The purpose of the authors in this study was to determine the effects of reasoning, use of models during testing, and sex type on posttest achievement in chemical bonding under controlled instruction. Eighty-four high school students taking chemistry were randomly assigned within their classes to models and no models groups for the posttest. Reasoning capabilities were assessed by the Piagetian Logical Operations Test (PLOT) (Staver & Gabel, JRST, Vol. 16, No. 6, 1979), prior to instruction. All students then received the same instruction on chemical bonding which included teacher demonstrations of concepts with three-dimensional molecular models, interspersed teacher questions during the introduction and development of concepts, student manipulation of three-dimensional molecular models during laboratory experiments, and text reading assignments on concepts prior to their instruction in class. The posttest on molecular geometry and shape contained three sections requiring memory and application (Bloom, Taxonomy of educational objective, handbook I: Cognitive domain. New York: David McKay, 1956). Data were analyzed by regression (Nie et al., Statistical package for the social sciences, 2nd ed. New York: McGraw-Hill, 1975). Results indicate that reasoning accounted for a significant portion (p < 0.05) of the variance in the total score, memory score, and application score, but not on the synthesis score of the posttest. Use of models and sex type did not account for a significant (p > 0.05) portion of the variance on total scores or any section of posttest. The three-way interaction of reasoning, model usage, and sex type accounted for a significant portion (p < 0.05) of the variance in total scores, and in the memory and application sections of the posttest. Discussion focused on the results, conclusions, and implications for science teaching.