Carbon now in terrestrial ecosystems regulates partitioning between soil organic C (SOC) and atmospheric CO2. Our objectives were to assess SOC dynamics using natural C-13 abundance in corn (Zea mays L., a C-4 species)-soybean [Glycine max (L.) Merr., a C-3 species] sequences. Fifteen treatments of continuous corn, continuous soybean, various sequences of corn and soybean, and fallow were initiated in 1981 at Lamberton, MN, on a Webster clay loam (fine-loamy, mixed, mesic Typic Haplaquoll). In 1991, soil and aboveground shoot samples from all treatments were analyzed for total organic C and delta(13)C. Carbon inputs, delta(13)C, and SOC were integrated into a two-pool model to evaluate C dynamics of corn and soybean. Total SOC was similar across all treatments after 10 yr; however, differences in soil delta(13)C occurred between continuous corn (delta(13)C = -17.2 parts per thousand) and continuous soybean (delta(13)C = -18.2 parts per thousand). Modeled C dynamics showed SOC decay rates of 0.011 yr(-1) for C-4-derived C and 0.007 yr(-1) for C-3- derived C, and humification rates of 0.16 yr(-1) for corn and O.11 yr(-1) for soybean. Decay and humification rates were slightly lower than those found in other Corn Belt studies. Levels of SOC were predicted to decline an additional 7 to 18% with current C inputs from either corn or soybean, respectively. Annual C additions required for SOC maintenance averaged 5.6 Mg C ha(-1), 1.4 to 2.1 times greater than previously reported estimates. Controlled variation in natural C-13 abundance in corn-soybean rotations during a 10-yr period adequately traced C dynamics.