Vol 466 | 1 July 2010 | doi:10.

1038/nature09166

LETTERS
Large colonial organisms with coordinated growth in oxygenated environments 2.1 Gyr ago
Abderrazak El Albani1, Stefan Bengtson2, Donald E. Canfield3, Andrey Bekker4, Roberto Macchiarelli5,6, Arnaud Mazurier7, Emma U. Hammarlund2,3,8, Philippe Boulvais9, Jean-Jacques Dupuy10, Claude Fontaine1, Franz T. Fursich11, Francois Gauthier-Lafaye12, Philippe Janvier13, Emmanuelle Javaux14, Frantz Ossa Ossa1, Anne-Catherine Pierson-Wickmann9, Armelle Riboulleau15, Paul Sardini1, Daniel Vachard15, Martin Whitehouse16 & Alain Meunier1

The evidence for macroscopic life during the Palaeoproterozoic era (2.51.6 Gyr ago) is controversial15. Except for the nearly 2-Gyrold coil-shaped fossil Grypania spiralis6,7, which may have been eukaryotic, evidence for morphological and taxonomic biodiversification of macroorganisms only occurs towards the beginning of the Mesoproterozoic era (1.61.0 Gyr)8. Here we report the discovery of centimetre-sized structures from the 2.1-Gyr-old black shales of the Palaeoproterozoic Francevillian B Formation in Gabon, which we interpret as highly organized and spatially discrete populations of colonial organisms. The structures are up to 12 cm in size and have characteristic shapes, with a simple but distinct ground pattern of flexible sheets and, usually, a permeating radial fabric. Geochemical analyses suggest that the sediments were deposited under an oxygenated water column. Carbon and sulphur isotopic data indicate that the structures were distinct biogenic objects, fossilized by pyritization early in the formation of the rock. The growth patterns deduced from the fossil morphologies suggest that the organisms showed cell-to-cell signalling and coordinated responses, as is commonly associated with multicellular organization9. The Gabon fossils, occurring after the 2.452.32-Gyr increase in atmospheric oxygen concentration10, may be seen as ancient representatives of multicellular life, which expanded so rapidly 1.5 Gyr later, in the Cambrian explosion. Our samples come from the Francevillian Group, which belongs to a well-recognized lithostratigraphic succession, outcropping across 35,000 km2 in southeastern Gabon11,12. This group is exposed in the intracratonic basins of Plateau des Abeilles, Lastoursville and Franceville (Fig. 1), and reaches a maximum thickness of about 2,000 m. The group consists of five unmetamorphosed and undeformed sedimentary formations, FA to FE, bounded by conformable surfaces11,12. The lower part of the sequence (FA Formation) comprises fluvial deposits of a low-stand system tract dominated by onshore-tocoastal sandstones. In the FB Formation, marine deltaic deposition is indicated by facies development and sedimentary structures such as load casts, water escape structures, cross-stratification and hummocky cross-stratification. Shallower water conditions are observed in the FC Formation, whereas subsequent deposits (FD and FE) show

intercalated volcanic and continental sediments accumulated during the ultimate filling phase of the basin (Supplementary Fig. 1).

Gabon

Volcanic complex of NGoutou

Okondja Basin Plateau des Abeilles Lastoursville Basin Ondili shoal Francevillian Basin Archean basement

50 km

Franceville

Volcanic rocks of NGoutou Phanaerozoic deposit Francevillian group

Gneiss of Okanja Gneiss of Ogoue Gneiss and granite (Archaean)

Figure 1 | Simplified geological map of Gabon. Showing the Francevillian basin (inset) and the location of the fossiliferous site (star) near the town of Franceville.

1 Laboratoire HYDRASA, UMR 6269 CNRS-INSU, Universite de Poitiers, 86022 Poitiers, France. 2Department of Palaeozoology, Swedish Museum of Natural History, Box 50007, SE104 05 Stockholm, Sweden. 3Nordic Center for Earth Evolution, DK-5230 Odense M, Denmark. 4Department of Geological Sciences, University of Manitoba, Manitoba, R3T 2N2 5 Canada. Departement Geosciences, Centre de Microtomographie, Universite de Poitiers, 86022 Poitiers, France. 6Departement de Prehistoire, UMR 7194 CNRS, Museum National dHistoire Naturelle, Paris, 75005, France. 7Societe Etudes Recherches Materiaux, CRI Biopole, 86000 Poitiers, France. 8Department of Geological Sciences, Stockholm University, SE9 10 ` 106 91 Stockholm, Sweden. Departement Geosciences, UMR 6118, Universite de Rennes, 35042 Rennes, France. Bureau de Recherches Geologiques et Minieres, 45060 Orleans, France. 11GeoZentrum Nordbayern, Universitat Erlangen, Fachgruppe Palaoumwelt, D 91054 Erlangen, Germany. 12Laboratoire dHydrologie et de Geochimie de Strasbourg, UMR 7517 13 14 CNRS, 67084 Strasbourg, France. Departement Histoire de la Terre, UMR 7207 CNRS, Museum National dHistoire Naturelle, Paris, 75005, France. Departement de Geologie, ` ` ` Unite de Recherche Paleobotanique-Paleopalynologie-Micropaleontologie, Universite de Liege, Sart-Tilman Liege 4000, Belgium. 15Laboratoire Geosystemes, FRE 3298 CNRS, 16 Universite Lille 1, 59655 Villeneuve dAscq, France. Laboratory for Isotope Geology, Swedish Museum of Natural History, Box 50007, SE-104 05 Stockholm, Sweden.

100
2010 Macmillan Publishers Limited. All rights reserved

NATURE | Vol 466 | 1 July 2010

LETTERS

The lower part of the investigated FB2 section, where FB2 is the upper part of FB, outcropping near Franceville, consists of sandstone beds deposited in channels near the fair-weather wave base in the low-energy environment of a prograding delta. The topmost part of this section consists of an oxidized and stromatolitic hardground surface. This is sharply overlain by a 5-m-thick deposit of finely laminated horizontal black shales, interbedded with thin siltstone layers, deposited by waning storm surge without any evidence for subaerial exposure (Supplementary Fig. 1). The age of the FB deposits is well constrained to 2,100 6 30 Myr (refs 1315), roughly contemporaneous with the ,2.222.10 Gyr Lomagundi marine positive carbon-isotope excursion16 (see Supplementary Information and Supplementary Fig. 1) and about 200250 Myr after the first significant rise in atmospheric oxygen concentration10. More than 250 pyritized specimens embedded within their sedimentary matrix were collected in situ from at least 18 thin horizons, identified within the FB2 black shale lithofacies (Supplementary Fig. 2). In some cases, the layers containing the specimens were locally coated with iron oxides, owing to oxidation of pyrite crystals. The specimens range in shape from elongated to nearly isodiametric forms, with occasional finger-like protrusions (Fig. 2, Supplementary Fig. 3). Their length and width range from 7 to 120 mm and from ,5 to 70 mm respectively, and their thickness varies from ,1 to 10 mm. We estimate a density of up to 40 specimens per m2, with forms of different sizes and shapes and disparate orientations occurring together (Supplementary Figs 3, 4). We used micro-computed tomography (micro-CT)-based threedimensional (3D) imaging to characterize the outer and inner morphologies of the structures (see Supplementary Information). Most specimens show a pattern of radial fabric at the outer edge of their undulate or lobate periphery (Figs 2, 3, 4ac, Supplementary Figs 58); this is often curved, so as to meet the outer rim at a roughly perpendicular angle. In some cases, the radial fabric does not reach the outer rim (Fig. 4d), whereas in others it is simply lacking. The central parts of the larger forms are commonly thrown into smooth, transverse folds, which do not reach the outer edge and which are externally expressed as wrinkling of the structure (Figs 3, 4bc,
a

Figure 3 | In situ macrofossil specimen from the FB2 Formation. a, Lower side of the fossil (top) with its impression left in the black shales (bottom), showing peripheral radial fabric and wrinkled appearance. b, Micro-CTbased virtual reconstruction (volume rendered in semi-transparency), showing radial fabric and two inner pyrite concretions. c, Longitudinal virtual section running close to the estimated central part of the specimen, evidencing the fold pattern. Scale bars, 1.0 cm.

Figure 2 | Examples of black shale bedding surfaces. a, b, Bearing macrofossils in colony form from the FB2 level. Scale bars, 1.0 cm.

