Grid sampling (one sample per 0.40 ha) in 12 cotton (Gossypium hirsutum L.) fields for stink bugs and boll injury caused by stink bug feeding was conducted in 2007 and 2008 in South Carolina and Georgia. The main species collected using the beat cloth method were the green stink bug, Acrosternum hilare (Say) (69%); brown stink bug, Euschistus servus (Say) (18%); and the southern green stink bug, Nezara viridula (L.) (14%), respectively. The inverted distance weighted interpolation method and Spatial Analysis by Distance IndicEs analyses showed spatial variability in both boll injury and stink bug densities. Four of five fields with field-average boll injury exceeding 15% had a combination of either soybean or peanut adjacent to the cotton field of interest. The Spatial Analysis by Distance IndicEs index of dispersion for stink bug densities and boll injury averaged over the season indicated significant aggregation at the 5% error rate in only one and four analyses (combinations of fields, species, and life stages) of a total of 54 in 2007 and 2008, respectively. By sampling date, overall indices of dispersion for boll injury data indicated significant aggregation in four of the 63 date-field combinations at the 5% error rate. Across years, overall indices of spatial association between boll injury and stink bug densities were significant in 17 of 69 analyses, with positive associations detected in seven of 12 fields; these data suggest that spatial distribution of stink bugs in cotton fields does not always coincide with boll injury.