Evaluation of Stromatolites from the 3.4 Ga Kromberg Formation, Barberton Greenstone Belt, South Africa Corey E

Home , Pilbara Craton, Raman spectroscopy, Warrawoona Group

Louisiana State University LSU Digital Commons

LSU Master's Theses Graduate School

2014 Evaluation of Stromatolites from the 3.4 Ga Kromberg Formation, Barberton Greenstone Belt, South Africa Corey E. Shircliff Louisiana State University and Agricultural and Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_theses Part of the Earth Sciences Commons

Recommended Citation Shircliff, Corey E., "Evaluation of Stromatolites from the 3.4 Ga Kromberg Formation, Barberton Greenstone Belt, South Africa" (2014). LSU Master's Theses. 1507. https://digitalcommons.lsu.edu/gradschool_theses/1507

This Thesis is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Master's Theses by an authorized graduate school editor of LSU Digital Commons. For more information, please contact [email protected]. EVALUATION OF STROMATOLITES FROM THE 3.4 GA KROMBERG FORMATION, BARBERTON GREENSTONE BELT, SOUTH AFRICA

A Thesis

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Master of Science

The Department of Geology and Geophysics

by Corey Elizabeth Shircliff B.S., Beloit College, 2011 May 2014

This thesis is dedicated to my parents, Jim and Jenny Shircliff. Thank you for inspiring

and encouraging me.

ii Acknowledgements

Although it is difficult to adequately thank them, the author is extremely grateful to Dr. Gary Byerly, her advisor, for the countless hours of assistance given, experience in the field, and honest guidance throughout her career as a graduate student. She would also like to thank Dr. Maud Walsh for valuable discussions, assistance in the field, and for the use of the samples she collected in the field.

Thanks are also due to Dr. Sophie Warny for guidance and valuable discussion as a committee member, Dr. Achimm Hermann, for use of his microscope, Jill Bambrick

Banks for her helpfulness with the handheld XRF, Dr. Donald Lowe for his assistance in the field, Rick Young for aid in the rock lab, Chris O’Loughlin and the LSU Raman laboratory, and to Heather Lee, who has been extremely helpful and patient with the author.

iii Table of Contents

Acknowledgements………………………………………………………………………………....iii

List of Tables…………………………………………………………………………………...vi

List of Figures………………………………………………………………………………………....vii

Abstract……………………………………………………………………………………………………………………………………..ix

1. Introduction………………………………………………………………………………..1

2. Geologic Setting………………………………………………………………...………4..7 2.1. Barberton Greenstone Belt……………………………………...…………..7 2.2. Kromberg Formation………………...……...………………………………..10

3. Methods…………………………………………………………………………………...17 3.1. Bulk Rock X-Ray Fluorescence………...…………………………………..17 3.2. Raman Spectroscopy…………………...…………………………………..17 3.3. ICP-Mass Spectrometry………………...…………………………………..14..18 3.4. δ13C Isotopes ………….………………...………………………………….14..18

4. Results……………………………………………………………………………………..19 4.1. Domical Type………………..…...……………………………………...... 20 4.2. Flat-Laminated Type……………...…………………………………………...25 4.3. Bulk Rock X-Ray Fluorescence…….………………………………………25..30 4.4. Raman Spectroscopy………………...…………………………………...…27..33 4.5. ICP-Mass Spectrometry….…………....……....…………………….……....35 4.6. δ13C Isotopes ………....……………....……………………………………..36..39

5. Discussion………………………………………………………………………………....41 5.1. Morphologic Features ……………....………………………………………..41 5.2. Chemistry……………………………………………………………………....47

6. Conclusion.………………………………………………………………………………...54

References……………………………………………………………………………………..56

Appendix A: Locations of Sections………………...………………………………………...62

Appendix B: Complete Measured Sections………………………………………………………...67..63

Appendix C: Expanded Methods……….…………………………………………………....76

Appendix D: Handheld XRF Results………………………………………………………...78

Vita…………………………………………………………………………………………...79

v List of Tables

Table 1. Methods and implications for identifying stromatolitic origin……………………………...4

Table 2. Summary of results…………………………………………………………………………………………………...19

Table 3. Major element percent composition-- XRF…………………………………………………………………………………………....31

Table 4. Trace element composition in ppm (XRF)…………………………………………………………………………..31

Table 5. Trace and rare earth element amounts (ppm)……………………………………………………....37

Table 6. Cerium and europium anomalies…………………………………………………………………………..38

Table 7. Ua and log FeO values for K1 samples…………………………………………………………………...39

Table 8. Carbon and carbon isotopes………………………………………………………………………………………………………....40

vi List of Figures

Figure 1. Generalized geologic map of the BGB………………………………………………………………….....8

Figure 2. Geologic map of the field area……………………………………………………………………………....9

Figure 3. Generalized stratigraphic column………………………….…………………………………………………....11

Figure 4. Generalized basal K1 lithofacies………………………………………………………………………...... 13

Figure 5. Generalized K1 measured sections………………………………………………………………………...15

Figure 6. Sample CES 10-13 (domical sample)……………………………………………………………………………..21

Figure 7. Raman shirft results for sample CES 10-13……………………………………………………………………...21

Figure 8. Domical stromatolite; light microscope image………………………………………………………………………………....23

Figure 9. Domical stromatolite trough; light microscope image……………………………………………………………..24

Figure 10. Interior cut surfaces of flat-laminated forms……………………………………………………………....26

Figure 11. Sample CES 4-2b………………………………………………………………………………………………...27

Figure 12. Raman shift results for sample CES 4-2b………………………………………………………………………....28

Figure 13. Fine resolution Raman shift results for sample CES 4-2b…………………………………....28

Figure 14. Ripped-off laminae package in sample CES 8-4……………………………………………………………………………....29

Figure 15. Sample CES 8-6; light microscope image………………………………………………………………………………..30

Figure 16. Uranium vs Nb, Pb, Zr, and Ti (ppm)…………………………………………………………………...32

Figure 17. TiO2 vs Nb…………………………………………………………………………………………………………………...32

Figure 18. Raman shift results for sample CES 8-6………………………………………………………………………………...33

Figure 19. Raman shift results for sample CES 8-4………………………………………………………………………………………....34

vii Figure 20. Rare earth element compositions compared to the Primitive Mantle………………………………………………..36

Figure 21. Spider diagram of K1 samples compared to the PM……………………………………………………………………..38

Figure 22. Authigenic Uranium (Ua) vs log FeO……………………………………………………………………..39

viii Abstract

Silicified sedimentary rocks from the 3.4 Ga Kromberg Formation of the

Barberton Greenstone Belt in South Africa contain laminated structures that have been identified as possible stromatolites in the field. Morphological evaluation and a variety of chemical analyses are presented here, in an effort to describe the samples in a sedimentary context and consider biogenicity of these laminated forms. Two major types of laminated structures were identified in the field – domical laminates and flat-laminated samples with little to no synoptic relief. The domical sample presents the best morphological evidence for biogenicity. There are several characteristics that suggest the deposition must be biologically mediated: dome slopes are greater than 40º and their crests have thickened laminae, varied fine-grained sand bimodal depositional patterns appear within the domes, with a high degree of laminae inheritance from the base of the sample to the top. The flat-laminated samples, while lacking domical morphology, do show high levels of lamina cohesion, mineralogic deposits in individual lamina, and, in most cases, a high degree of laminae inheritance. Raman spectroscopy indicates that the laminae in the domical and flat-laminated samples are carbonaceous, with strong disordered and ordered carbon peaks appropriate for indigenous carbon in these greenschist facies. Although the carbonaceous matter is less than 1% of the rock, samples from the lower K1 Member of the Kromberg Formation were analyzed for δ13C, and the values range from -29‰ to -39‰, which is consistent with the isotopic signatures of autotrophic microbes. Rare earth element (REE) analyses indicate that the depositional environment was marine and anoxic. With all the evidence taken together, the author suggests it is more plausible for the domical sample to be biogenic. Additionally, it is

ix likely that the flat-laminated samples are also biogenic, even though there is no strong resemblance to modern stromatolites. However, they do resemble modern microbial mats, further supporting a biogenic interpretation.

x 1. Introduction

For much of the last century, scientists have been trying to explain the evolution of life on early Earth, particularly when and how our microbial ancestors developed.

However, the issue is a challenge to fully comprehend because so little evidence remains.

Fortunately, there are some ancient sedimentary outcrops where investigations are yielding new possibilities – or at least stimulating some debate (Schopf et al., 2007 for a review of these locations). The study of sedimentary rocks reveals much information about the environment of deposition that potentially includes evidence for the existence of ancient microbial life. This thesis looks at one type of structure found in the some of the oldest and well-preserved sedimentary rocks on Earth – stromatolites. An examination of these putative biogenic structures could shed light on how early life evolved and help clarify the timing of this process.

Part of the reason it is difficult to evaluate stromatolites in the Archean is because the definition of a stromatolite is not well-constrained, and has changed over the last century (Awramik and Grey, 2005). When the term was first used in publication in 1908 by Ernst Kalkowsky, it referred to laminated limestone that was microbially formed. A succinct definition of what a stromatolite has come to mean today is by Schopf (2006): a stromatolites is “an accretionary sedimentary structure, commonly thinly layered, megascopic and calcareous, interpreted to have been produced by the activities of mat- building communities of mucilage-secreting micro-organisms, mainly photoautotrophic prokaryotes”. The benefit of this definition is that it is applicable to stromatolites of any age, unlike some of its precursors (Schopf, 2006).

1 The oldest stromatolites, 3.5-3.3 Ga, have been investigated in both the Western

Australian and South African sedimentary and metasedimentary sequences (ex: Lowe,

1980; Walter et al, 1980; Walter, 1983; Walsh and Lowe, 1985; Byerly et al, 1986;

Beukes and Lowe, 1989; Hofmann, 2000; Allwood et al., 2009;). Despite the abundance of research, there is still some debate on how to determine biogenicity of a layered structure that appears to be a fossil stromatolite. Most of this dispute focuses on how far stromatolites can be unambiguously traced into the fossil record. Although stromatolites are found in some of the earliest sedimentary rocks, they are billions of years old and thus have undergone diagenetic and metamorphic alteration. Because microbes build stromatolites, microfossils would be expected in these structures. To date, none have been found in association with ancient fossil stromatolites. However, the extensive recrystallization present in the stromatolites obscures some of the original textures, which may account for the paucity of microfossils that once existed. Even without microfossil evidence, many ancient stromatolites exhibit morphological similarities to modern stromatolites (Allwood et al., 2009; Petroff et al., 2010). Still, there has been some debate on the origin of reported ancient stromatolites, with suggestions of abiogenic stromatolite growth (Lowe, 1994; Grotzinger and Rothman, 1996; Brasier et al., 2002). The first widely accepted research on Paleo-Archean stromatolites was published in the early

1980s. Several publications identified stromatolites and/or algal mats in the South

African Barberton Greenstone Belt (BGB) as well as the Pilbara craton of Western

Australia (Walter et al, 1980; Lowe, 1983; Walter, 1983; Walsh and Lowe, 1985; Byerly et al, 1986; Beukes and Lowe, 1989; Walsh, 1992). However, in 1994 Lowe argued that the textures described in some previously published papers could have been produced

2 abiogenically. These papers include the study of Western Australian stromatolites by both

Walter (1980) and Lowe (1983) and the study of BGB stromatolites by Byerly et al

(1986). Lowe claimed the oldest stromatolite with unquestionable evidence was of the

3,000 Myr Pongola Supergroup, South Africa (Beukes and Lowe, 1989).

