The purpose of this study was to compare the conventional lecture-discussion method and an illustrated programed textbook method when teaching a unit of instruction on the basic concepts of metallurgy. The control group used a portion of a conventional textbook accompanied by lecture, chalkboard illustration, and class discussion. The experimental group received their instruction solely from four illustrated programed textbooks. The data were obtained from 124 college students enrolled in lower division metals classes in the Industrial Arts and Technology Department at Fresno State College in California. Homogeneity of the classes used was determined by a standardized criterion test. The classes were then divided into two groups by arbitrarily assigning a method of instruction to each class. The null hypothesis that there were no significant differences between the two teaching methods was not accepted on the basis of the analysis of the pretest and posttest scores. However, the differences between the groups became negligible 5 weeks after the conclusion of the experiment. Conclusions indicate that programed instruction is as effective as conventional teaching techniques and schools should make use of programed instruction. In addition, programed instruction can provide solutions to problems of teaching groups with widely varying ability groups. (Author/GB)