(cid:0)(cid:2)(cid:2)(cid:3)(cid:4)(cid:5)(cid:3)(cid:6) (cid:7)(cid:8)(cid:9)(cid:10)(cid:11)(cid:2)(cid:12)(cid:13)(cid:4)(cid:5) 3.46GaApexchert‘microfossils’reinterpretedasmineralartefactsproduced duringphyllosilicateexfoliation David Wacey, Martin Saunders, Charlie Kong, Alexander Brasier, Mar- tinBrasier PII: S1342-937X(15)00190-2 DOI: doi: 10.1016/j.gr.2015.07.010 Reference: GR1485 Toappearin: GondwanaResearch Receiveddate: 30April2015 Reviseddate: 15July2015 Accepteddate: 26July2015 Please cite this article as: Wacey, David, Saunders, Martin, Kong, Charlie, Brasier, Alexander, Brasier, Martin, 3.46 Ga Apex chert ‘microfossils’ reinterpreted as mineral artefacts produced during phyllosilicate exfoliation, Gondwana Research (2015), doi: 10.1016/j.gr.2015.07.010 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. ACCEPTED MANUSCRIPT 3.46 Ga Apex chert ‘microfossils’ reinterpreted as mineral artefacts produced during phyllosilicate exfoliation David Waceya,b,1, Martin Saundersc, Charlie Kongd, Alexander BraTsiere, and Martin Brasierf† P I R aSchool of Earth Sciences, University of Bristol, Queen’s Road, Bristol, BS8 1RJ, UK. bAustralian Research Council Centre of Excellence for CCore to Crust Fluid Systems, & Centre for Microscopy Characterisation and AnalSysis, The University of Western Australia, 35 Stirling Highway, Crawley, WA 600U9, Australia. cCentre for Microscopy Characterisation and NAnalysis, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. A d Electron Microscopy Unit, University of New South Wales, Kingsford, NSW 2052, M Australia e School of Geosciences, Meston BDuilding, University of Aberdeen, Old Aberdeen, AB24 3UE, UK E f†Department of Earth Sciences, University of Oxford, South Parks Road, Oxford, T OX1 3AN, UK. P E 1Corresponding ACuthor: David Wacey C School of EAarth Sciences University of Bristol Life Sciences Building 24 Tyndall Avenue Bristol, BS8 1TQ Email: [email protected] Tel: 0117 954 5400 Fax: 0117 954 5420 ACCEPTED MANUSCRIPT ABSTRACT Filamentous microstructures from the 3.46 billion year (Ga)-old Apex chert of Western Australia have been interpreted as remnants of Earth’s oldest cellular T life, but their purported biological nature has been robustly Pquestioned on I numerous occasions. Despite recent claims to the contraRry, the controversy surrounding these famous microstructures remains uCnresolved. S U Here we interrogate new material from the original ‘microfossil site’ using high N spatial resolution electron microscopy to decode the detailed morphology and A chemistry of the Apex filaments. LighMt microscopy shows that our newly discovered filaments are identical to the previously described ‘microfossil’ D holotypes and paratypes. Scanning and transmission electron microscopy data E show that the filaments coTmprise chains of potassium- and barium-rich P phyllosilicates, interleaved with carbon, minor quartz and iron oxides. E Morphological features previously cited as evidence for cell compartments and C dividing cells are shown to be carbon-coated stacks of phyllosilicate crystals. C Three-dimAensional filament reconstructions reveal non-rounded cross sections and examples of branching incompatible with a filamentous prokaryotic origin for these structures. When examined at the nano-scale, the Apex filaments exhibit no biological morphology nor bear any resemblance to younger bona fide carbonaceous microfossils. Instead, available evidence indicates that the microstructures formed during fluid-flow events that facilitated the hydration, heating and ACCEPTED MANUSCRIPT exfoliation of potassium mica flakes, plus the redistribution and adsorption of barium, iron and carbon within an active hydrothermal system. T Keywords: Apex Chert, Pilbara Craton, Microfossils, PseudofoPssils, Archean Life I R 1. Introduction C S The Apex microfossil debate is one of the longest running and highest profile U controversies in palaeobiology and evolution. At its heart are filamentous N microstructures found in black chert veins within the 3.46 Ga Apex Basalt near A Marble Bar in the Pilbara Craton of WeMstern Australia (Schopf and Packer, 1987; Schopf, 1993). On one side of the d ebate is the claim that these microstructures D represent at