It is now increasingly recognized that the Late Devonian black shales of the eastern US have a much more complex depositional history than assumed just 10 years ago. Detailed sedimentologic/stratigraphic studies of various units have shown that water depth may have not been uniformly deep as previously assumed, and the idea that black shales accumulated in the anoxic waters of a stratified basin has been all but abandoned. Fifty samples, taken over the 9.5 meter thickness of the Chattanooga Shale were analyzed for C, S, and a range of major and trace elements, in order to examine the anoxia question from a geochemical perspective. One geochemical index, DOP, suggests anoxia for essentially the entire time of Chattanooga accumulation, whereas others (e.g. Fe/Ti, Mo/Ti) suggest intermittent episodes of bottom water anoxia. Anoxia indicated by any of these indexes are in conflict with observations of bioturbation and sedimentary features. Pyrite enrichment due to intermittent reworking explains the consistently high DOP values. A recently introduced petrographic measure of anoxic condition, the size distribution of pyrite framboids, was applied to the same sample set that was used for geochemical analysis. Framboid size distributions in all samples indicate dysoxic conditions at the sea floor. However, when framboid sizes are examined for individual laminae (approx. 50 ƒÝm) we see that anoxic and dysoxic size distributions can alternate with each other. The latter situation is typical for those intervals where Fe/Ti, Mo/Ti ratios suggest anoxic conditions. It appears therefore that even during anoxic periods as indicated by bulk geochemistry, anoxia were short-lived events that may have reoccurred every few years. The anoxic "signal" seen in geochemical proxies probably arises through time averaging of intermittent anoxic "spikes".