Thick-bedded turbidite

Slump

A 350 350 350 350 350 Coarse, sandy debrite

FA 6 Seismite A A Massive sandstone Fermeuse Fm. A A A,Ch, I? 300 300 300 300 300 A, B, Ch, H, Pa Covered interval

Ripple orientation

250 250 250 250 250 Frond orientation Fault A

37 A - Aspidella 200 FA 5 200 200 200 200 B - Bradgatia Cd - Charniodiscus Ch - Charnia A F - Unidentified frond Fm - Fractofusus misrai I - Ivesheadia repassey Fm. 150 T 150 150 150 150 Ha - Hadryniscala A, Cd Hi - Hiemalora Pa - Parviscopa FA 4 Pe - Pectinifrons Pr - Primocandelabrum

100 100 100 100 100

FA 3 A A, Cd, Ch, F, Fm, Ha, Hi, I, L, Pa, Pe, Pr AA, F A, Cd, Ch Cd, Ch A, H,Ch, F, I?

50 50 50 50 50 Mistaken Point Fm. FA 2 A

m 0 m 0 m 0 Clay V Silt C. Sand m 0 M. Sand FA 1 m 0 F . Sand .f. Sand 0.1 1 10 100cm 0 20 40 60 80 100+cm 0 20 40 60 80 100

Ash bed thickness Turbidite thickness % TABC % TDE Drook Fm.

Figure 2.6. Composite stratigraphic section based on measured sections in Fig. 2.4, showing vertical variation in lithology, paleocurrent orientations, ash

content, turbidite thickness, Bouma Sequence TABC/TDE ratios, and fossils, separated into the six facies associations. This facies association is present only in the lowermost 5 m of the measured section north of Catalina Harbour, corresponding with the top of the Shepherd Point Member of the Drook

Formation (Fig. 2.3; O'Brien and King 2005). Strata in this member below the base of the measured section are of similar character. This association consists entirely of mudstone: primarily thinly laminated F4, and some thin beds of F1 muddy turbidites. There are some slight, decimetre-scale, symmetrical undulations in the bed (i.e., inferred seismoslumps of F6), especially toward the top. Volcanic ash was not observed in this association.

2.5.2 FA2: variegated, disturbed, Goodland Point association

FA2 includes most of the measured section north of Catalina Harbour: 51 m of section F, as well as the lowermost 22 m of the measured section south of Catalina Harbour in sections A and B (Fig. 2.4). This association falls within the Goodland Point Member, which has been correlated with the lower part of the Mistaken Point Formation (Fig. 2.3; O'Brien and King

2005). This association consists primarily of thin- to medium-bedded turbidites, both muddy (F1,

F2) and sandy (F3), with thin beds and laminae of interturbidite mudstone (F4) and ash (F5).

Within this association, the lowermost occurrence of ash is observed, 16 m above the base of the

Goodland Point Member. Turbidites within this association show varying degrees of disturbance, ranging from tabular F1, F2, and F3 turbidites to the undulating and homogenized probable seismoturbidites of F6 (Fig. 2.5E). Both the amount of sand (i.e., F3 turbidites) and apparent degree of disturbance are greater in the upper part of the Goodland Point Member in the area south of Catalina Harbour. FA 2 in the upper 11 m the Goodland Point association was measured in two coastal sections separated laterally by less than 500 m. While certain distinctive beds can be traced between the two sections, overall there is very little lateral continuity of beds or units

between them, possibly due to faulting. Sandstone in this part of the section is typically light

gray to beige with a greenish hue, whereas mudstone is dark and gray-green. Pervasive

silicification is characteristic of this association.

2.5.3 FA3: thick-bedded, mudstone- and ash-rich, Murphy's Cove association

The 36 m thick Murphy's Cove Member, which overlies the Goodland Point Member and correlates with the upper Mistaken Point Formation (Fig. 2.3), consists entirely of FA3. This association is the most mudstone-rich part of the succession, and is dominated by dark blue–gray or green–gray, medium- to thick-bedded F1 turbidites and F4 interturbite deposits. Only one rippled F2 bed was observed, near the base of the member, due to the general lack of sandstone.

Ash layers (F5) are abundant and thick relative to the rest of the section (Fig. 2.5B); the ash beds

of FA3 are commonly 10 cm or more thick, and are up to 1.1 min thickness. In the lowest 10 m

of the Murphy's Cove Member, FA3 contain the most fossiliferous interval of the measured

section, including a small surface densely packed with Aspidella fossils (Fig. 2.2A), and the

bedding plane known as the Discovery Surface (Fig. 2.2B; FS7 in Fig. 2.4) that is covered with

hundreds of fossils, mainly of Charniodiscus and Aspidella (Hofmann et al. 2008). Because there

is almost no sandstone in this part of the section to contrast with the mudstone, it is not possible

to identify deformed beds (if present) as in FA2. Some thick (decimetre-scale), structureless

mudstone intervals within this association that are ascribed to TE- or TF-division turbidites and interturbidites may actually be disturbed homogenites belonging to F6, in which the original beds planes or laminae have been obscured by liquefaction. Discerning such muddy homogenites from structureless TE and TF beds is difficult.

2.5.4 FA4: thick-bedded, mudstone-rich, coarsening-upward, Catalina association

FA4 corresponds to the Catalina Member, lower Trepassey Formation (Fig. 2.3). In the measured section, this member is 50 m thick, but a fault 6 m below the top may have cut out section that is present elsewhere in the Catalina Dome. Like the underlying units of FA3, FA4 consists primarily of muddy, distal turbidites. In FA4, these are mostly thick-bedded F1, with some F2 and rare F3 that increase in abundance toward the top of the member, imparting an overall upward-coarsening aspect to the upper 20 m of the Catalina Member. Mudstone in FA4 is dark blue–gray to brown, and sandstone a lighter gray or brown. Numerous F5 ash beds and laminae exist through this facies association, but most are only millimetres thick, and do not

approach the thicknesses of many of the ash beds in FA3.

2.5.5 FA5: thick-bedded, sandstone-rich, Port Union association

FA5 corresponds to the 94 m thick Port Union Member, Trepassey Formation, the most

sand-rich interval in the measured section. This facies association consists primarily of thick-

bedded, sandy turbidites (F3) alternating with thinner-bedded, relatively fine-grained deposits resembling FA4 (F1–F3 turbidites; Fig. 2.5D). Sandstone in FA5 is typically light gray to light brown, whereas mudstone is blue–gray to brown. Ash beds (F5) are present throughout, but are significantly more widely spaced than the underlying strata. Although the Port Union Member is correlated with the upper Trepassey Formation (Fig. 2.3; O'Brien and King 2005), sandy turbidite beds of the thickness and coarseness found in this facies association are unknown from the Trepassey Formation on the Avalon Peninsula. Within FA5, near the base of the member, there is a 2 m thick, muddy slumped bed (F7), which represents the first evidence of deposition on a slope in the succession. Any seismic activity that occurred during the deposition of this unit

could have resulted in failures upslope, creating slumps or typical turbidites, rather than

seismoturbidites or homogenites. The anomalously thick (4 m) massive sandstone bed of facies

F9 is likely the result of a particularly large seismic event resulting in a high-energy grain flow or

turbidity current. Other sandstone-rich intervals in FA5, especially toward the top of the member,

contain stacked metre-thick TA turbidite beds with erosional bases, and locally, mudstone rip-up

clasts. Fossils are least common in this facies association, with only rare Aspidella in the

mudstone-rich intervals.

2.5.6 FA6: thin-bedded, mudstone-rich, slumped, Fermeuse association

The thickest stratigraphic unit measured for this study, the lowest 178 m of the Fermeuse

Formation, corresponds to FA6. This association is dominated by monotonous, blue–gray to

brown, mostly thin-bedded, mudstone-rich turbidites (F1 and F2 with rare F3 beds, and thin beds

or laminae of F4 and F5 throughout) that are episodically interrupted by metre-scale mudstone- rich slump beds (F7) (Fig. 2.5A). The presence of ash beds that preserve rangeomorph fossils in the Fermeuse Formation is unique to this area, none having been observed in correlative strata on the Avalon Peninsula. The only occurrence of F8, the 2.5 m thick, very coarse debrite (Fig. 2.5F) is present within FA6. As with the thick F9 sandstone in FA5, this deposit may record a particularly intense seismic event. The muddy F7 slumps that occur throughout this association, which are composed of beds of a similar character to their surrounding strata, are the product of slope failures, which because of their great abundance indicate deposition on a slope.

2.6 Ripple-mark paleocurrents

Paleocurrent directions are a key tool for understanding large-scale basin geography and

evolution. In the Ediacaran Avalonian succession, there is a transition from broadly eastward to

broadly southward paleoflow in turbidites, as determined from TC ripple marks (Wood et al.

2003; Ichaso et al. 2007). This shift has been attributed to a two-phase tectonic model, whereby the subduction of oceanic crust that created a series of volcanic arcs to the west, which was the source of the eastward-directed flows, transitioned diachronously into a strike-slip tectonic regime (Murphy and Nance 1989; Murphy et al. 1999; Nance et al. 2002), which created a highland to the north that was the source of the southerly flows. In the southeastern Avalon

Peninsula (Mistaken Point area), the east to south paleocurrent transition occurs at the transition between the Trepassey and Fermeuse formations (Wood et al. 2003), whereas in the northwest

(Spaniard's Bay area) it occurs lower in the succession, within the Mistaken Point Formation

(Ichaso et al. 2007).

A change in paleocurrent direction is also evident in the Catalina sections on the

Bonavista Peninsula (Figs. 2.4, 2.6, 2.7), occurring approximately halfway through the Port

Union Member of the Trepassey Formation. The paleocurrent shift is gradual, beginning between

40 and 50 m above the base of the Port Union Member. Ripples below that level indicate paleocurrent flow to the east, ripples between 50 m above the base of the Port Union Member and 10 m above the base of the Fermeuse Formation are oriented to the southeast, and above that level through most of the Fermeuse Formation the ripples are aligned strongly to the south- southeast (Fig. 2.7). The shift occurs lower in the Catalina succession than it does in the

Mistaken Point area, consistent with its location farther north and therefore closer to the uplift created by strike-slip faulting (Calon 2001). Anomalously, however, it occurs higher in the

Catalina Dome succession than it does in the Spaniard's Bay area (Fig. 2.8), but this difference

0 0

270 90 270 90

180 180 F ermeuse Formation vector mean of frond Fermeuse Formation ripples, n=15 orientations of Fs12. Other fossil surfaces with

fronds in the Fermeuse Formation (Fs10 and Fs11) were locally rotated by slumping. 0

270 90 No frond fossils found in the Port Union Member

180 Upper Port Union Member ripples n=6

0 0

270 90 270 90

180 180

Catalina and lower Port Union Member Catalina Member vector mean of frond ripples, n=6 orientations, only one fossil surface with fronds

0 0

270 90 270 90

180 180 Goodland Point and Murphy’s Cove ripples, Goodland Point and Murphy’s Cove members, n=10 vector means of frond orientations by surface Figure 2.7. Rose diagrams showing the distribution of ripple mark and fossil frond paleocurrent directions in the measured section of the Catalina Dome. n, number of samples.

43 Catalina Spaniard’s Bay Mistaken Point

Fermeuse

Trepassey

Mistaken Point

Briscal/ Drook

Average ripple mark paleocurrent direction Transition from phase 1 to phase 2 tectonics

Figure 2.8. Comparison of turbidite ripple mark paleocurrent orientations in the Conception and St. John's groups in the Catalina, Spaniard's Bay, and Mistaken Point areas. Paleocurrent data from Gardiner and Hiscott (1988); Narbonne et al. (2001); Wood et al. (2003); Ichaso et al. (2007), and this work. Note the stratigraphic diachroneity of the transition from eastward to southward paleoflow that occurs in each area.

may have been caused by its position farther west where it is likely to have been influenced for a

longer period of time by flows derived from the arc to the west. Precise radiometric dates are

needed to constrain these relationships in terms of absolute time, rather than just stratigraphic

position.

2.7 Frond-fossil paleocurrents

Mistaken Point fronds such as Charnia and Charniodiscus were tethered to the seafloor at their bases with stalks that were flexible enough to allow them to bend with the current (Fig.