least eleven species of filamentous prokaryote microfossils comprising E some of the earliest morphoTlogical evidence of life on Earth (Schopf and Packer, P 1987; Schopf, 1992, 1993, 1999, 2006; Schopf et al., 2007; Schopf and Kudryavtsev E 2009, 2012). An opposing view is that these microstructures are not microfossils, C merely blobs of carbon, fortuitously arranged in roughly filamentous patterns around C crystal bounAdaries (Brasier et al., 2002, 2004, 2005, 2006, 2011, 2015). Adding complexity to the debate are separate reports from elsewhere in the Apex chert of other non-biological microfossil-like artefacts and later carbonaceous contaminants (Pinti et al., 2009; Marshall et al., 2011; Olcott Marshall et al., 2012; Sforna et al., 2014), inconclusive chemical studies attempting to assess the biogenictiy of Apex chert carbon (De Gregorio and Sharp, 2006; De Gregorio et al., 2009), plus questions over the suitability of the confocal laser Raman microspectroscopy technique favoured by the proponents of a biological origin (Pasteris and Wopenka, 2002, 2003; Marshall and Olcott Marshall, 2013). Below we summarise the history of the study of ACCEPTED MANUSCRIPT the Apex chert microstructures before presenting our new high spatial resolution electron microscopy data. T 1.1. History of the Apex microfossil debate P I The Apex chert ‘microfossils’ first entered the literature in 1R987 (Schopf and Packer, 1987) and were described in detail in the early 1990’s (SCchopf, 1992, 1993). They S were formally classified as fossils of uncertain biological affinity, ‘Bacteria Incertae U Sedis’. However, phrases in Schopf (1993) such as ‘I interpret…..the remaining seven N species….as probable cyanobacteria’ and ‘…suggest that cyanobacterial oxygen- A producing photosynthesizers may alreadMy have been extant this early in Earth history’ heavily implied a cyanobacterial aff inity. This implication entered into many D textbooks and popular science books written in subsequent years. With it came the E inference of a relatively advTanced level of prokaryote evolution at 3.5 Ga, plus an P early origin of photosynthetic oxygen production on Earth. Little attempt was made to E test these dogmatic assumptions until the authenticity of the ‘microfossils’ themselves C was challenged almost a decade later (Dalton, 2002). C A There were a number of subtle details in the initial reports (Schopf and Packer, 1987; Schopf, 1992, 1993) that are not wholly consistent with an interpretation of the filaments as microfossils, yet a decade or more passed without a thorough re- examination of the type material. Firstly, the claimed taxonomic diversity is particularly vast, being more diverse than 92% of all other Precambrian filamentous fossil assemblages. Comparable diversity is not seen until some 1500 Ma later, for example in the Gunflint Formation of Canada (Barghoorn and Tyler, 1965). Such taxonomic diversity in a deposit of this great age would require an early origin and ACCEPTED MANUSCRIPT diversification, then very slow evolution, of filamentous microbes. A number of the figured microfossils are rather light in colour, with yellow, orange and light brown examples (see colour images in Brasier et al., 2011). This is in contrast to other T reports of early Archean carbon that illustrate a dark brown to blPack colour, and hence I raises questions about the true age and carbonaceous compoRsition of some of the Apex microstructures. The filaments do not exhibit bioloCgical behaviour; instead they S are solitary, irregular and randomly orientated. Questions were also raised about the U selectivity of the data chosen for publication, with suggestions that more complex N objects exhibiting branching, a trait that would not have occurred in such primitive A organisms, were withheld from publicatMion (Packer quoted in Dalton, 2002). Although reported as ‘thin-sections’ (Schopf and Packer, 1987; Schopf 1993), the thicknesses of D the rock slices containing the type specimens are not standard (30 µm) but range from E 193 to 380 µm (Brasier et aTl., 2005). The structures illustrated by Schopf and Packer P (1987) and Schopf (1993) cannot be imaged in a single depth plain: this requires E montaging of several