The incidence of vaginal birth after cesarean (VBAC) and characteristics of VBAC births are investigated using 1980-85 National Hospital Discharge Survey Data collected by the National Center for Health Statistics. Only 3.4 per cent of mothers with previous cesarean delivery had VBAC in their subsequent 1980 delivery; this increased to 6.6 per cent in 1985. Because VBAC is a relatively infrequent event, 1980-85 data were combined and indicate that in this period 4.9 per cent of mothers with previous cesarean had a vaginal birth in their subsequent delivery. Combined 1980-85 VBAC rates are under 10 per cent for all age, race, marital status, region, hospital size, hospital ownership, and expected source of payment groups. Between 1980 and 1985, over 1.4 million repeat cesareans were performed for mothers having a live birth. Evidence suggests that potentially over 500,000 of these repeat cesareans could have been VBACs (over and above the 74,000 VBACs which occurred). VBAC mothers' mean length of hospital stay is 3.2 days, which compares closely with 3.0 days for other vaginal deliveries, but both contrast sharply with 5.6 days for repeat cesareans and 6.0 days for primary cesareans. Except for the uterine scar from the previous cesarean, VBAC mothers appear to have about the same history and frequency of complications as mothers with other vaginal deliveries. If the 500,000 repeat cesareans had been VBACs, surgical fees and costs for 1.2 million days of hospital stay would have been averted over the 1980-85 period.