2.2B; Seilacher 1992, 1999), and their final orientations on bedding surfaces provides a valuable source of information about background current flow between turbidite events (Wood et al. 2003;

Ichaso et al. 2007). Fronds preserved on any one surface are strongly aligned, but the orientation

varies between bedding planes throughout the section. The E-surface and some other surfaces in

the Mistaken Point Formation at Mistaken Point show frond alignment parallel with turbidite

flow direction as inferred from the current ripples in the adjacent turbidite beds, but most

surfaces show fronds oriented at right angles to the downslope turbidite flow direction, reflecting

the presence of contour-parallel currents between turbidite events (Wood et al. 2003).

Oriented fronds are common in the Catalina Dome sections. Two fossil surfaces with

oriented fronds in the Fermeuse Formation (between 50 and 60 m above the base of the

formation; FS10 and FS11, Fig. 2.4) have been rotated by synsedimentary slumping and are not

included in this analysis, nor are the equally anomalous paleocurrents from ripple marks in this

interval. Frond orientations for the other nine surfaces show no downslope alignment of fronds

(i.e., none are parallel to the paleoflow indicated by the turbidites; Fig. 2.7), and the dominant alignment of eight of the nine fossil surfaces is contour-parallel (i.e., essentially orthogonal to

turbidite ripples). Five of these surfaces show frond alignments 90° counterclockwise to the

downslope direction, similar to that reported by Wood et al. (2003) from Mistaken Point. This

contour current direction is consistent with the Coriolis effect-induced deflection that would

occur in the southern hemisphere (Wood et al. 2003), which is the position inferred for Avalonia

(e.g., Li et al. 2008). However, three of the surfaces show contour-parallel alignment in the

opposite sense, with the fronds oriented 90° clockwise from the turbidite downslope direction.

One surface, the abundantly fossiliferous “Discovery surface” (FS7, Fig. 2.4), shows fronds that are oriented in an upslope direction.

The dispersion of frond orientations varies considerably between surfaces. On most surfaces the fronds show tight clustering around a preferred direction, as is the case on the

Discovery Surface FS7, whereas the spread of orientations on some surfaces is fairly large. For

example, four fronds on FS5, which yield an average paleoflow to the north, have an 80° spread of orientations.

We infer from these observations that, although there was a dominant contour current with a flow direction similar to that described from the Mistaken Point area, its influence was relatively weak and intermittent, with other bottom currents also influencing frond orientations.

The bottom currents responsible for the frond orientations on the four fossil surfaces that don't follow this general trend may have been the result of tidal flow or internal waves (cf. Dykstra

2012). Dissolved nutrients are believed to have been important in sustaining the organisms of the

Avalonian Ediacaran assemblage (Laflamme et al. 2009; Sperling et al. 2011), and the frond

orientation on the Discovery Surface may be consistent with upwelling and the maintenance of

the dense, thriving ecosystem preserved on this surface.

2.8 Paleoenvironment and basin evolution

The abundant and thick turbidites throughout this succession, and the absence of wave-

generated sedimentary structures, support the interpretation that the Catalina Dome succession accumulated in a deep marine depositional environment — a view consistent with the findings of previous studies of the equivalent strata on the Avalon Peninsula (e.g., Misra 1971; King 1990;

Myrow 1995; Narbonne et al. 2001; Wood et al. 2003; Ichaso et al. 2007). As discussed in the

preceding text, a broadly similar interpretation to that of previous studies, of a transition from

no-slope (i.e., basin floor) to slope deposition within a forearc basin, can be applied, but notable

differences from the Avalon Peninsula sections include the presence of sandy turbidites in the

upper Trepassey Formation, and the persistence of ash beds throughout the Fermeuse Formation.

All formations in the Catalina Dome area are thinner than correlative formations on the Avalon

Peninsula by about 75% (Fig. 2.3). This difference may be due to differential seafloor

subsidence, which is typical in forearc settings (Xie and Heller 2009).

The dominantly hemipelagic mudstone (F4), and absence of volcanogenic facies (F5) in

the Shepherd Point Member of the Drook Formation, suggest that the member may predate the

emergence of the part of the volcanic arc that would have served as the source for the sand and

ash. In contrast, in the Mistaken Point area, thick-bedded turbidites, ash beds, and associated

Ediacaran fossils extend far down into the Drook Formation (Narbonne and Gehling 2003), with

which the Shepherd Point Member has been correlated (O'Brien and King 2005). This may have

been due to differences in the prevailing wind direction, or may imply some level of diachroneity

from south to north in the emergence of the arc.

The highly variegated Goodland Point Member of the Mistaken Point Formation contains

beds that have been affected by soft-sediment deformation, commonly as subtle and pronounced

47 undulations in the bedding. These undulations are largely symmetrical, lacking any evidence of the lateral movement expected with slumping on a depositional slope. A flat basin floor depositional environment for this unit agrees with depositional interpretation of the Mistaken

Point Formation on the Avalon Peninsula (Wood et al. 2003; Ichaso et al. 2007) with which this member has been correlated. Compared with the rest of the succession, FA2 in the Goodland

Point Member is particularly variable, with both mudstone-rich and sandstone-rich turbidites ranging from centimetres to decimetres thick. It is rich in ash, and contains significantly more sandstone than the strata above and below. These characteristics suggest increased tectonic and volcanic activity in the nearby arc.

The overlying Murphy's Cove Member of the Mistaken Point Formation is marked by a shift from variegated mudstone and sandstone to a monotonous muddy interval that contains almost no sand and abundant, thick, centimetre- and decimetre-scale ash beds. This higher volume and thickness of ash compared with equivalent Avalon Peninsula strata is likely due to a greater proximity to the volcanic arc source to the west, though it could also be explained by differences in wind direction. The boundary between the Murphy's Cove Member and the overlying Catalina Member of the Trepassey Formation is not pronounced in the Catalina Dome area; the most apparent difference between the two members is that the ash beds in the higher strata are generally thinner. Presumably this change is the result of a decline in volcanic activity, since this change in the ash layers is not accompanied by any other notable change in the character of the strata.

The facies associations of both of the above members (i.e., FA3 and FA4) are characterized by unusually thick muddy turbidites. It is typical for sandy, proximal turbidites to form thick beds and muddy, distal turbidites to be thinner (Arnott 2010). In studies of the

48 correlative strata on the Avalon Peninsula (Wood et al. 2003, Ichaso et al. 2007), similarly thick- bedded muddy turbidites were noted in the Spaniard's Bay area, about 100 km north-northwest of

Mistaken Point (Fig. 2.1). Ichaso et al. (2007) suggested that this was due to ponding caused by a topographic high between the northwestern and southeastern Avalon Peninsula, potentially created by topography associated with the Harbour Main volcanics that crop out between the two sites. Because the study area falls on the same side of the proposed “Harbour Main High” as

Spaniard's Bay, this ponding hypothesis can also be applied to the Catalina Dome area.

The Catalina Member (FA4), which has been correlated with the lower Trepassey

Formation, resembles the Trepassey Formation on the Avalon Peninsula in that it consists mainly of muddy turbidites and thin ash beds. No slump beds were observed within the Catalina

Member. However, the Port Union Member (FA5), which is correlated with the upper Trepassey

Formation, is quite different from any exposure of the Trepassey Formation observed outside the

Catalina Dome, requiring a somewhat different paleoenvironmental interpretation. A muddy slump bed near the base of the member is the first clear stratigraphic evidence of slope deposition. Above that slump bed, thick, sandy turbidites and amalgamated TA beds with layers of muddy rip-up clasts become more abundant upward. This assemblage suggests that deposition occurred in a submarine channel or channel-lobe–frontal-splay environment (cf. Arnott 2010).

The similarity in ripple mark paleocurrent orientations toward the top of the member and in the overlying Fermeuse Formation support this interpretation. Sea-level fall alone is an insufficient explanation for this increase in coarser facies because this explanation would suggest that an increase in coarseness would likely be evident in the Trepassey Formation elsewhere whereas, in fact, this is not observed. The geographic limitation of a turbidite channel accounts for the

absence of such an increase in sandstone content in the Trepassey Formation elsewhere, where

deposition might have occurred on a smooth, unchannelized slope or basin floor.

The part of the Fermeuse Formation in the Catalina Dome measured for this study is

similar to the Fermeuse on the Avalon Peninsula, aside from the persistent presence of thin ash

beds. Thin muddy turbidites dominate the formation, with thick slump beds, together indicating a

distal, fairly smooth slope environment. These deposits are the most consistent and monotonous

of the succession, with the one notable exception of the single occurrence of F8, the 2.5 m thick,

very coarse sandy debrite 70 m above the base of the Fermeuse Formation (i.e., FA6). The

coarseness of the matrix of this unit is difficult to explain, as even the sandiest, highest energy

intervals of FA5 (i.e., the Port Union Member) do not approach its grain size, but it is likely that

it was derived from a very proximal setting.

2.9 Paleobiological implications

Ediacaran fossil occurrences on the Avalon Peninsula can be subdivided into two stratigraphically distinct assemblages, each marked by a difference in taxonomic composition and preservation (Narbonne et al. 2001; Narbonne 2005). Preservation under beds of volcanic

ash throughout the Conception Group yielded diverse assemblages with abundant rangeomorph

fossils, including those in the famous beds at Mistaken Point. By contrast, the younger Fermeuse

Formation contains only a low-diversity assemblage dominated by the discoid fossil Aspidella

preserved under beds of sandstone. Narbonne (2005) concluded that the transition from the

rangeomorph-dominated assemblages of the Conception Group to the Aspidella-dominated

assemblages of the Fermeuse Formation did not represent biological evolution, but instead

reflected a shift in preservational conditions due to the scarcity of ash beds in the St. John's

Group of the Avalon Peninsula.

The studied sections at Catalina in the Bonavista Peninsula provide an ideal test for this

hypothesis, in that ash beds and rangeomorph fossils both persist well into the Fermeuse

Formation (Hofmann et al. 2008). Ash beds in the Fermeuse Formation in Catalina Dome are

underlain by rangeomorph fossils similar to those occurring beneath beds of volcanic ash in the

underlying Conception Group, whereas Fermeuse strata that do not contain ash beds are similar to those of the Avalon Peninsula in containing only Aspidella fossils. This observation also

supports the interpretation of Aspidella as a taphomorph holdfast of largely unpreserved

rangeomorph fronds (Gehling et al. 2000), rather than as the base of cnidarian polyps (Jenkins

1992) or discoid bacterial colonies (Grazhdankin and Gerdes 2007).

2.10 Conclusions

Discovery of Ediacaran sections in the Catalina Dome on the Bonavista Peninsula

effectively doubled the known strike length of the deep-water basin in which the Mistaken Point

biota lived (O'Brien and King 2004a, 2004b, 2005; Hofmann et al. 2008). Detailed stratigraphic

and sedimentological studies herein provide a key northern data point for interpretations of the

basin configuration and the ecology of the Mistaken Point biota. It is concluded that the

succession throughout the entire area accumulated in a deep marine, aphotic paleoenvironment

that shallowed upward from a flat basin floor to a basin-bounding slope, with ash beds supplies

from an adjacent volcanic arc. The higher volume of both sandstone and ash in the study area

relative to the strata exposed to the east and south in the Avalon Peninsula is consistent with its

position closer to the arc, which was the source of both the ash and terrigenous sediment in the

lower half of the succession. A shift in paleocurrent regime, with flow changing from eastward to

southerly, similar to what is seen in strata on the Avalon Peninsula, supports a two-phase tectonic

history hypothesized for the Late Neoproterozoic (Ediacaran) rocks of the Avalon Zone (Murphy and Nance 1989; Murphy et al. 1999; Nance et al. 2002). However, the change occurs somewhat

higher stratigraphically than might be expected on the basis of data further south, assuming

southward progradation of the phase-two slope. The association of this paleocurrent transition

within possible submarine channel deposits suggests either that the hypothesized tectonic

transition occurs lower than it appears or that the study area falls in the northwest corner of the

basin where sediment derivation from the arc persisted longer than it did further to the east. The

implication of the stratigraphic diachroneity of the tectonic transition can only be confirmed by

the acquisition of a higher-resolution set of geochronological data throughout the area. The

absence of ash and fossils in the lowest part of the measured section, and their presence well into

the Fermeuse Formation, implies further diachroneity in the growth and exposure of different

parts of the volcanic arc, and (or) complex wind patterns that might have changed through time.