Some research has suggested abiogenesis for Archean stromatolites (Lowe, 1994;

Grotzinger and Rothman, 1996; Brasier et al., 2002). With that said, convincing data have been presented over the last decade for stromatolites with biologic origin (e.g. van

Kranendonk et al., 2003; Allwood et al., 2009; Petroff et al., 2010). Much of this research focuses on the morphological attributes of the stromatolites and the environment of deposition in which the stromatolites grew and were preserved. The presence of additional fossil life forms (such as microbial mats and microfossils) within the same rock units as stromatolites is sometimes cited as a supporting line of evidence for the existence of biogenic stromatolites. However, the existence of other microbial evidence has also been questioned. In 2003 and 2004, two papers were published by Garcia-Ruiz et al. and Brasier et al., respectively, both of which questioned the evidence for Archean microfossils (Garcia-Ruiz et al., 2003) and for determining biogenicity of any ancient structures (Brasier et al., 2004). After these papers, research began to focus on multiple aspects of microbial mats and/or stromatolites, such as morphology, carbon isotopes, trace element chemistry, Raman spectroscopy, and other techniques to determine biogenicity in Archean sediments (Table 1) (Tice and Lowe, 2006; Allwood et al., 2010;

Sugahara and Sugitani, 2010; Marshall et al., 2012).

Although morphological evidence is typically the focus of the biogenicity debate, chemical evidence is frequently used to support claims of biogenicity (eg. Tice and Lowe

3 2006a; Schopf et al., 2007; Allwood et al., 2009). In their review on organic matter in the

Kromberg, Tice and Lowe (2006a) provide multiple chemical analyses as supporting a biogenic claim for microbial mat structures. In addition to looking at specific chemistry of the organic matter, they also used REEs, and bulk rock chemistry to help understand the environment in which these mats may have been deposited. Schopf and others (2007) review the occurrences of stromatolites in the Precambrian, and discuss the varied chemical methods, such as laser-Raman and carbon isotopic evaluation , which are used to support claims of biogenicity in fossil stromatolites. More recently, Allwood and others (2009) focused on the morphology of the fossil stromatolites, but interpreted the environment of deposition as a shallow carbonate platform based on REE analysis. Based on the literature, a variety of chemical analyses were used here to both examine the stromatolitic structures and understand the environment in which the lower K1 was deposited (Table 1).

Table 1. Methods and implications for identifying stromatolitic origin. Method Implications References Values of -25 ± 10 ‰ are consistent Hayes et al., δ13C Isotopes with carbon isotopic fractionation by 2002; Schopf, autotrophic microbes 2006 Organic Geochemistry Clarifies if carbonaceous matter is Ueno et al., laser-Raman syndepositional; indicates if 2001a; Schopf carbonaceous matter is kerogen-like et al., 2005 Sediment deposition consistent with Buick et al., Thickened lamina over dome crests adhering on exposed microbial surface 1983 High angled domes consistent with Hofmann et al., Morphology Angle of repose greater than 40° stromatolite morphology and not 1999 abiogenic convex structures More regularly laminated in the structures Pattern consistent with stromatolites Hofmann et al., than in intermound areas rather than sedimentary layering 1999

4 As Tice and Lowe (2006) discuss, there is no ‘smoking gun’ when studying ancient samples and looking for evidence of biologic activity. In other words, no single morphological feature that can only be formed biologically and no one chemical analysis that represents biogenicity in ancient sedimentary rock. Adding to this problem is the issue that characterizing stromatolites is not an easy task. Allwood et al. (2009) discuss stromatolites that have complex and varied growth modes and form a variety of structures.

Their work shows the difficulty in establishing both a simple definition of a stromatolite and a characterization of its appearance in the fossil record. The morphological complexity of stromatolites is a large part of the reason that the debate over the biogenicity of ancient layered structures continues.

Many scientists studying ancient stromatolites have attempted to create a list of criteria that must be fulfilled in order to classify a sample as a stromatolite (Buick et al.,

1981; Walter, 1983; Hofmann et al., 1999). This is a difficult task; modern stromatolites are diverse, but not widespread. Additionally, the Precambrian had a much more diversified and widespread population, which makes a simple classification more complicated, particularly when basing some criteria on growth modes of modern stromatolites (Awramik and Grey, 2005). With this said, the criteria presented by

Hofmann and others (1999) is highlighted here, due to their consideration of Archean forms in their list of criteria. Their criteria are specifically tailored to coniform stromatolites from the 3.45 Ga Pilbara craton of Western Australia. The criteria are as follows: (1) there should be greater uniformity if laminae in the coniform structure compared to laminae in the intermound regions which were subjected to more variable environmental conditions; (2) they are not the product of slumping or sideways

5 compression; (3) the continuity of laminae across different structures is difficult to attribute to chemical precipitation; (4) the arrangement and spacing of the structures indicate growth under uniform conditions within the basin for limited intervals of time; and (5) the slopes of the cones is higher than 40º, which is far greater than the angle of repose for loose sediment. The geological content of these features areconsistent with them being biogenic, and they formed in a near-shore marine environment (Hofmann et al., 1999; Awramik and Grey, 2005).Multiple lines of evidence are necessary to prove the biogenicity of ancient fossil life forms, with each increasing the likelihood of biogenicity and subsequently excluding any abiogenic explanation that could be used to explain one type of evidence alone. This research focuses on holistically approaching the question of biogenicity in these rocks by using evidence such as morphological features, field relationships, chemical composition, and mineral composition.

6 2. Geologic Setting

2.1 Barberton Greenstone Belt

Although ancient stromatolites have been reported in both the South African

Barberton Greenstone Belt (BGB) as well as the Pilbara Craton of Western Australia, the focus of this research is on those from the BGB, located in eastern South Africa (Fig. 1).

The BGB is a volcanic and sedimentary sequence of rocks formed during the Paleo-

Archean (Fig. 2). The rocks from the BGB sequence are well-preserved and moderately well-exposed throughout the BGB. This is especially notable because the BGB contains some of the oldest well-preserved and minimally altered sedimentary rocks in the world, and is thus a good place to look for fossil life that may be preserved in these sedimentary sequences.

The BGB stratigraphy is contained within the Barberton Supergroup, with the

Onverwacht, Fig Tree, and Moodies Groups, (Viljoen and Viljoen, 1969; Lowe and

Byerly, 2007) (Figs. 1-3). The Onverwacht, the oldest of the groups, consists mostly of mafic and ultramafic volcanic sequences (Lowe and Byerly, 1999). It is divided into seven formations: the Sandspruit, Theespruit, Komati, Hooggenoeg, Kromberg, Mendon, and Weltevedren (Figs. 1-3). The Inyoka Fault, which is the major fault of the area, divides the BGB into distinct northern and southern domains. The Weltevedren is the only formation from the Onverwacht to be found solely north of the Inyoka fault; the other six are in the southern domain of the BGB. According to Lowe and Byerly (1999), the upper Mendon Formation is probably similar in age to the Weltevedren Formation.

Although much of the 8-10 km thick Onverwacht Group is primarily volcanic rock, the

Onverwacht is especially important for this research because it also has thick chert

7 sequences that represent the sedimentary deposition of the time (Lowe and Byerly, 1999).

These thick cherts are probably a result of low-temperature metasomatic alteration; low

temperature fluid with dissolved silica upwelled to the sediment-water interface, where it

then silicified the newly-deposited sediment and formed an impenetrable chert cap that

helped preserve the features of the original seafloor (Knauth and Lowe, 1978; 2003) It is

within these cherts that we find stromatolites and other evidence of life (ex: Byerly et al.,

1986; Walsh, 1992; Tice and Lowe, 2006).

Figure 1. Generalized geologic map of the BGB. The research area is highlighted by the dashed box. K1 is at the base of the blue Kromberg. Modified from Lowe et al., 2012.

Modified unit. K1 2012. al., et from Lowe Kromberg the in Samples collected Geologic 2. area. were map field the of Figure

9 2.2. Kromberg Formation

The Kromberg Formation is divided into three members- K1, K2, and K3 (Fig. 3).

K1, the focus of this study, consists of thick, layered chert units as well as a laterally continuous evaporite layer at the base, described in detail by Fisher Worrell (1985) and

Lowe and Fisher Worrell (1999). Although the original mineralogy of these sedimentary rocks has been obscured by near-total silica replacement, much of the original structures of these beds, as well as the evaporites, remain. The base of K1 has been dated to approximately 3,416 Ma (Kroner et al., 1991), and, although there is little age control within the Kromberg, the chert unit at the top, K3c, which separates it from the Mendon

Formation, has been dated to approximately 3,334 Ma (Byerly et al., 1996). Located on top of K1, K2 consists of volcaniclastics, and at the top of the Kromberg, K3 is a basaltic member (Ransom et al., 1999). On the west limb of the Onverwacht Anticline, most of

K1 is the Buck Reef Chert, comprised of alternating sections of black and white banded chert and ferruginous chert.

The base of K1 is more complex, represented in Fig. 4 (Fisher Worrell, 1985;

Lowe and Fisher Worrell, 1999). Fisher Worrell (1985) mapped the base of K1 and described the rock types and depositional setting (Fig. 4). Her conclusions were that the sediments at base of the Kromberg in the central portions of the western limb of the

Onverwacht Anticline were formed in a shallow marine setting.

The Hooggenoeg Formation, directly below the Kromberg, records the waning volcanism of the time and has a thick volcaniclastic sequence at the top (Lowe and

Byerly, 1999). It is in these shallow-water transitional sections that the stromatolites are

Figure 3. Generalized stratigraphic column of the southern domain of the BGB (Modified from Lowe et al., 2003; Thompson-Stiegler et al., 2011; Decker, 2013)

11 preserved. The contact between the Hooggenoeg and the Kromberg also record a transition from a subaerial to a subaqueous environment (Lowe and Fisher Worrell, 1999).

Along the central portion of the west limb of the Onverwacht Anticline K1 lies unconformably on H6, deeply eroded into a hypabyssal dacitic pluton. The sandstones and conglomerates at the base of K1 (Lithofacies 1 in Fig. 4) are typically felsic volcaniclastic material, weathered from H6. A silicified evaporite unit, termed K1e, includes between 5 and 40 meters of silicified, laminated, and wave rippled shallow water sediments at the base of K1, but only on west-central portions of the west limb of the anticline (Lowe and Fisher Worrell, 1999). K1e includes Lithofacies 2, 3, and 4 (Fig.

