A timed test of swallowing capacity has been designed for use in patients with neurogenic dysphagia. Swallowing speed (ml/s) has been demonstrated to have high intra- and inter- rater and test- retest reliability, and to be essentially independent of flavour or temperature. "Guideline" normal values were established in individuals without a swallowing disorder: swallowing speed was less in females than males and declined in both groups with age. The validity of a swallowing speed less than 10 ml/s as an index of abnormal swallowing was tested by comparison with the complaint of abnormal swallowing in a group of 81 neurological patients. Swallowing speed had a sensitivity of 96% and specificity of 69%: some apparent false positive responses were found in patients with disordered swallowing, mainly due to multiple sclerosis. Using a standard questionnaire and examination a similar pattern of symptoms and signs were statistically associated with both the clinical complaint of abnormal swallowing and swallowing speed. It is concluded that swallowing speed is a reliable and valid index for assessing disordered swallowing in neurological patients and may be of value in monitoring response to therapy.