Supplementary Fig. 68, 11). Laminae of the host shale are draped around the folds (Supplementary Fig. 13), showing that the folding occurred before compaction. X-ray diffraction analyses show no mineralogical difference between the clay matrices in the specimens and the host shale (Supplementary Fig. 14, Supplementary Table 1). The larger specimens often also contain a central pyrite body (Figs 3, 4c, d, Supplementary Figs 8, 11), which is developed differently. Sometimes it forms a median layer within the folded sheet (Supplementary Figs 68), but it is more often nodular (Figs 3b, 4cd, Supplementary Figs 8, 11), sometimes deflecting the transverse folds (Fig. 4c, Supplementary Fig. 11). We measured topographic thickness along geometrically homologous virtual sections, which indicated progressive thinning towards the periphery (Supplementary Figs 9, 10). Differences in X-ray attenuation within the specimens are largely due to the differential distribution of octahedral pyrite crystals. The peripheral radial fabric is characterized by pyrite-free regions expressed in the microfabric as canals or slits (Supplementary Figs 12, 15). Secondary-ionization mass-spectrometric analysis of the pyrite reveals very light d34S values of about 225% to 230% in the fossilized sheets, with the central pyrite nodule tending towards heavier values of 5% to 15%, particularly in the outer margins (Fig. 5, Supplementary Table 2). The sheet, which represents the main body of the fossilized structure, was therefore pyritized during early diagenesis, when sulphate reducers were in direct contact with the effectively unlimited sulphate pool of the overlying water column. The high fractionations suggest sulphate concentrations in excess of 200 mM (ref. 17) (Supplementary Fig. 17). The pyritized nodules apparently formed later, from pore fluids more depleted in sulphate, and the pattern of sulphur isotopes suggests that
101

2010 Macmillan Publishers Limited. All rights reserved

LETTERS

NATURE | Vol 466 | 1 July 2010

15

34S

1 mm 30

Figure 4 | Micro-CT-based reconstructions and virtual sections of four specimens from the FB2 macrofossil record of Gabon. Samples show a disparity of forms based on: external size and shape characteristics; peripheral radial microfabric (missing in view d); patterns of topographic thickness distribution; general inner structural organization, including occurrence of folds (seen in views b and c) and of a nodular pyrite concretion in the central part of the fossil (absent in views a and b). a, Original specimen. b, Volume rendering in semi-transparency. c, Transverse (axial) two-dimensional section. d, Longitudinal section running close to the estimated central part of the specimen. Scale bars, 5 mm. Specimens from top to bottom: G-FB2-f-mst1.1, G-FB2-f-mst2.1, G-FB2-f-mst3.1, G-FB2-fmst4.1.

Figure 5 | Section through specimen G-FB2-f-mst4.3. d34S values (coloured spots, see scale) are measured in the central pyrite nodule (centre) and surrounding sheet material (top and bottom) by secondary-ionization mass spectrometry. See Supplementary Information.

pyritization began at the centre and continued towards the outer margins, during which process the remaining sulphate became progressively more depleted in light isotopes. The sulphur isotope patterns thus support the interpretation that the pyritized sheets represent early diagenesis of original biological fabric, whereas the occasional central lump of pyrite is a later, post-burial, diagenetic feature that is not likely to reflect original morphology. The differences in the organic carbon d13C content recorded between five specimens and their associated host shale sediment also support the fossilized structures representing a distinct organic entity (Supplementary Table 3). Plants and biomineralized animal tissues of the Phanaerozoic eon are commonly pyritized; pyritization of soft tissues is rare but typically results in faithful replication. This preservation is thought to be favoured by a low content of organic molecules and high content of reactive iron in the pore-waters18. We find no evidence to support an inorganic origin of the structures from the FB2 black shale level, whether concretions resulting from epitaxic/crystal growth processes, or features of diagenetic, sedimentary, hydrothermal, or tectonic origin. There is a superficial resemblance between the Gabon structures and the Ediacaran dubiofossil Mawsonites spriggi, which has been interpreted as a sand volcano interacting with biomats19; however, this interpretation accounts for neither the fine internal radial fabric nor the inner fold
102

pattern of the Gabon fossils, and there is no structural evidence of sediment injection in association with the fossils. The Gabon fossils also resemble radially growing pyrite or marcasite crystals, or pyrite suns, which are occasionally found in Phanaerozoic shales. However, a micro-CT-based comparison of the inner structures clearly shows that the pyrite suns have a much more regular and linear radial fabric than the Gabon specimens, and that this fabric extends all the way to the centre of the structure, without any evidence of flexible folding (Supplementary Fig. 16). Indeed, we are unaware of any inorganic processes that can generate the style of flexible folding and irregular radial fabric that we observe in the Gabon fossils (Fig. 4). The accumulated evidence suggests that the structures are biogenic. The fold pattern seen in the centre of most of the specimens indicates deformation of a flexible sheet, implying an originally cohesive structure of organic composition. The radial fabric is commonly deflected to meet the rim of the specimen, suggesting that the original material was growing by peripheral accretion of flexible organic matter. We conclude that the Gabon structures fulfil the general criteria of biogenicity applied to fossil-like forms in the early rock record20 (Supplementary Table 4). The presence of abundant organic matter in the FB Formation21,22 (Supplementary Table 5), including steranes of eukaryotic origin23, is consistent with this interpretation. We consider it most likely that these structures represent fossilized colonial organisms. Bacterial colonies growing on surfaces are known to coordinate their behaviour, resulting in regular shapes and distinct fabrics9; radial fabrics are common, and are thought to be due to repulsive chemotaxis24. Most studies of bacterial colony growth have been done on monocultures in Petri dishes, where colonies exceed centimetre size9. In nature, fairy-ring colonies, formed by cyanobacteria and diatoms and reaching a diameter of 15 cm, have been

2010 Macmillan Publishers Limited. All rights reserved

NATURE | Vol 466 | 1 July 2010

LETTERS

reported25. Nonetheless, structures similar to those from Gabon are unknown in the available fossil record and, because of their complex inner structural morphology and the sterane signature in the FB rocks, it is also possible that they represent colonial eukaryotes. Microbial mat-forming communities, including organisms whose phototactic behaviour modifies the mat shape, are inferred to have been prevalent in marine and lacustrine environments from the early Archaean eon26. Because of their sediment-binding properties, such mats often leave characteristic structures in carbonates and siliciclastic rocks. Such structures, however, including those formed in shales and mudstones, do not resemble the Gabon fossils27. Colonies with regular fabric resulting from coordinated-growth behaviour, as we infer for the Gabon fossils, represent a degree of organization different to that of such mat-forming communities. They require cell-tocell signalling and coordinated responses, akin to that required for multicellular organization9. The Gabon fossils represent the earliest evidence for such signalling and coordinated-growth behaviour on the scale of macroorganisms. One fundamental selective advantage of multicellularity is large size28, but ambient oxygen levels must be high enough to allow aerobic organisms to grow large. Our iron-speciation analyses reveal low ratios of highly reactive iron to total iron (FeHR/FeT)29, consistent with sediment deposition under an oxygenated water column (Supplementary Fig. 18). This implies that these fossil organisms, living on the sediment surface, were likely to engage in aerobic respiration. This is consistent with the timing of deposition, some 200 to 250 Myr after the first accumulation of oxygen into the atmosphere10,30. Although we cannot determine the precise nature and affinities of the 2.1-Gyr macroorganisms from the Francevillian B Formation of Gabon, we interpret these fossils as ancient representatives of multicellular life, which expanded so rapidly 1.5 Gyr later.
METHODS SUMMARY
We assessed textural relations between the pyritized sheet and the shale matrix embedding the macrofossils on sections, using a Nikon Eclipse E600. We carried out scanning electron microscopy on a JEOL 5600 LV, equipped with an Oxford EDX for mineralogical analyses. We obtained X-ray diffraction patterns from randomly-oriented powders and oriented preparations using a PANalytical XPert diffractometer (Ni-filtered Cu-Ka radiation), equipped with an accelerator detector (2u 2h analysis angle). We ran high-resolution micro-CT on X8050-16 Viscom AG equipment. We made reconstructions using DigiCT v.2.3 (Digisens), 64-bit version, running on a 2.5 GHz Dell T7400 Precision Windows XP 64 workstation with 32 GB of DDR RAM and two NVIDIA graphic cards (Quadro FX 5600 and Telsa C870). We carried out virtual sections and 3D rendering on AVIZO v.5 (Mercury Computer Systems). We carried out SRXTM tomographic microscopy at the X02DA TOMCAT beamline of the Swiss Light Source at the Paul Scherrer Institute (http://www.psi.ch/). We studied organic matter using Rock-Eval III pyrolysis (Oil Show Analyzer). We took isotopic measurements (d13Ccarb) on a VG Sira 10 triple collector mass spectrometer. We investigated iron speciation using the sequential extraction protocol, and determined sulphide concentrations by the chromium reduction method (CRM). We measured the concentration of iron in all iron fractions, except for pyrite, by atomic absorption spectrometry. We measured the d34S composition of bulk rock on Ag2S precipitates from samples of the sulphide that was liberated by CRM. We added about 200 mg to a tin cup with V2O5 and combusted it using a Thermo elemental analyser coupled via a Conflow III interface to a Thermo Delta V Plus mass spectrometer. We analysed S isotopes (32S, 33S and 34S) by secondary-ionization mass spectrometry using a Cameca IMS1270e7. For further details of sample treatment and analytical procedures, see Supplementary Information.
Received 29 March; accepted 4 May 2010.
1. 2. 3. Seilacher, A., Bose, P. K. & Pfluger, F. Triploblastic animals more than 1 billion years ago: trace fossil evidence from India. Science 282, 8083 (1998). Knoll, A. H., Javaux, E. J., Hewitt, D. & Cohen, P. Eukaryotic organisms in Proterozoic oceans. Philos. Trans. R. Soc. London B 361, 10231038 (2006). Bengtson, S., Rasmussen, B. & Krapez, B. The Paleoproterozoic megascopic Stirling biota. Paleobiology 33, 351381 (2007).

