We conducted a meta-analysis of 15 studies on maternal prenatal smoking and ASD risk in offspring. Using a random-effects model, we found no evidence of an association (summary OR 1.02, 95% CI 0.93-1.12). Stratifying by study design, birth year, type of healthcare system, and adjustment for socioeconomic status or psychiatric history did not alter the findings. There was evidence that ascertaining exposure at the time of birth produced a lower summary OR than when this information was gathered after birth. There was no evidence of publication bias. Non-differential exposure misclassification was shown to have the potential for negligible influence on the results. We found no evidence to support a measurable association between maternal prenatal smoking and ASD in offspring.