A 3-month pilot project was undertaken at the University of Texas to gain experience in administering the Cynthia Buchanan Language Program (Buchanan, 1967) and to test its effectiveness in making meaningful changes in the language development of disadvantaged Mexican-American preschoolers. A group of 114 Mexican-American children were chosen as experimental subjects who would receive instruction from the Buchanan Program, while another group of 101 subjects served as the control. It was hypothesized that while both groups would make significant gains in language development, the rate of gain of the experimental group would be significantly greater than that of the control group. Both groups were pre- and posttested with the Metropolitan Readingess Test, the Murphy-Durrell Reading Readiness Analysis, the Gates Reading Readiness Test, and the Lee-Clark Reading Readiness Test. The results solidly supported the hypothesis. Next, an analysis of covariance was run on the data to determine whether or not the results were generalizable to all levels of beginning scores. The results of the analysis indicated that they were not generalizable. (MH)