In a comparative study of screening tests for visual acuity in young children, two groups of children were presented with three different tests. In the younger group (three to five years) three single opto-type tests were compared: Sheridan's five and seven letter matching test, the Fooks test and the E card test. The Sheridan test gave the best results and the E card test was found to be unsuitable for this age-group. The Fooks is an attractive test but was less sensitive in the detection of defects than the Sheridan test. In the older group (five to seven years) a single opto-type test, the Sheridan-Gardiner test, was compared with the E chart and the Snellen chart. The Sheridan-Gardiner test was found to have limitations in the detection of defects, including amblyopia, therefore results obtained by this method should not be considered to be directly comparable with those of the Snellen chart. The E chart gave good results in the detection of defects and there was less lateral confusion than had been expected. It has disadvantages, but there is still a place for its use in the screening of normal school-age children. The Snellen chart was the most effective test in the detection of defects, but not all the children were able to co-operate in its use. It is concluded that where it is possible to use the Snellen chart it should always be the method of choice.