4). It is along K1ethat many of the stromatolitic samples presented here were collected.

The stromatolites typically were found on top of a bed of polymictic sandstone or conglomerate; this differs from the sandstone of Lithofacies 1 because it has grains that are dacitic, chery, and even komaatiitic, while Lithofacies one is composed primarily of felsic grains. Directly above the stromatolites was laminated and massive black chert

(Lithofacies 6). A summation of the field relationships of the stromatolitic sections is represented in Figure 5.

On the central portion of the western limb, normal faulting became more common, causing half-grabens up to 1.5 km in width to form (Lowe and Fisher Worrell,

1999). As a result, the black and white chert unit is thicker in this area – up to 200 km, and the K1e unit reaches its maximum thickness here – between 38 and 40 meters (Lowe and Fisher Worrell, 1999). Since these thicker sequences of evaporites are more common and have more complexity than their eastern, thinner counterparts, these half-grabens served as small, local basins for the accumulation of the thick evaporite beds (Lowe and

FisherFigure W 4.orrell, Generalized 1999). Towardbasal K1 the lithofacies, axis of the described anticline, by these Fisher cherts Worrell begin (1985) to thin, and and Lowe and Fisher Worrell, 1999).

are interbedded with thick volcanic units (Viljoen and Viljoen, 1969). No samples were collected from this area; the focus of this research is on the central and western portions of the western limb of the Onverwacht anticline.

The upper portions of K1c transition into the Buck Reef Chert along the western limb of the Onverwacht Anticline, which has been interpreted as being deposited in calmer and possibly deeper waters than the basal sediments of K1c and K1e (Lowe,

1982; Lowe and Fisher Worrell, 1999; Tice and Lowe, 2006).

13 Since so many of the original evaporitic and fine detrital textures of K1e and K1c

remain intact, it is thus not surprising that evidence for microbial life has also been

recorded and preserved in these rocks. Layers of fine carbonaceous material resembling

microbial mats were reported in K1 by Walsh (1992). Although rare, filamentous

microfossils were found in association with these mats (Walsh, 1992). Also reported in

that research were ellipsoidal and spindle microfossils in cherts of the basal Kromberg

(Walsh, 1992). In 2009, microfossils that had similar ellipsoidal shapes and sizes were

reported in similarly-aged cherts from the Pilbara Craton of Western Australia (Sugitani

et al, 2009). Using 3-D image reconstruction, these Australian microfossils were found to

be flanged in distinct patterns, rather than simple ellipsoids, and were also reported as

having double cell walls.

In addition to the large microfossils reported by Walsh, a SEM was used to

identify spherical and rod-shaped forms less than 5 µm, which were interpreted as

probable prokaryotes- most likely a form of bacteria (Westall et al., 2001). Additionally,

this research presented a negative carbon isotope value (δ13C) of -27 per mil, which has

been interpreted as a biologic fingerprint (Schidlowski, 1988; 2001; Mojzis et al., 1996).

However, as discussed in a review of carbonaceous matter in the pre-3.0 Ga Archean,

Tice and Lowe (2006) explain that due to the work of Horita and Berndt (1999) and van

Zuilen et al. (2002), it is not possible to assume a negative carbon isotope value

exclusively represents life: both of the papers present abiologic processes that have

similar carbon isotopic fractionations.

Although Tice and Lowe (2006) do not find individual microfossils in their work

in K1 of the Kromberg Formation, they present compelling evidence for biogenicity of

A 14

microbial mats using light microscopy, X-ray fluorescence of major and trace elements, total organic carbon, δ13C of carbonaceous material, point counting, and Raman spectrometry, specifically drawing comparisons to low-relief microbial mats which occur in modern Yellowstone hot springs (Lowe et al., 2001). While Tice and Lowe (2006) canvassed the entire thickness of K1 for their research, rather than the basal portions of

K1 on which this research focuses, the measured section from their research is fewer than

500 m east of sample location 8 in this research (Appendix A, C).

The samples presented here were collected along the basal portions of Kromberg, in K1 and K1e, on the west-central portion of the western limb of the Onverwacht

Figure 5. Generalized K1measured sections, from Fisher Worrell (1985) (black circles), and this research (red squares).

15 anticline (Fig. 4, Fig. 5, Appendix A). Each sample collected came from a section that was measured and described in the field (Appendix B). Although measured sections were described without reference to any preexisting lithofacies, an attempt was later made to organize the measured sections to align with those described by Fisher Worrell (1985)

(Fig. 4). It became apparent in the field, and upon organization of the measured sections, that the stromatolites occur in a specific, laterally congruent portion of the lower K1, sandwiched between Lithofacies 5 and Lithofacies 6 (Fig. 5)

16 3. Methods

3.1. Bulk Rock X-Ray Fluorescence

Samples were selected for analysis by Maud Walsh and Gary Byerly in 2011.

Although these samples are from the basal portions of K1 and represent samples in which stromatolitic patterns were visible, they are not the specific stromatolitic samples collected by the author. Eight sample analyses were selected to be presented in this research. The samples were sent to Washington State University (WSU) GeoAnalytical

Laboratory for bulk rock analysis using their ThermoARL Advant’XP+ sequential X-ray fluorescence spectrometer. At WSU, samples were prepared for analysis. They were first ground to a very fine powder, weighed with di-lithium tetraborate flux at a 2:1 flux: rock ratio, and then fused at 1000ºC in a muffle oven. The bead is then cooled, reground, refused, and then polished to create a flat surface for analyses. The samples were analyzed for both major and trace elements. Full results are presented in Appendix D.

For detailed methods, see the WSU lab site at: http://www.sees.wsu.edu/Geolab/equipment/xrf.html.

3.2. Raman Spectroscopy

Raman spectra were collected in the Engineering Sciences department at

Louisiana State University (LSU) using a Horiba Labram spectroscopy unit with a 632.81 nm diode red laser. The samples analyzed were a combination of both polished and rough surface thin sections ground to a thickness of the standard 30 µm. Additionally, to be sure the thin section adhesive was not corrupting the data, rough surface blocks that the thin

17 sections were cut from were used as a control. The first series of analyses were not baseline corrected, but the later series were, as suggested by the lab tech.

3.3. ICP-Mass Spectrometry

Samples were analyzed in an Agilent model 4500 ICP-MS at WSU. Prior to analyses, the samples were powdered and mixed with lithium tetraborate flux and fused into a bead in a 1000° C muffle furnace. The sample is then reground, diluted and finally analyzed. For full methods please visit http://environment.wsu.edu/facilities/geolab/technotes/icp-ms_method.html.

3.4. δ13C Isotopes

Samples were analyzed at Texas A&M University using a Costech Elemental

Analyzer (CEA). The samples were first ground into a homogenous powder, and then run through the CEA. For full procedures, visit http://sibs.tamu.edu/services/.

18 4. Results

Several samples display textures consistent with stromatolites at the mesoscopic and microscopic levels. These textures consist of domed laminations with a high degree of inheritance, angles of repose on laminated domes that are higher than that of natural sediment deposition, evidence of sediment trapping in troughs between domes, and mineral deposits along laterally continuous laminae.

In addition to morphological characteristics, chemical analyses provide valuable data. The Raman spectroscopy reveals the mineralogy of the carbonaceous material, as well as some minerals that were unidentifiable in thin section, such as anatase and rutile.

Other chemical analyses on laterally-equivalent samples indicate negative carbon isotope values as well as Rare Earth Element relative abundances. Although the samples are mostly quartz, X-Ray diffraction measures the presence of other major and trace elements.

The results of the morphological and chemical analyses are presented in Table 2 and discussed below.

Table 2. Summary of results including morphology, Raman, REE data, Primitive Mantle (PM) relative amount, Ua range, and the average δ13C.

19 4.1. Domical Type

When domed laminations mimic the shape of the most basal dome in a series, the set of domed laminae is considered to have high inheritance; each lamina inherits the morphology of the lamina below, typically with subtle geometrical changes in each lamina. The best example of inheritance in this research is the domed stromatolite sample,

CES 10-13 (Fig. 6). The slabbed surface of the sample clearly shows the inheritance from the basal dome to the top of the domes, (Fig. 6). The character of the domes changes subtly from the base to the top of the sample. For example, at the base of the domical sample, domes are relatively tall and steep-sided compared to their counterparts at the top of the section. Basal domes are typically about twice as tall as the width of each dome - but the domes at the top of the stromatolite sequence are wider than they are tall, typically with positive relief no more than 0.5 cm but width of 3-3.8 cm (Fig. 6). The positive relief of domes above the substrate on which they grow is known as synoptic relief.

The domical sample consists of several laterally-linked, adjacent columns of laminae. These laminae are made of mostly microcrystalline quartz with fine organic material that forms thin, laterally continuous laminations; the mineralogical components were identified using Raman spectroscopy (Fig. 7). The sample becomes more regularly laminated up-section through the stromatolite, and the domes coalesce with each other with each successive layer (Fig. 6).

Figure 6. Sample CES 10-13 (Domical sample). (A). Weathered surface of the sample; (B). Line drawing traces the prominent laminae in the sample. Abrupt breaks in horizontal lines represent a quartz vein which is a secondary feature that obscures the laminae where it cuts through the sample. Several distinct domical laminated columns are visible, which merge toward the top of the sample.

Figure 7. Uncorrected Raman shift results for sample CES 10-13. The wave number and mineral abbreviation are labeled on the crest of each peak and are as follows: T=tourmaline, M=muscovite, Q=quartz, D=disordered carbon, O=ordered carbon. Each colored line represents spectra results from different areas of the sample.

21 When examining these samples at a higher magnification, features not seen in handsample become apparent. One example of this is the ability to accurately measure dome steepness. The domes, while flat on the tops, have laminae that approach 90º in steepness to the horizontal (Fig. 8). These laminae are pale grey, extremely fine-grained, and thin toward the margins of the domes (Fig. 8, Fig. 9). Although the grains that are found between laminae are as large as 60 µm in diameter, the grey laminae, which are composed of fine-grained carbonaceous material and microcrystalline quartz (Fig. 7), appear to have trapped finer grains than the parts of the dome that do not have the fine grey laminae. Lamina are as thin as 2 µm and can be as thick as 8 µm in the troughs, but thicken to as wide as 300 µm over the pinnacle of the dome.

The laminae in the troughs have no grains interbedded within them, but in the dome, laminae bind grains between layers, accounting for the thicker laminations (Fig. 8,

Fig. 9). The largest grains in the sample are found trapped in the troughs between domes

(Fig. 8, Fig. 9) and reach sizes up to 1 mm, although the average size of the grains is closer to about 100-200 µm. These grains represent an assortment of rounded and angular fragments, and may have once been a variety of minerals, as evidenced by cubic and rectangular grains; most of these have been replaced by silica. However, some of the largest clasts consist of partially graphitized carbon (Fig. 7). The pattern of laminae and sediment grains within the domes is also notable; there appears to be a bimodal depositional pattern that is defined by alternating layers of very fine and fine grains. As noted in Figure 9, the very fine grained material is closely associated with the portions of the dome that have densely-occurring laminae.

