Soil-surface CO(2) flux (F(s)) is an important component in prairie C budgets. Although grazing Is common in grasslands, its effects on F(s) have not been well documented. Three clipping treatments: (i) early-season clipping (EC); (ii) full-season clipping (FC); and (iii) no clipping (NC); which represented two grazing strategies and a control, were applied to plots in a tallgrass prairie in northeastern Kansas, USA. Measurements of F(s) were made with a portable gas-exchange system at weekly to monthly intervals for 1 yr. Concurrent measurements of soil temperature and volumetric soil water content at 0.1 m were obtained with dual-probe heat-rapacity sensors. Measurements of F(s) also were obtained in grazed pastures. F(s) ranged annually from 8.8 x 10(-3) mg m(-2) s(-1) during the winter to 0.51 mg m(-2) s(-1) during the summer, following the patterns of soil temperature and canopy growth and phenology. Clipping typically reduced F(s) 21 to 49% by the second day after clipping despite higher soil temperatures in clipped plots. Cumulative annual F(s) were 4.94 4.04, and 4.11 kg m(-2) yr(-1) in NC, EC, and FC treatments, respectively; thus, clipping reduced annual F(s) by 17.5%. Differences in F(s) between EC and FC were minimal, suggesting that different grazing strategies had little additional impact on annual F(s). Daily F(s) in grazed pastures was 20 to 37% less than F(s) in ungrazed pastures. Results suggest that grazing moderates F(s) during the growing season by reducing canopy photosynthesis and slowing translocation of carbon to the rhizosphere.