Different methods have been used for concentric (CONC) isokinetic testing of ankle dorsiflexion (DF) and plantar flexion (PF). However, little information is available on either the reliability of these protocols or eccentric (ECC) torque production. As well, previous studies utilized non-weight-bearing test positions. The purpose of this study was to develop a reliable method of testing CONC and ECC DF and PF in a weight-bearing position. One group of adults, including older men and women were tested on two occasions with a standing protocol. Another group of older and younger men and women were compared between the standing method and a supine protocol. For both positions the velocity was 30 degrees.s-1 for the CONC and ECC actions, and the range of motion was 10 degrees DF to 20 degrees PF. Reliability coefficients for peak torque (PT) were acceptable and ranged from 0.65 to 0.90. The two testing positions were significantly related (r = 0.84-0.91), but CONC PT in standing were greater than supine. Therefore, this method of testing isokinetic DF and PF in a standing position has acceptable reliability and produces results consistent with those acquired with the supine protocol. It provides a means of comparing groups and examining intervention strategies while weight-bearing.