The experiment reported on in this paper investigated primarily the overall effectiveness in raising the language achievement of black children of the pattern practice approach and the use of overall contrasts between standard and nonstandard dialects within the pattern practice approach. The experiment was conducted in three third-grade and three kindergarten classes in an elementary school district in the San Francisco Bay area. Seventy-three black third graders and 68 black kindergartners were used as subjects. Although there is considerable variation in their speech patterns, their speech is by and large representative of lower- to middle-class black dialect. The six classes were randomly divided into groups for three different treatments. The treatment period lasted for five weeks, with 15 to 20 minutes a day devoted to intensive language training. In Treatment A an attempt was made first to establish the concept of a difference between standard and nonstandard speech. In Treatment B the materials and approach were the same as in A, but no attempt was made to introduce a concept of standard as opposed to nonstandard speech. Treatment C was a control activity in which the participating teachers were free to devise their own language arts activities. (Authors/JM)