Post-traumatic stress disorder (PTSD) and related psychopathology were examined in 95 female rape victims beginning soon after the assault (mean = 12.64 days). Subjects were assessed weekly for 12 weeks. Ninety-four percent of women met symptomatic criteria for PTSD at Assessment 1, decreasing to 65% at Assessment 4 (mean = 35 days postassault), and 47% at Assessment 12 (mean = 94 days postassault). PTSD and related psychopathology decreased sharply between Assessments 1 and 4 for all women. Women whose PTSD persisted throughout the 3-month study did not show improvement after the fourth assessment; women who did not meet criteria for PTSD 3 months postassault showed steady improvement over time. This pattern was evidenced even after initial PTSD severity was statistically controlled. Moreover, PTSD status at 3 months postassault could be predicted with a high degree of accuracy by two brief self-report measures administered at the first assessment. The implications of the present findings and directions for future research are discussed.