A pelagic food web model was formulated with the goal of developing a quantitative understanding of the relationship between total production, export production, and environmental variables in marine ecosystems. The model assumes that primary production is partitioned through both large and small phytoplankton and that the food web adjusts to changes in the rate of allochthonous nutrient inputs in a way that maximizes stability, i.e., the ability of the system to return to steady state following a perturbation. The results of the modeling exercise indicate that ef ratios, defined as new production/totalproduction=exportproduction/totalproduction, are relatively insensitive to total production rates at temperatures greater than ~25¿C and lie in the range 0.1--0.2. At moderate to high total production rates, ef ratios are insensitive to total production and negatively correlated with temperature. The maximum ef ratios are ~0.67 at high rates of production and temperatures of 0¿--10¿C. At temperatures less than ~20¿C, there is a transition from low ef ratios to relatively high ef ratios as total production increases from low to moderate values. This transition accounts for the hyperbolic relationship often presumed to exist between ef ratios and total production. At low rates of production the model predicts a negative correlation between production and ef ratios, a result consistent with data collected at station ALOHA (22¿45'N, 158¿W) in the North Pacific subtropical gyre. The predictions of the model are in excellent agreement with results reported from the Joint global Ocean Flux Study (JGOFS) and from other field work. In these studies, there is virtually no correlation between total production and ef ratios, but temperature alone accounts for 86% of the variance in the ef ratios. Model predictions of the absolute and relative abundance of autotrophic and heterotrophic microorganisms are in excellent agreement with data reported from field studies. Combining the ef ratio model with estimates of ocean temperature and photosynthetic rates derived from satellite data indicates that export production on a global scale is ~20% of net photosynthesis. The results of the model have important implications for the impact of climate change on export production, particularly with respect to temperature effects. ¿ 2000 American Geophysical Union