The prevalence of probable seismoturbidites in the lower levels of the Catalina Dome succession,

and the anomalously coarse and thick debrite (F8) and massive sandstone (F9) beds higher in the succession, hint at more common and stronger tectonic activity in this part of the basin than

further to the east.

Weak bottom currents were active between turbidity currents, with paleoflow

predominantly to the north in present-day coordinates, consistent with the contour current

direction observed from the same stratigraphic interval on the Avalon Peninsula. Other bottom

currents, perhaps driven by internal waves or tides, sometimes flowed to the south or west (i.e.,

upslope). Currents from any of these directions could have supplied the benthic biota with

52 nutrients needed to sustain life. The persistence of ash beds into the Fermeuse Formation provided the taphonomic conditions that were not present in equivalent strata on the Avalon

Peninsula. This finding sheds light on the controversial affinity of the Ediacaran discoid fossil

Aspidella, supporting the hypothesis that this taxon represents preserved holdfasts of frondose taxa such as Charniodiscus.

CHAPTER THREE:

Two new small fronds from Mistaken Point,

Newfoundland

3.1 ABSTRACT

Small, stemmed frond fossils are common at Mistaken Point, and have previously been informally referred to as “dusters”, but due to their small size and consequent relatively poor preservation they have not yet been described taxonomically. Two new genera are herein defined on the basis of their unique constructions: the mop-like rangeomorph Priscipeniculus hofmanni n. gen. n. sp., and a flabellate, lobate frond Broccoliformis alti n. gen. n. sp. Several other previously described taxa had also been included under the umbrella of “dusters”. Collectively, these taxa show that the low epibenthic tier that small fronds occupied was more diverse than previously realized, with multiple taxa converging on the stemmed, small-frond body plan.

3.2 Introduction

Macroscopic Ediacaran fossils range from 579 to 541 million years old and are the oldest known large, architecturally complex biota in the fossil record (Narbonne 2011). Ediacara-type taxa consist of soft-bodied organisms preserved as casts or molds, typically on the bases on event beds induced by storms, turbidity currents, or volcanic eruptions. Preservational conditions characteristic of the Ediacaran sea floor, such as the prevalence of microbial mats and a lack of pervasive bioturbation (Seilacher 1999; Gehling 1999), allowed for their preservation in numerous fossil localities around the world (Narbonne 2005; Fedonkin et al. 2007; Xiao and

Laflamme 2009; Erwin et al. 2011). There has been a great deal of controversy around the

affinities of the Ediacara biota with a wide variety of interpretations. In general, however, focus

has shifted from classifying the Ediacara biota as a whole to evaluating each Ediacaran taxon separately, and at present the prevailing view is that many or most Ediacaran taxa represent extinct clades at a high taxonomic level along with some stem-group animals, including potential

sponges and mollusks (Fedonkin et al. 2007; Xiao and Laflamme 2009; Sperling et al. 2011;

Erwin et al., 2011; Laflamme et al. 2013). Considerable controversy still exists for the affinities

of many Ediacaran taxa.

Ediacaran fronds in particular were traditionally interpreted as pennatulacean cnidarians

(e.g. Sprigg 1947, 1949; Glaessner and Wade 1966; Anderson and Conway Morris 1982; Jenkins

1985, 1992; Boynton and Ford 1995) or ctenophores (Gürich 1933; Dzik 2002), but other

interpretations have included fungi or fungi-grade organisms (Peterson et al. 2003), elaborate

colonies of prokaryotes (Steiner and Reitner 2001), terrestrial lichens (Retallack 1994, 2013a) or

organisms belonging to an extinct high-level clade, either at the kingdom level (Seilacher 1992) or the phylum level within the metazoa (Pflug 1970, 1972; Buss and Seilacher 1994).

Interpretations of Ediacaran fronds as cnidarian or ctenophore filter feeders have been discarded due to the absence of openings large enough to accommodate filter feeding polyps or other feeding structures (Narbonne 2004). Interpretations that require photosynthesis have been discarded for all taxa found in Newfoundland on the basis of numerous sedimentological studies of the strata in which the fossils are found in that area, which conclude that the depositional

environment was deep marine, well below the photic zone (Misra 1971; Myrow 1995; Wood et

al. 2003; Ichaso et al. 2007, Mason et al. 2013). Geochemical evidence is not consistent with a

chemosynthesis-based ecosystem (Canfield et al. 2007). The current leading hypothesis is that

they are stem-group animals that represent an early “failed experiment” in metazoan evolution

(Narbonne 2005). This interpretation is supported by paleoecological analysis (Clapham and

Narbonne 2002; Clapham et al. 2003), which finds that in the census populations preserved in

situ in the Mistaken Point area (Figs. 3.1, 3.2), the biota show epifaunal tiering and other

B A 100 km

Newfoundland

46° 10' N Drook Formation 25 Cape Race Road Fig. 3-1B N

repassey T Formation Mistaken Point Formation Briscal Formation Fermeuse 2 km Formation Mistaken Point E-Surface

53°10' N

Figure 3.1. (A) Location map showing the study area in the southeast of the island of Newfoundland. The Avalon Terrane is in dark gray. (B) Geologic map of the Mistaken Point area, with the Mistaken Point E-surface marked with a star. The G-surface is exposed metres away from the E-surface. Adapted from Clapham et al. 2003.

57 Signal Hill Gp. (>3000m) U-Pb (Ma) Renews Head Fm. 500m s Gp. Fermeuse Fm. St. John’

Trepassey Fm. 565 +/- 3 Mistaken Point Mistaken Point Fm. E-Surface

Briscal Fm.

579 +/- 1

Drook Fm. Conception Gp.

582 +/- 1 Gaskiers Fm. 584 +/- 1

Mall Bay Fm.

Harbour Main Gp. (base not exposed)

Figure 3.2. Ediacaran stratigraphy of the eastern Avalon Peninsula. U–Pb dates from Benus (1988) and Narbonne et al. (2012).

58 ecological patterns similar to those observed in metazoan filter feeding communities of the

Phanerozoic (Narbonne 2005).

The Mistaken Point area is exceptional for its abundance and diversity of deep marine

Ediacaran fossils preserved beneath ash beds. The vast majority of taxa found there have been

studied and described in great detail (e.g. Gehling and Narbonne 2007; Laflamme et al. 2007;

Bamforth et al. 2008; Liu et al. 2011) and used in paleoecological studies (Clapham and

Narbonne 2002; Clapham et al. 2003); however, many of the stemmed small frond fossils have previously received relatively little attention, which this study seeks to redress.

3.3 Ediacaran frond morphology

The frond body plan is common among Ediacaran assemblages all over the world

(Laflamme and Narbonne 2008b), including South Australia (Glaessner and Daily 1959, Jenkins

and Gehling 1978), Namibia (Pflug 1970, 1972; Narbonne et al. 1997; Vickers-Rich et al. 2013),

the White Sea area of Russia (Fedonkin 1985), central England (Boynton and Ford 1995, Wilby

et al. 2011) and eastern Newfoundland, Canada (Misra 1971; Narbonne and Gehling 2003;

Laflamme et al. 2004; Laflamme et al. 2007; Hofmann et al. 2008; Narbonne et al. 2009). This

paper will follow the morphological terminology proposed by Laflamme and Narbonne (2008a,

2008b).

Ediacaran fronds are generally composed of three main parts (Fig. 3.3): a basal holdfast,

typically bulbous or disc-shaped; a stem and/or central stalk (the stalk being an extension of the

stem into the petalodium); and a leaf-like, morphologically complex petalodium (Laflamme and

Narbonne 2008a, 2008b), the ornamented leaf-like section generally composed of two or more

Petalodium Petaloid

Central stalk

Secondary branches

Tertiary branches

Basal petalodium Primary angle branches Stem

Holdfast

Figure 3.3. Generalized morphology of a stemmed frond. The three main components are the holdfast, stem, and the petalodium. Primary branches may be attached to a central stalk, as in the right half of the figure, or to the distal end of the stem, as in the left half of the figure.

60 'petaloids' (Laflamme and Narbonne 2008a, 2008b), each composed of repeating modules arranged in branches attached to the stalk or stem. The petalodium, as the most morphologically

complex and variable part of the frond, is most important for identification and classification

(Jenkins and Gehling 1978). The largest branches connected directly to the stem or stalk are

termed primary branches, any smaller branches that emerge from those branches are termed

secondary branches, and even smaller branches attached to the secondary branches are termed

tertiary branches. The frond body plan characteristic of many Ediacaran taxa is likely an example

of convergent evolution due to shared ecology – in this case, a common need to elevate above

the substrate, the same reason that frond-shaped morphologies are observed today across several

kingdoms: plants, animals, and fungi (Laflamme and Narbonne 2008b). Laflamme and Narbonne

(2008a, 2008b) describe four distinct branching architectures which they regarded as

apomorphies for four high-level taxonomic groups within the Ediacara biota. Three of these are

known from Mistaken Point: Arborea-type branching, which consists of parallel primary

branches attached to a central stalk at right angles with teardrop-shaped secondary branches;

Rangea-type branching, which is fractal – self-similar over at least three orders of magnitude,

and constructed from individual 'rangeomorph frondlets'; and Charnia-type branching, which is

subcategory of Rangea-type, composed of sigmoidal primary branches with a zigzagging central

axis, and secondary modular elements that are composed of tertiary rangeomorph elements. The

latter two belong to a high level clade called the Rangeomorpha (Pflug 1972; Jenkins 1985;

Narbonne 2004; Laflamme et al. 2013), and the first constitutes a clade called the

Arboreomorpha (Laflamme et al. 2013). Rangeomorphs along with the arboreomorph genus

Charniodiscus Ford, 1958 dominate the deep sea Ediacaran assemblages of eastern

Newfoundland (Narbonne 2004; Hofmann et al. 2008; Narbonne et al. 2009) and central England

(Boynton and Ford 1995; Wilby et al. 2011), and are known sparingly from younger and shallower Ediacaran localities (Fedonkin et al. 2007).

Laflamme et al. (2009) concluded that Ediacaran fronds may have absorbed dissolved organic carbon osmotrophically, and that the fractal branching of rangeomorph elements would have increased surface-area-to-volume ratios sufficiently for this to be a viable strategy. This mode of feeding is used today mainly among bacteria, but the larger pool of available dissolved organic carbon in the Ediacaran combined with the high surface-area-to-volume ratio would have made this a viable feeding strategy for macroscopic Ediacaran organisms (Laflamme et al. 2009).

It is therefore believed that the tiering observed in the rangeomorph-dominated assemblages in eastern Newfoundland and central England evolved due to competition to take maximum advantage of the dissolved organic carbon in the water column of their environments, and that this was the ecological pressure that led to the convergent evolution of a frond body plan among multiple clades in the Ediacaran (Laflamme et al. 2008b).

3.4 Geological setting

Much of eastern Newfoundland, including the Avalon, Bonavista, and Burin peninsulas, is part of the Avalon Terrane (O'Brien et al. 1983; Myrow 1995)(Fig. 3.1), which consists of

Neoproterozoic and early Paleozoic strata that were deposited on and adjacent to the microcontinent Avalonia, a volcanic arc off the coast of Gondwana (Nance et al. 2002;

Pisarevsky et al. 2011). Architecturally complex fossils have been found in three main locations in Ediacaran strata of the Avalon Terrane of Eastern Newfoundland (Fig. 3.1): the Mistaken Point area on the southeastern tip of the Avalon Peninsula (see review in Narbonne et al. 2012), the

Spaniard's Bay area in the northwest of the Avalon Peninsula (Narbonne 2004; Ichaso et al. 2007;

Flude and Narbonne 2008: Narbonne et al. 2009), and the Catalina Dome on the eastern coast of

the Bonavista Peninsula (O'Brien and King 2004a; Hofmann et al. 2008; Mason et al. 2013).

Ediacaran fossils occur commonly to abundantly in the Drook, Briscal, and Mistaken Point formations of the Conception Group, and the Trepassey and Fermeuse formations of the St.