Figure 8. Domical stromatolite; light microscope image. The coarse grained trough between the stromatolite domes (left half of the image) and the edge of one stromatolite dome (right half of the image). The angles of lamina or sets of lamina are measured and indicated with the approximate angle. Image is from the transmitted light microscope.

The areas between sequences of dense laminae are characterized by having sparse lamina occurrences with larger, fine-grained material. In Figure 8 it is possible to see this pattern on the right side of the figure: the 68º symbol marks the top of a lamina- directly above and below this lamina are the coarser- grained areas. Figure 9 is an image of a trough between two domes with the margins of domes on either side.

Figure 9. Domical stromatolite trough; light microscope image. Dense lamina in the trough get wider and sparser in the domes with increased distance between individual lamina. The large subangular black grains are a soft black mineral, identified via Raman spectroscopy to be carbonaceous.

Again, the larger clasts are trapped in trough, whereas finer-grained sediment comprises the domes and displays the alternating bimodal pattern. It is also possible to see that the laminae thin on the margins of each dome (close to and within the troughs) and the distance between individual lamina grows greater with distance from the trough.

The angles on the slopes of these two domes also approach 90º.

24 4.2. Flat-Laminated Type

The remaining samples described here are markedly different from the domical sample.

These samples have layered features but display much less synoptic relief and overall poorly-defined levels of inheritance (Fig. 10, Fig 11). Several samples have grey-brown weathering. The samples in Figure 10 are examples of the various flat-laminated forms, and show some evidence of inheritance and synoptic relief. It is important to note that, while the samples featured in figures 10 and 11 are all from the base of K1, they represent samples from each of the measured sections represented on the map, and thus can be several kilometers in lateral distance from each other. (Figs. 2 and 5, Appendix A).

In two of the flat-laminated samples, there are black, cubic or pseudocubic minerals identified as TiO2 (anatase and/or rutile) by the Raman spectra that are associated with some of the laminations (Fig. 12, Fig. 13).

Similarly to the domical sample, all the laminae are extremely fine-grained, but there is a notable absence of coarse-grained clastic material like that in the troughs of the domical sample. No grains greater than 40 µm are visible within the laminated portions of the flat-laminated samples, another difference between the domical and flat-laminated samples. The texture of the laminations is also dissimilar; the flat-laminated samples are more densely laminated than in the domical samples. This is true with the sample in Fig.

11, as well as with all of the other low-relief samples.

Additionally, there are several instances where the laminae appear to have been bent, as seen with the undulating, rippled effect at the base of the laminae in Fig. 11 or the bending of the laminae in Fig. 14, but have still managed to largely stay intact.

Figure 10. Interior cut surfaces of several samples exhibiting stromatolitic characteristics. (A). Sample CES 4-2b displays grey-brown weathered laminae that have some relief and inheritance. It is directly on top of a grit bed that is strongly associated with the stromatolitic samples; (B). Sample CES 8-6 shows high degrees of inheritance with subtle relief; (C). Sample CES 8-4 consists of one dome with positive relief and with a high level of laminae inheritance. All black scale bars are 1 cm.

Although the cause of this deformation is not known, it is possible that the sample in Fig. 11 was influenced by the growth of evaporite crystals before lithification, while the laminae in Fig. 14 were a result of forming in an environment with high-energy events that ripped up packages of laminae from the substrate and redeposited them as deformed structures prior to lithification.

Although the laminated samples with low relief are more numerous than the domical sample, there are unique properties about some of the laminated samples, which

26 perhaps reflect a slight difference in the environment of formation. One sample from section 8 has a more deformed texture than the other samples (Fig. 14). . Although there have been lamina or sets of lamina that appear to have been ripped or dislocated from the main body of laminae, these isolated strips of laminae remain largely intact.

Figure 11. Sample CES 4-2b. Black arrows denote locations of black pseudo-cubic minerals which are associated with individual lamina. In the lower center of the grey laminae, there is deformation in which the laminae form z-shaped curves with no fracture or breakage of the laminae. This indicates high levels of cohesion in the sediment. Reflected light microscope image.

27 Figure 12. Uncorrected Raman Shift results for sample CES 4-2b. The wave number and mineral abbreviation are labeled on the crest of each peak and are as follows: A=anatase, Q=quartz, D=disordered carbon, O=ordered carbon. Each colored line represents spectra results from different areas of the sample.

14000

12000

10000

8000

6000

4000 Intensity Intensity (Normalized)

2000

0 100 200 300 400 500 600 700 800 Raman Shift (cm-1)

Figure 13. Uncorrected Raman Shift results for sample CES 4-2b, from wavenumbers 100-800. Small peaks belonging to tourmaline and pyrite are visible when looked at in finer resolution. The wave number and mineral abbreviation are labeled on the crest of each peak and are as follows: A=anatase, Q=quartz, T=tourmaline, and P=pyrite. Each colored line represents spectra results from different areas of the sample.

Figure 14. Ripped-off laminae package in sample CES 8-4. The bending of the laminae around an axis without breakage demonstrates strong levels of sediment cohesion. Transmitted light microscope image.

Several samples exhibit laminae which are bent around an axis without breakage or fracture.

Another sample from section 8 is finely laminated and has pockets of fine-grained clastic material. Additionally, the sample is extremely dense and darkly colored and the

Raman results indicate a strong presence of carbonaceous material, as well as quartz and minor rutile (Fig. 15).

Figure 15. Sample CES 8-6. Reflected light microscope image. Isopachous- laminations with laterally discontinuous pockets of sediment.

4.3. Bulk Rock X-Ray Fluorescence

The results of X-Ray Fluorescence (XRF) analyses confirm what was observed in thin section – the stromatolitic samples of K1 are predominantly quartz (Table 3). Other major element oxides identified in small quantities ware TiO2, and Al2O3. The remaining element oxides presented in Table 4 had minor (less than 1% in all but one case) occurrences. It is also important to quantify the amounts of trace elements (Table 5).

Comparing the trace element amounts shows several elements share a close relationship.

For example, the relationship between trace element niobium and TiO2 (Fig. 16) has an

R2 value of 0.9505. As seen in Figure 17, U also shares a high R2 with Ti as well as Nb.

30 The amount of TiO2 in these samples in general is relatively high (Table 3). Sugitani and others (1996) found a similar pattern in Archean Pilbara cherts that consisted of greater than 90% SiO2, and enriched TiO2 values relative to Al2O3.

Table 3. Major element percent composition--XRF

Table 4. Trace element composition, X-ray fluorescence (ppm).

Figure 16. Uranium plotted against Nb, Pb, Zr, and Ti. U and Ti have the closest relationship, followed closely by that of U and Nb.

16.0

14.0

12.0

10.0

8.0

Nb Nb (ppm) 6.0

4.0

2.0 y = 3.7714x R² = 0.9505 0.0 0.000 1.000 2.000 3.000 4.000 5.000

TiO2 (%)

Figure 17. TiO plotted against Nb (% vs. ppm). 2

32 4.4. Raman Spectroscopy

When viewing the samples collected in thin section, there were several minerals that were unidentifiable. Additionally, although the black fine-grained material present in all of the samples was suspected to be carbonaceous, it was uncertain whether all the black matter was carbon, and whether the carbonaceous grains consisted of unordered carbon, such as that found in kerogen, or ordered carbon, like that in graphite.

Using the Raman spectra results, it was possible to see that virtually all samples examined contained some amounts of carbon, in both the ordered (O) and disordered (D) states (Fig. 12, Fig. 18, Fig. 19). Some spectra collected represented a singular carbonaceous grain with little else around it other than silica, such as the green spectra in

Figure 7. Other analyses of finer-grained portions of the sample yielded a conglomerate

Figure 18. Baseline corrected Raman Shift results for sample CES 8-6, a flat- laminated sample. The wave number and mineral abbreviation are labeled on the crest of each peak and are as follows: A=anatase, R=rutile, P=pyrite, D=disordered carbon, O=ordered carbon. Each separate line represents spectra results from different areas of the sample.

33 of data peaks, representing multiple minerals in the area analyzed, such as in Figures 12,

13 and 18, which are all from flat-laminated samples.

The disordered and ordered carbon peaks appear from 1330-1345 cm-1 and 1595-

1613 cm-1, respectively. Although these carbonaceous peaks appear in every analysis, the intensity of the peaks varies throughout the samples. For example, sample CES 8-6 (Fig.

18) seems to be concentrated in carbonaceous material compared other mineral peaks, with a quartz peak at 459 cm-1 and the remaining minor peaks at 195 cm-1 and 636 cm-1 being anatase. However, CES 8-4 (Fig. 19), less than one meter below CES 8-6 (Fig. 18), has less intense carbon peaks and a variety of peaks in lower wavenumbers, representing minerals such as rutile, anatase, quartz, tourmaline, and pyrite.

Figure 19. Baseline corrected Raman Shift results for sample CES 8-4. The wave number and mineral abbreviation are labeled on the crest of each peak and are as follows: A=anatase, T=tourmaline, R=rutile, P=pyrite, Q=quartz, D=disordered carbon, O=ordered carbon. Each colored line represents spectra results from different areas of the sample.

34 In CES 8-6 (Fig. 18), the major peaks at 240 cm-1, 440 cm-1, and 608 cm-1 are rutile peaks (Downs, 2006, for all reference spectra).

It is important to note that the disordered carbon peak always has higher intensity than the ordered peak. Although it is not possible to calculate relative percentages within the sample from the Raman spectra, larger, broader peaks typically indicate higher amounts of the mineral are present.

4.5. ICP-Mass Spectrometry

The cherts from K1 are enriched in light rare earth elements relative to the heavy rare earth elements (Fig. 20), as compared to the Primitive Mantle (PM). There are two notable anomalies in the data: a negative cerium anomaly in three samples and a positive europium anomaly in all but one sample. These anomalies are shown in Table 6. It is possible that the cerium anomaly is a result of modern weathering because the anomaly is not present or close to one in most of the samples. This is probably due to the presence of a small component of felsic volcanic material in the detrital component of these cherts.

In addition to looking at the REEs, a spider diagram was created to understand whether certain elements were enriched or depleted relative to the PM (Fig.

21). In the spider diagram, the samples fall into three distinct groups. One group consists of samples 65-11s, 62-11s, and 62-5. The ‘s’ indicates a stromatolitic sample and it is notable that the two stromatolitic samples have extremely similar data. They are enriched relative to the PM in the trace elements, with the exception of three elements: P, Y, and

Yb. Samples 64-1 and 62-20 are distinctly different; although they are enriched relative to the PM in U, Pb, and Ta, all other elements are below 1, depleted compared to the PM.

35 These two samples also have the lowest concentrations on the REE diagram (Fig. 20).

The remaining samples are of intermediate enrichment; about half of the elements are enriched, while the other half are depleted compared to the PM. It is notable that U and

Ta are the only elements that are enriched compared to the PM in all samples, even the most depleted ones. Conversely, P, Y, Yb, and Gd in most cases are depleted relative to the PM in every sample. Uranium, particularly authigenic uranium (Ua = U-Th/3), is a useful way to identify redox reactions (Wignall and Myers, 1988; Tice and Lowe, 2006b).

