ABSTRACT.Thirty‐one children with asthma took part in a 10‐week period of training comprised of either swimming, a specially planned dry land exercise or a combination of both.There was no difference in the degree of bronchospasm induced by the different forms of exercise, nor was there a consistent change in the clinical pattern of asthma with either type of activity.No reduction in the bronchospastic response to exercise was noted after the 10 weeks of training but there was less tachycardia. An improvement in the mean maximum expiratory pressure occurred in both the dry land and swimming groups indicating improved chest wall function.It was concluded that provided asthma was adequately managed an appropriately planned dry land exercise programme resulted in similar advantages to swimming training without added risk of exercise induced ast