Research on the effects of two modes of sentence combining instruction on writing skills was conducted from 22 September through 17 December 1976, at Evans School in Tempe, Arizona. Subjects were 40 students in two fifth grade classes designated the individualized class and the group class. The individualized class followed a sentence combining curriculum using a program of self-study constructed with the following components: (1) a brief reading for each lesson, establishing the context for the introduction of the sentence-combining exercises, (2) comprehension questions, which were used to provide a check of the adequacy of students' reading of the material, (3) sentence-combining exercises, and (4) sentence-decomposing exercises. The group class followed the same curriculum, but the sentence-combining exercises were presented in a slightly different form: (1) choral reading of a prose selection, (2) presentation of the concept for practice through the use of blackboard examples, (3) choral practice of sentence combining, and (4) writing of combined sentences from oral cues. The two teachers involved in the study rotated during the project. The results indicated that scores from writing samples, which included mean length of T-units and mean number of multiclause T-units compared to single-clause T-units, favored the group class. No significant differences existed between the two groups on reading scores obtained by a miscue analysis. (MKM)