4.

5. 6. 7.

8.

9. 10. 11. 12.

13.

14.

15. 16.

17. 18.

19.

20. 21.

22.

23.

24.

25. 26.

27. 28. 29. 30.

Lamb, D. M., Awramik, S. M. & Zhu, S. Paleoproterozoic compression-like structures from the Changzhougou Formation, China: eukaryotes or clasts? Precambr. Res. 154, 236247 (2007). Rasmussen, B., Fletcher, I. R., Brocks, J. J. & Kilburn, M. R. Reassessing the first appearance of eukaryotes and cyanobacteria. Nature 455, 11011105 (2008). Han, T.-M. & Runnegar, B. Megascopic eukaryotic algae from the 2.1-billion-yearold Negaunee Iron-Formation, Michigan. Science 257, 232235 (1992). Schneider, D. A., Bickford, M. E., Cannon, W. F., Schulz, K. J. & Hamilton, M. A. Age of volcanic rocks and syndepositional iron formations, Marquette Range Supergroup: implications for the tectonic setting of Paleoproterozoic iron formations of the Lake Superior region. Can. J. Earth Sci. 39, 9991012 (2002). Bengtson, S., Belivanova, V., Rasmussen, B. & Whitehouse, M. The controversial Cambrian fossils of the Vindhyan are real but more than a billion years older. Proc. Natl Acad. Sci. USA 106, 77297734 (2009). Shapiro, J. A. & Dworkin, M. (eds) Bacteria as Multicellular Organisms (Oxford Univ. Press, 1997). Bekker, A. et al. Dating the rise of atmospheric oxygen. Nature 427, 117120 (2004). Gauthier-Lafaye, F. & Weber, F. The Francevillian (Palaeoproterozoic) uranium ore deposits of Gabon. Chem. Geol. 84, 22672285 (1989). Gauthier-Lafaye, F. & Weber, F. Natural nuclear fission reactors: time constraints for occurrence, and their relation to uranium and manganese deposits and to the evolution of the atmosphere. Precambr. Res. 120, 81100 (2003). Bros, R., Stille, P., Gauthier-Lafaye, F., Weber, F. & Clauer, N. Sm-Nd isotopic dating of Proterozoic clay material: an example from the Francevillian sedimentary series, Gabon. Earth Planet. Sci. Lett. 113, 207218 (1992). Hoori, K., Hidaka, H. & Gauthier-Lafaye, F. U-Pb geochronology and geochemistry of zircon from the Franceville series at Bidoudouma, Gabon. The 15th Annual Goldschmidt Conference. (2005). Gancarz, A. J. in The Natural Fission Reactors: Annual International Atomic Energy Agency Conference 513520 (TC-119/40, IAEA, 1978). Bekker, A. et al. Fractionation between inorganic and organic carbon during the Lomagundi (2.222.1 Ga) carbon isotope excursion. Earth Planet. Sci. Lett. 271, 278291 (2008). Habicht, K. S., Gade, M., Thamdrup, B., Berg, P. & Canfield, D. E. Calibration of sulfate levels in the Archean ocean. Science 298, 23722374 (2002). Farrell, U. C., Martin, M. J., Hagadorn, J. W., Whiteley, T. & Briggs, D. E. G. Beyond Beechers Trilobite Bed: widespread pyritization of soft tissues in the Late Ordovician Taconic foreland basin. Geology 37, 907910 (2009). Seilacher, A., Buatois, L. & Mangano, M. G. Trace fossils in the EdiacaranCambrian transition: behavioral diversification, ecological turnover and environmental shift. Palaeogeogr. Palaeoclimatol. Palaeoecol. 227, 323356 (2005). Wacey, D. Early Life on Earth: A Practical Guide (Springer, 2009). Cortial, F., Gauthier-Lafaye, F. & Lacrampe-Couloume, G. Oberlin, A. & Weber, F. Characterization of organic matter associated with uranium deposits in the Francevillian formation of Gabon (lower proterozoic). Org. Geochem. 15, 7385 (1990). Mossman, D. J., Gauthier-Lafaye, F. & Jackson, S. Carbonaceous substances associated with the Paleoproterozoic natural nuclear fission reactors of Oklo, Gabon: paragenesis, thermal maturation and carbon isotopic and trace element composition. Precambr. Res. 106, 135148 (2001). Dutkiewicz, A., George, S. C., Mossman, D. J., Ridley, J. & Volk, H. Oil and its biomarkers associated with the Palaeoproterozoic Oklo natural fission reactors, Gabon. Chem. Geol. 244, 130154 (2007). Ben-Jacob, E. Bacterial self-organization: co-enhancement of complexification and adaptability in a dynamic environment. Philos. Transact. Ser. A 361, 12831312 (2003). Grazhdankin, D. & Gerdes, G. Ediacaran microbial colonies. Lethaia 40, 201210 (2007). Allwood, A. C., Walter, M. R., Kamber, B. S., Marshall, C. P. & Burch, I. W. Stromatolite reef from the Early Archaean era of Australia. Nature 441, 714718 (2006). Schieber, J. in Atlas of Microbial Mat Features Preserved within the Clastic Rock Record (eds Schieber, J. et al.) 117134 (Elsevier, 2007). Bonner, J. T. First Signals: The Evolution of Development (Princeton Univ. Press, 2000). Canfield, D. E. et al. Ferruginous conditions dominated later Neoproterozoic deepwater chemistry. Science 321, 949952 (2008). Frei, R., Gaucher, C., Poulton, S. W. & Canfield, D. E. Fluctuations in Precambrian atmospheric oxygenation recorded by chromium isotopes. Nature 461, 250254 (2009).

Supplementary Information is linked to the online version of the paper at www.nature.com/nature. Acknowledgements We thank the Ministry of Mines, Oil, Energy and Hydraulic Resources and the General Direction of Mines and Geology of Gabon for collaboration and assistance, and the French Embassy at Libreville and the French Ministry for Foreign Affairs for support. We thank F. Mayaga-Mikolo, D. Beaufort, B. Cost, D. Thieblemont, F. Pambo and H. Sigmund for discussions. For assistance in Gabon and France, we acknowledge S. Accolas, T. Bonifait, B. Braconnier, N. Dauger, F. Duru, D. Fabry, F. Haessler, M. Jouve, G. Letort, D. Paquet, J.-C. Parneix, D. Proust, M. Stampanoni and X. Valentin. We also acknowledge the 103

2010 Macmillan Publishers Limited. All rights reserved

LETTERS
Institut Francais du Petrole, the Swiss Light Source (TOMCAT beamline) at the Paul Scherrer Institute, and the Centre de Microtomographie at the University of Poitiers (CdMT). Nordsim is operated under an agreement of the Joint Committee of the Nordic Research Councils for Natural Sciences (NOS-N), with further funding from the Knut and Alice Wallenberg Foundation; this is Nordsim contribution 256. Research was supported by the French CNRS-INSU, the Bureau ` de Recherches Geologiques et Minieres (BRGM), the Danish National Research Foundation and the Swedish Research Council. Author Contributions A.E.A. conceived and headed the project. A.E.A., S.B., D.E.C., E.H., A.B., R.M., J.-J.D., P.J. and A.Meunier designed research. A.E.A., A.Mazurier, E.H., F.O.O. and P.S. did field research. A.E.A. and F.O.O. analysed sedimentology. A.E.A., S.B., F.T.F., P.S. and D.V. analysed morphology. A.E.A., S.B., R.M. and A.Mazurier carried out microtomographic analyses. E.J. analysed palynology.