Ua calculations indicate that in several of the samples, U is extremely enriched, with Ua values above 1 (Table 7, Fig. 22). Any Ua values above 0.20 are much higher than those reported by Tice and Lowe (2006b) in their Kromberg sections.

10.000

MW64-4 1.000 MW 64-1 MW 65-11c MW 62-11c MW 62-11s MW 62-20 0.100 MW 62-5 MW 65-11s

0.010 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Figure 20. Rare earth element concentrations in K1 cherts, compared to the Primitive Mantle.

Trace and rare earth element amounts in earth and element ppm. rare Trace

5. Table

Table 6. Cerium and europium anomalies. If the value is above 1, is it a positive anomaly; below 1 is a negative anomaly.

1000

100

65-11s

10 62-11s 62-5 64-4 65-11c 1 62-11c 62-20 64-1 0.1

0.01 U Pb Th Nb Ta La P Gd Hf Zr Ti Y Yb ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm

Figure 21. Spider Diagram of K1 samples compared to the Primitive Mantle. If value is above 1, it is enriched, and if it is below 1, it is depleted.

0 -0.2 -0.4

-0.6 -0.8 -1

log FeO log FeO %) (wt -1.2 -1.4 -1.6 -1.8 0 1 2 3 4 5 6 7 8 Ua (ppm)

Figure 22. Authigenic uranium (Ua) plotted against log FeO. The data most closely represents a shallow shelf setting, especially the log FeO. Ua values above 0.20 are much higher than typical shelf values.

Table 7. Authigenic Uranium (Ua) and log FeO values for K1 samples.

4.6. δ13C Isotopes

The carbon isotopic data as well as the total percent carbon for samples in K1 are presented in Table 8. Although portions of these samples are black and were thought to

39 have high carbon content, it is notable that no samples contained enough carbon to measure even half of a percent of the sample. The carbon isotope values ranged from

-29‰ to -39‰, with an average of -33.9‰.

Table 8. Total carbon percentage of each sample and the carbon isotopic signature recorded per mil (‰).

40 5. Discussion

Multiple morphological observations and chemical and mineralogical analyses come together here to form a story about the formation of the basal portions of K1. An important part of this story is the genesis of these rocks; it is important to examine these samples for potential biologic signatures, both morphologic and chemical. There have been a variety of approaches to prove or disprove the biogenicity of potential fossils (for discussion, see Tice and Lowe, 2006). Here, the author favors a holistic approach which systematically builds a plausible scenario in which these samples could have formed with implications for biogenicity. This approach has been used in the case of other Archean stromatolites (Hofmann, 2000; Allwood et al., 2006; Allwood et al., 2009) and has been suggested as the obvious approach for future research of these ancient, possibly biogenic rocks by Awramik and Grey (2005) and Tice and Lowe (2006), since there is no singular method to undoubtedly prove biogenesis in ancient stromatolites. In this discussion, the focus is first on the morphological characteristics and their implications, and subsequently the chemistry of K1 and the inferences made from those data.

5.1. Morphologic Features

In their publication on stromatolite diversity in the 3.43 Ga Strelley Pool formation of the Pilbara craton of Western Australia, Allwood and others describe

“encrusting/domical stromatolites” (Allwood et al., 2009). These stromatolites are described as having separate domal features at the beginning of their growth period, and they then gradually merge together with distance from the base of stromatolite growth.

They are preserved in both chert and carbonate material, and are copious in the Strelley

41 Pool. While only one such domical sample was identified in the lower K1 for this research (sample CES 10-13), it presents the best morphological evidence of stromatolitic origin. The remaining flat laminated samples that, in some cases, do have some positive synoptic relief are not to be ruled out; although the morphological evidence is not profound, there are other factors to consider.

It has been suggested that fossil stromatolites should have some positive relief, and in the cases where this relief is present, the laminae should thicken over crests or flexures (Buick et al., 1981). The domical sample most clearly exhibits this thickening over crests (best seen in Figs. 7and 8). The steepness of the slopes of the stromatolites indicates that the sediment adhered or stuck to the sides of stromatolites; the natural angle of repose for coarse sand-sized grains is, at most, 40º (Carrigy, 1970).

As noted earlier, the domical stromatolite sample has an extreme difference in grain size between the trough and dome (best seen in Fig. 7). The bimodal pattern within the domes is observed and discussed by Reid et al. (2000) in modern stromatolites. They discussed stromatolites which have alternating laminations- a fine-grainedlayer that represents encrusting prokaryotic communities living on the surface of the stromatolite during times of relative quiescence in the water column, and the thicker gaps between the fine-grain encrustedlamina that are composed of grains which get trapped by the sticky cell exopolymers during times of activity and sedimentation in the water column. This pattern is further supported by observations in the work of Pope et al. (2000), who describe similar patterns in fossil stromatolites.

The stromatolites that do not have pronounced domical features, such as those in

Figure 6, have a different depositional story. As noted by Pope et al. (2000), there is a

42 large body of research that suggests that stromatolites with isopachous layering, or uniformly thick layering, probably are formed by in-situ precipitation of minerals, which then act as a substrate for the next layer of the stromatolite to form. This is probably the growth mode of the flat-laminated samples. These minerals were determined to be TiO2, mostly anatase but with some rutile, by Raman spectroscopy. Since these anatase grains are laterally continuous in layers that were deformed during or shortly after deposition, and it is possible to see these TiO2 grains in their respective lamina through the deformed portion of the sample, it is then possible to conclude that these mineral grains were formed synsedimentarily, thus reflecting on the chemical processes in that lamination at that time. Additionally, although it is not possible to see these minerals in every sample, the Raman spectra of the samples show that anatase is present in all the flat-laminated samples, and rutile in some samples. Notably, the domical sample lacked any titanium minerals, but did have muscovite, which was not seen in any other samples.

Although the spectra indicate anatase and some rutile in the flat-laminated types, it is possible that these are products of extreme chemical weathering. Ilmenite can be weathered to anatase and rutile; this is especially true in acidic settings (Fitzpatrick and

Chittleborough, 2002). However, there was no evidence of ilmenite. Authigenic anatase is typically a product of weathered titanite, but rutile is also a possibility (Malengreau et al., 1995). Prior to any mineral formation, the question of where the titanium originates from must be examined. Titanium can be incorporated in the crystalline structure of many silicate minerals, and potentially released during diagenesis (Morad and Aldehan, 1982).

Both pyroxene and amphibole can have quantities of TiO2 (up to 6%) which are similar to the levels found in the flat-laminated samples (Deer, Howie, and Zussman, 1962).

43 Biotite can have up to 12% TiO2 (Hayama, 1959). There are certainly plenty of places for the relatively high percentages of TiO2 in these samples to weather from.

If the anatase and rutile in these rocks were detrital grains, it is useful understand how they could have been deposited. The layered anatase and rutile in several of the samples (fig. 11) could have been trapped by the sticky exopolymer coating of microbes.

Alternatively, the grains may have precipitated from the water column. After deposition, the basal K1 may have been subaerially exposed for a time, in which extreme weathering occurred within the sediment or primordial soil, perhaps causing the enrichment of anatase and rutile as the clay in the original sediment became unstable via weathering and released additional titanium into the environment. This must have been shortly after deposition and before the silicification of the basal K1- essentially syndepositionally- as the silicifaction would have hindered this process. The interpretation of the lower

Kromberg as a tidal flat setting fits well into the pattern observed here; this would permit intermittent subaerial exposure.

Multiple types of stromatolites have been presented in this research, rather than a singular morphotype. As Allwood et al. discussed (2009), the variability in stromatolite morphology probably does not reflect a diverse microbe pool. Rather, it is the same microbe, or similar, that forms a stromatolite that has varying morphology, which is dependent on the environment in which it forms. Wave energy, water depth, and clastic input are all variables that can affect stromatolite growth, and thus morphology. With that said, it is possible that the domical stromatolites formed in higher energy water than those of the other, less domical forms. Lowe and Fisher-Worrell (1999) suggested a low- gradient alluvial system, such as a braided floodplain or sandflat, for the cherts and

44 evaporites at the base of the Kromberg. The domed stromatolite sample is from section

10, the section farthest to the west of the study area and does not have any evaporites in the section, whereas all the other samples presented here- those with the isopachous laminae- are from sections (4, 12, and 8) which have evaporites in close association to the potentially stromatolitic samples (Appendix A). It is likely that these low-relief stromatolites formed in water that was in a less clastically-dominated portion of this shallow alluvial setting, close to the calm evaporitic areas Lowe and Fisher Worrell

(1999) discuss. The domical stromatolite, however, may have formed closer to aerially- or subaerially-exposed sediment, where debris and periodic high energy events in the water column effected the sediment deposition. This may have been volcaniclastic sediment from the underlying Hooggenoeg. Volcaniclastic debris was first discussed by

Fisher Worrell (1985) and then expanded on by Lowe and Fisher-Worrell (1999) as being a likely source of clastic input into the sediments of the lower Kromberg. Although the samples are all in laterally equivalent portions of K1 (Fig. 5), it is probable that there was some variability in water depth and wave energy across the study area, as indicated by the change in mineralogy and grain size from the western to eastern portions of the lower

Kromberg.

If the criteria by Buick and others (1981), Walter (1983) and Hofmann and others

(1999) are applied to the samples presented in this research, only one sample passes the test. The criteria are a good starting place when evaluating these structures, but Buick and others (1981) and Walter (1983) base their criteria on analogues that are too modern; stromatolites were the most diverse in the Proterozoic (Awramik, 1992; Semikhatov and

Raaben, 1996) so it is important to consider multiple growth modes and morphologies.

45 Hofmann and others (1999) focus on Archean stromatolites in their criteria, but the focus is on conical forms. Not surprisingly, only the domical sample passes Hofmann’s checklist; as the only sample with dome slopes greater than 40º, it is the only sample that fulfills all criterion. This sample also displays uniform laminae in the domical areas of the sample, but has fewer laminae in the troughs, fulfilling the first criterion. The thinning of the laminae on the slopes of the domes proves that they were not a result of sideways compression- if that were the case there would not be such a large difference between laminae on the dome pinnacles and troughs, fulfilling the next criterion. Since the Raman spectra provide a carbonaceous origin with distinct D and O peaks for the laminae, it is likely that the laminae are not a result of precipitation of chemical grains

(Tice and Lowe, 2006). And lastly, the uniform sizes and mineralogy of the domes, as well as the bimodal intra-dome depositional pattern in all domes, and the coalescence of domes with more regular laminations in the sample together show that the domes all grew at the same time in the same basin. Of all the samples presented here, this domical sample is the only one to fulfill all of these criteria and thus is the most likely to be biogenic based on morphology alone. However, in their paper on stromatolite diversity in the

Archean, Allwood and others (2009) present forms which they describe as biogenic that do not have steep-sided dome slopes, and are overall low-relief, relatively flat samples, similar to the flat-laminated forms presented here. So, although Hofmann’s (1999) criteria are valuable, it may not be inclusive to multiple forms of stromatolites.