John's Group (Fig. 3.1A, 3.1B). The small frond specimens included in this study were found in

the Mistaken Point Formation in the Mistaken Point Ecological Reserve, Avalon Peninsula.

The Mistaken Point Formation is the uppermost formation in the Conception Group, part

of a deep marine siliciclastic succession that overlies the mainly igneous rocks of the Harbour

Main Group (King 1988, 1990). The Mistaken Point Formation is dominated by mudstone-rich

siliciclastic turbidites deposited at the toe-of-slope in a forearc basin (Misra 1971; Wood et al.

2003; Ichaso et al. 2007; Mason et al. 2013). The turbidites are primarily Bouma T(C)DE beds

(Bouma 1962), with ‘TF’ interturbidite laminated mudstone deposits from the low energy

intervals between turbidity currents (cf. Hesse 1975; Wood et al. 2003). The absence of wave- generated structures or evidence of exposure in several kilometres of stacked turbidites in the succession implies that the depositional environment was at great depth, from which it has been inferred that organisms living in that environment could not have relied on photosynthesis (Wood et al. 2003; Narbonne 2005; Ichaso et al. 2007). In addition to the mudstone-rich siliciclastic turbidites, the succession is punctuated by thin beds of volcanic ash from the adjacent island arc deposited within the TF divisions, which are responsible for the preservation of the soft-bodied

organisms beneath them (Seilacher 1999; Wood et al. 2003; Ichaso et al. 2007) in what Narbonne

(2005) termed Conception-style preservation. Because ash layers are less resistant than the

silicified mudstone, the rocks tend to weather along the planes of ash beds and expose the

surface of the mudstone immediately underneath. The fossils are preserved as both positive and

negative epireliefs on these surfaces (Narbonne 2005). Frond-shaped organisms, which were

tethered to the sea floor but otherwise were flexible in the water column, were typically oriented

in a common direction with other fronds on the same rock surface, either parallel with the

downslope direction inferred from Bouma TC-division turbidite current ripples or in a contour-

parallel orientation 90 degrees counterclockwise from the downslope direction (Wood et al.

2003).

3.5 The Mistaken Point assemblage

The Mistaken Point Ecological Reserve, southeastern Avalon Peninsula, Newfoundland,

is famous for its exceptional Ediacaran fossils which are exposed on numerous broad, flat

surfaces along the coastal cliffs (Misra 1971; Narbonne et al. 2007). These surfaces are

effectively snapshots of the Ediacaran sea floor, with fossils of the soft-bodied biota preserved in

relief by episodic burial by volcanic ash. The succession (Fig. 3.2) contains many significant

fossil horizons, including the oldest known complex macroscopic fossils, found in the Drook

Formation (Narbonne and Gehling 2003), dated at 578.8 +/- 0.5 Ma (van Kronendonk et al.

2008; Narbonne et al. 2012a). These include the rangeomorph fronds Trepassia Narbonne,

Laflamme, Greentree and Trusler 2009 (see also Narbonne and Gehling 2003); Charnia Ford

1958 (see also Laflamme et al. 2007), and a possible specimen of Charniodiscus (see Liu et al.

2012); along with the possible early sponge (Sperling et al. 2011) Thectardis Clapham,

Narbonne, Gehling, Greentree, and Anderson 2004. The oldest known stemmed fronds,

Culmofrons Laflamme, Flude, and Narbonne 2012, are found in the overlying Briscal Formation, along with the branching rangeomorph Bradgatia Boynton and Ford, 1995 (Flude and Narbonne

2008) and the spindle-shaped rangeomorph Fractofusus Gehling and Narbonne 2007. Above

that, the Mistaken Point Formation is the most richly fossiliferous part of the succession, with

diverse assemblages of rangeomorphs and other Ediacaran taxa preserved at Gull Rock Cove and

Mistaken Point itself (Wood et al. 2003; Clapham et al. 2003), including the well-known and well-studied Mistaken Point E-surface, dated at 565 +/- 3 Ma (Benus 1988; Narbonne et al.

2012a), on which most of the fossils in this study are found. The overlying Trepassey Formation also contains fossil assemblages preserved by ash (Wood et al. 2003; Bamforth et al. 2008).

Above that formation, ash deposition had ceased in this part of Newfoundland, so the taphonomic window allowing for preservation of architecturally complex Ediacaran fossils was closed (Gehling et al. 2000; Wood et al. 2003; Narbonne 2005; Mason et al. 2013).

Over the last decade, numerous taxa have been described from the Mistaken Point area and the E-surface in particular, as well as elsewhere in the ecological reserve and elsewhere in

Newfoundland. These have included taxa known from Charnwood Forest, England and new taxa, totaling 26 types (Narbonne et al. 2012b). Many of these taxa have been rangeomorphs: organisms constructed of numerous similar, fractally branching rangeomorph ‘modules’ assembled into a variety of different body plans (Narbonne 2004), such as the reclining, spindle- shaped Fractofusus (Gehling and Narbonne 2007), comb-shaped Pectinifrons Bamforth,

Narbonne, and Anderson, 2008, bush-shaped Bradgatia (Flude and Narbonne 2008), stemless elongate frond Charnia (Laflamme et al. 2007), and the stemmed, tulip-shaped frond Culmofrons

(Laflamme et al. 2012). Mistaken Point taxa that do not belong to the Rangeomorpha include

Thectardis (Clapham et al. 2004) and the stemmed arboreomorph frond with a discoid holdfast

Charniodiscus (Laflamme et al. 2004). All known taxa were benthic and immobile, although Liu et al. (2010) reported a single level of putative trace fossils implying surficial locomotion from

Mistaken Point.

Virtually all fossils in the Mistaken Point area have been named, with the notable exception of many of the small fronds. Many small fronds are simply juveniles of known taxa;

Liu et al. (2012) reported the discovery of particularly tiny juvenile fronds from the Drook

Formation, most of which can be identified as Charnia, Trepassia, or possibly Charniodiscus.

But some small fronds exist in the Mistaken Point Formation that bear little resemblance to any named taxa. Hofmann et al. described two distinctive small frond taxa from the Bonavista

Peninsula, Primocandelabrum Hofmann, O'Brien, and King 2008; and Parviscopa Hofmann,

O'Brien, and King, 2008, that bore no resemblance to any larger Ediacaran taxa. In previous paleoecological work at Mistaken Point (Clapham and Narbonne 2002; Clapham et al. 2003), many small, stemmed fronds were grouped together and informally called ‘dusters’, but closer inspection finds significant disparity within that informal grouping, including representatives of the taxon Primocandelabrum from the Bonavista Peninsula.

Interpretation of the small fronds has been problematic due to taphonomy. Because of the mode of preservation of the Mistaken Point biota, the resolution of the fossils is dependent upon the grain size of the ash beneath which they are preserved. This mode of preservation presents a problem for the study of the smallest fossils in the assemblage in particular, because their features are often obscured proportionally more than the larger specimens. Other Ediacaran fronds are classified heavily on the basis of their secondary and tertiary branching, which are not discernible in most small frond specimens. Because of this difficulty, the small fronds are some of the last fossils of the Mistaken Point E-surface to be described systematically, and only the best-preserved specimens can be identified with confidence.

3.6 Methods

Removal of fossils from the Mistaken Point Ecological Reserve is strictly prohibited, so

paleontological study of the Mistaken Point biota relies heavily on field photographs, latex

molds, and casts of the original fossils. A 70 square metre cast of the Mistaken Point E-surface

was made by Research Casting International in 2009 and 2010 in association with the provincial

government of Newfoundland and Labrador, the Royal Ontario Museum, Johnson GeoCentre,

Queen's University, and the University of Oxford. The complete cast is located in Research

Casting International's facility in Trenton, Ontario, and this was an invaluable resource as it

allowed thorough inspection of a large portion of the Mistaken Point E-surface – on which most

of the small frond fossils included in this study are found – in a controlled indoor environment.

251 fossils were numbered and photographed from this cast, although many were too poorly

preserved and ambiguous to be identified in this study. Additional field photographs were taken

in Newfoundland of small fronds on parts of the E-surface that were not included in the casting project, as well as of two fossils from the G-surface. Dozens of latex molds of these and other fossils made by the authors and other Queen's University researchers were also used as resources and photographed for comparison. One original specimen that was previously collected from the

E-surface by the Royal Ontario Museum, ROM 38641, was also included in this study.

Strata in the Mistaken Point area have been subjected to tectonic shortening of about 40% on most of the fossiliferous surfaces. This deformation is especially apparent in fossils with originally circular features such as frond holdfasts, which are circular in every undeformed

Ediacaran assemblage in the world (Ford 1958; Jenkins and Gehling 1978; Boynton and Ford

1995, Wilby et al. 2011). In the Mistaken Point area, these fossils have become oblong, all in the

same direction and the same amount on a given bedding surface, elongate in the direction of

cleavage (Seilacher 1999; Wood et al. 2003). To account for this deformation, each photograph

was retrodeformed (i.e., stretched in the direction of shortening until oblong holdfasts become circular; Seilacher 1999; Wood et al. 2003; Ichaso 2007). In specimens without a holdfast to use

as a direct indicator of the degree of shortening, an average retrodeformational ratio from across

the surface was used.

Once retrodeformed, each specimen was measured for a variety of dimensions: holdfast

diameter, stem length, stem width, petalodium length, petalodium width, frond length, and basal

petalodium angle. Ratios between these measurements were compared to find morphometric

differences in overall shape of the fronds. Petalodium branching architecture was also carefully

observed in sufficiently well-preserved specimens.

Approximately 270 fossils in total were numbered and photographed. Many were

identified as or tentatively assigned to known Ediacaran taxa, including Charnia, Culmofrons,

Bradgatia, Charniodiscus, Avalofractus Narbonne, Laflamme, Greentree, and Trusler 2009,

Beothukis Brasier and Antcliffe 2009, and Primocandelabrum. Others were too ambiguous to

identify as either existing taxa or newly defined taxa. For many potential small frond fossils from

the E-surface only the holdfast and stem/stalk were preserved, with no evidence of the

petalodium.

In all, five clear, well-preserved small fronds were determined to belong to two new

monotypic genera. Measurements of these specimens are summarized in Table 3.1. Other small

frond specimens with similar shapes may belong to these taxa, but due to poor preservation, such

identifications are uncertain.

Table 3.1. Measurements of retrodeformed specimens of Priscipeniculus and Broccoliformis. All measurements in centimetres except for petalodium basal angle, which is measured in degrees.

Specimen Holdfast Stem Stem Petalodium Petalodium Total frond Petalodium Type Figure Surface number diameter length width length width length basal angle 1 Priscipeniculus 3.4A, 3.4C E 1.18 2.87 0.29 5.51 5 9.11 133 2 Priscipeniculus – G 3.58 5.74 1.84 7.58 7.47 13.42 133 3 Priscipeniculus 3.4E E 1.31 3.16 0.46 5.08 5.24 8.09 107 4 Broccoliformis 3.4B, 3.4D E 3.2 3.92 0.89 3.56 7.92 7.48 175 5 Broccoliformis 3.4F E 2.52 2.87 0.7 4.41 6.65 7.42 175 69

3.7 Systematic paleontology

Group RANGEOMORPHA Pflug 1972

Genus PRISCIPENICULUS new genus

“Feather dusters” Wood et al. 2003 Figure 9, Narbonne 2005 Figure 5e

Type species. – Priscipeniculus hofmanni new species, by monotypy.

Diagnosis. – As per species.

Etymology – From the Latin “prisci” meaning “old”, and “peniculus” meaning “duster” or

“mop”.

PRISCIPENICULUS HOFMANNI new species

Figures 3.4A, 3.4C, 3.4E

Holotype. – On the Mistaken Point E-surface. Plastotype on the E-surface cast at the Royal

Ontario Museum, ROM 60065.1.

Diagnosis. – Cm-scale frondose fossil with a roughly deltoid petalodium attached to a bulbous or

discoidal holdfast by a cylindrical stem. The petalodium consists of at least nine thin primary

branches attached at the base of the petalodium extending upward distally from the base of the

petalodium. Where well-preserved, rangeomorph architecture is subtly visible. No axial stalk is

evident. Branches in the foreground can be seen in their entirety but background branches are

partially obscured beneath foreground branches, implying a bushy, mop-like shape of the organism before compaction.

