The Schedule for Oral Motor Assessment (SOMA) was developed for the purpose of objectively rating the oral-motor skills of preverbal children, with a view to identifying areas of deficient abilities that could have clinical significance. The instrument can be administered without special equipment, by a trained observer. Oral-motor function is assessed across a range of food textures and fluids. Ratings of oral-motor skills are largely made post hoc by analysis of a videorecording of the test administration. The test-retest and interrater reliability of the instrument have been shown to be excellent. Criterion validity was investigated by means of a novel 'seeded cluster analysis' procedure in which 127 young children were assessed, most of whom were between 8 and 24 months of age. Ten percent of the sample had known abnormal oral-motor function in association with cerebral palsy (ages between 12 and 42 months). Not only was criterion validity satisfactorily established by the analysis but an abbreviated version of the SOMA--suitable for screening purposes--was developed. This has been shown to have a positive predictive validity greater than 90% and sensitivity greater than 85% for the detection of infants with clinically significant oral-motor dysfunction.