Abstract. In the first part of this review, I discuss how we can predict the direct short-term effect of enhanced CO2 on photosynthetic rate in C3 terrestrial plants. To do this, I consider: (1) to what extent enhanced CO2 will stimulate or relieve demand on partial processes like carboxylation, light harvesting and electron transport, the Calvin cycle, and end-product synthesis; and (2) the extent to which these various processes actually control the rate of photosynthesis. I conclude that control is usually shared between Rubisco (which responds sensitively to CO2) and other components (which respond less sensitively), and that photosynthesis will be stimulated by 25–75% when the CO2 concentration is doubled from 35 to 70 Pa. This is in good agreement with the published responses. In the next part of the review, I discuss the evidence that most plants undergo a gradual inhibition of photosynthesis during acclimation to enhanced CO2. I argue that this is related to an inadequate demand for carbohydrate in the remainder of the plant. Differences in the long-term response to CO2 may be explained by differences in the sink-source status of plants, depending upon the species, the developmental stage, and the developmental conditions. In the third part of the review, I consider the biochemical mechanisms which are involved in ‘sink’ regulation of photosynthesis. Accumulating carbohydrate could lead to a direct inhibition of photosynthesis, involving mechanical damage by large starch grains or Pi-limitation due to inhibition of sucrose synthesis. I argue that Pi is important in the short-term regulation of partitioning to sucrose and starch, but that its contribution to ‘sink’ regulation has not yet been conclusively demonstrated. Indirect or ‘adaptive’ regulation of photosynthesis is probably more important, involving decreases in amounts of key photosynthetic enzymes, including Rubisco. This decreases the rate of photosynthesis, and potentially would allow resources (e.g. amino acids) to be remobilized from the leaves and reinvested in sink growth to readjust the sink-source balance. In the final part of the review, I argue that similar changes of Rubisco and, possibly, other proteins are probably also involved during acclimation to high CO2.