This study focused on Robert Gagne's method for curricular development, which consists of structuring a knowledge domain into a learning hierarchy. Two methods of generating learning hierarchies and two different sequencings of these hierarchies were compared and their effects were measured. Four programmed texts were developed from two different hierarchical arrangements of the subject matter considered--mathematics and science. A Gagne hierarchy and a modal child hierarchy were used. A randomized version of both sets was also constructed. The texts were administered to fifth-grade students who had vailed a pre-test for knowledge of the two selected tasks. Five experimental conditions were established. Four of the groups used one of the hierarchy plans developed--the Gagne hierarchy; the randomized Gagne hierarchy; the child hierarchy; or the randomized child hierarchy. Group 5 received the dependent measures but were not taught the tasks. An immediate post-test and a delayed post-test were administered. Additional tasks developed by both students and teachers were used in determining relationships between methods and the generation of a learning hierarchy. Among the results were: significant differential effects of methods of hierarchical organization of learning material; lack of significant effect of sequencing of subordinate tasks; and lack of significant differential effect on learning the terminal objectives. (Author/EB)