NATURE | Vol 466 | 1 July 2010

A.E.A., C.F., F.O.O. and A.Meunier analysed mineralogy. S.B., D.E.C., A.B., E.H., P.B., A.-C.P.-W., A.R. and M.W. carried out isotope and geochemical analyses. F.G.-L. provided geological samples. A.E.A., S.B., D.E.C., A.B., R.M., A.Mazurier, E.H., P.B., C.F., F.T.F., F.G.-L., P.J., E.J., F.O.O., A.-C.P.-W., A.R., D.V., M.W. and A.Meunier analysed data. A.E.A., S.B., D.E.C., R.M. and E.H. wrote the main part of the manuscript. A.B., A.Mazurier, P.B., J.-J.D., C.F., F.T.F., F.G.-L., P.J., E.J., A.-C.P.-W., A.R., D.V., M.W. and A.Meunier provided critical input to the manuscript. Author Information The repository of the fossils is the Department of Geosciences, University of Poitiers, France. Reprints and permissions information is available at www.nature.com/reprints. The authors declare no competing financial interests. Readers are welcome to comment on the online version of this article at www.nature.com/nature. Correspondence and requests for materials should be addressed to A.E.A. (abder.albani@univ-poitiers.fr).

104
2010 Macmillan Publishers Limited. All rights reserved

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

1 - The Francevillian Group and the FB2 section

Figure S1 | a, Lithostratigraphy of the Palaeoproterozoic Francevillian group with the five Formations (FA to FE) and age constraints shown. b, Detailed stratigraphy of the FB2 section outcropping near the town of Franceville, showing the macrofossil-bearing levels (black shales). c, Stratigraphic evolution of 13Ccarb values and of total organic carbon (TOC, wt. %) contents (see Tab. S5).

www.nature.com/nature

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Two Sm-Nd isochrons on clay minerals formed during the very early diagenesis give the FB Formation an age of 2099 115 million years (Ma) (Bros et al., 1992). Zircons from the tuff in the overlying FD Formation provided a 2083 5 Ma SHRIMP U-Pb syndepositional age (Hoori et al., 2005). The Oklo diagenetic uranium deposit at the FA-FB boundary has been dated at 2050 30 Ma by using U-Pb method (Gancarz, 1978). Taken together, these data indicate the depositional age for the Francevillian B Formation near 2100 30 Ma. This is supported by the highly positive 13C values of the carbonate fraction of the FB2 black shales (Fig. S1), the range of which (+5.5 to +9.6) is consistent with deposition during the ~2.22-2.10 Ga Lomagundi seawater positive carbon isotope excursion (Bekker et al., 2008).

www.nature.com/nature

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

2 - The site and the macrofossils

Figure S2 | a, The Francevillian B2 site outcropping near the town of Franceville, Gabon. b, The transition between the cemented sandstone beds and the black shales. c, The 5 m-thick finely-laminated fossiliferous black shales. Scale bar, 50 cm.

www.nature.com/nature

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S3 | Black shale bedding surface bearing macrofossils from the FB2 level photographed in situ. The disparity of spatially close and serially repeated forms is evident. Scale bar, 10 cm.

www.nature.com/nature

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S4 | a, Black shale surface from the FB2 level bearing macrofossils. In this case, density approximates 40 specimens/m2. b, c, d, Details showing the disparate orientation of differently sized and shaped structures (as indicated by arrows). The surface is locally coated with iron oxides. Contact deformation of the surrounding black shale sediment with the structures is occasionally seen. Scale bar, 10 cm.

www.nature.com/nature

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S5 | Macrofossil specimen (G-FB2-f-mst1.3) from the FB2 Formation. a, Lower side of the specimen (top) with its counterpart (impression) left in the black shales (bottom). b, Micro-CT-based 2D reconstruction (volume virtually rendered in coloured semi-transparency; for technical informations, see section 3 below) showing peripheral radial fabric. Scale bars, 1 cm.

www.nature.com/nature

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S6 | Macrofossil specimen (G-FB2-f-mst2.4) from the FB2 Formation. a, Lower side of the specimen (right) with its counterpart (left) showing peripheral radial fabric and folded appearance. b, Micro-CT-based 2D (left) and 3D (right) virtual reconstructions showing radial fabric and the inner fold pattern. c, Longitudinal section running close to the estimated central part of the specimen evidencing the extent of the central fold. Scale bars, 1 cm.

www.nature.com/nature

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S7 | Macrofossil specimen (G-FB2-f-mst2.2) from the FB2 Formation. a, Lower (right) and upper (centre) sides of the specimen with its counterpart (left) showing wrinkled appearance and, to a minor extent, peripheral radial fabric (notable on the lower side). b, Micro-CT-based 2D (left) and 3D (right) virtual reconstructions showing radial fabric and a particularly complex inner fold pattern. c, Longitudinal section evidencing the multiple, heterogeneous folds. Scale bars, 1 cm.

www.nature.com/nature

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S8 | Macrofossil specimen (G-FB2-f-mst3.10) from the FB2 Formation. a, Upper side of the specimen showing peripheral radial fabric and folded appearance. b, Micro-CT-based 2D (left) and 3D (right) virtual reconstructions showing radial fabric, folds, and the pyrite concretion filling the central part. c, Longitudinal section evidencing the fold pattern and the position of the pyritic body. Scale bars, 1 cm.

www.nature.com/nature

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

3 - Microtomography (micro-CT and SR-micro-CT) The macrofossils have been studied by means of high-resolution microtomography (microCT) at the Centre de Microtomographie of the University of Poitiers, France. The equipment used, a X8050-16 Viscom AG (1004x1004 camera), is a multi-scale X-ray inspection system which allows the analysis of variable sized objects up to a diameter of 300 mm. Depending on the specimens, scan parameters were 100 to 120 kV, 0.5 to 0.8 mA current, 1500 to 1800 views, 32 integrations by view, 16-bit tif images. Reconstructions have been done at the Etudes Recherches Matriaux company (www.erm-poitiers.fr) using the software DigiCT v.2.3 (Digisens) 64-bit version running on a 2.5 GHz Dell T7400 Precision Windows XP 64 workstation with 32 GB of DDR RAM and two NVIDIA graphic cards (Quadro FX 5600 and Telsa C870). This workstation allows using of the Digisens acceleration plug-ins for tomographic reconstruction, SnapCT, which takes advantage of the GPU processing capabilities. The final volumes were reconstructed in 992x992 formats with isotropic voxels. The spatial resolution varied from 17 to 83 m3. Virtual sections, 3D rendering and animations (see the attached four movies produced by A.Ma.) were performed by means of AVIZO v.5 (Mercury Computer Systems Inc.) 64-bit version. Two specimens (G-FB2-f-mst1.1 and mst3.8) have been also studied by means of synchrotron radiation X-ray microtomography (SR-micro-CT) at the tomography station of the Materials Science beamline (TOMCAT; http://sls.web.psi.ch/view.php/beamlines/tomcat/index.html) of the Swiss Light Source set at the Paul Scherrer Institute (http://www.psi.ch/), Villigen, Switzerland. In order to optimize the signal-to-contrast ratio, the X-ray energy was set at 40 keV, because of the mineral composition (e.g., mostly pyrite; Fig. S15). The magnification of the X-ray microscope was 10x or 20x. On-chip binning (2x) was selected to improve the signal-to-noise ratio for the 20x scans. The projections (1001 or 1501, depending on magnification; 550 to 700ms exposure time) were made over 180, and thus online post-processed and rearranged into flat- and darkfield-corrected sinograms. Reconstructions were performed on a 16-nodes Linux PC farm using highly optimized filtered back-projection routines. The reconstructed voxels are isotropic and have a size of 0.74 m. Topographic thickness variation has been assessed on the most suitable specimens along geometrically homologous (usually "parasagittal") virtual sections. The results provide a coherent pattern of symmetrical structural heterogeneity. Two examples are shown in Figs. S9 and S10.

www.nature.com/nature

10

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S9 | Micro-CT-based topographic thickness variation (in m) and cross-sectional outline assessed along three virtual sections (different colours) parallel to the main axis in the specimen G-FB2-f-mst1.1. Note the close pattern shown by the two lateral sections (blue and red) and the distinct outline of the thicker intermediate section (green) running closer to the central part of the fossil. Scale bars 5 mm.

www.nature.com/nature

11

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S10 | Micro-CT-based topographic thickness variation (in m) and cross-sectional outline assessed along two virtual sections running through the central part of specimen GFB2-f-mst2.4. The external profile of the specimen suggests an inwards compaction of the original structure involving one of its two long margins. Note the resemblance between the two profiles and the relative position, shape and extent of the two folds. Scale bars, 5 mm.

www.nature.com/nature

12

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S11 | Micro-CT-based imaging of the structural fold patterns shown by seven selected specimens from the FB2 Formation (axial virtual sections). a, G-FB2-f-mst2.1. b, G-FB2-fmst2.2. c, G-FB2-f-mst2.4. d, G-FB2-f-mst3.1. e, G-FB2-f-mst3.2. f, G-FB2-f-mst3.4. g, GFB2-f-mst3.10. Scale bars, 5 mm.

www.nature.com/nature

13

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S12 | SR-micro-CT-based 3D reconstruction of a lateral portion of a specimen from the FB2 Formation (G-FB2-f-mst.3.8) showing radial microfabric. The long axis of the investigated sample (from upper left to lower right) is approximately parallel to the long axis of the specimen. The upper and lower surfaces of the sample correspond to the two opposite surfaces of the specimen; the other surfaces (two cylindrical and two flat) are artificial, formed by the boundaries of the SR-micro-CT scan. Note the distribution of the octahedral pyrite crystals and their absence in a sediment-filled area forming narrow slits through the structure. Crystal size decreases towards periphery. Scale bar, 100 m.

www.nature.com/nature

14

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

4 - Composition of macrofossils Textural relations between a pyritized sheet and the sediment A section normal to bedding (Fig. S13) capturing the microstructure of the shale matrix shows that the shale laminae are draped around the folds of the pyritized sheet. This confirms that the structures are pre-compactional folds rather than features formed during late diagenesis.

