A 2-yr study was conducted in South Carolina wheat fields to describe spatial and temporal dynamics of stink bugs sampled with sweep nets. In 2008, the main phytophagous species were Euschistus servus (Say) (35.8, 36.7%), Oebalus pugnax (F.) (35.0, 6.0%), Nezara viridula L., (15.2, 46.3%), and Thyanta custator (F.) (14.0, 10.5%) for adults and nymphs, respectively. In 2009, the main phytophagous species were O. pugnax (39.8, 37.8%), E. servus (38.4, 30.0%), N. viridula (14.6, 29.5%), and T. custator (6.8, 2.8%). χ2 goodness-of-fit tests indicated that all species (both adults and nymphs) had observed distributions that did not match the Poisson distribution. Indices of dispersion (I_D) across sampling dates were significantly >1 for all species and life stages, also suggesting a nonrandom distribution. Slopes of Taylor's power law were significantly greater (P < 0.05) than a value of 1 for only adult O. pugnax and nymph T. custator. Coefficients β of patchiness regressions were significantly >1 in all cases except for E. servus adults and T. custator nymphs. The inverted distance weighted interpolation method showed considerable levels of spatial variability in densities within fields. Sampling along the edge of fields showed a significant distance from field border effect (P < 0.05) on all adults in both years except for N. viridula. High densities in wheat adjacent to a susceptible crop such as corn suggests that localized control methods in wheat may be effective in mitigation of stink bugs and damage in corn.