PURPOSE: To investigate feasibility and effectiveness of an individually-directed, group strength-training programme on knee muscle strength after stroke. METHOD: Ten volunteers (62 +/- 11 years, mean +/- SD), 6 - 12 months after first-ever unilateral stroke, walking independently with or without aids were recruited. Using an A1-B-A2 design, 3 sets of baseline measures were taken at 2 weekly intervals; volunteers then attended twice weekly sessions of low intensity progressive strengthening exercises and were assessed after each series of 8 sessions to a maximum of 24 sessions; post training, measures were repeated after 4 - 6 weeks. Measures included isometric and concentric knee extensor muscle strength and 10 m walking velocity. RESULTS: Strength of knee extensor muscles was improved after training (ANOVA, p < 0.05). On cessation of training, isometric strength increased by 58 +/- 19% and concentric strength at 30 degrees /s by 51 +/- 14%; walking velocity quickened from 0.47 +/- 0.06 m x s-1 to 0.57 +/- 0.08 m x s-1 (t = -3.31, p < 0.01). These gains were maintained 4 - 6 weeks after completion of training. CONCLUSIONS: These findings support the use of low intensity strength training after stroke and confirm published evidence. It was feasible for one therapist to deliver the training programmes for 4 - 6 participants at a time; an important feature when resources are limited.