Figure S13 | Section of a specimen from the FB2 Formation (G-FB2-f-mst2.6) normal to bedding, photographed in reflected, plane-polarized light. a, Complete section of the pyritized sheet showing general relationships between folded sheet and shale lamination. b, Detail of fold concavity showing that shale laminae do not transverse into sheet structure. Scale bars, 2.5 mm.

www.nature.com/nature

15

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

X Ray Diffraction As shown in Figure S14a, three samples were extracted from the host sediment (1) and from the edge (2) and the centre (3) of a pyritic body of a specimen from the FB2 Formation (G-FB2-f-mst2.1). The clay fraction was concentrated using centrifugation after gentle grinding. XRD patterns were obtained from randomly-oriented powders (Fig. S14b) and oriented preparations (Fig. S14c) using a PANalytical XPert diffractometer (Ni-filtered CuK radiation) equipped with accelerator detector (2 2 analysis angle). Analytical conditions were as follows: 40 kV, 40 mA; 0.020 2 scanning step 45 sec counting time per step; 2.565.0 2 and 2.0-30.0 scanning angular ranges for randomly oriented powders and oriented mounts, respectively. As seen in Table S1 (below), there is no significant difference in clay mineralogy between the three sampled spots.

www.nature.com/nature

16

www.nature.com/nature
Intensit (cps)

doi: 10.1038/nature09166

Intensit (cps)

10000 9600 9200 8800 8400 8000 7600 7200 6800 6400 1 6000 5600 5200 4800 4400 4000 3600 3200 2 2800 2400 2000 1600 1200 800 400 30 10000 9600 9200 8800 8400 8000 7600 7200 6800 6400 1 6000 5600 5200 4800 4400 4000 3600 3200 2 2800 2400 2000 1600 1200 800 400 30 5
- 14,20 Ch - 10,60 ML - 10,00 I - 7,08 Ch 14,20 Ch

2 10 15
- 10,60 ML - 10,00 I - 5,01 I - 4,72 - Ch I 4,49
- 4,18 Go - 3,72 Do - 3,55 Ch - 3,34 Qz - 3,13 Py - 4,26 Qz

4 20 25
- 7,08 Ch - 3,20 I - 2,99 I
- 2,89 Do - 2,58 C + I - 2,45 Go - 2,42 Py - 2,25 C + I - 2,21 Py - 2,45 Q - 2,71 Py + Go - 2,58 Ch + I

6 30

10

12

14

Position (2*thta)

Position (2*thta)

c
35
- 4,26 Qz

16 40
- 4,72 Ch

18 45
- 5,01 I
- 1,82 Q + C - 1,72 Go - 1,69 Go - 1,63 Py - 1,56 Py + Go + C -1,50 Py + Go + C -1,45 Py + Go - 1,91 Py + Go

- 2,28 Q - 1,82 Q + Ch - 2,19 Do - 2,12 Q - 2,02 Do - 1,98 Q

20 50
- 1,80 Py + Go

- 4,18 Go

22 55
- 3,55 Ch
- 3,34 Qz - 3,334 I

- 1,82 Q + Ch - 1,78 Do - 1,66 Q + Ch- 1,67 Q

24 60 65

26

-1,54 Q + Do -1,49 Ch + I + Do -1,45 Q

28

- 3,13 Py

30

SUPPLEMENTARY INFORMATION

17

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S14 | Mineralogical composition of the samples extracted from the pyritized body and its host sediment of specimen G-FB2-f-mst2.1 from the FB2 Formation. a, Sampled sites; 1: host sediment (arrow indicates the counterprint of the specimen); 2, edge; 3, central part of the specimen. Scale bars, 1 cm. b, XRD patterns of randomly oriented powders. c, XRD patterns of oriented preparations in the air-dried state. Chlorite (Ch), Illite/Smectite mixed-layer (ML); Illite (I); Quartz (Q), Goethite (Go), Pyrite (Py), Dolomite (Do).
sediment (see Fig. 14a-1) bulk mineral components (see Fig. 14b) Quartz Pyrite Goethite Dolomite Chlorite mixed-layer (Illite/Smectite) Illite phyllosilicates of clay fraction (see Fig. 14c) Chlorite mixed-layer (Illite/Smectite) Illite specimen (see Fig. 14a-2-3) edge ++++ +++ ++ + ++ + + centre ++ ++ ++ ++ + +

+ ++ + +

+++ ++ +++

+++ ++ +++

+++ ++ ++

Table S1 | Mineralogical composition of the G-FB2-f-mst2.1 specimen from the FB2 Formation. Relative quantities are appreciated using the intensity of the representative XRD peaks. Scanning Electron Microscopy (SEM) and Energy Dispersive Analysis System (EDX) Beside the structural investigations performed at infra-micrometric level by means of SRmicro-CT (Fig. S12), the mineral composition of the specimens, notably at the level of their periphery, has been analysed at the Department of Geosciences at the University of Poitiers by using a scanning electron microscope (SEM) JEOL 5600 LV equipped with an Oxford energy dispersive analysis system (EDX). The content of major elements has been normalized using silicate standards. Analytical conditions were 15 kV, probe current 6.10-10 A. An example of SEM-EDX results (specimen G-FB2-f-mst2.1) is provided in Fig. S15.

www.nature.com/nature

18

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S15 | SEM-EDX analysis of the lobate outer edge of a specimen from the FB2 Formation (G-FB2-f-mst2.1) showing radial microfabric. a, General view (left) and details of the areas within (lobe) and between (sediment) two lobate peripheral structures. b, Elemental composition of the lobe showing high content of sulphur (S) and iron (Fe). c, Elemental composition of the intermediate zone filled by clayey sediments.

www.nature.com/nature

19

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Secondary Ionization Mass Spectrometry (SIMS) Specimen G-FB2-f-mst4.3 was mounted in a 25 mm epoxy disc, coated with 30 nm gold and inserted into the ion microprobe sample holder together with pre-polished and coated pieces of sulfide standards (Crowe & Vaughan, 1996): Ruttan pyrite (34SCDT = +1.2 ), Balmat pyrite (34SCDT = +15.1 ), and an in house mass independently fractionated standard, 248474, from the Isua greenstone belt (34SCDT = +2 ; 33S =+ 3.2 ). S isotopes (32S, 33S and 34S) were analysed using a Cameca IMS1270e7 (SIMS) located at the Swedish Museum of Natural History, Stockholm (NordSIMS facility). Analytical methods and instrument parameters were similar to those used by Whitehouse et al. (2005) and Kamber and Whitehouse (2007). Briefly at +10 kV, ca. 2 nA Cs+ primary beam was operated in critical focussing (Gaussian) mode with a 5 m raster to generate a ca. 10 m diameter analytical area. The primary beam was operated together with a normal incidence electron flooding gun for charge compensation, -10kV secondary beam, software (Cameca CIPS version 5) automated centering of field aperture and simultaneous detection in three Faraday detectors at an effective mass resolution of ca. 4500 on the 33S channel, sufficient to resolve
33

S from

32 1

S H. The NMR field controller was utilised to ensure reproducibility of the

secondary ion signal during the analytical session. The pyrite isotope standards were analysed several times, bracketing analyses of sulfides of unknown isotopic composition, with the Ruttan pyrite used as the primary standard and the other two standards as monitors. External precision on the 34SCDT values is ca. +0.6 (2 std deviations) based on the Ruttan standard analyses. The results are shown in Tab. S2.
32 34

Sample ID surrounding sheet material_2-1 surrounding sheet material_2-2 surrounding sheet material_2-3 surrounding sheet material_2-4 surrounding sheet material_2-5 surrounding sheet material_2-6 surrounding sheet material_2-7 surrounding sheet material_2-8 surrounding sheet material_2-9 surrounding sheet material_2-10 surrounding sheet material_1-1 surrounding sheet material_1-2 surrounding sheet material_1-3 surrounding sheet material_1-4 surrounding sheet material_1-5 surrounding sheet material_1-6

