Data was gathered in conjunction with another study regarding the relationships between classroom interaction and constructive student attitudes. Using Flanders' categories, observations were made in two sections (the highest 20 percent and lowest 20 percent) of a departmentalized sixth grade. Each of four teachers (science, math, social studies, language arts) were observed with each section during 10 complete class sessions, five during the first and five during the last month of the semester. Data was entered onto a 10 x 10 matrix and interpretive techniques suggested by Flanders and by Soar were tested against results of the pupil affect as measured by a questionnaire administered at the end of the semester. Significant differences between teachers were obtained on Flanders' revised i/d ratio, on Soar's flexibility count, on the 7-7 (extended teacher criticism) and the 7-9 (teacher criticism followed by student initiation) cell frequencies, and on a number of other analytic combinations. The i/d ratio did not correlate with pupil affect as would be predicted--the highest i/d classroom did not have the most favorable affect nor the lowest the poorest affect. The flexibility count was most likely affected by achievement level of classroom in its relation to the emotional climate. Low achievers preferred low flexible interactions; high achievers seemed less affected by teacher flexibility. Combined 7-7 and 7-9 counts seems to be correlated with affect. (JS)