The records of 42 women with Bell's palsy during pregnancy, and of 91 nonpregnant women, whose dats of onset of Bell's palsy and of the preceding menstrual cycle were precisely known, were studied for factors that might show relation between pregnancy or the menstrual cycle and Bell's palsy. Of the 42 cases in pregnancy, 31 occurred in the third trimester, five in the first two weeks postpartum, and six in the first two trimesters combined. Our calculated frequency of Bell's palsy in pregnant women is 45.1/100,000 births; for nonpregnant women of the same age group the calculated incidence is 17.4/100,000 per year. No causative relation was found between toxemia, hypertension or primigravidity, and Bell's palsy. Over 60% of the cases in nonpregnant women occurred in the first 14 days of the menstrual cycle with peaks on the first and seventh days and near ovulation. No clear evidence for an etiologic relationship was seen with edema or hormonal changes in either pregnancy or the menstrual cycle. A number of factors in pregnancy and the menstrual cycle suggested an etiologic role for herpes simplex virus reactivation in Bell's palsy. There was no evidence that prednisone treatment is contraindicated during pregnancy.