Objective: To evaluate the efficacy of a self-collected vaginal introital specimen, obtained by women using only an instruction booklet as a guide, for the detection of Chlamydia trachomatis using polymerase chain reaction (PCR). Methods: Comparison analysis in an Urban hospital clinic, using a convenience sample of 101 women undergoing screening or treatment for C trachomatis infection. Subjects were given an illustrated instruction booklet and were asked to collect a single specimen from the vaginal introitus using the Amplicor collection kit. Cervical and introital specimens then were obtained by clinicians. The presence of C trachomatis in patient-collected versus clinician-collected samples was determined by PCR. Discordant specimens were retested by PCR, using primer pairs to the C trachomatis major outer membrane protein gene, and were screened for the presence of PCR assay inhibitors. Results: Ninety-nine subjects completed the self-collection. Sixty-two percent were adolescents and 17% spoke English as their second language. Forty-three percent had never used tampons, and 36% had never looked at their own genitals. Twenty-eight cervical specimens (28%), 32 clinician-collected introital specimens (32%), and 33 self-collected introital specimens (33%) were positive for C trachomatis infection. All clinician-collected introital specimens that were positive for C trachomatis and all cervical specimens that were positive for C trachomatis were positive on self-test. Compared with clinician-collected introital specimens, self-collected specimens had a sensitivity of 100% and a specificity of 98.5%. Compared with cervical specimens, self-collected specimens had a sensitivity of 100% and a specificity of 93.4%. After discrepant analysis, self-test had a sensitivity of 100% and a specificity of 94.6% for detection of C trachomatis cervical infection. Conclusion: Self-collection of introital samples is both sensitive and specific in screening for C trachomatis.