A random controlled trial of the relaxing effects of an aromatherapy massage on disordered behaviour in dementia was conducted. Twenty-one patients were randomly allocated into one of three conditions, aromatherapy and massage (AM), conversation and aromatherapy (CA) and massage only (M). AM showed the greatest reduction in the frequency of excessive motor behaviour of all three conditions. This reached statistical significance between the hours of three and four pm (p < 0.05). Post hoc analysis suggested that at this time of day the AM consistently reduced motor behaviour when compared with CA (p = 0.05). This provides preliminary evidence of a measurable sedative effect of aromatherapy massage on dementia within a robust scientific paradigm. Further research is recommended with an expanded sample size. Copyright © 2001 John Wiley & Sons, Ltd.