Description. – The holotype (Figs. 3.4A, 3.4C), from the Mistaken Point E-surface, is a 9.1 cm- long frond with a small, bulbous holdfast, 1.2 cm in diameter, attached to a cylindrical stem 2.9 cm long and 3 mm thick. At the end of the stem, many (at least twelve visible in this specimen)

A B

C D

E F

71 Figure 3.4. (previous) Photographs of new small frond taxa. Scale bars 2 cm. A) Unretrodeformed holotype fossil of Priscipeniculus hofmanni from the Mistaken Point E-surface. B) Unretrodeformed holotype fossil of Broccoliformis alti from the Mistaken Point E-surface. C) Retrodeformed latex mold of the holotype of Priscipeniculus hofmanni; white arrow points to a primary branch with visible secondary branching. D) Retrodeformed latex mold of the holotype of Broccoliformis alti. E) Retrodeformed field photograph of an additional specimen of Priscipeniculus hofmanni on the Mistaken Point E-surface. F) Retrodeformed photograph of an additional specimen of Broccoliformis alti from the E-surface cast at the Research Casting International facility in Trenton, Ontario.

72 thin (1 mm or less) primary branches emerge from the base of the petalodium, forming a deltoid petalodium, 5.5 cm long and 5 cm wide, with a 133° angle at the base, and an 80° angle at a pointed tip on the distal end of the petalodium. Additional primary branches emerge from behind the branches of which the connections to the base can be seen. On two or three primary branches, subtle secondary branching is visible, implying rangeomorph affinity.

In addition to the holotype, there are two more specimens that are well-preserved (Fig.

3.4E), one also from the E-surface and the other from the slightly stratigraphically higher G- surface. Several other small frond fossils from Mistaken Point may belong to the species but are not preserved clearly enough to be certain. The G-surface specimen is somewhat larger than the two E-surface specimens, with a total length of 13.4 cm. Its holdfast, with a diameter of 3.5 cm, is also larger relative to its size. It is also the specimen of Priscipeniculus hofmanni with the least number of visible primary branches: only nine. However, it does have a very similar shape and overall morphology to the holotype. The third specimen is very similar to the holotype, only with a slightly shorter stem and a more acute basal petalodium angle, 107°, unlike the other two specimens with angles of 133°. Both of the additional specimens are somewhat more rounded than the pointed, deltoid shape of the holotype. Secondary branching is not visible on either.

Etymology. – Named for the late Hans Hofmann who contributed greatly to our understanding of Ediacaran paleontology.

Comparisons. – Priscipeniculus is similar to Primocandelabrum in that all of the primary branches are connected to the stem at the base of the petalodium. However, the branches of

Priscipeniculus are thinner and more numerous, and the petalodium shape is more deltoid with a point at the distal end, compared to the inverse triangle shape of Primocandelabrum. The

branches of Primocandelabrum are arranged as a two-dimensional fan as opposed to the three-

dimensional mop-like shape of Priscipeniculus.

The petalodium of Priscipeniculus has some similarities to Bradgatia in that there are many primary branches all attached together at the base, but Bradgatia lacks the stem and holdfast of Priscipeniculus and shows a great deal more secondary and tertiary branching that forms individual ‘petals’ in a way that Priscipeniculus does not.

An as yet unnamed stemmed frond from the Ediacaran deposits of Charnwood Forest,

Central England, informally referred to as the “dumbbell-like frond” (Wilby et al. 2011), shows some similarity with Priscipeniculus, with an orbicular petalodium attached to a stem with a holdfast. However, the dumbbell frond's holdfast is quite distinct from the small bulbous holdfast of Priscipeniculus, as it is a large disc, larger even than the frond's petalodium, with many concentric rings. Within the dumbbell frond's petalodium, the branching pattern appears somewhat similar to that of Priscipeniculus, with no defined central axis and multiple primary branches attached at the base of the petalodium, but beyond that the branching appears more complex in the dumbbell frond. It may be a closely related taxon.

Remarks. – The three-dimensional, mop-like architecture of Priscipeniculus makes it unique among the Mistaken Point biota. The ‘mop’ of primary branches that make up the petalodium would likely have been an effective means of straining flowing water for nutrients in a different but analogous way to other fronds.

Group uncertain

Genus BROCCOLIFORMIS new genus

Type species. Broccoliformis alti new species, by monotypy.

Diagnosis. – As per species.

Etymology – From the Italian “broccolo” referring to the vegetable, and the Latin “formis”

meaning shape.

BROCCOLIFORMIS ALTI new species

Figures 3.4B, 3.4D, 3.4F

Holotype. – On the Mistaken Point E-surface. Plastotype at the Royal Ontario Museum, ROM

62623.

Diagnosis. – Irregularly lobate flabellate frond connected to a short cylindrical stem that attaches

at the opposite end to a disk. Internal features consist of poorly developed lobes radiating in a

semi-circular fan-shaped array from the center of the diameter, with an angle at the base of the

petalodium near 180 degrees.

Etymology. – From the Latin “altus” meaning “deep”, referring to the deep marine environment

of the organism.

Description. – The holotype (Fig. 3.4B, 3.4D), from the Mistaken Point E-surface, is a flabellate

frond characterized by a large, bulbous holdfast (3.2 cm diameter), stem (0.9 cm wide and 3.9 cm

long), and a roughly semi-circular (3.5 cm long and 7.9 cm wide) petalodium. The flabellate

petalodium has a sharp outer margin, and internal features of the petalodium consist of irregular,

poorly developed lobes radiating in a fan-shaped array from the base of the petalodium where it

attaches to the stem. The angle at the base of the petalodium is 175°, close to a horizontal line,

and the petalodium itself is wider than it is long. Compared to other fronds, the holdfast, which

appears as a thickened bulb continuous with the stem at its base, it notably large relative to the

total length of the frond.

The other known specimen of Broccoliformis (Fig. 3.4F), also from the E-surface, is

slightly smaller but overall quite similar to the holotype. It also has a relatively large holdfast

(2.5 cm), and thick stem (0.4 cm wide, 2.9 cm long). This specimen has similar overall

petalodium morphology to the holotype, including the flabellate shape with a 175° angle of the

base of the petalodium, and greater petalodium width than length (4.4 cm length 6.7 cm width).

Comparisons. – In contrast to most other fronds with the exception of some Primocandelabrum,

the petalodium of Broccoliformis is wider than it is long. Its petalodium has the widest basal

angle of any Mistaken Point stemmed organism. Like a number of other Ediacaran fronds, including Priscipeniculus as well as Avalofractus, Culmofrons, and Primocandelabrum,

Broccoliformis has a round, bulbous holdfast and a cylindrical stem. The stem is somewhat thick compared to that of other small fronds, and the diameter of the holdfast is relatively large compared to the total length. Unlike other Mistaken Point fronds, the petalodium of

Broccoliformis does not exhibit modular construction or obvious branches. Instead,

Broccoliformis has an irregular lobate morphology.

The lobate structure of the petalodium bears a resemblance to other lobate fossils from the Mistaken Point biota known as ivesheadiomorphs (Liu et al. 2011), a group with includes the genera Ivesheadia Boynton and Ford, 1995, 1996, and Blackbrookia Boynton and Ford 1995, although Liu et al. (2011) regard all ivesheadiomorphs as taphomorphs of other taxa and therefore as variants of each other. The somewhat oblong shape and raised perimeter of the petalodium of the holotype are particularly reminiscent of the taxon Blackbrookia, which is primarily known from Charnwood Forest, England (Boynton and Ford 1995; Wilby et al. 2011) and is also reported from Bonavista Peninsula (Hofmann et al. 2008). The frond morphology of

Broccoliformis, in contrast with the irregular, unstructured shapes of Blackbrookia and

Ivesheadia, may be useful in interpreting other lobate Ediacaran taxa.

Remarks. – The lobate, irregular Ivesheadiomorpha (Boynton and Ford 1995; Liu et al. 2011)

are a controversial group of taxa from the Ediacaran biota of eastern Newfoundland and

Charnwood Forest. When Boynton and Ford first described them in 1995, they interpreted them as cnidarian-grade organisms, but this view has fallen out of favour. Three alternative explanations have recently been put forward: Laflamme et al. (2011) suggested that they are microbial colonies; Liu et al. (2011) argued that they are taphomorphs of other Ediacaran taxa representing advanced stages of decay; and Wilby et al. (2011) put forward the idea that they are the result of upright fronds causing sediment deposition beneath them during turbidity current flow.

Unfortunately, none of these explanations for ivesheadiomorphs can be used to interpret

Broccoliformis. If the morphological similarity between Broccoliformis and Blackbrookia is homologous, the existence of a stem and holdfast attached to the lobate structure is inconsistent with the microbial interpretation of Laflamme et al. (2011) or the sediment deposition explanation of Wilby et al. (2011). Similarly, the flabellate shape of the Broccoliformis is unique among Mistaken Point fronds, making it difficult to reconcile as a taphonomic degradation of some more common Mistaken Point frond as inferred for Ivesheadia by Liu et al. (2011). An alternative explanation is to echo the original interpretation of Boynton and Ford (1995) that ivesheadiomorphs are tissue-grade organisms themselves, albeit apparently lacking any of the apomorphies of the Cnidaria. This interpretation would suggest that perhaps another clade of

Ediacaran organisms evolved convergently into a frondose body plan in order to extend higher into the water column to compete for nutrients.

3.8 Other small fronds at Mistaken Point

Apart from these two newly described small fronds, there are several other taxa that share

a broadly similar morphology and that would have occupied the same ecological tier at Mistaken

Point. These include juveniles of larger fronds present on the surface: Charnia, Charniodiscus,

and Beothukis, as well as taxa only known as small fronds from elsewhere in eastern

Newfoundland: Primocandelabrum (Fig. 3.5B) from the Bonavista Peninsula (Hofmann et al.

2008), and Avalofractus (Fig. 3.5A) from Spaniard’s Bay. Some may be juveniles of the recently

described Culmofrons, which is found lower in the succession in the Mistaken Point area

(Laflamme et al. 2012). Many of the fossils that Clapham and Narbonne (2002) and Clapham et

al. (2003) referred to as ‘dusters’ are now known to be Primocandelabrum or Beothukis, while

others may be juvenile Culmofrons.

Primocandelabrum is a stemmed frond with a petalodium shaped like an inverted triangle

constructed by few thick primary branches attached at the base of the petalodium. It was first

described from the Mistaken Point and Trepassey formations of the Bonavista Peninsula

(Hofmann et al. 2008). Hofmann et al. (2008, p. 212) believed that many of the small fronds at

Mistaken Point represent specimens of Primocandelabrum sp. but did not illustrate or cite any

specific specimens they attributed to this taxon. The tentaculate holdfast of the species

Primocandelabrum hiemaloranum Hofmann, O’Brien, and King, 2008 has not been observed at

Mistaken Point, but one specimen from the Mistaken Point E-surface shows subtle secondary branches attached to its thick primary branches, suggesting rangeomorph architecture (Fig.

3.5B). On the Mistaken Point E-surface at Watern Cove, there is a large frond that resembles

Primocandelabrum (Fig. 3.5C) with two thick primary branches attached at the base of the

A Figure 3.5. (A) Avalofractus from the Mistaken Point E-surface.White arrow points to a primary branch that shows secondary branching. (B) Primocandelabrum from the Mistaken Point E-surface. White arrow points to a primary branch that shows subtle secondary branching. (C) Large frond from the E-surface at Watern Cove; has a similar shape to other fossils identified as Primocandelabrum, but the two thick secondary branches are unique. Scale bars are 2 cm.

79 petalodium, but uniquely to this specimen, each of these branches also has a thick secondary

branch attached that points inward toward the central axis at approximately a 60° angle. The total

length of the frond is 12.7 cm, with a particularly wide petalodium, measuring 5.6 cm in length

and 12.3 in width. This specimen may be a separate species of Primocandelabrum, or

considering its greater size than most Primocandelabrum, perhaps a more mature specimen.

