This paper studies theoretically and empirically the geographic transmission of trade shocks over the territory of a country. Increases in labor demand in a location raise local wages and draw workers away from employment in neighboring locations: those locations experience a reduction in labor supply and an increase in prevailing wages even if not initially affected, or not engaged in the production of tradeable goods; adjustment in their wages affect in turn other close-by locations. In addition, increases in prevailing wages in a location affect all the industries producing there: other locations active in the same industries gain then market shares and experience an increase in labor demand even when they are far apart. I develop a model capable of incorporating realistic geographic features and isolate theoretically the different components of this diffusion. The model is general enough to also allow the study of transmission of localized immigration and productivity shocks. I estimate its main components with data on US commuting patterns and sectoral employment. I illustrate the impact of reductions in trade frictions in a sector on locations active and inactive in it, and the consequences of productivity growth on nominal wages of workers vs. real wages of residents. The model delivers insights on the consequences of ignoring commuting ?flows in analyses of local labor markets.