A study is designed to assess the relative effectiveness of two types of teacher supervision--that by a colleague-supervisor team and that by a single supervisor--both using videotape in a series of microteaching sequences. In each of four teach-reteach sequences, a specific teaching behavior is selected for emphasis in the supervisory conference. The first two sequences are reported here: orienting students to the learning task and reinforcing student responses. Subjects were interns in a master of arts in teaching program. Three teams of two interns and the supervisor comprised the experimental group; five interns, each working individually with the supervisor, the control group. The same five-step conference format was followed by both groups: orientation to the conference, provision of feedback, selection of a focus, provision of discrimination training, and provision of closure. Performance criteria were developed and used to code microteaching videotapes to obtain change scores between teach and reteach sessions. The hypothesis was tested using the Mann Whitney U statistical test. Colleague-supervised interns performed a significantly greater number of specific teaching behaviors in the first sequence. In the second they performed significantly better on three of eight verbal and three of ten nonverbal reinforcing behaviors. (The supervisory sequence is described, and performance criteria and teaching behaviors for pre-instructional set and for student reinforcement are appended.) [Not available in hardcopy due to marginal legibility of the original document.] (JS)