The statistical basis underlying the use of several visual acuity charts is examined. A new chart which offers theoretical and practical advantages for visual acuity screening is described.