A survey asking teachers to rate the suitability of their laboratory for instruction in independent study, small group instruction, and large group instruction modes was sent to random samples of Michigan Class B and C high schools identified from a preliminary study as having biology laboratories with either a split lecture-laboratory design, a perimeter tables design, a central-fixed or a central-movable tables design. A repeated measures covariance design, with recency of laboratory construction as the covariate, indicated that the split design was perceived as more effective for each instructional mode than were the other three designs. The most frequent recommendations for increasing the effectiveness of the designs included more functional tables, ventilation systems, student stations, and room darkening facilities; more space for group and individual activities; more service outlets; and increased storage space. (Author/AL)