S cps (x106) 740 729 730 739 737 748 747 753 754 714 733 746 735 605 733 738

S/32S

sig % 0.029 0.030 0.031 0.030 0.028 0.030 0.030 0.025 0.028 0.023 0.029 0.028 0.026 0.027 0.026 0.029

34SCDT -27.39 -27.46 -27.85 -27.07 -24.03 -24.63 -28.05 -25.42 -27.50 -28.81 -27.34 -27.05 -27.88 -26.23 -28.15 -27.30

1 s.d. 0.32 0.33 0.34 0.33 0.31 0.33 0.33 0.28 0.32 0.27 0.32 0.32 0.30 0.30 0.29 0.32

0.042879 0.042876 0.042859 0.042893 0.043027 0.043001 0.042850 0.042966 0.042874 0.042816 0.042881 0.042894 0.042858 0.042930 0.042845 0.042883

www.nature.com/nature

20

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

surrounding sheet material_1-7 surrounding sheet material_1-8 surrounding sheet material_1-9 surrounding sheet material_1-10 surrounding sheet material_1-11 surrounding sheet material_1-12 surrounding sheet material_1-13 surrounding sheet material_3-1 surrounding sheet material_3-2 surrounding sheet material_3-3 surrounding sheet material_3-4 surrounding sheet material_3-5 surrounding sheet material_3-6 surrounding sheet material_3-7 central pyrite nodule_1-1 central pyrite nodule_1-2 central pyrite nodule_1-3 central pyrite nodule_1-4 central pyrite nodule_1-5 central pyrite nodule_1-6 central pyrite nodule_1-7 central pyrite nodule_1-8 central pyrite nodule_1-9 central pyrite nodule_1-10 central pyrite nodule_2-1 central pyrite nodule_2-2 central pyrite nodule_2-3 central pyrite nodule_2-4 central pyrite nodule_2-5 central pyrite nodule_2-6 central pyrite nodule_2-7 central pyrite nodule_2-8 central pyrite nodule_2-9 central pyrite nodule_2-10 central pyrite nodule_3-1 central pyrite nodule_3-2 central pyrite nodule_3-3 central pyrite nodule_3-4 central pyrite nodule_3-5 central pyrite nodule_3-6 central pyrite nodule_3-7 central pyrite nodule_3-8 central pyrite nodule_3-9 central pyrite nodule_3-10 central pyrite nodule_3-11 central pyrite nodule_4-1 central pyrite nodule_4-2 central pyrite nodule_4-3 central pyrite nodule_4-4 central pyrite nodule_4-5 central pyrite nodule_4-6 central pyrite nodule_4-7 central pyrite nodule_4-8 central pyrite nodule_4-9

758 739 738 717 725 702 691 802 749 738 745 760 750 741 736 738 744 761 738 732 731 727 727 707 647 717 726 732 738 954 745 748 745 738 756 749 742 737 726 716 725 726 735 741 746 700 724 735 746 752 754 751 752 739

0.042955 0.042863 0.042861 0.042889 0.042900 0.042817 0.042785 0.042858 0.042904 0.042912 0.043023 0.043020 0.043012 0.043014 0.044267 0.043833 0.043559 0.043452 0.043386 0.043343 0.043270 0.043270 0.043093 0.043232 0.043269 0.043012 0.043126 0.043040 0.043141 0.042895 0.043052 0.043026 0.043063 0.042857 0.044589 0.044295 0.044084 0.043795 0.043405 0.043255 0.043345 0.043549 0.043975 0.044301 0.044572 0.044059 0.043944 0.043753 0.043458 0.043313 0.043017 0.042876 0.042970 0.042889

0.027 0.028 0.027 0.027 0.027 0.026 0.026 0.025 0.022 0.029 0.027 0.026 0.028 0.027 0.025 0.029 0.028 0.027 0.028 0.028 0.028 0.030 0.030 0.030 0.032 0.025 0.031 0.028 0.023 0.014 0.027 0.027 0.028 0.034 0.025 0.026 0.027 0.027 0.028 0.031 0.027 0.027 0.025 0.027 0.028 0.028 0.028 0.028 0.014 0.028 0.019 0.030 0.025 0.027

-25.68 -27.76 -27.81 -27.16 -26.91 -28.79 -29.52 -27.88 -26.83 -26.64 -24.12 -24.18 -24.38 -24.32 4.10 -5.76 -11.97 -14.40 -15.90 -16.86 -18.52 -18.52 -22.54 -19.38 -18.54 -24.37 -21.79 -23.73 -21.44 -27.03 -23.47 -24.05 -23.21 -27.89 11.40 4.73 -0.06 -6.62 -15.46 -18.87 -16.81 -12.20 -2.53 4.87 11.02 -0.62 -3.23 -7.57 -14.25 -17.54 -24.26 -27.46 -25.33 -27.17

0.30 0.32 0.31 0.31 0.30 0.30 0.30 0.29 0.26 0.32 0.30 0.30 0.32 0.31 0.29 0.32 0.31 0.30 0.31 0.31 0.31 0.33 0.33 0.33 0.35 0.29 0.34 0.31 0.27 0.20 0.31 0.31 0.31 0.37 0.29 0.30 0.31 0.31 0.31 0.34 0.31 0.30 0.29 0.31 0.31 0.31 0.32 0.32 0.20 0.32 0.24 0.33 0.29 0.31

www.nature.com/nature

21

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

central pyrite nodule_4-10 central pyrite nodule_5-1 central pyrite nodule_5-2 central pyrite nodule_5-3 central pyrite nodule_5-4 central pyrite nodule_5-5 central pyrite nodule_5-6 central pyrite nodule_5-7 central pyrite nodule_5-8 central pyrite nodule_5-9 central pyrite nodule_5-10 central pyrite nodule_5-11 central pyrite nodule_5-12 central pyrite nodule_5-13 central pyrite nodule_5-14 central pyrite nodule_5-15 central pyrite nodule_5-16 central pyrite nodule_5-17 central pyrite nodule_5-18 central pyrite nodule_5-19 central pyrite nodule_5-20 central pyrite nodule_5-21 central pyrite nodule_5-22 central pyrite nodule_5-23 central pyrite nodule_5-24 central pyrite nodule_5-25 central pyrite nodule_5-26 central pyrite nodule_5-27 central pyrite nodule_5-28 central pyrite nodule_5-29 central pyrite nodule_5-30 central pyrite nodule_5-31 central pyrite nodule_5-32 central pyrite nodule_5-33 central pyrite nodule_5-34 central pyrite nodule_5-35 central pyrite nodule_5-36 central pyrite nodule_5-37 central pyrite nodule_5-38 surrounding sheet material_4-1 surrounding sheet material_4-2 surrounding sheet material_4-3 surrounding sheet material_4-4 surrounding sheet material_4-5 surrounding sheet material_4-6 surrounding sheet material_4-7 surrounding sheet material_4-8 surrounding sheet material_4-9 surrounding sheet material_4-10 surrounding sheet material_4-11 central pyrite nodule_6-1 central pyrite nodule_6-2 central pyrite nodule_6-3 central pyrite nodule_6-4

731 626 676 685 693 703 711 728 733 608 576 624 654 646 687 695 701 733 728 680 467 621 678 688 689 695 697 696 706 704 696 700 706 710 717 714 729 725 724 723 718 453 736 697 721 733 726 723 703 467 662 703 684 692

0.043004 0.044195 0.044005 0.043718 0.043520 0.043374 0.043279 0.043221 0.043142 0.043968 0.043063 0.043158 0.043223 0.043321 0.043152 0.043153 0.044723 0.043439 0.043060 0.043650 0.043262 0.043104 0.043132 0.044086 0.044355 0.043106 0.043416 0.043323 0.043254 0.043284 0.043304 0.043495 0.043642 0.043880 0.044159 0.044263 0.044421 0.044571 0.044681 0.042950 0.042933 0.042918 0.042931 0.042815 0.043065 0.042931 0.043036 0.043050 0.042923 0.042789 0.042955 0.042927 0.043084 0.043056

0.027 0.033 0.029 0.027 0.028 0.028 0.029 0.026 0.024 0.017 0.032 0.032 0.032 0.039 0.027 0.027 0.021 0.026 0.024 0.014 0.034 0.028 0.027 0.032 0.026 0.028 0.030 0.025 0.027 0.027 0.030 0.029 0.027 0.028 0.025 0.027 0.026 0.026 0.025 0.026 0.029 0.013 0.028 0.018 0.028 0.027 0.027 0.025 0.026 0.009 0.029 0.034 0.025 0.027

