Purpose: To examine a tobacco prevention elementary school project completed in 2000 in one rural state in the southeastern USA. Design/methodology/approach: Initially, 721 public school fourth through sixth grade teachers across Mississippi attended (for reimbursement) a tobacco prevention workshop that included curriculum-integrated material; 658 completed a follow-up questionnaire (91.3 percent return rate). In addition, 1,762 of their students completed an anti-tobacco pre-test and 1,723 (868 girls, 853 boys) completed an anti-tobacco post-test (97.8 percent return rate). Findings: Students improved from pre- to post-test; girls scored better than boys. African American students and teachers completed more lessons than White students and teachers. Sixth graders completed fewer lessons and scored lower than fourth and fifth graders. Research limitations/implications: The test had a yes-no format and a low ceiling, so many students answered correctly at pretest. Also, teachers were mostly white women, though they were demographically similar to teachers across the state, other groups would expand generalizability. Further, a true control group could evaluate the possibility that results were due to a secular trend. Practical implications: Not only were teachers positive about a tobacco prevention workshop that included curriculum integrated materials, but also their students scored healthier after having tobacco prevention lessons. Further, those who received the most lessons improved the most, indicating a dose-specific learning curve. (Contains 2 tables.)