In an attempt to create a scenario for the cause of the glacial to interglacial CO2 change recorded in air trapped in polar ice, we call on an increase in the alkalinity of polar surface waters. In this way we circumvent a major deficiency of the polar nutrient scenarios of Sarmiento and Toggweiler (1984), Siegenthaler and Wenk (1984) and Knox and McElroy (1984). Namely, our scenario does not require a drop in the nutrient content of polar surface waters in conformity with the demonstration by Boyle (1988a, b) that the cadmium content of planktonic foraminifera from polar regions did not decrease from late glacial to Holocene time. The rise in alkalinity required by our model is a natural consequence of the demise, during glacial time, of N...