Hematite-coated microfossils: primary ecological fingerprint or taphonomic oddity of the Paleoproterozoic?

Microfossils belonging to the 1.88-billion-year-old 'Gunflint-biota' are preserved as carbonaceous and hematitic filaments and spheres that are believed to represent ancient chemolithoautotrophic Fe(II) oxidizing bacteria that grew above a chemocline where ferruginous seawater upwelled into shallow, oxygenated waters. This 'biological' model posits that hematite formed during burial from dewatering of the precursor ferric oxyhydroxides that encrusted Fe(II)-oxidizing bacteria. Here, we present an alternate 'taphonomic' model in which iron-rich groundwaters discharged into buried stromatolites; thus, the mineralization reactions are more informative of diagenetic processes than they are for primary marine conditions. We sampled centimeter-scale columnar stromatolites from both the lower and upper stromatolite horizons of the Biwabik and Gunflint formations, across a range of metamorphic gradients including unaltered to prehnite-pumpellyite taconite, supergene altered ore, and amphibolite-pyroxene grade contact-metamorphic zones. Fossils are rare to very rare and comprise curved filaments that exist in clusters with similar orientations. The filaments from throughout the Biwabik are similar to well-preserved carbonaceous Gunflintia from Ontario. Spheres of Huroniospora are also found in both formations. Microfossils from the least altered sections are preserved as carbon. Prehnite-pumpellyite samples are composed of either carbon or hematite (Fe2 O3 ). Within the contact aureole, filaments are densely coated by magnetite (Fe3 O4 ); the highest grade samples are secondarily oxidized to martite. The consistency in stromatolite microstructure and lithofacies throughout the metamorphic grades suggests they formed under similar environmental conditions. Post-depositional alteration led to replacement of the carbon by iron oxide. The facies association, filament distribution, and lack of branching and attached spherical cells argue against Gunflintia being a direct analogue to common marine, chemolithoautotrophic Fe(II)-oxidizing bacteria. Instead, we propose that the presence of hematite-coated microfossils is a reflection of taphonomic processes and does not necessarily reflect the byproduct of an original microbial ecosystem.