Because of the well-documented morbidity of radical hysterectomy, it would appear desirable to isolate a subset of patients at low risk of parametrial spread or disease recurrence who might thereby be candidates for less radical surgical therapy. To this end, the records of all patients undergoing radical hysterectomy for cervical carcinoma at our institution between 1956 and 1985 were reviewed. Of the 387 patients treated for squamous carcinoma clinically confined to the cervix, 83 (21.4%) had tumors with depth of invasion greater than 3 mm (stage IB) but volume of tumor less than or equal to that of a sphere 2 cm in diameter (4.19 cm3) and no tumor in angiolymphatic spaces. Of the 83 patients, none had parametrial nodal metastasis. Median follow-up of this subgroup was 9.8 years, and the Kaplan-Meier estimate of 5-year disease-free survival was 97.6% (95% confidence interval, 94.3-100%). We consider patients in the subset described above to be candidates for modified radical hysterectomy and pelvic lymphadenectomy, with a decrease in morbidity associated with the less radical dissection, a low risk of failure to excise occult parametrial tumor, and the expectation of an excellent chance of long-term disease-free survival.