Studied was the interaction between curriculum treatment and selected aptitude variables as predictors of learning and transfer scores. Visualization ability was measured by a paper folding test; general reasoning ability was measured by the Necessary Arithmetic Operations Test. Geometry students (191) were assigned to treatment groups for two class sessions of self-instruction using programed materials. One treatment emphasized graphs; the other (analytical) emphasized the multiplicative properties of signed numbers. Following treatment, the students were given a ten-item learning test and an eight-item transfer test. The statistical results for interaction of the two abilities with learning was inconclusive. A significant interaction with transfer was considered inconclusive because of low statistical reliability for the transfer test. (RS)