A study was conducted to assess the validity and reliability of R. Moss's "Classroom Analysis of Teachers' Theoretical Orientation to Reading" in order to provide an observation system specifically intended for assessing a teacher's reading instruction as a function of theoretical orientation. Five reading experts were chosen to individually teach reading lessons to a group of second and third graders. The subjects instructed the youngsters using reading approaches with three main orientations: grapho-phonics, skills, and whole-language. The lessons ranged in length from 12 to 28 minutes. The videotapes of these lessons were then individually presented to two classes of undergraduate education majors who used a slightly modified Moss observational system to rate the teachers' instructional behaviors. The original instrument was lengthened from four to six subcategories. The reliability analyses suggested that the instrument could yield stable estimates of theoretical orientation. In general the items on the validity assessment performed as expected. However, some items were correlated with factors other than those that were theoretically predicted. (HOD)