Most studies of responses of insects to elevated carbon dioxide have been made using short-term exposures to treated food plants and have involved measurements of responses in growth, reproduction, food consumption and efficiencies of conversion at specific stages in the life cycle. These will be reviewed in the light of longer-term studies recently published where whole generations have been reared in chambers with simultaneous treatment of plants and where insects have been free to select their food and microenvironment. Factors such as seasonal change in plants, choice of food plant, mode of feeding, timing of exposure, temperature, the role of natural enemies are considered and the whole placed in the context of other aspects of climate change. It is concluded that in studies to date, the only feeding guild in which some species have shown increases in population density in elevated carbon dioxide are the phloem feeders. Chewing insects (both free-living,and mining) generally have shown no change or reduction in abundance, though relative abundance may be greatly affected. Compensatory feeding is common in these groups.