A 3-year Title III project in Union, New Jersey involved screening of 869 incoming kindergarten children in 1967 and 840 children in 1968, subsequent identification of children's perceptual difficulties, and remediation. The children were screened in areas of perceptual-motor match, auditory dynamics, associative processes, and gross-motor coordination. Data on socioeconomic background, birth, and developmental history were collected through parent interviews and a questionnaire. Children who received scores in the lowest 5% of one or more developmental areas or whose total score was in the lowest 10% bracket were selected for intensive training from the 1967 class. The same selection procedure was used for the 1968 class; through children in the lowest 20% level were included. First year children received training through diagnostic teaching for 1/2 hour of perceptual activities four days per week. Also, the perception teachers worked with each kindergarten class 1 day per week and were observed by kindergarten teachers. During the second year only experimental groups were given intensive training. Four research studies were conducted during each year. First year project results indicated that a significant mean difference existed between pretest and posttest results for the 172 experimental children receiving intensive training as compared with 500 children receiving an enrichment program only. Comparison of pretest and posttest results for both the 1967 and 1968 classes indicated that the intensive training was most effective in areas of visual-motor integration, and in certain aspects of gross-motor development. (MC)