A decade and one half have passed since scientists in Grenoble and in Bern demonstrated that the CO2 content of the atmosphere was substantially lower during glacial time. Yet a totally satisfactory explanation still eludes us. The hypotheses fall into two main categories. In the first group are those which call on changes in the strength of the so-called biological pump. Glacial plant growth somehow led to a larger removal of CO2 from the surface ocean. In the second category are those which call on cyclic storage of CaCO3 in marine sediments. For some reason CaCO3 was preferentially dissolved during glacial time leading to a higher oceanic carbonate ion content. The problem is that all the proposals fail in one way or another to meet the observational constraints. Recent findings have encouraged proponents in each camp. Field experiments in the Galapagos area leave no doubt that the late John Martin's iron fertilization scheme works. This confirmation encourages those who propose that the high glacial dust fluxes provided the iron necessary to utilize NO3 and PO4 currently going to waste in areas of intense upwelling. Paleo pH reconstructions based on boron isotope measurements in foraminifera shells by Sanyal and Hemming lend support to a CaCO3 camp proposal made by Archer and Maier-Reimer that the glacial CO2 lowering was driven by a larger CO2 production through respiration in the upper few centimeters of deep sea sediments. While both proposals are tantalizing neither is without very serious drawbacks. Hence the debate is likely to rage for years to come.