Holland (1984) suggests that the contents of redox-sensitive elements, such as Mo, in marine shales deposited under an oxic atmosphere should be higher than those deposited under an anoxic atmosphere and should show positive correlations with the organic carbon (C_org) contents. This is because oxidative weathering of terrestrial rocks is expected to promote dissolution of such elements and to increase their concentrations in seawater. Therefore, Mo and C_org contents of marine shales may be useful indicators of oxygen level in the contemporaneous atmosphere.

We have carried out systematic analyses of more than 100 drillcore samples of Archean black shales in 9 drillholes in South Africa and Australia, and reviewed literature data on more than 100 samples of C_org-rich shales/sediments of many different ages. We have found the Mo-C_org relationships in modern marine sediments show a great variability, that those in the Archean shales (C_org: 0 - 1 wt. % and Mo: 0 - 5 ppm) are essentially the same as those in the Phanerozoic shales, and that high Mo content are typically associated with high sulfide-S content.

We have also computed the dissolution rates of pyrite, in which most Mo in rock resides, as a function of atmospheric oxygen level. All pyrite grains less than 1 cm³ in volume are likely to be completely oxidized in less than 10⁴ years, if the atmospheric oxygen level is more than 10^-6 atm (=5¥10^-4 % PAL). That is, in contrast to a popular view, the riverine Mo flux has most likely been constant since Archean, regardless of the atmospheric oxygen level.

Above results suggest that the geochemical cycles of the redox-sensitive elements have been essentially the same since the Archean, and that the Phanerozoic-style oceanic redox structure (i.e., generally oxic ocean with localized anoxic/euxinic basins where bacterial sulfate reduction is active) may have already developed in the Archean.