Evaluating the annual sources and sinks of carbon from land-use change helps con- strain other terms in the global carbon cycle and may help countries choose how to comply with commitments for reduced emissions. This paper presents the results of recent analyses of land-use change in China and tropical Asia. The original forest areas are estimated to have cov- ered 54610 6 ha in tropical Asia and 42510 6 ha in China. By 1850, 44% of China's forests had been cleared, and another 27% was lost between 1850 and 1980, leaving China with 13% forest cover (29% of the initial forest area). Tropical Asia is estimated to have lost 26% of its initial forest cover before 1850 and another 33% after 1850. The annual emissions of carbon from the two re- gions reflect the different histories over the last 150 years, with China's emissions peaking in the late 1950s (at 0.20.5 Pg Ca −1 ) and tropical Asia's emissions peaking in 1990s (at 1.0 Pg Ca −1 ). Despite the fact that most deforestation has been for new agricultural land, the majority of the lands cleared from forests in China are no longer croplands, but fallow or degraded shrublands. Unlike croplands, the origins of these other lands are poorly documented, and thus add consider- able uncertainty to estimates of flux before the 1980s. Nevertheless, carbon emissions from China seem to have decreased since the 1960s to nearly zero at present. In contrast, emissions of car- bon from tropical Asia were higher in the 1990s than that at any time in the past. Changes in land use are important in the global carbon cycle. They explain a large part of the carbon sink in northern mid-latitude lands and indicate a large source in the tropics. The better these fluxes from land-use change are defined, the better constrained are the magnitude, location, temporal pattern, and mechanisms for other carbon sinks, and thus the more predictable is the be- havior of future sinks. A recent analysis of land-use change in China and recent revisions to the FAO's estimates of tropical deforestation offer opportunities for revising earlier estimates of car- bon flux for large portions of Asia. This paper presents new estimates of the annual flux of carbon from land-use change in tropical Asia and China. The estimates cover the period 18502000 and show contrasting histories for the two regions.