photographs. Inconsistencies between illustrations of the same C specimens also occur (Schopf and Packer, 1987; Schopf, 1993); for example, Fig. 3i C from SchopAf (1993) is the same specimen as Fig. 1a from Schopf and Packer (1987) but the former omits what appears to be a fold in the filament seen in the top right of the latter. In 2002, the Apex debate gathered pace with the publication of back-to-back papers in Nature (Brasier et al., 2002; Schopf et al., 2002). Brasier et al. (2002) presented new geological mapping data, a petrographic re-examination of the type material and observations from additional material collected from the original ‘microfossil’ locality. This revealed further inconsistencies in the earlier reports. For example, it ACCEPTED MANUSCRIPT had been claimed that the microfossils occurred in a sedimentary bedded chert unit and that all of the microfossils occurred in early rounded sedimentary clasts (Schopf, 1993). In contrast, Brasier et al. (2002) showed that the microstructures occurred in T black chert veins that intrude the lower part of the Apex Basalt. PFurthermore, Brasier I et al. (2002) showed that the microstructures co-occurred wRith a suite of minerals and textures characteristic of a hydrothermal setting, and thaCt near-identical micro- S structures also occurred in later generations of chert matrix as well as other rock types U in the immediate vicinity of the black chert veins. Using a computer-assisted montage N imaging approach, Brasier et al. (2002) also showed that parts of the type A microstructures had been left out of theM original manually-montaged photomicrographs (Schopf and Pack er, 1987; Schopf, 1993). This additional D information appeared to show filament branching and distribution of carbon around E ghosts of mineral crystals. TThe combined evidence led Brasier et al. (2002) to P conclude that the ‘microfossils’ were in fact carbonaceous mineral rims that formed E around recrystallizing grain margins during a complex series of hydrothermal events. C C Schopf et aAl. (2002) dismissed the petrographic contextual arguments presented by Brasier et al. (2002). They countered with confocal laser Raman microspectroscopic data that demonstrated a kerogenous carbonaceous composition for the microfossils. These authors claimed that correlation of this kerogenous chemistry with morphologically identifiable features characteristic of microorganisms confirmed their earlier conclusion that the Apex microstructures were bona fide microfossils. Laser Raman specialists, however, felt that Schopf et al. (2002) had over-interpreted the data and went on to demonstrate that laser Raman cannot unambiguously distinguish carbon of a biological precursor from that of an abiotic nature (Pasteris and Wopenka, ACCEPTED MANUSCRIPT 2002, 2003). Since these seminal papers, the debate has swung back and forth with Schopf and colleagues using laser Raman and confocal laser scanning microscopy (CLSM) in an attempt to reassert the biogenicity of the microfossils (e.g., Schopf, T 2006; Schopf et al., 2007; Schopf and Kudryavtsev 2009, 2012),P while Brasier and I colleagues have presented more detailed geological mappinRg and petrography in support of their mineral rim pseudofossil hypothesis (e.gC., Brasier et al., 2004, 2005, S 2006, 2011). Schopf and colleagues now concede that the geological setting is U hydrothermal rather than sedimentary and have moved away from a cyanobacterial N interpretation for the ‘microfossils’, suggesting instead that the ‘microfossils’ likely A represent ‘remnants of thermophilic micMrobes, preserved in situ and perhaps permineralized in hydrothermally m illed and rounded organic rich clasts’ (Schopf and D Kudryavtsev, 2012). E T P In latter years the debate has been blurred somewhat by descriptions of further suites E of abiogenic microfossil-like objects from various parts of the Apex chert (Pinti et al., C 2009; Marshall et al., 2011), and a multitude of chemical analyses of Apex carbon (De C Gregorio anAd Sharp, 2006; De Gregorio et al. 2009; Olcott Marshall et al., 2012; Sforna et al., 2014). There is no doubt that these studies have provided valuable data concerning the formation and subsequent alteration environments of the Apex