Three blocks of classroom materials were developed, taught, and evaluated in ninth grade physical science classes of the San Antonio Independent School District. Concepts involving measurement, ratio, and mapping were taught by laboratory investigations to an experimental group of 27 classes (N = 527). Twenty-one control classes (N = 399) studied the unchanged physical science course. It was hypothesized that there was no difference in the degree to which the selected mathematics concepts were learned by the two groups, and an instrument was developed to test this hypothesis. Significant gains in knowledge of mathematical concepts were made by both experimental and control groups. Regular and low ability students in the experimental groups made gains significantly greater than those in the control groups. (BC)