Using a nonlinear autoregressive distributive lag framework, the paper examines the dynamic effects of gasoline price, gasoline price volatility, income, and transit service coverage and frequency on public transit ridership in six U.S. cities (New York City, Chicago, Los Angeles, Boston, San Francisco, and Cleveland). The results indicate that, in the long run, rising gasoline prices increase transit ridership in all cities. More importantly, transit riders react differently, depending on the direction of gasoline price movements. The variable corresponding to price increases is found to have a larger elasticity than that for price decreases in five out of six cases, supporting gasoline price asymmetry. In Chicago, Los Angeles, Boston, and Cleveland, a small but statistically significant long-run effect of gasoline price volatility on transit ridership was found, suggesting that gasoline price uncertainty is an important factor affecting transit ridership in these cities. In the short run, transit service coverage is found to be the key determinant of transit ridership, implying that expanding transit service coverage can boost public transit ridership in the short term.