Background and Purpose—Epidemiological studies suggest an association between stroke incidence and stroke mortality and long-term exposure to particulate matter (PM) air pollution. However, the magnitude of the association is still unclear. Methods—We searched the Pubmed citation database for epidemiological studies and reviews on stroke and PM exposure. Then, we carried out a meta-analysis to quantify the pooled association between stroke incidence and mortality and long-term exposure to PM. Meta-analyses were performed for stroke events and stroke mortality and for PM10 and PM2.5 separately and jointly. Results—We identified 20 studies, including a total of >10 million people, on long-term PM exposure and stroke event or stroke mortality. For exposure to PM10 (including estimated exposure to PM10 from studies using PM2.5), the pooled hazard ratio for each 10-μg/m3 increment in PM10 was 1.061 (95% confidence interval, 1.018–1.105) and 1.080 (0.992–1.177) for overall stroke events and stroke mortality, respectively. A stratified analysis by continent revealed that the association between stroke and long-term PM10 exposure was positive in North America (1.062 [1.015–1.110]) and Europe (1.057 [0.973–1.148]), but studies in Asia (1.010 [0.885–1.153]) showed a high degree of heterogeneity. Considering exposure to PM2.5 (Europe and North America combined), the hazard ratios for a 5-μg/m3 increment were 1.064 (1.021–1.109) and 1.125 (1.007–1.256) for stroke events and mortality, respectively. Conclusions—The scientific evidence of the past decade identifies long-term exposure to PM, and PM2.5 in particular, as a risk factor for stroke. However, we found some currently unexplained geographical variability in this association.