A stereoscopic test was developed for the visual screening of preschool children. The TNO test for stereoscopic vision utilized the same principle as the Julesz random-dot stereogram, and provided a simple and unequivocal test criterion understood even by young children. The results of a comprative evaluation of the TNO and Titmus stereoscopic tests proved the TNO test to be the more reliable of the two, particularly in the 2- to 4-year-old age range. Failure to pass this test at the 240 seconds of arc disparity level yielded an excellent screening criterion, as attested by the results of a validation experiment employing 81 patients (2 to 7 years old) with known visual health records. Furthermore, the screening results obtained from 129 preschool children (2 to 5 years old) tested in the classroom by a nonprofessional examiner, suggest that, under these more realistic conditions, the TNO test yields at least 60% less overreferrals than the Titmus test.