46 5.2. Chemistry

Studies that question the biogenicity of proposed microfossils and stromatolites frequently discuss the uncertainty of when the carbonaceous material in the fossil-bearing rocks became incorporated into the rock (Garcia-Ruiz et al., 2003; Brasier et al., 2004).

As Buick et al (1981) and Walter (1983) discuss, to arrive at a conclusion of biogenicity for stromatolites or microfossils, it must be proven that the structures formed at the time of deposition, and that any chemical evidence provided represents the in-situ deposition, rather than material that may have been incorporated into the rocks at a later time. The

Raman spectroscopy data presented in this thesis confirms that there is carbonaceous material (CM), and that this CM, which defines the laminae in all the samples discussed here, was formed at the same time as sedimentary deposition which formed what we now know as K1. The chemical analyses on the K1 samples show that virtually no chert has more than 1% of carbonaceous matter, in fact, the values are typically closer to zero than to one (Table 8).

The carbonaceous peaks from the Raman analyses provide evidence for the formation of in-situ carbonaceous material. The D, or disordered, peak represents the kerogen-like carbonaceous material, essentially carbon with no crystalline structure. The

O, or ordered, peaks represent the graphite-like carbon present. The relationship between these two peaks provides telling information about formation history (Tice and Lowe,

2004).

It has been suggested that any biogenic structures should have well-developed D and O peaks in order to be considered a synsedimentary life form (Tice et al., 2004). In their research, Tice et al. found that the D and O peaks for the carbonaceous material in

47 much of the sedimentary rocks of the Onverwacht were well-developed and fell in a certain range of wave numbers, which indicated that the carbonaceous material analyzed was deposited prior to the regional metamorphism that affected much of the Onverwacht.

Although their results and those presented here are not directly comparable due to the use of different sized laser diodes, the patterns of the D and O peaks are similar. The fact that every Raman spectra result has distinct D and O peaks, rather than one broad peak, agree with the results of Tice et al. (2004), who conclude the carbonaceous matter was in situ and was heated to 300-400º C, at most.

The rare earth element (REE) data provides useful information about the environment of deposition of the lower Kromberg. The Archean ocean is characterized by having a positive Eu-anomaly (Danielson et al., 1992; Farquhar et al., 2000). As seen in Figure 20, this is also the case with the K1 samples. Although positive Eu anomalies in

Archean cherts have previously been interpreted to represent a hydrothermal water source, it is more likely in K1 to represent the typical Eu-enriched Archean ocean (Tice and

Lowe, 2006b). When this REE data is compared with that of Tice and Lowe (2006b), the entire K1 chert section has a positive Eu anomaly, even though it has been interpreted to represent a deepening-upward sequence (Lowe and Fisher Worrell, 1999). The prevalence of the Eu anomaly throughout a section of variable environments indicates that localized hydrothermal input had little to do with the anomaly (Tice and Lowe,

2006b).

The REE pattern here is different from most Archean cherts of similar age

(Sugitani, 1992; Tice and Lowe, 2006b; Allwood et al., 2009). However, there are some similarities. In one instance in the Western Australian Pilbara terrain, similarly-aged

48 biogenic samples were evaluated for REE composition (Sugahara et al., 2010). The pattern reported for the cherts closely associated with evaporites was an enrichment of

Light REEs (LREEs) and a slight positive europium anomaly (Sugahara et al., 2010).

Tice and Lowe (2006b) saw a similar pattern when they compared REE data throughout the K1. In all of K1 except for the base, they found that there is an enrichment of the heavy REEs, rather than the light; only the base of K1 had the enrichment of

LREEs. Additionally, every sample had positive Europium anomalies. They also reported sporadic cerium anomalies, and attributed the intermittent pattern to modern weathering.

Since the REEs and trace elements are normalized to the primitive mantle, it is possible to determine whether they are enriched or depleted relative to the PM. This is important, because it is then possible to determine the type of rock the clastic material weathered out of. Unfortunately, this pattern is not reflected in the underlying

Hooggenoeg igneous rocks, so it is not possible to conclude the sediments weathered from a Hooggenoeg source. The spider diagram (Fig. 21) shows the extreme enrichment of U, Pb, and Ti compared to the PM. Although they did not analyze for uranium,

Sugitani and others (1996) found that their Archean cherts from the Pilbara terrain were also extremely enriched in Pb, similar to those reported here. They also had high levels of titanium, and concluded that the TiO2 values indicated extreme chemical weathering of the source rock. It is difficult to say why these samples are so enriched in U, Pb, and Ti, but by looking at the relationship between some of these elements in the rock, it is possible to understand which elements the concentration process effected. The relationship between U and Ti and U and Nb is strong (Fig. 16), as is the relationship between Nb and TiO2 (Fig. 17). Additionally, the relationship between U and Pb and U

49 and Zr is moderately strong.. If the titanium weathered out of Hooggenoeg pyroxenes, amphiboles, or biotites, it is likely that so did the Nb and Zr, because the relationship indicates a shared process. There are many questions still about the processes that caused these elemental and mineralogical abundances.

One possible explanation for the enriched U in these samples is diagenesis of the sediment syndepositionally, or at least prior to silicification. Although this does not explain the relationship of U to the other enriched elements, it has been posited that the most likely enrichment scenario is via anoxic marine deposition (Tice and Lowe, 2006b).

This is measured by comparing the Ua values to log FeO values (Fig. 22, Table 7).

While all of the FeO values are consistent with the findings of Tice and Lowe (2006b), the same cannot be said for the Ua values. The Ua values less than 0.2, along with the log

FeO values, plot in a region recognized by Tice and Lowe (2006b) to be consistent with a shallow shelf setting. The remaining five samples have Ua values too high to be consistent with any environmental setting, and have most likely been affected by some additional enrichment process. Although over half of the values aren’t reliable for interpreting depositional environment, the remaining three values lend credence to the depositional story discussed by Lowe and Fisher Worrell (1999), where the base of the

Kromberg forms in a shallow marine setting and deepens up-sequence.

Although this explanation is certainly plausible, the values reported by

Tice and Lowe (2006b) are much lower than the Ua values reported here; their values did not exceed 0.2, whereas five of the eight samples analyzed here had values between 0.8 and 7.0. This indicates that uranium availability was different during the deposition of the most basal portions of K1. The samples analyzed here are from the very base of the

50 Kromberg, while those analyzed by Tice and Lowe are throughout the chert section of K1, and with the exception of one of their samples, not from the basal K1. During the

Archean, the ocean was low in uranium (Li et al., 2013). An enrichment of uranium must have occurred during the time that the basal K1 was deposited. A Ua value of 0 would indicate that all uranium was detrital, but since the basal K1 samples have values approaching 7.0, another process must have enriched the uranium in the sediment at the time of deposition.

Authigenic uranium is measured by subtracting Th/3 from U because this is the value typically seen in igneous rocks- if there is an excess of U, it must be authigenic because it exceeds the ratios that would be present in detrital grains only. The modern ocean has higher U than the Archean ocean had (Li et al., 2013); this U is U(VI), a soluble form. Particulate, non-soluble U is U(IV). When water becomes oxidated, more

U(VI) can dissolved in the solution, but if the water becomes anoxic, the U would become the insoluble U(IV), and be deposited as fine-grained U. Because of the high U content in these basal K1 stromatolitic cherts, with values much higher than that of

Archean seawater, the U must be U(IV). Additionally, although the U is extremely enriched in these samples, the Th is not, which means that whichever process caused the uranium to become enriched in the lower Kromberg also dissociated the Th from the U, creating a departure from the typical ratio of Th/U=3.

Whether this unusual U enrichment was caused by microbes or some other process is not clear, but given the carbon isotope values (table 8) as well as the stromatolitic structures presented here, microbial mats (Tice and Lowe, 2006a), and microfossils (Walsh, 1992) reported in the basal Kromberg, it is not unlikely that

51 microbes are responsible for this oxidation, which is only seen at the base of K1. The microfossils reported by Walsh (1992) are also at the base of K1.

In the past, carbon isotopes have been used to prove biogenicity in metasedimentary rocks in which little sedimentary fabric is preserved (Mojzis et al., 1996

However, the use of carbon isotopes as a simple means of discovering biogenicity in rocks has been debated in the literature (ex: van Zuilen et al., 2002; Wacey, 2009), so it is important to consider carbon isotope values as supporting data only. Although there has been controversy surrounding carbon isotopes values as life proxies, it is still a valuable tool to use when looking at Archean sedimentary or metasedimentary samples (Horita and Berndt, 1999, Schidlowski, 2001; van Zuilen et al., 2002;Tice and Lowe, 2006).

Carbonaceous matter is readily available in a variety of Archean sedimentary facies

(Lowe, 1983; Buick and Dunlop, 1990; Lowe and Fisher-Worrell, 1999; Allwood et al.,

2010). Although negative isotope values in the range typically produced by organisms have been synthesized abiogenically (Horita and Berndt, 1999; van Zuilen et al., 2002), it can also be formed biologically. The fact that the carbon isotope data in K1 ranges from

-29‰ to -39‰ per mil, which is congruent with a biologic isotopic signature (Mojzis et al., 1996; Schidlowski, 2001), helps to support a biogenic interpretation for these samples.

The story of the lower K1 is becoming clearer. Laminated, carbonaceous forms that are potentially biogenic were most likely deposited in an anoxic shallow marine setting that was, in some areas, more clastically influenced than in others, varying stromatolite morphology across the lower Kromberg. The sediments experienced extreme weathering, which may have been what enriched the lower K1 in U, Pb, and TiO2, and the high Ua values could indicated microbial activity as an oxidizing mechanism in these

52 sediments. Two distinct types of potential stromatolites are examined, and could represent the reaction of a primordial microbe to different environments- domical stromatolites formed in areas with less intense weathering and closer to terrestrial input.

Although it is not possible to say with certainty, it is more likely than not that the laminated forms presented here, especially the domical sample, are biogenic.

53 6. Conclusion

By using a holistic, multi-evidentiary approach, it is possible to suggest that the samples presented here are biogenic. The morphology of both types of stromatolites in this research – domical and flat-laminated– is unique to fossil and modern stromatolites.

The domical sample with dome angles higher than that of the natural angle of repose of sediment represents a clastically influenced shallow marine environment with periodic high-energy events that deposit large grains in the troughs between individual domes.

The flat-laminated, low relief forms are rich in extremely fine-grained anatase and/or rutile, minerals that were probably formed as a result of extreme weathering of Ti-rich source minerals such as pyroxene or biotite, and later, diagenesis. The flat-laminated forms display high levels of sediment cohesion, which could indicate microbial influence.

The carbon isotope data of samples from K1 are well within the range of the fingerprint left by autotrophic microbes.

