Effects of spaced, repeated retrieval practice and test-potentiated learning on mathematical knowledge and reasoning in an authentic educational setting are investigated in the study. Research participants were a cohort of second-year pre-service mathematics teachers. A revised taxonomy table was utilized to measure the knowledge and reasoning proficiency of participants. Findings indicate that the intervention was effective in enhancing categories, where familiar algorithmic reasoning and procedural knowledge were required. It was less effective with categories requiring flexible and creative use of conceptual and procedural knowledge. Theories of storage and retrieval strength and conceptual knowledge are used to explain the findings.