chert and the carbon contained within. However, these microfossil-like artefacts do not show close similarity to the type Apex ‘microfossils’, and the chemical analyses were not performed on the ‘microfossils’. This has led some to question their relevance to the (re)interpretation of the type Apex ‘microfossils’ (Schopf and Kudryavtsev, 2012) and, if anything, has made the original claims easier to defend. Indeed recently, Schopf and Kudryavtsev (2012) claimed ‘a resolution of the controversy’ in favour of ACCEPTED MANUSCRIPT a bona fide microfossil origin for the Apex chert microstructures, although no additional data to support this hypothesis was presented in that paper. However, it is clear that the controversy has not been resolved, highlighted by the vigorous debate T that immediately followed publication of that paper (Marshall anPd Olcott Marshall, I 2013; Olcott Marshall and Marshall, 2013; Pinti et al., 2013R; Schopf and Kudryavtsev, 2013). C S U Here we return to the original microfossil locality and use high spatial resolution N electron microscopy to show precisely what minerals and textures comprise A filamentous microstructures equivalent Mto the ‘microfossil’ holotypes and paratypes. This allows us to demonstrate that t he purported microfossil filaments are not cellular D in nature and do not constitute morphologically preserved prokaryotes. Instead, they E comprise sheets of phyllosiTlicates with carbon sandwiched in between. Our hypothesis P of a phyllosilicate origin for the Apex ‘microfossils’ was first presented in Brasier et E al. (2015) and this was supported by a subset of the data reported here. This C contribution greatly expands on the evidence presented in Brasier et al. (2015) with C analyses ofA many more filaments, plus three-dimensional data and comparative Raman data not previously published. 2. Materials and methods 2.1. Sample locality and geological setting The c. 3.46 Ga Apex Basalt is found in the East Pilbara granite greenstone terrane of the Pilbara craton, Western Australia (Fig. 1). It is part of the Warrawoona Group, a 10-15 km thick volcano-sedimentary succession dominated by extrusive volcanic rocks with minor interstratified chert, barite, carbonate and volcaniclastic units. The ACCEPTED MANUSCRIPT ‘microfossils’ were initially reported from the Apex chert (informal name), the lowermost of several thin stratiform chert horizons within the Apex Basalt in the vicinity of Chinaman Creek, approximately 5 km west of the township of Marble Bar T (Schopf and Packer, 1987; Schopf, 1993). However, the geologiPcal setting of the I Apex ‘microfossils’ has been reinvestigated, with the resultRs presented in detail in Brasier et al. (2005) and Brasier et al. (2011). In summaCry, this new geological S mapping has shown that the ‘microfossils’ do not come from a sedimentary stratiform U chert layer as initially claimed (Schopf, 1993) but from a hydrothermally influenced N subsurface vein system, some 100 m below the stratiform Apex chert (Fig. 1c-d). This A context has been supported by subsequeMnt independent studies (Van Kranendonk, 2006; Pinti et al., 2009) and a hydro thermal geological context for the ‘microfossils’ D is now widely accepted (Schopf and Kudryavtsev, 2012). The reinterpreted E geological setting does not Trule out the presence of life in these rocks because P microorganisms are common in modern hydrothermal environments (e.g., Jannasch E and Wirsen, 1981). It does, however, bring added complexity to the interpretation of C any microstructures contained within such rocks, not least because of the possibility C of the preseAnce of abiogenic carbon generation in hydrothermal environments (e.g., Berndt et al., 1996) and the ease with which elements and minerals can be altered and transported by hydrothermal fluids. The material studied here (sample CHIN-3) comes from the original ‘microfossil’ locality (Fig. 1c-d) and was collected in 2001 (see also locality CC4 of Brasier et al., 2011). Analysis of previously published type specimens is compromised by the 193- 380 m thick preparations (Brasier et al., 2005) making optical petrography difficult. There is also an understandable prohibition (by the Natural History Museum, London)