A STATISTICAL analysis of the correlation of x-ray cephalopelvimetry to obstetric outcome has just been completed. The records of 350 patients measured by the Ball method and delivered on the wards of the Sloane Hospital for Women were studied. The accuracy of the method was found to be high — between 98 and 100 per cent — when disproportion was absent (negative accuracy), but only fair — 90 per cent at the inlet and 73 per cent at the midpelvis — when disproportion was present (positive accuracy). The greatest fault was the large number of borderline x-ray diagnoses in which . . .