Beothukis has a spatulate shape that in some specimens makes it appear as if it might

have a stem between the holdfast and the petalodium, which allows for confusion with the true

‘dusters’ Priscipeniculus, especially where small and poorly preserved. Other stemmed small

fronds on the Mistaken Point E-surface bear a resemblance to Culmofrons from Lower Mistaken

Point and the Briscal Formation, but they are small and not sufficiently well-preserved to classify

them definitively to this genus. Juvenile Charnia and Charniodiscus fronds also share this low tier.

3.9 Discussion

When Clapham and Narbonne (2002) and Clapham et al. (2003) first studied the paleoecology of Mistaken Point, many of the taxa included in their analysis were still undescribed and unnamed, but most, such as the ‘spindle’, now called Fractofusus, and the

‘pectinate’, now called Pectinifrons, were nonetheless readily discernible due to their unique

constructions. However, the smaller taxa were more problematic when approached informally,

and a wide variety of small frond fossils that share a broadly similar organism-scale shape were

grouped together as ‘dusters’, including these two new taxa, Priscipeniculus and Broccoliformis, as well as the relatively recently described taxa Primocandelabrum (Hofmann et al. 2008),

Beothukis (Brasier and Antcliffe 2009), Avalofractus (Narbonne et al. 2009), and Culmofrons

(Laflamme et al. 2012).

Consequently it is worth noting that the lower tier that the small fronds occupy has more

diversity than previously realized in paleoecological analysis. At Mistaken Point, there are a

number of different small fronds, many of which are not known as larger taxa, and most of which

have a broadly similar basic body shape. Small fronds from the E-surface show significant overlap in terms of stem length vs. total frond length, and petalodium length vs. width (Fig. 3.6).

This observation is in contrast with Spaniard's Bay, where the various types of small frond are

distinct from each other based on the same criteria (Narbonne et al. 2009). The Spaniard's Bay

types plot in three distinct groups, while the Mistaken Point taxa show significant overlap, which

makes differentiating and identifying the numerous poorly preserved specimens challenging.

Primocandelabrum in particular has a large morphometric range that both Priscipeniculus and

Broccoliformis fall within, although their unique constructions allow them to be differentiated.

While there is not great disparity in the overall shape of the small, stemmed fronds, there is considerable diversity in petalodium architecture. It can therefore be concluded that this tier was

a competitive one, with significant convergent evolution of multiple taxa.

In addition to the small fronds that were lumped together as ‘dusters’, other, less similar

taxa also shared the same tier, slightly elevated from the substrate, adding to its biodiversity.

These taxa included the bush-like rangeomorph Bradgatia (Flude and Narbonne 2008), the

triangular potential sponge Thectardis (Clapham et al. 2004), and the comb-like rangeomorph

Pectinifrons (Bamforth et al. 2008), as well as juveniles of taxa such as Charniodiscus and

Charnia.

0.8

0.7

0.6

0.5

0.4

0.3 Stem length/Frond length

0.2

0.1

0 0 0.5 1 1.5 2 2.5 3

Petalodium width/Petalodium length LEGEND

Priscipeniculus hofmanni Zones for Spaniard’s Bay fronds Broccoliformis alti Avalofractus Avalofractus Primocandelabrum Trepassia Other unidentified small fronds Beothukis

Figure 3.6. Comparison of the morphometric distribution of Mistaken Point small, stemmed fronds with the small fronds describe by Narbonne et al. (2009) in the Spaniard’s Bay area. Distribution of Mistaken Point taxa on the chart is represented by empty ellipses while distribution of Spaniard’s Bay taxa is represented by patterned ellipses. Mistaken Point taxa show significantly more overlap, meaning the overall body shape of the fronds shows convergence across multiple frond architectures

82 Figure 3.4. (previous) Photographs of new small frond taxa. Scale bars 2 cm. A) Unretrodeformed holotype fossil of Priscipeniculus hofmanni from the Mistaken Point E-surface. B) Unretrodeformed holotype fossil of Broccoliformis alti from the Mistaken Point E-surface. C) Retrodeformed latex mold of the holotype of Priscipeniculus hofmanni; white arrow points to a primary branch with visible secondary branching. D) Retrodeformed latex mold of the holotype of Broccoliformis alti. E) Retrodeformed field photograph of an additional specimen of Priscipeniculus hofmanni on the Mistaken Point E-surface. F) Retrodeformed photograph of an additional specimen of Broccoliformis alti from the E-surface cast at the Research Casting International facility in Trenton, Ontario.

71 3.10 Conclusions

Although the small frond fossils of Mistaken Point are difficult to interpret and to identify

due to the limitations of their taphonomy and overall similarity in first order shape, two new taxa

have been described on the basis of their unique petalodium constructions: the mop-like

Priscipeniculus hofmanni, and lobate, flabellate Broccoliformis alti.

With description of the small frond fossils previously informally referred to together as

‘dusters’, the last major gap in the taxonomic description of the known Mistaken Point

assemblage is filled, excluding fossils that are currently too unclear and poorly preserved to

describe meaningfully. Future discoveries of similar fossils with better preservation in

Newfoundland, England, or elsewhere, may shed light on remaining unnamed fossils at Mistaken

Point, and the two taxa described in this study based on a small number of specimens with sub-

optimal preservation. Now that the decade-long endeavor to describe all of the major fossils of

the Mistaken Point assemblage is essentially complete, it may be enlightening to revisit detailed paleoecological study of the Mistaken Point biota, keeping in mind the greater taxonomic disparity and competition now known to have existed at this tier of the ecosystem.

CHAPTER FOUR:

General Conclusions

4.1 Objectives of Thesis

The study of the Ediacaran of Eastern Newfoundland, Canada has become an active area of research in recent decades. In that time, numerous paleontological and sedimentological studies have contributed substantially to the understanding of the early deep marine ecosystems that were home to some of the oldest known macroscopic, complex organisms and their paleobiology. Studies of the sedimentology of the Ediacaran of the Bonavista Peninsula and of the taxonomy of the small fronds of Mistaken Point fill significant previously existing gaps in this framework. Many interesting questions remain concerning the Mistaken Point Assemblage, but a more complete picture of the depositional basin and of the biodiversity of the biota provides a more refined perspective for approaching future work.

4.2 Sedimentology of the Catalina Dome

Sedimentological study of the upper Conception and lower St. John's groups of the

Catalina Dome, Bonavista Peninsula reaffirmed many of the conclusions of previous studies of the correlative strata on the Avalon Peninsula (Wood et al. 2003; Ichaso et al. 2007). These include a deep marine depositional environment, turbidite ponding west of the “Harbour Main

High”, a transition from no-slope to slope deposition, and a turbidite paleocurrent shift from broadly eastward to southward – although that shift occurred diachronously across the basin.

The sandstone-rich Port Union Member of the Trepassey Formation of the Catalina

Dome is distinct from the Trepassey Formation of the Avalon Peninsula, and probably represents

85 a local turbidite channel. Another significant difference found between the Catalina Dome succession and correlative strata on the Avalon Peninsula is diachroneity of volcanic ash deposition, with no ash beds found in the Drook or lowermost Mistaken Point formations of the

Catalina Dome, while ash beds there persist into the Fermeuse Formation, where there are none present in the Avalon Peninsula Fermeuse. The latter observation is significant to the interpretation of the problematic discoid fossil Aspidella terranovica Billings 1872, which is abundant in the Fermeuse Formation. Complex, Mistaken Point-style communities are preserved beneath the ash beds of the Fermeuse Formation of the Catalina Dome, supporting the interpretation that Aspidella is a taphomorph of otherwise unpreserved fronds, representing only the holdfast.

Chapter 3 was published in January, 2013 (Mason et al. 2013). Nearly synchronous with this publication, Retallack published a radical re-interpretation of all Ediacaran fossils and strata worldwide as terrestrial in origin (Retallack 2013a). Retallack (2012, 2013b) specifically extended this re-interpretation to include the Ediacaran of Newfoundland. Retallack’s conclusions are incompatible with the overwhelming evidence for deep-sea deposition presented in Chapter 3, and have also been criticized on the basis of sedimentological and biological observations in Australia and China (Xiao and Knauth, 2013; Xiao et al. 2013).

4.3 Small fronds of Mistaken Point

Among the small frond fossils of Mistaken Point that can be confidently identified, many are juveniles of larger fronds known from the E-surface such as Beothukis, Charniodiscus, and

Charnia, as well as fronds belonging to taxa described from other areas of Newfoundland, i.e.,

Primocandelabrum Hofmann, O'Brien and King 2008 and Avalofractus Narbonne, Laflamme,

Greentree, and Trusler 2009, each of which is only known from small specimens similar in size

to those at Mistaken Point. In addition to numerous other fossils that are too poorly preserved to

identify, two newly described monotypic genera can be discerned that show unique constructions

not otherwise known from the Ediacara biota: Priscipeniculus hofmanni and Broccoliformis alti.

Priscipeniculus is a deltoid, three-dimensional, mop-like rangeomorph frond with many primary branches attached at the base of the petalodium. Broccoliformis is a flabellate, lobate frond with a petalodium somewhat resembling the ivesheadiomorph taxon Blackbrookia, which may be of significance to interpretation of Blackbrookia and other ivesheadiomorphs.

Many of these small frond fossils, including juveniles of existing taxa, members of taxa described elsewhere, and the two newly described taxa, were previously assigned the informal name of “duster” or “featherduster”. The small, stemmed fronds share a similar overall body plan, with a petalodium attached to a stem with a discoid or bulbous holdfast, but the taxa show substantial diversity of petalodium shape and construction. This is an example of convergent evolution, which reveals that biodiversity at the lower elevated tier of the Mistaken Point biota is greater than previously realized.

REFERENCES

Anderson, M.M., and Conway Morris, S. 1982. A review, with descriptions of four unusual

forms, of the soft-bodied fauna of the Conception and St. John’s groups (Late

Precambrian), Avalon Peninsula, Newfoundland. Proceedings of the Third North

American Paleontological Convention, 1: 1–8.

Anderson, M.M., and Misra, S.B. 1968. Fossils found in pre-Cambrian Conception Group of

Southeastern Newfoundland. Nature, 220: 680–681.

Arnott, R.W.C. 2010. Deep-Marine Sediments and Sedimentary Systems. In Facies Models 4.

Edited by N.P. James and R.W. Dalrymple. The Geological Association of Canada. pp.

295–322.

Bamforth, E.L., Narbonne, G.M., and Anderson, M.M. 2008. Growth and ecology of an

Ediacaran multibranched rangeomorph from the Mistaken Point Assemblage,

Newfoundland. Journal of Paleontology, 82: 763–777.

Benus, A.P. 1988. Sedimentological context of a deep-water Ediacaran fauna (Mistaken Point,

Avalon Zone, eastern Newfoundland), In Trace Fossils, Small Shelly Fossils, and the

Precambrian-Cambrian Boundary. Edited by E. Landing, G. M. Narbonne, and P. M.

Myrow. New York State Museum and Geological Survey Bulletin, pp. 8–9.

Bezerra, F.H.R., Fonseca, V.P., and Filho. F.P.L. 2001. Seismites: origin, criteria for identification

and examples from the Quaternary record of Northeastern Brazil. Pesquisas em

Geociencias, 28: 205–212.

Billings, E. 1872. On some fossils of the primordial rocks of Newfoundland. Canadian Naturalist

and Quarterly Journal of Science, 6:465–479.

Bouma, A.H. 1962. Sedimentology of some flysch deposits. Elsevier, Amsterdam, The

Netherlands.

Boynton, H.E., and Ford, T.D. 1995. Ediacaran fossils from the Precambrian (Charnian

Supergroup) of Charnwood Forest, Leicestershire, England. Mercian Geologist, 13: 165–

182.

Boynton, H.E., and Ford, T.D. 1996. Ediacaran fossils from the Precambrian (Charnian

Supergroup) of Charnwood Forest, Leicestershire, England – revision of nomenclature.

Mercian Geologist, 14: 3.

Brasier, M.D., and Antcliffe, J. 2009. Evolutionary relationships within the Avalonian Ediacara

biota: new insights from laser analysis. Journal of the Geological Society, London, 166:

363–84.