-24.56 2.27 -2.04 -8.56 -13.04 -16.35 -18.50 -19.81 -21.61 -2.88 -23.42 -21.26 -19.77 -17.55 -21.39 -21.36 14.23 -14.89 -23.47 -10.09 -18.89 -22.48 -21.84 -0.21 5.89 -22.43 -15.39 -17.51 -19.07 -18.40 -17.94 -13.61 -10.28 -4.87 1.46 3.80 7.39 10.78 13.30 -25.98 -26.35 -26.70 -26.39 -29.02 -23.35 -26.40 -24.03 -23.70 -26.59 -29.63 -25.86 -26.49 -22.93 -23.56

0.31 0.38 0.34 0.33 0.33 0.34 0.35 0.32 0.30 0.26 0.37 0.37 0.37 0.43 0.33 0.33 0.28 0.32 0.31 0.23 0.39 0.34 0.33 0.38 0.32 0.34 0.35 0.31 0.33 0.33 0.36 0.35 0.33 0.33 0.31 0.33 0.32 0.32 0.31 0.32 0.34 0.23 0.34 0.26 0.33 0.33 0.33 0.32 0.32 0.21 0.34 0.39 0.31 0.33

www.nature.com/nature

22

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

central pyrite nodule_6-5 central pyrite nodule_6-6 central pyrite nodule_6-8 central pyrite nodule_6-9 central pyrite nodule_6-10 central pyrite nodule_6-11 central pyrite nodule_6-12 central pyrite nodule_6-13 central pyrite nodule_6-14 central pyrite nodule_6-15 central pyrite nodule_6-16 central pyrite nodule_6-17 central pyrite nodule_6-18 central pyrite nodule_6-19 central pyrite nodule_6-20 central pyrite nodule_6-21 central pyrite nodule_6-22 central pyrite nodule_6-23 central pyrite nodule_6-24 central pyrite nodule_6-25 central pyrite nodule_6-26 central pyrite nodule_6-27 central pyrite nodule_6-28 central pyrite nodule_6-29 central pyrite nodule_6-30 central pyrite nodule_6-31

657 677 699 697 702 704 551 710 712 723 723 719 712 675 704 671 697 700 699 689 706 704 600 704 689 685

0.043564 0.043336 0.044268 0.042924 0.042978 0.044503 0.043590 0.043084 0.043125 0.043093 0.043208 0.043065 0.043185 0.044654 0.043583 0.043557 0.043270 0.043327 0.043305 0.043292 0.043371 0.042915 0.043549 0.043561 0.042987 0.043567

0.025 0.026 0.021 0.028 0.027 0.023 0.021 0.026 0.025 0.027 0.028 0.024 0.023 0.028 0.023 0.047 0.025 0.024 0.026 0.025 0.038 0.028 0.006 0.026 0.030 0.028

-12.05 -17.21 3.92 -26.57 -25.34 9.25 -11.45 -22.94 -22.00 -22.72 -20.12 -23.37 -20.65 12.66 -11.62 -12.20 -18.72 -17.42 -17.91 -18.22 -16.42 -26.77 -12.39 -12.12 -25.13 -11.98

0.31 0.32 0.28 0.34 0.33 0.29 0.28 0.32 0.31 0.33 0.34 0.30 0.30 0.34 0.30 0.51 0.31 0.30 0.32 0.31 0.43 0.34 0.20 0.32 0.36 0.34

Table S2 | 34S values (average 1 s.d.) obtained by means of Secondary Ionization Mass Spectrometry (SIMS) on a longitudinal section running through the central part of a specimen from the FB2 Formation (G-FB2-f-mst4.3) and sampling at 150 sites its central pyrite nodule and the surrounding sheet material, till the marginal edge of the specimen (see also Fig. 5 in the main text). Mass Spectrometry Five fossilized specimens from the FB2 Formation and their associated host shale sediment were analysed for 13C content at the Institute for Geology and Geochemistry (IGG), Stockholm University (courtesy of H. Sigmund). The average results are shown in Tab. S3.
13C () sediment -27.83 -28.97 -32.72 -11.74 -22.44 13C () specimen -33.32 -34.65 -33.35 -27.16 -30.77

Sample ID s1 s2 s3 s4 s5

diff. () 5.49 5.68 0.63 15.42 8.33

www.nature.com/nature

23

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Table S3 | 13C values ( V-PDB) measured on five specimens from the FB2 Formation and their associated host shale sediment.

www.nature.com/nature

24

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

5 - Criteria of biogenicity

Figure S16 | Micro-CT-based axial virtual sections showing the inner structural organization of a specimen from the FB2 Formation compared with a non-biological object. a, A pyrite sun formed by radially growing crystals similar to colloform pyrite texture (rev. in Barrie et al., 2009). b, The peripheral radial fabric and the central folds characterizing the specimen GFB2-f-mst2.1. Scale bars, 1 cm.

www.nature.com/nature

25

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Criteria of antiquity A Structures must occur in rocks of known provenance (in situ) B Structures must occur in rocks of established age, ideally dated directly by radiometric techniques C Structures must be indigenous and syngenetic with the primary fabric of the host rock, i.e., they must be physically embedded within the rock, not introduced by post-depositional fluids D Structures should not occur in high-grade metamorphic rocks E The geological context of the host rock must be well understood at regional scale F Fossils should not be significantly different in colour from that of the rock matrix G There should be evidence for organo-sedimentary interaction Criteria of biogenicity H Structures should exhibit biological morphology that can be related to extant cells, structures or activities I More than a single step of biology-like processing should be evident. These steps may take the form of : biominerals (pyrite) geochemical fractionations of isotopes (sulphur and carbon) specific organic compounds J Structures should occur within a geological context that is plausible for life, i.e., at temperatures and pressures that extant organisms are known to survive K - Structures should fit within a plausible evolutionary context L - Structures should be abundant and ideally occur in a multi-component assemblage. Ideally they should show colonial/community behaviour

FB2 proposed fossils +++ +++ +++ +++ +++ ++ + ++ +++ +++ ++ +++ + ++

Table S4 | Condition shown by the 2.1 Ga FB2 black shales outcropping near Franceville, Gabon, and their fossilized macroscopic structures with respect to some among the most common criteria considered to assess antiquity and biogenicity (adapted from Wacey 2009). Assessment criteria ranging from "fully met condition" (+++) to "condition not met" (), through "uncertain condition" (~). +++ = condition fully met; ++ = condition mostly met; + = condition partly met.

www.nature.com/nature

26

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

6 - Geochemical and isotopic data (bulk sediment) Total organic carbon content and carbon isotopes The origin of the organic matter (OM) cannot be identified from the bulk data obtained from Rock-Eval pyrolysis because of the high maturity of the samples. Several authors have inferred algal and cyanobacterial origins for organic matter in the black shales of the FB Formation (Cortial et al., 1990; Mossman et al., 2001). A contribution from cyanobacteria and marine algae to the OM of the FB Formation is also supported by biomarkers (Dutkiewicz et al., 2007). Notably, a eukaryote signature indicated by steranes has been proposed to be responsible for very high OM content in the FB Formation (Dutkiewicz et al., 2007). The total organic carbon (TOC) content (wt. %) of three samples from the lower sandstone beds of the FB2 unit and 10 samples from the laminated black shales of the FB2 fossiliferous horizon are shown in Tab. S5. Carbon isotopic values for 10 black shale samples are shown using standard notation in units of relative to V-PDB standard in Tab. S5. The 13C values of the carbonate fraction (13Ccarb) were measured on CO2 released by reaction of 50 to 300 mg of fine-grained powder with 100% H3PO4 at 50C (Stable Isotopes Laboratory, Department of Geosciences at the University of Rennes 1). H2S released during reaction between sulphide grains and H3PO4 was eliminated by reaction with Ag3PO4 at ca. 60C for 5 minutes. Isotopic measurements were done on a VG Sira 10 triple collector mass spectrometer. Repeated analyses of internal lab standard Prolabo Rennes gave a mean 13C value of -9.720.03 (1s, n=14), close to the accepted value of -9.692. No correction was added to the measured values, and the analytical uncertainty is estimated lower than 0.1. The studied carbonate-containing rocks are characterized by carbonate content between 2.9 and 11.4 wt. % and 13Ccarb values between +5.5 and +9.6 V-PDB (Fig. S1c), which are typical for marine carbonates deposited during the 2.22-2.10 Ga Lomagundi seawater positive carbon isotope excursion (Schidlowski et al., 1975; Bekker et al., 2008).

www.nature.com/nature

27

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Sample FB9-02 FB9-04 FB9-06 FB9-08 FB9-09 FB9-10 FB9-11 FB9-12 FB9-13 FB9-14 FB9-15 FB9-16 FB9-17

13Ccarb ( V-PDB)

