Sixty-seven Mexican-American children were administered a special 7-week Head Start language training program during the summer of 1967. Three basic treatments were used, and there was a control group. Two teachers were used, thus raising the number of groups to eight. The three basic treatments involved a structured English language training program; in one group, Spanish was the language of instruction; in a second group, English was the instructional language; and in the third, both languages were used. The control groups received the usual preschool art and music activities. Tests were administered at the beginning of the program, at the end, and the next spring. It was found that (1) since the groups were initially of varying ability, final differences in performance could have been due to this initial difference; (2) the teacher factor, sex factor, and age factor contributed nothing to the results; (3) the structured language treatments did not produce better scores than the control treatment; and (4) the bilingual treatment was not significantly superior to the Spanish or English treatment. (WD)