Although the two types of stromatolites probably represent subtly different environments of formation, the REE data and Ua data suggest these laminated structures formed in shallow marine waters that were most likely anoxic, but with some localized oxidating activity. Both types of stromatolites are also defined by laminae that are made of fine-grained carbonaceous material - both the ordered and disordered types. This indicates that the carbonaceous matter was formed before the regional metamorphism which affected the BGB after the Kromberg was deposited, and is not a more modern source. Lastly, although the focus of this research was not to seek out cellular microfossils, some have been identified in the lower K1 in laterally equivalent sections,

54 which further supports the likelihood of the existence of microbes during the deposition of the lower K1.

Although no one line of evidence alone can undoubtedly prove the biologic beginnings of stromatolitic fossils, taken together, the author suggests that it is more likely than not that the domical sample is a fossil stromatolite. Although the morphological evidence for the flat-laminated forms is not as robust, it is not unlikely that the formation of those forms was mediated by primordial microbes.

55 References

Allwood, A.C., Walter, M.R., Kamber, B.S., Marshall, C.P., and Burch, I.W., 2006. Stromatolite reef from the Early Archean era of Australia. Nature, v. 441, pp. 714-718.

Allwood, A.C., Grotzinger, J.P., Knoll, A.H., Burch, I.W., Anderson, M.S., Coleman, M.L., and Kanik, I., 2009. Controls on development and diversity of Early Archean stromatolites. PNAS, v. 106 (24), pp. 9548-9555.

Allwood, A.C., Kamber, B.S., Walter, M.R., Burch, I.W., and Kanik, I., 2010. Trace elements record depositional history of an Early Archean stromatolitic carbonate platform. Chemical Geology, v. 270 (1-4), pp. 148-163.

Armstrong, R.A., Compston, W., de Wit, M.J., Williams, I.S., 1990. The stratigraphy of the 3.5-3.2 Ga Barberton Greenstone Belt revisited: a single zircon ion microprobe study. Earth and Planetary Science Letters, v. 101 (1), pp. 90-106.

Awramik, S.M., and Grey, K., 2005. Stromatolites: biogenicity, biosignatures, and bioconfusion. SPIE, v. 5906, pp. 1-9.

Beukes, N.J., and Lowe, D.R., 1989. Environmental control on diverse stromatolite morphologies in the 3000 Myr Pongola Supergroup, South Africa. Sedimentology, v. 36, pp. 383-397.

Brasier, M.D., Green, O.R., Jephcoat, A.P., Kleppe, A.K., Van Kranendonk, M.J., Lindsay, J.F., Steele, A., and Grassineau, N.V., 2002. Questioning the evidence for Earth's oldest fossils. Nature v. 416, pp. 76-81.

Brasier, M.D., Green, O.R., Lindsay, J. F., and Steele, A., 2004. Earth's oldest (approximately 3.5 Ga) fossil and the "Early Eden hypothesis": questioning the evidence. Origins of Life and Evolution of the Biosphere v. 34, pp. 257-269.

Buick, R., Dunlop, J., Groves, D., 1981. Stromatolite recognition in ancient rocks: an appraisal of irregularly laminated structures in an Early Archean chert-barite unit from North Pole, Western Australia. Alcheringa, v. 5, pp. 161-181.

Buick, R., and Dunlop, J.S.R., 1990. Evaporitic sediments of Early Archaean age from the Warrawoona Group, North Pole, Western Australia. Sedimentology, v. 37 (2), pp. 247-277.

Byerly, G.R., Lowe, D.R., and Walsh, M.M., 1986. Stromatolites from the 3,300-3,500- Myr Swaziland Supergroup, Barberton Mountain Land, South Africa. Nature, v. 319, pp. 489-491.

Byerly, G.R., Kroner, A., Lowe, D.R., Todt, W., Walsh, M.M., 1996. Prolonged magmatism and time constraints for sediment deposition in the early Archean Barberton

56 greenstone belt: evidence from the Upper Onverwacht and Fig Tree groups. Precambrian Research, v. 78 (1-3), pp. 125-138.

Byerly, G.R., Lowe, D.R., Wooden, J.L., Xie, X., 2002. An Archean impact layer from the Pilbara and Kaapvaal cratons. Science, v. 297, pp. 1325-1327.

Carrigy, M.A., 1970. Experiments on the angles of repose of granular materials. Sedimentology, v. 14, pp. 147-158.

Danielson, A., MÖller, P., Dulski, P., 1992. The europium anomalies in banded iron formations and the thermal history of the oceanic crust. Chemical Geology, v. 97, pp. 89- 100.

Dann, J.C., 2000. The 3.5 Ga Komati Formation, Barberton Greenstone Belt, South Africa, Part I: new maps and magmatic architecture. South African Journal of Geology, v. 103, pp. 47-68.

De Wit, M.J., Armstrong, R., Hart, R.J., and Wilson, A.H., 1987. Felsic igneous rocks within the 3.3-3.5 Ga Barberton Greenstone Belt: high crustal level equivalents of the surrounding Tonalite-Trondhjemite Terrain, emplaced during thrusting. Tectonics, v. 6, pp. 529-549.

Decker, N.B., 2013. High resolution tephra and U/Pb chronology of the 3.33-3.26 Ga Mendon Formation, Barberton Greenstone Belt, South Africa. M.S. thesis, Baton Rouge, LA: Louisiana State University, 66 p.

Deer, W.A., Howie, R.A., and Zussman, S., 1962. Rock-forming minerals, vol. 3. Longmans, 270 pp.

Downs R.T., 2006. The RRUFF Project: an integrated study of the chemistry, crystallography, Raman and infrared spectroscopy of minerals. Program and Abstracts of the 19th General Meeting of the International Mineralogical Association in Kobe, Japan. O03-1

Farquhar, J., Bao, H., Thiemens, M., 2000. Atmospheric influence of Earth’s earliest sulfur cycle. Science, v. 289, pp. 756-758.

Fisher Worrell, G., 1985. Sedimentology and mineralogy of silicified evaporites in the basal Kromberg Formation, South Africa. M.S. thesis, Baton Rouge, LA: Louisiana State University, 152 p.

Fitzpatrick, R.W., and Chittleborough, D.J., 2002. Titanium and zirconium minerals. In ‘Soil mineralogy with environmental applications.’Eds. J.B. Dixon and D.G. Schulze. Soil Science Society of America Book Series No. 7 (Soil Science Society of America: Madison, WI).

57 Garcia-Ruiz, J.M., Hyde, S.T., Carnerup, A.M., Christy, A.G., Van Kranendonk, M.J., Welham, N.J., 2003. Self-assembled silica-carbonate structures and detection of ancient microfossils. Science, v. 302, pp. 1194-1197.

Grotzinger, J.P., and Rothman, D.H., 1996. An abiotic model for stromatolite morphogenesis. Nature, v. 383, pp. 423-425.

Hayama, Y., 1959. Some considerations on the colour of biotite and its relation to metamorphism. Journal of the Geological Society of Japan, v. 65, pp. 21-30.

Hofmann, H.J., 2000. Archean stromatolites as microbial archives. In: Riding, R.E., Awramik, S.M. (Eds.), Microbial Sediments. Springer-Verlag, Berlin, Heidelberg, pp. 315-327.

Hoffman, H.J., Grey, K., Hickman, A.H., and Thorpe, R.I., 1999. Origin of 3.45 Ga coniform stromatolites in Warrawoona Group, Western Australia. GSA Bulletin, v. 111, pp. 1256-1262.

Horita, J., and Berndt, M.E., 1999. Abiogenic methane formation and isotopic fractionation under hydrothermal conditions. Science, v. 285, pp. 1055-1057.

Kroner, A., Byerly, G.R., and Lowe, D.R., 1991. Chronology of early Archean granite- greenstone evolution in the Barberton Mountain land, South Africa, based on precise dating by a single zircon evaporation. Earth and Planetary Science Reviews, v. 103, pp. 41-54.

Li, W., Czaja, A.D., Van Kranendonk, M.J., Beard, B.L., Roden, E.E., Johnson, C.M., 2013. An anoxic, Fe(II)-rich, U-poor ocean 3.46 billion years ago. Geochemica et Cosmochemica Acta, v. 120, pp. 65-79.

Lowe, D.R., 1980. Stromatolites 3,400-Myr Old from the Archean of Western Australia. Nature, v. 284, pp. 441-443.

Lowe, D.R., 1982. Comparative sedimentology of the principal volcanic sequences of Archean greenstone belts in South Africa, Western Australia, and Canada: implications for crustal evolution. Precambrian Research v. 17, pp. 1-29.

Lowe, D.R., 1983. Restricted shallow-water sedimentation of early Archean stromatolitic and evaporitic strata of the Strelley Pool Chert, Pilbara Block, Western Australia. Precambrian Research, v. 19, pp. 239-283.

Lowe, D.R., 1994. Abiological origin of described stromatolites older than 3.2 Ga. Geology, v. 22, pp. 387-390.

58 Lowe, D.R., and Byerly, G.R., 1999. Stratigraphy of the west-central part of the Barberton greenstone belt, South Africa. Special Paper-Geological Society of America, v. 329, pp. 1-36.

Lowe, D.R., Byerly, G.R., Kyte, F.T., Shukolyukov, A., Asaro, F., and Krull, A., 2003. Spherule beds 3.47-3.24 billion years old in the Barberton Greenstone Belt, South Africa: a record of large meteorite impacts and their influence on early crustal and biological evolution. Astrobiology, v. 3, pp. 7-48.

Lowe, D.R., and Byerly, G.R., 2007. Geology of the Barberton Greenstone Belt, South Africa in “Earth’s Oldest Rocks”, ed. M. van Kranendonk, Elsevier, pp. 481-526.

Lowe, D.R., and Fisher Worrell, G., 1999. Sedimentology, mineralogy, and implications of silicified evaporites in the Kromberg Formation, Barberton Greenstone Belt, South Africa. In: Lowe, D.R., Byerly, G.R., (Eds.), Geologic Evolution of the Barberton Greenstone Belt, South Africa. Geological Society of America Special Paper, v. 239. Boulder, Colorado, pp. 167-188.

Lowe, D.R., Byerly, G.R., Heubeck, C., Geological Society of, A., 2012. Geologic map of the west-central Barberton Greenstone Belt, South Africa. Geological Society of America, Boulder, Colo.

Malengreau, N., Muller, J., Calas, G., 1995. Spectroscopic approach for investigating the status and mobility of Ti in kaolinitic minerals. Clays and Clay Minerals, v.43, pp. 615- 621.

Marshall, A.O., Emry, J.R., and Marshall, C.P., 2012. Multiple generations of carbon in the Apex Chert and implications for preservation of microfossils. Astrobiology, v. 12 (2), pp. 160-166.

Mojzis, S.J., Arrhenius, G., McKeegan, K.D., Harrison, T.M., Nutman, A.P., and Friend, C.R., 1996. Evidence for life on Earth before 3,800 million years ago. Nature, v. 386, pp. 665-668.

Morad, S., and Aldehan, A.A., 1982. Authigenesis of titanium minerals in two Proterozoic sedimentary rocks from southern and central Sweden. Journal of Sedimentary Petrology, v. 52, pp. 1295-1305.

