I found something about Hematite and Bacteria:
http://www.lpi.usra.edu/meetings/lpsc2001/pdf/1438.pdf"This study shows that bacterial structures can be
preserved in Fe-oxides. Bacterial remnants from ironrich
regions of the Gunflint Iron Formation show a wide
range of preservational states, and it is likely that
degradation of the structures occurred before
fossilization of the microorganisms.
The abundance and ubiquity of fossilized polymeric
substances in these samples suggests that a possible
relationship between bacteria, Fe-oxides, and these
substances may exist. Further, there may be a strong
relationship between microbial mats and the
precipitation of iron oxides, leading to the genesis of
banded iron formations."
http://faculty.eas.ualberta.ca/konhauser/CJES-VP-2003.pdf"The immobilization of silica and iron by the bacteria Bacillus subtilis was monitored in controlled microcosms
to elucidate the role iron may play in aiding bacterial silicification in hot springs. Silica and iron immobilization was
monitored as a function of bacterial concentration, iron concentration, and silica concentration (both undersaturated and
oversaturated with respect to amorphous silica). Results demonstrate that bacterial cells do immobilize more Fe than
bacteria-free systems in solutions with iron concentrations ≤50 ppm Fe. However, as iron concentrations increase, the
difference between Fe immobilization in bacterial and bacteria-free systems decreases as non-bacterially mediated precipitation
processes dominate. Additionally, bacterial systems that had immobilized more Fe compared with bacteria-free systems
did not immobilize more silica than bacteria-free systems. By comparing molar ratios of (silica in solution)/(bacterially
bound Fe), it is evident that insufficient iron is bound to the bacterial surface to act as an effective salt bridge for silica
sorption. This appears to be because much of the iron is immobilized by non-bacterially mediated precipitation of
phases such as Fe(OH)3 and poorly ordered hydrous iron silicates. It follows that in silica-enriched hot springs, silica
and iron immobilization processes are significantly dominated by non-bacterially mediated precipitation. Any bacterially
mediated processes are exceedingly small and outside the resolution of these experiments."