This study examines aptitude treatment effects in an inquiry/learning cycle based physical science class for elementary education majors. The aptitude was formal reasoning ability and the students were arranged into three groups: high, middle, and low ability reasoners. The treatment was method of forming groups to work in the laboratory. Students in each of three classes were grouped according to reasoning ability. In one class the laboratory groups were homogeneous, i.e., students of similar reasoning ability were grouped together. In the second class the students were grouped heterogeneously, i.e., students of different reasoning ability were grouped together. In the third class, the student choice pattern, the students chose their own partners. The findings were that there were no aptitude treatment interaction for achievement or for gain in formal reasoning ability, that grouping students of similar cognitive ability together for laboratory work in the class was more effective in terms of science achievement than grouping students of differing cognitive ability together or than allowing students to choose their own partners, and that students at different levels of reasoning ability experienced differential gains in that ability over the semester.