Buss, L.W., and Seilacher, A. 1994. The Phylum Vendobionta: a sister group of the Eumetazoa?

Paleobiology 20: 1–4.

Calon, T. 2001. Late Precambrian sedimentation and orogenesis of the Avalon Peninsula, eastern

Avalon Zone, Newfoundland. Field Trip A9/B8, Geological Association of Canada –

Mineralogical Association of Canada Joint Annual Meeting, St. John's, NL.

Canfield, D.E., Poulton, S.W., and Narbonne, G.M. 2007. Late Neoproterozoic deep ocean

oxygenation and the rise of animal life. Science, 315: 92-95.

Cantero, M.I., Cantelli, A., Pirmez, C., Balachandar, S., Mohrig, D., Hickson, T.A., Yeh, T.,

Naruse, H. and Parker, G. 2012. Emplacement of massive turbidites linked to extinction

of turbulence inturbidity currents. Nature Geoscience, 5: 42–45.

Clapham, M.E. and Narbonne, G.M. 2002. Ediacaran epifaunal tiering. Geology, 30: 527-544.

Clapham, M.E., Narbonne, G.M., and Gehling, J.G. 2003. Paleoecology of the oldest- known

animal communities: Ediacaran assemblages at Mistaken Point, Newfoundland.

Paleobiology, 29: 527-544.

Clapham, M.E., Narbonne, G.M., Gehling, J.G., Greentree, C., and Anderson, M.M. 2004.

Thectardis avalonensis: A new Ediacaran fossil from Mistaken Point, Newfoundland.

Journal of Paleontology, 78: 1031-1036.

Dalrymple, R.W., and Narbonne G.M. 1996. Continental slope sedimentation in the Sheepbed

Formation (Neoproterozoic, Windermere Supergroup), Mackenzie Mountains, N.W.T.

Canadian Journal of Earth Sciences, 33: 848–862.

Darroch, S.A.F., Laflamme, M., and Clapham, M.E. 2013. Population structure of the oldest

known macroscopic communities from Mistaken Point, Newfoundland. Paleobiology, 39:

591-608.

Darwin, C. 1859. On the Origin of Species. John Murray, London.

Druschke, P., Hanson, A.D., and Wells, M.L. 2009. Structural, stratigraphic, and geochronologic

evidence for extension predating Palaeogene volcanism in the Sevier hinterland, east-

central Nevada. International Geology Review. 51: 743–775.

Dykstra, M. 2012. Deep water tidal sedimentology, In Principles of Tidal Sedimentology. Edited

by Davis, R.A. and Dalrymple, R.W. Springer. pp. 371–395.

Dzik, J. 2002. Possible ctenophoran affinities of the Precambrian ‘seapen’ Rangea. Journal of

Morphology, 252: 315–334.

Erwin, D.H., Laflamme, M., Tweedt, S.M., Sperling, E.A., Pisani, D., and Peterson, K.J. 2011.

The Cambrian conundrum: early divergence and later ecological success in the early

history of animals. Science, 334: 1091–1097.

Eyles, N., and Eyles, C.H. 1989. Glacially-influenced deep-marine sedimentation of the Late

Precambrian Gaskiers Formation, Newfoundland, Canada. Sedimentology, 36: 601–620.

Fedonkin, M.A. 1985. Systematic description of the Vendian metazoa. p. 71-120. In The Vendian

System: Volume 1. : Paleontology. Edited by B.S. Solokov and A.B. Iwanowski.

Springer-Verlag, Berlin.

Fedonkin, M.A., Gehling, J.G., Grey, K., Narbonne, G.M., and Vickers-Rich, P. 2007. The Rise

of Animals: Evolution and Diversification of the Kingdom Animalia. Johns Hopkins

University Press, Baltimore, Maryland.

Flude, L.I., and Narbonne, G.M. 2008. Taphonomy and ontogeny of a multibranched Ediacaran

fossil: Bradgatia from the Avalon Peninsula of Newfoundland. Canadian Journal of Earth

Sciences, 45: 1095–1109.

Ford, T.D. 1958. Pre-Cambrian fossils from Charnwood Forest. Proceedings of the Yorkshire

Geological Society, 31: 211–217.

Fuller, C.W., Willett, S.D., and Brandon, M.T. 2006. Formation of forearc basins and their

influence on subduction zone earthquakes. Geology, 34: 65–68.

Gardiner, S., and Hiscott, R.N. 1988. Deep-water facies and depositional setting of the Lower

Conception Group (Hadrynian), southern Avalon Peninsula, Newfoundland. Canadian

Journal of Earth Sciences, 25: 1579–1594.

Gehling, J.G. 1999. Microbial mats in terminal Neoproterozoic siliciclastics: Ediacaran death

masks. Palaios, 14: 40-57.

Gehling, J.G., Narbonne, G.M., and Anderson, M.M. 2000. The first named Ediacaran body

fossil, Aspidella terranovica. Palaeontology, 43: 427-456.

Gehling, J.G., Droser, M.L., Jensen, S., and Runnegar, B.N. 2005. Ediacara organisms: Relating

form to function. In Evolving form and function: Fossils and development. Edited by

D.E. Briggs. Yale University Peabody Museum of National History, New Haven,

Connecticut. pp. 43–66.

Gehling, J.G., and Narbonne, G.M. 2007. Spindle-shaped Ediacara fossils from the Mistaken

Point assemblage, Avalon Zone, Newfoundland. Canadian Journal of Earth Sciences, 44:

367-387.

Glaessner, M.F., and Daily, B. 1959. The geology and Late Precambrian fauna of the Ediacara

fossil reserve. Records of the South Australian Museum, 13: 369–407.

Glaessner, M.F., and Wade, M. 1966. The late Precambrian fossils from Ediacara, South

Australia. Palaeontology, 9: 599–628.

Grazhdankin, D. 2004. Patterns of distribution in the Ediacaran biotas: facies versus

biogeography and evolution. Paleobiology, 30: 203–221.

Grazhdankin, D., and Gerdes, G. 2007. Ediacaran microbial colonies. Lethaia, 40:201–210.

Gürich, G. 1930. Die bislang altesten Spuren von Organismen in Sudafrika. International

Geological Congress (XV), Pretoria, Union of South Africa, Die altesten Fossilien Sud-

Afrikas, 2: 670–680.

Gürich, G. 1933. Die Kuibis-Fossilien der Nama-Formation von Südwestafrika.

Palaeontologische Zeitschrift, 15: 137–154.

Hesse, R. 1975. Turbiditic and non-turbiditic mudstone of Cretaceous flysch sections of the east

Alps and other basins. Sedimentology, 22: 307-416.

Hoffman, P.F., Kaufman, A.J., Halverson, G.P., and Schrag, D.P. 1998. A Neoproterozoic

snowball earth. Science, 281: 1342-13346.

Hoffman, P.F., and Schrag, D.P. 2002. The snowball Earth hypothesis: testing the limits of global

change. Terra Nova, 14: 129-155.

Hoffman, P.F., and Li, Z.X. 2009. A palaeogeographic context for Neoproterozoic glaciation.

Palaegeography, Palaeoclimatology, Palaeoecology, 277: 158-172.

Hofmann, H.J., O'Brien, S.J., and King, A.F. 2008. Ediacaran biota on Bonavista Peninsula,

Newfoundland, Canada. Journal of Paleontology, 82: 1–36.

Ichaso, A.A., Dalrymple, R.W., and Narbonne, G.M. 2007. Paleoenvironmental and basin

analysis of the late Neoproterozoic (Ediacaran) upper Conception and St. John's groups,

west Conception Bay, Newfoundland. Canadian Journal of Earth Sciences, 44: 25–41.

Jenkins, R.J.F. 1985. The enigmatic Ediacaran (late Precambrian genus) Rangea and related

forms. Paleobiology, 11: 336–355.

Jenkins, R.J.F. and Gehling, J.G. 1978. A review of the frond-like fossils of the Ediacara

assemblage. Records of the South Australian Museum, 17: 347–359.

Jenkins, R.J. 1992. Functional and ecological aspects of Ediacaran assemblages. In Origin and

early evolution of the metazoa. Edited by J.H. Lipps and P.W. Signor. Plenum Press, New

York. pp. 131–176.

Jenness, S.E. 1963. Terra Nova and Bonavista map-areas (2D E1/2 2C). Geological Survey of

Canada Memoir 327.

King, A.F. 1988. Geology of the Avalon Peninsula, Newfoundland (parts of 1K, 1L, 1M, 1N, and

2C). Newfoundland Department of Mines, Mineral Development Division, Map 88-01,

scale 1:250 000.

King, A.F. 1990. Geology of the St. John's area. Report 90-2, Newfoundland Department of

Mines and Energy, Geological Survey Branch.

Knoll, A.H., M. R. Walter, G. M. Narbonne, and N. Christie-Blick. 2004. A new period for the

geologic time scale. Science, 305: 621-622.

Knoll, A.H., Walter, M.F., Narbonne, G.M., and Christie-Blick, N. 2006. The Ediacaran Period:

A new addition to the geologic time scale. Lethaia, 39: 13-30.

Knoll, A.H., and Carroll, S.B. 1999. The early evolution of animals: Emerging views from

comparative biology and geology. Science, 284: 2129-2137.

Krogh, T.E., Strong, D.F., O’Brien, S.J., and Papezik, V. 1988. Precise U–Pb dates from zircons

in the Avalon Terrane of Newfoundland. Canadian Journal of Earth Sciences, 25: 442–

453.

Laflamme, M., Narbonne, G.M., and Anderson, M.M. 2004. Morphometrics of the Ediacaran

frond Charniodiscus from the Mistaken Point Formation, Newfoundland. Journal of

Paleontology, 78: 827-837.

Laflamme, M., Narbonne, G.M., Greentree, C., and Anderson, M.M. 2007. Morphology and

taphonomy of the Ediacaran frond: Charnia from the Avalon Peninsula of Newfoundland,

p 237–257. In The Rise and Fall of the Ediacaran Biota. Edited by P. Vickers- Rich and P.

Komarower. Geological Society, London, Special Publications 286.

Laflamme, M., and Narbonne, G.M. 2008a. Competition in a Precambrian world: Palaeoecology

and functional biology of Ediacaran fronds. Geology Today, 24: 182-187.

Laflamme, M., and Narbonne, G.M. 2008b. Ediacaran fronds. Palaeogeography,

Palaeoclimatology, Palaeoecology, 258: 162–179.

Laflamme, M., Xiao, S., and Kowalewski, M. 2009. Osmotrophy in modular Ediacara

organisms. Proceedings of the National Academy of Sciences of the United States of

America, 106: 14438–14443.

Laflamme, M., Schiffbauer, J.D., and Narbonne, G.M. 2011. Deep-water Microbially Induced

Sedimentary Structures (MISS) in deep time: the Ediacaran fossil Ivesheadia, In

Microbial Mats in Siliciclastic Depositional Systems Through Time. Edited by N. K.

Noffke, H. Chafetz. SEPM Special Publication 101. pp. 111–123.

Laflamme, M., Flude, L.I. and Narbonne, G.M. 2012. Ecological tiering and the evolution of a

stem: the oldest stemmed frond from the Ediacaran of Newfoundland, Canada. Journal of

Paleontology, 86: 193–200.

Laflamme, M., Darroch, S.A.F, Tweedt, S.M., Peterson, K.J., and Erwin D.H. 2013. The end of

the Ediacara biota: Extinction, biotic replacement, or Cheshire Cat? Gondwana Research,

23: 558-573.

Li, X.Z., Bogdanova, S.V., Collins, A.S., Davidson, B., De Waele, R.E., Ernst, I.C., Fitzsimons,

W., Fuck, R.A., Gladkochub, D.P., Jacobs, J., Karlstrom, K.E., Lu, S., Pease, V.,

Pisarevsky, S.A., Thrane, K., and Vernikovsky, V. 2008. Assembly, configuration, and

break-up history of Rodinia: A synthesis. Precambrian Research, 160: 179–210.

Liu, A.G., McIlroy, D., and Brasier, M.D. 2010. First evidence for locomotion in the Ediacara

biota from the 565 Ma Mistaken Point Formation, Newfoundland. Geology, 38: 123–126.

