Enigmatic is a word that often comes up in discussions about Proterozoic molar tooth carbonate structures (MTS). But when unusual features such as these are common in rocks of a particular age, there is almost always an important message waiting to be discovered. In this case, the observed temporal patterns for MTS likely track first-order trends in evolving compositions in the oceans during Earth’s middle history when CO2 in the atmosphere and carbonate saturation in the ocean were high but declining and oxygen (O2) in the ocean-atmosphere system was on the rise. A new paper by Tang et al. (this volume) gives us a new way to think about MTS origins, and nested within their model are wide ranging implicit and explicit linkages to Earth surface evolution as life was becoming more complex in a slow march toward the world we know today.

The Ordovician (Hirnantian; 445 Ma) hosts the second most severe mass extinction in Earth history, coinciding with Gondwanan glaciation and a growing body of geochemical evidence for marine anoxia. It remains unclear whether global cooling, expanded oxygen-deficiency, or a combination drove the Late Ordovician Mass Extinction (LOME). Here, we present new paired iodine and sulfur isotope geochemical data from three globally distributed carbonate successions to constrain changes in local and global marine redox conditions. Iodine records suggest locally anoxic conditions were potentially pervasive on shallow carbonate shelves, while sulfur isotopes suggest a reduction in global euxinic (anoxic and sulfidic) conditions. Late Katian sulfate-sulfur isotope data show a large negative excursion that initiated during elevated sea level and continued through peak Hirnantian glaciation. Geochemical box modeling suggests a combination of decreasing pyrite burial and increasing weathering are required to drive the observed negative excursion. This reduction of pyrite burial suggests a ~3% decrease of global seafloor euxinia during the Late Ordovician. The sulfur datasets spanning the late Hirnantian–early Silurian provide further evidence that this trend was followed by increases in euxinia which coincided with eustatic sea-level rise during subsequent deglaciation. A persistence of shelf anoxia against a backdrop of waning then waxing global euxinia was linked to the two LOME pulses. These results place important constraints on both local and global marine redox conditions throughout the Late Ordovician and suggest that non-sulfidic shelfal anoxia—along with glacioeustatic sea level and climatic cooling—were important factors leading to the LOME.