TOC (wt. %) (rock-eval) 0.12 0.39 0.39

7.3 6.8 5.6 5.6 5.9 7.0 5.5 6.5 9.6 6.1

2.30 1.79 3.41 1.98 2.00 3.41 2.20 2.12 2.47 1.72

carbonate (wt. % carb) (calcimetry) 0 (sandstone) 0 (sandstone) 0 (sandstone) 2.9 6.4 8.2 10.0 11.4 5.2 7.1 7.4 3.3 8.4

Table S5 | 13Ccarb values ( V-PDB), total organic carbon (TOC, wt. %), and carbonate content of samples from the FB2 section (cf. Fig. S1c). Isotopes of sulphur The 34S composition of bulk rock, was measured at the Nordic Center for Earth Evolution at the University of Southern Denmark, Odense, on Ag2S precipitates from 22 samples of the sulphide that was liberated by the CRS method (Newton et al., 1995; Lyons, 1997). About 200 g were added to a tin cup with V2O5 and combusted using a Thermo elemental analyzer coupled via a Conflow III interface to a Thermo Delta V Plus mass spectrometer. Sulphur isotope compositions are expressed as permil () deviations from V-CDT using the conventional delta notation (Fig. S17) with a standard deviation of about 0.2.

www.nature.com/nature

28

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Figure S17 | 34S values ( V-CDT) measured on 22 samples from the lower (FB1) and the upper (FB2) parts of the Francevillian B Formation. Iron Iron speciation was investigated using the sequential extraction protocol (Poulton & Canfield, 2005) and sulphide concentrations were determined by the chromium reduction method (CRS) (Zhabina & Volkov, 1978; Canfield et al., 1986). The concentration of iron in all iron fractions, except for pyrite, were measured by atomic absorption spectrometry (AAS), while pyrite content was estimated from the sulphide concentrations by Cr reduction assuming a pyrite stoichiometry (Fe2S) in the sediment. Reproducibility was above 95% for all phases but magnetite (Femag), where concentrations were close to zero and we consider the error to be within the detection level of the AAS. The highly reactive iron (FeHR) is considered to be the diagenetically active iron and is defined as the sum of carbonate-associated iron (Fecarb), ferric oxides (Feox) and Femag, plus all the iron that has been converted to sulphide phases in the water column or sediment (Fepy) (Canfield et al., 1992; Poulton et al., 2004); so FeHR = Fecarb + Feox + Femag + Fepy. The highly reactive iron is then related to total iron content of the sediment (FeT) to form a redox indicator, as sediments deposited below an anoxic water column usually have a FeHR/FeT ratio above 0.38 (Raiswell & Canfield, 1998; Poulton & Raiswell, 2002; Canfield et al., 2008). We applied this iron speciation technique to 24 well-preserved rock samples. In our dataset there are four samples with FeHR/FeT >0.38, indicating anoxia, while the other 20 samples are consistent with oxygenated water-column conditions (Fig. S18). The analyzed samples were

www.nature.com/nature

29

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

not deposited by turbidities, which could distort an anoxic iron signal (Raiswell & Canfield, 1998), thus we interpret this sample set to have been deposited in a predominantly oxic environment. FeHR/FeT
0,70

0,60

0,50

0,40

0,30

0,20

0,10

0,00

FB2

FB1

Figure S18 | Ratio of highly reactive to total Fe (FeHR/FeT) measured in 24 samples from the lower (FB1) and the upper (FB2) units of the Francevillian B Formation.

www.nature.com/nature

30

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Cited references
Barrie C.D., Boyce A.J., Boyle A.P., Williams P.J., Blake K., Ogawara T., Akai J. & Prior D.J. 2009. Growth control in colloform pyrite. Am. Mineral. 94, 415-429. Bekker A., Holmden C., Beukes N.J., Kenig F., Eglinton B. & Patterson W.P. 2008. Fractionation between inorganic and organic carbon during the Lomagundi (2.22-2.1 Ga) carbon isotope excursion. Earth Planet. Sci. Letters 271, 278-291. Bros R., Stille P., Gauthier-Lafaye F., Weber F. & Clauer N. 1992. Sm-Nd isotopic dating of Proterozoic clay material: an example from the Francevillian sedimentary series, Gabon. Earth and Planet. Sc. Letters 113, 207-218. Canfield D.E., Poulton S.W., Knoll A.H., Narbonne G.M., Ross G., Goldberg T. & Strauss H. 2008. Ferruginous conditions dominated later Neoproterozoic deep-water chemistry. Science 321, 949952. Canfield D.E., Raiswell R. & Bottrell S. 1992. The reactivity of sedimentary iron minerals towards sulfide. Am. J. Sci. 292, 659-683. Canfield D.E., Raiswell R., Westrich J.T., Reaves C.M. & Berner R.A. 1986. The use of chromium reduction in the analysis of reduced sulfur in sediments and shales. Chem. Geol. 54, 149-155. Cortial F., Gauthier-Lafaye F., Lacrampe-Couloume G. Oberlin A. & Weber F. 1990. Characterization of organic matter associated with uranium deposits in the Francevillian formation of Gabon (lower Proterozoic). Organic Geochem. 15, 73-85. Crowe D.E. &Vaughan R.G. 1996. Characterisation and use of isotopically homogeneous standards for in situ laser microprobe analysis of 34S/32S ratios. Am. Mineral. 81, 187-193. Dutkiewicz A., George S.C., Mossman D.J., Ridley J. & Volk H. 2007. Oil and its biomarkers associated with the Palaeoproterozoic Oklo natural fission reactors, Gabon. Chem. Geol. 244, 130154. Gancarz A.J. 1978. U-Pb age (2.05x109 years) of the Oklo uranium deposit. The Natural Fission Reactors. IAEA Symp. Proc., Vienne, 513-520. Hoori K., Hidaka H. & Gauthier-Lafaye F. 2005. U-Pb geochronology and geochemistry of zircon from the Franceville series at Bidoudouma, Gabon. The 15th Annual Goldschmidt Conference. Moscow, Idaho, USA. Kamber B.S. & Whitehouse M.J. 2007. Micro-scale sulphur isotope evidence for sulphur cycling in the late Archean shallow ocean. Geobiol. 5, 5-17. Lyons T.W. 1997. Sulfur isotopic trends and pathways of iron sulfide formation in upper Holocene sediments of the anoxic Black Sea. Geochim. Cosmochim. Acta 61, 3367-3382. Mossman D.J., Gauthier-Lafaye F. & Jackson S. 2001. Carbonaceous substances associated with the Paleoproterozoic natural nuclear fission reactors of Oklo, Gabon: Paragenesis, thermal maturation and carbon isotopic and trace element composition. Precamb. Research 106, 135-148.

www.nature.com/nature

31

doi: 10.1038/nature09166

SUPPLEMENTARY INFORMATION

Newton R.J., Bottrell S.H., Dean S.P., Hatfield D. & Raiswell R. 1995. An evaluation of the use of the chromous chloride reduction method for isotopic analyses of pyrite in rocks and sediment. Chem. Geol. (Isot. Geosci. Sect.) 125, 317-320. Poulton S.W. & Canfield D.E. 2005. Development of a sequential extraction procedure for iron: implications for iron partitioning in continentally-derived particulates. Chem. Geol. 214, 209-221. Poulton S.W., Krom M.D. & Raiswell R. 2004. A revised scheme for the reactivity of iron (oxyhydr) oxide minerals towards dissolved sulfide. Geochim. Cosmochim. Acta 68, 3703-3715. Poulton S.W. & Raiswell R. 2002. The low-temperature geochemical cycle of iron: from continental fluxes to marine sediment deposition. Am. J. Sci. 302, 774-805. Raiswell R. & Canfield D.E. 1998. Sources of iron for pyrite formation in marine sediments. Am. J. Sci. 298, 219-245. Schidlowski M., Eichmann R. & Junge C.E. 1975. Precambrian sedimentary carbonates: carbon and oxygen isotope geochemistry and implications for the terrestrial oxygen budget. Precambr. Res. 2, 1-69. Wacey, D. Early Life on Earth. A Practical Guide (Springer, 2009). Whitehouse M.J., Kamber B.S., Fedo C.M. & Lepland A. 2005. The importance of combined Pb and S isotope data from early Archaean rocks, southwest Greenland, for the interpretation of S-isotope signatures. Chem. Geol. 222, 112-131. Zhabina N.N. & Volkov I.I. 1978. A method of determination of various sulfur compounds in sea sediments and rocks. In (W.E. Krumbein, ed.) Env. Biogeochem. Geomicrob. v.3. Methods, Metals and Assessment. Ann Arbor Science, pp. 735-745.

Movie 1 | Specimen G-FB2-f mst1.1. Movie 2 | Specimen G-FB2-f mst2.1. Movie 3 | Specimen G-FB2-f mst3.1. Movie 4 | Specimen G-FB2-f mst4.1.

www.nature.com/nature

32