Objective: To investigate the associations between early life exposure to particulate matter with an aerodynamic diameter less than 2.5 μm (PM 2.5 ) and the risk of autism spectrum disorder (ASD) among school aged children. Methods: A total of 165 children with ASD and 165 age and gender matched typical development (TD) children were recruited. Children s basic information were obtained via questionnaires, and the severity of ASD symptoms was assessed with Social Responsiveness Scale (SRS). Early life PM 2.5 exposure (preconception, entire pregnancy, and the first two years after birth) were extracted from the Tracking Air Pollution in China (TAP) datasets. Conditional Logistic regression and generalized linear model were used to evaluate the associations of early life exposure to PM 2.5 with the risk and the ASD severity symptoms, respectively. Results: The PM 2.5 exposure of ASD group during preconception[(55.08±9.34)μg/m 3], entire pregnancy[(50.44±8.71)μg/m 3], the first year after birth [(45.04± 8.25 )μg/m 3] and the second year after birth [(40.19±7.12)μg/m 3] were significant higher than those in TD children [(47.66± 7.63 , 44.19±7.16, 38.95±6.07, 35.76±5.65)μg/m 3]( t =7.94, 7.13, 7.70, 6.32, P <0.05). After adjusting for potential confounding, each increase of 1 μg/m 3 in PM 2.5 was associated with higher risk of ASD during preconception ( OR=1.21, 95%CI =1.13-1.29), entire pregnancy( OR=1.18, 95%CI =1.11-1.26), the first year after birth ( OR=1.30, 95%CI =1.18-1.43) and the second year after birth ( OR=1.29, 95%CI =1.17-1.42). No similar results were observed regarding the analyses of SRS total and sub scale scores( P >0.05). Conclusion Early life exposure to PM 2.5 is relate to the risk of ASD, these findings indicated that more attention should be paid to ambient PM pollution in the early life prevention and control of ASD.