Liu, A.G., McIlroy, D., Antcliffe, J.B., and Brasier, M.D. 2011. Effaced preservation in the

Ediacara biota and its implications for the early macrofossil record. Palaeontology, 54:

607–630.

Liu, A.G., McIlroy, D., Matthews, J.J., and Brasier, M.D. 2012. A juvenile assemblage of

Ediacaran fronds from the Drook Formation, Newfoundland. Journal of the Geological

Society of London 169: 395–403.

Mason, S.J., Narbonne, G.M., Dalrymple, R.W., and O'Brien, S.J.. 2013. Paleoenvironmental

analysis of Ediacaran strata in the Catalina Dome, Bonavista Peninsula, Newfoundland.

Canadian Journal of Earth Sciences. 50: 197–212.

Misra, S.B. 1971. Stratigraphy and depositional history of the late Precambrian coelenterate-

bearing rock, southeastern Newfoundland. Geological Society of America Bulletin, 82:

979–987.

Montenat, C., Barrier, P., Ott d'Estevoum, P., and Hibsch, C. 2007. Seismites: An attempt at

critical analysis and classification. Sedimentary Geology, 196: 5–30.

Murphy, J.B., and Nance, R.D. 1989. Model for the evolution if the Avalonian-Cadomian belt.

Geology, 17: 735–738.

Murphy, J.B., Keppie, J.D., Dostal, J., and Nance, R.D. 1999. Neoproterozoic – early Paleozoic

evolution of Avalonia. In Laurentia—Gondwana connections before Pangea. Edited by V.

A. Ramos and J. D. Keppie. Geological Society of America, Special Paper 336: 253–266.

Mutti, E., Ricci Lucchi, F., Seguret, M. and Zanzucchi, G. 1984. Seismoturbidites: a new group

of resedimented deposits. Marine Geology, 55: 103–116.

Myrow, P.M. 1995. Neoproterozoic rocks of the Newfoundland Avalon zone. Precambrian

Research, 7: 123–136.

Myrow, P.M., and Kaufman, A.J. 1999. A newly discovered cap carbonate above Varanger-age

glacial deposits in Newfoundland, Canada. Journal of Sedimentary Geology, 69: 784-793.

Myrow, P.M., Fischer, W., and Goodge, J.W. 2002. Wave-modified turbidites: combined-flow

shoreline and shelf deposits, Cambrian, Central Transantarctic Mountains. Journal of

Sedimentary Research, 72: 641–656.

Nance, R.D., Murphy, J.B., and Keppie, J.D. 2002. A Cordilleran model for the evolution of

Avalonia. Tectonophysics, 352: 11–31.

Narbonne, G.M., Saylor, B.Z., and Grotzinger, J.P. 1997. The youngest Ediacaran fossils from

Southern Africa. Journal of Paleontology, 71: 953–967.

Narbonne, G.M., Dalrymple, R.W., Gehling, J.G., Wood, D.A., Clapham, M.E., and Sala, R.A.

2001. Neoproterozoic fossils and environments of the Avalon Peninsula, Newfoundland.

Field Trip B5, Geological Association of Canada – Mineralogical Association of Canada

Joint Annual Meeting, St. John's, NL.

Narbonne, G.M. 2004. Modular construction of complex early Ediacaran life forms. Science,

305: 1141-1144.

Narbonne, G.M. 2005. The Ediacara biota: Neoproterozoic origin of animals and their

ecosystems. Annual Reviews of Earth and Planetary Sciences, 33: 13.1-13.22.

Narbonne, G.M., and Gehling, J.G. 2003. Life after Snowball: the oldest complex Ediacaran

fossils. Geology, 31: 27-30.

Narbonne, G.M., Gehling, J.G., and Vickers-Rich, P. 2007. The misty coasts of Newfoundland.

In The Rise of Animals: Evolution and diversification of the Kingdom Animalia. Edited

by M.A. Fedonkin, J.G. Gehling, K. Grey, G.M. Narbonne, and P. Vickers-Rich. John

Hopkins University Press, Baltimore, Maryland. pp. 53–68.

Narbonne, G.M., Laflamme, M., Greentree, C. and Trusler, P. 2009. Reconstructing a lost world:

Ediacaran rangeomorphs from Spaniard's Bay, Newfoundland. Journal of Paleontology,

83: 503–523.

Narbonne, G.M. 2010. Ocean chemistry and early animals. Science, 328: 53-54.

Narbonne, G.M. 2011. When life got big. Nature, 470: 339–340.

Narbonne, G.M., Xiao, S., and Shields, G.A. 2012a. The Ediacaran Period. In Geological

Timescale 2012. Edited by F. Gradstein, J. Ogg, and G. Ogg. Elsevier, pp. 427-449.

Narbonne, G.M., Laflamme, M., and Thomas, R. 2012b. When life got big: Ediacaran glaciation,

oxidation, and the Mistaken Point biota of Newfoundland. GAC-MAC Joint Annual

Meeting Field Trip Guidebook – B1.

Normore, L.S. 2010. Geology of the Bonavista map-area (NTS 2C/11), Newfoundland. Current

Reasearch Report 10-1, Newfoundland and Labrador Department of Natural Resources

Geological Survey, pp. 281–301.

O'Brien, S.J., Wardle, R.J., and King, A.F. 1983. The Avalon zone: a Pan-African terrane in the

Appalachian Orogen of Canada. Geological Journal, 18: 195–222.

O'Brien, S.J., O'Niell, P.P., King, A.F., and Blackwood, R.F. 1988. Eastern margin of the

Newfoundland Appalachians – a cross-section of the Avalon and Gander zones.

Geological Association of Canada/Mineralogical Association of Canada Field Trip

Guidebook, Trip B4, pp. 59–62.

O'Brien, S.J., and King, A.F. 2002. Neoproterozoic stratigraphy of the Bonavista Peninsula:

preliminary results, regional correlations and implications for sediment-hosted stratiform

copper exploration in the Newfoundland Avalon Zone. Current Research Report 02-01,

Newfoundland Department of Mines and Energy – Geological Survey, pp. 29–244.

O'Brien, S.J., and King, A.F. 2004a. Ediacaran fossils from the Bonavista Peninsula (Avalon

Zone), Newfoundland: Preliminary descriptions and implications for regional correlation.

Newfoundland Department of Mines and Energy—Geological Survey Report 04-01:203–

212.

O'Brien, S.J., and King, A.F. 2004b. Late Neoproterozoic to earliest Paleozoic stratigraphy of

Avalon Zone in the Bonavista Peninsula, Newfoundland: An update. Current Research

Report 04-01, Newfoundland Department of Mines and Energy – Geological Survey, pp.

213–224.

O'Brien, S.J., and King, A.F. 2005. Late Neoproterozoic (Ediacaran) stratigraphy of Avalon Zone

sedimentary rocks, Bonavista Peninsula, Newfoundland, Current Research Report 05-1,

Newfoundland and Labrador Department of Natural Resources – Geological Survey, pp.

101–113.

O'Brien, S.J., King, A.F. and Hofmann, H.J. 2006. Lithostratigraphic and biostratigraphic studies

on the eastern Bonavista Peninsula: an update. Current Research Report: 06-1,

Newfoundland and Labrador Department of Natural Resources – Geological Survey, pp.

257–263.

Peterson, K.J., Waggoner, B. and Hagadorn, J.W. 2003. A fungal analog for Newfoundland

Ediacaran fossils? Integrated and Comparative Biology, 43: 127–136.

Pflug, H.D. 1970. Zur Fauna der Nama-Schichten in Südwest-Afrika. II. Rangeidae, Bau und

Systematische Zugehörigkeit. Palaeontographica, A135: 198–231.

Pflug, H.D., 1972. Systematik der jung-präkambrischen Petalonamae. Paläontologische

Zeitschrift, 46: 56–67.

Pisarevsky, S.A., McCausland, P.J.A., Hodych, J.P., O'Brien, S.J., Tait, J.A., and Murphy, J.B.

2011. Paleomagnetic study of the late Neoproterozoic Bull Arm and Crown Hill

formations (Musgravetown Group) of eastern Newfoundland: implications for Avalonia

and West Gondwana paleogeography. Canadian Journal of Earth Sciences, 49: 308–327.

Retallack, G.J. 1994. Were the Ediacaran fossils lichens? Paleobiology, 20: 523–544.

Retallack, G.J. 2012. Ediacaran paleosols from the Mistaken Point Formation of Newfoundland.

Proceedings of the Oregon Academy of Science. 71: 33.

Retallack, G.J. 2013a. Ediacaran life on land. Nature, 492: 89-92.

Retallack, G.J. 2013b. Ediacaran Gaskiers glaciation of Newfoundland reconsidered. Journal of

the Geological Society of London, 170: 19-36.

Seilacher, A. 1969. Fault-graded beds interpreted as seismites. Sedimentology, 13: 155–159.

Seilacher, A. 1989. Vendozoa; organismic construction in the Proterozoic biosphere. Lethaia, 22:

229–239.

Seilacher, A. 1992. Vendobionta and Psammocorallia: lost constructions of Precambrian

evolution. Geological Society of London, Special Publications, 149: 607–613.

Seilacher, A. 1999. Biomat-related lifestyles in the Precambrian. Palaios, 14: 86–93.

Sperling, E.A., Peterson, K.J., and Laflamme, M. 2011. Rangeomorphs, Thectardis (Porifera?)

and dissolved organic carbon in the Ediacaran ocean. Geobiology, 9: 24–33.

Sprigg, R.C. 1947. Early Cambrian (?) jellyfishes from the Flinders ranges, South Australia.

Transactions of the Royal Society of South Australia, 71: 212–224.

100

Sprigg, R.C. 1949. Early Cambrian “jellyfishes” of Ediacara, South Australia and Mount John,

Kimberley District, Western Australia. Transactions of the Royal Society of South

Australia, 73: 72–99.

Steiner, M., and Reitner, J. 2001. Evidence of organic structures in Ediacara-type fossils and

associated microbial mats. Geology, 29: 1119–1122.

Van Kronendonk, M.J., Gehling, J.G., and Shields, G. 2008. Precambrian. In The Concise

Geologic Time Scale. Edited by J. G. Ogg, G. Ogg, and F.M. Gradstein. Cambridge

University Press, Cambridge, U.K. pp. 23–36.

Vickers-Rich, P., Ivantsov, A.Y, Trusler, P.W., Narbonne, G.M., Hall, M., Wilson, S.A.,

Greentree, C., Fedonkin, M.A., Elliott, D.A., Hoffmann, K.H., and Schneider, G.I.C.

2013. Reconstructing Rangea: New discoveries from the Ediacaran of Southern Namibia.

Journal of Paleontology, 87: 1-15.

Vrolijk, P.J., and Southard, J.B. 1997. Experiments on rapid deposition of sand from high-

velocity flows. Geoscience Canada, 24: 45–54.

Wilby, P.R., Carney, J.N., and Howe, M.P.A. 2011. A rich Ediacaran assemblage from eastern

Avalonia: evidence of early widespread diversity in the deep ocean. Geology, 39: 655–

658.

Williams, H. and King, A.F. 1979. Trepassey map area, Newfoundland. Geological Survey of

Canada Memoir 389.

Wood, D.A., Dalrymple, R.W., Narbonne, G.M., Gehling, J.G., and Clapham, M.E. 2003.

Environmental analysis of the late Neoproterozoic Mistaken Point and Trepassey

Formations, southeastern Newfoundland: paleobiological and tectonic implication.

Canadian Journal of Earth Sciences, 40: 1375-1391.

101

Xiao, S., and Knauth, L.P. 2013. Fossils come in to land. Nature, 493: 28-29.

Xiao, S., and Laflamme, M. 2009. On the eve of animal radiation: phylogeny, ecology, and

evolution of the Ediacara biota. Trends in Ecology & Evolution, 24: 31–40.

Xiao, S., Droser, M., Gehling, J.G., Hughes, I.V., Wan, B., Chen, Z., and Yuan, X. 2013.

Affirming life aquatic for the Ediacara biota in China and Australia. Geology, 41: 1095-

1098.

Xie, X., and Heller, P.L. 2009. Plate tectonics and basin subsidence history. Geological Society

of America Bulletin, 121: 55–64.

102