Petroff, A.P., Sim, M.S., Maslov, A., Krupenin, M., Rothman, D.H., and Bosak, T., 2010. Biophysical basis for the geometry of conical stromatolites. PNAS, published ahead of print May 17 2010.

Pope, M.C., Grotzinger, J.P., and Schreiber, B.C., 2000. Evaporitic subtidal stromatolites produced by in situ precipitation: textures, facies associations, and temporal significance. Journal of Sedimentary Research, v. 70, pp. 1139-1151.

59 Ransom, B., Byerly, G.R., and Lowe, D.R., 1999. Subaqueous to subaerial Archean ultramafic phreatomagmatic volcanism, Kromberg Formation, Barberton Greenstone Belt, South Africa. Special Paper-Geological Society of America, v. 329, pp. 151-166.

Reid, R.P., Visscher, P.T., Decho, A.W., Stolz, J.F., Bebout, B.M., Dupraz, C., Macintyre, L.G., Paerl, H.W., Pinckney, J.L., Prufert-Bebout, L., Steppe, T.F., and DesMarais, D.J., 2000. The role of microbes in accretion, lamination, and early lithification of modern marine stromatolites. Nature, v. 406, pp. 989-992.

Schidlowski, M., 1988. A 3,800-million-year isotopic record of life from carbon in sedimentary rocks. Nature, v. 333, pp. 213-318.

Schidlowski, M., 2001. Carbon isotopes as biogeochemical recorders of life over 3.8 Ga of Earth history: evolution of a concept. Precambrian Research, v. 106, pp. 117-134.

Schopf, J.W., Kudryavtsev, A.B., Agresti, D.G., Czaja, A.D., and Wdowiak, T.J., 2005. Raman imagery: a new approach to assess the geochemical maturity and biogenicity of permineralized Precambrian fossils. Astrobiology, v. 5, pp. 333-371.

Schopf, J.W., Kudryavtsev, A.B., Czaja, A.D., Tripathi, A.B., 2007. Evidence of Archean life: stromatolites and microfossils. Precambrian Research v. 158, pp. 141-155.

Sugahara, H., Sugitani, K., Mimura, K., Yamashita, F., and Yamamato, K., 2010. A systematic rare-earth elements and yttrium study of Archean cherts at the Mount Goldsworthy greenstone belt in the Pilbara Craton: Implications for the origin of microfossil-bearing black cherts. Precambrian Research, v. 177, pp. 73-87

Sugitani, K., 1992. Geochemical characteristics of Archean cherts and other sedimentary rocks in the Pilbara Block, Western Australia; evidence for Archean water enriched in hydrothermally-derived iron and silica. Precambrian Research, v. 57, pp. 21-47. Sugitani, K., et al 2009.

Sugitani, K., Horiuchi, Y., Adachi, M., and Sugisaki, R., 1996. Anomalously low Al2O3/TiO2 values for Archean cherts from the Pilbara Block, Western Australia – possible evidence for extensive chemical weathering on the early Earth. Precambrian Research, v. 80, pp. 49-76.

Thompson Stiegler, M., Lowe, D.R., and Byerly, G.R., 2011. Fragmentation and dispersal of komatiitic pyroclastics in the 3.5-3.2 Ga Onverwacht Group, Barberton Greenstone Belt, South Africa. GSA Bulletin, v. 123, pp. 1112-1126.

Tice, M.M., Bostick, B.C., and Lowe, D.R., 2004. Thermal history of the 3.5-3.2 Ga Onverwacht and Fig Tree Groups, Barberton Greenston belt, South Africa, inferred by Raman microspectroscopy of carbonaceous material. Geology, v. 32, pp. 37-40.

60 Tice, M.M., and Lowe, D.R., 2006. The origin of carbonaceous matter in pre-3.0 Ga greenstone terrains: a review and new evidence from the 3.42 Ga Buck Reef Chert. Earth-Science Reviews, v. 76, pp. 259-300.

Tice, M.M., and Lowe, D.R., 2006b. Hydrogen-based carbon fixation in the earliest known photosynthetic organisms. Geology, v. 34, pp. 37-40. Van Kranendonk, M.J., Webb, G.E., Kamber, B.S., 2003. Geological and trace element evidence for a marine sedimentary environment of deposition and biogenicity of 3.45 Ga stromatolitic carbonates in the Pilbara Craton, and support for a reducing Archaean ocean. Geobiology, v. 1, pp. 91-108. van Zuilen, M.A., Lepland, A., and Arrhenius, G., 2002. Reassessing the evidence for the earliest traces of life. Nature, v. 418, pp. 627-630.

Viljoen, M. J. and R. P. Viljoen, 1969. An introduction to the geology of the Barberton granite-greenstone terrain. Upper Mantle Project, The Geological Society of South Africa Special Publication. v. 2, pp. 9-28.

Wacey, D., 2009. Early life on Earth: a practical guide. Springer, 284 p.

Walsh, M.M., 1992. Microfossils and possible microfossils from the early Archean Onverwacht Group, Barberton Mountain Land, South Africa. Precambrian Research v. 54 (2-4), pp. 271-293.

Walsh, M.M., Lowe, D.R., 1985. Filamentous microfossils from the 3500 Myr-old Onverwacht Group, Barberton Mountain Land, South Africa. Nature, v. 314, pp. 530-532.

Walter, M.R, 1983. Archean stromatolites: evidence of Earth’s earliest benthos. In: Schopf, J.W. (Ed.), Earth’s Earliest Biosphere. Princeton University Press, NJ, pp. 187- 213.

Walter, M.R., Buick, R., Dunlop, J.S.R., 1980. Stromatolites 3,400-3,500 Myr old from the North Pole area, Western Australia. Nature, v. 284, pp. 443-445.

Westall, F., 2001. Early Archean fossil bacteria and biofilms in hydrothermally- influenced sediments from the Barberton Greenstone Belt, South Africa. Precambrian Research, v. 106, pp. 93-116.

Wignall, P.B., and Myers, K.J., 1988. Interpreting benthic oxygen levels in mudrocks: a new approach. Geology, v. 16, pp. 452-455.

61 Appendix A: Locations of Sections

Sample Locations Section Latitude Longitude 1 S 25 55 32 E 30 51 10 2 * * 3 S 25 55 30 E 30 51 35 4 S 25 55 25 E 30 52 49 5 S 25 55 49 E 30 54 57 6 S 25 55 32 E 30 54 14 7 S 26 00 21 E 31 01 30 8 S 25 55 33 E 30 54 31 9 * * 10 S 25 55 51 E 30 50 31 11 S 25 55 24 E 30 51 50 12 S 25 55 32 E 30 51 46 13 * *

*GPS coordinates not available. See below map for approximate sample locations.

62 Appendix B: Complete Measured Sections

75 Appendix C: Expanded Methods

Transmitted Light Microscopy

Samples bearing stromatolite-like features were slabbed and cut to size on water and oil saws. Thin sections were then ordered from Spectrum Petrographics, Inc. at a 30

µm thickness. These thin sections were studied and photographed at 4, 10, and 100x resolutions.

Reflected Light Microscopy

Due to the small lens of the transmitted light microscope, a Leica M125 reflected light microscope was used to photograph large images of the thin sections as well as look at mineral shapes and sizes that were obfuscated by fine-grained carbonaceous material in the transmitted light microscope.

X-Ray Fluorescence

Although this data was determined unusable due to the heterogeneous nature of the samples, it was still useful for determining overall major element abundances, such as

Si, Ti, and Al (Appendix D).

In order to evaluate the chemical composition, an Innov-X Systems Delta Mining

Premium (DP 6000) handheld XRF device was used to determine the amounts of elements within the samples at the LSU XRF laboratory. Each sample was spot-analyzed in multiple dime-sized locations on a flat slab, previously cut from the sample. These values were then averaged unless there was a large discrepancy between the values, in which case the XRF analysis for that sample was removed. In addition to running

76 analyses on samples collected for this research, samples from prior research with known chemical values were also analyzed, to determine the accuracy of the XRF device, and in order to correct handheld XRF values to more appropriately reflect the bulk composition of the rock sample.

Hydrofluoric Acid Etching

Because many sample slabs appeared to have black-colored, structureless visages,

48% hydrofluoric acid (HFA) was used to etch the surfaces to look for hidden patterns.

The slabs were put onto raised plastic strips, about 2 cm x 2 cm x 8 cm in size, to prevent etching on the bottom of the slab. The 48% HFA was administered using a dropping tool attached to the top of the HFA bottle, both of which were made of plastic. The slabs were etched four at a time in large, flat plastic bins inside of an HFA- certified-use fume hood at Louisiana State University. Each slab took between 40 minutes and two hours to etch.

77 Appendix D: Handheld XRF Results

Table of Averaged Results

Normalized Major Elements (wt. %)

CES 4-2 CES 8-4 CES 8-5 CES 8-5a CES 8-6 CES 10-13 SiO2 97.92 99.36 98.11 99.11 98.25 99.91 TiO2 1.77 0.40 0.09 0.10 0.17 0.05 Al2O3 0.19 0.00 0.48 0.59 0.00 0.00 FeO 0.11 0.24 1.29 0.19 1.56 0.07 MnO 0.01 0.00 0.03 0.01 0.01 0.00 MgO* 0.00 0.00 0.00 0.00 0.00 0.00 CaO 0.00 0.00 0.00 0.00 0.00 0.00 K2O 0.00 0.00 0.00 0.00 0.00 0.00 P2O5 0.00 0.00 0.00 0.00 0.00 0.00

Unnormalized Trace Elements (ppm)

Ni 0 17 98 93 35 15 Cr 112 31 29 124 15 21 V 49 46 9 13 5 3 Rb 2 7 0 7 0 1 Sr 3 0 0 0 1 0 Zr 64 36 4 3 4 7 Y 2 3 1.5 0 8 0 Cu 13 9 31 8 24 12 Zn 2 0 0 0 0 77 As 2 9 2 0 13 11 Pb 16 0 0 0 0 0

78 Vita

Corey Elizabeth Shircliff, daughter of Jim and Jenny Shircliff, was born in

Louisville, Kentucky. In 2007, she graduated from the James Graham Brown School in

Louisville, and was accepted at Beloit College, in Beloit, Wisconsin. After first declaring a major in sociology, Corey quickly realized geology was more likeable, and switched into the geology program, where she completed a study abroad semester at the University of Otago in New Zealand, attended Beloit’s field camp, and completed a senior thesis under Dr. Carl Mendelson in Quaternary Palynology of the Colorado Front Range.

After deciding to apply to graduate school, Corey was accepted with a teaching assistantship to Louisiana State University to work with Dr. Gary Byerly. In August of

2011, she began her tenure at LSU. Corey spent three weeks in South Africa during her first summer at LSU, and interned at Marathon Oil Corporation during her second summer. After Corey finishes her graduate program at Louisiana State University, she will be moving to Houston, Texas, to begin work as a geologist at Marathon Oil

Corporation.