Recent breast cancer treatment studies conducted in large urban settings have reported racial disparities in the appropriate use of adjuvant chemotherapy. This article presents the first focused evaluation of black-white differences in receipt and completion of chemotherapy for breast cancer in a primarily rural region of the United States. We performed chart abstraction on initial therapy received by 868 women diagnosed with Stages I, IIA, IIB, or IIIA breast cancer in 2001-2003 in southwest Georgia (SWGA). For chemotherapy, information collected included treatment plan, dates of delivery, concordance between therapy planned and received, and date and reasons for end of treatment. The patient's age at diagnosis, race, marital status, insurance coverage, hormone receptor status, comorbidities, socioeconomic status, urban/rural status, treatment site, and distance to the site were also collected. Following univariate analyses, we used multivariable logistic regression modeling to examine the impact of race on the likelihood of (1) receiving chemotherapy and (2) completing planned chemotherapy. For patients terminating chemotherapy prematurely, the reasons were documented. The results showed that the unadjusted black-white difference in receipt of chemotherapy (48.3 vs. 36.0%) was significant, but in the multivariable analysis the black-white odds ratio (OR = 1.18) was not. While the unadjusted black-white difference (92.0 vs. 87.8%) in completing chemotherapy was not significant, in multivariable models black race was positively associated with completing care (p ranging from 0.032 to 0.087 and OR, correspondingly, from 2.16 to 2.64). The impact of race on completing chemotherapy was influenced by marital status, with a significant black-white difference for patients not married (OR = 4.67), but no difference for those married (OR = 1.06). We find compelling racial differences in this largely rural region-with black breast cancer patients receiving or completing chemotherapy at rates that equal or exceed white patients. Further investigation is warranted, both in SWGA and in other rural regions.