The math curriculum of the Montessori system of education for children ages 3-12 is distinctive, incorporating multiple manipulatives and educational practices which have theoretical and empirical support in research. However, studies investigating the math achievement and learning of Montessori students and alumni have not consistently found Montessori programmes to be more effective than conventional or other programmes. Through a detailed review of such studies, we find that a Montessori advantage in math is more likely when programmes adhere to important principles of Montessori education, when students have had longer immersion in Montessori programmes, and when assessments are more conceptual in nature. We suggest that future research should take into account programme fidelity and enrolment duration, and outline other directions for future research.