The causes of the Protestant Reformation have long been debated. This paper attempts to revive and econometrically test the theory that the spread of the Reformation is linked to the spread of the printing press. The proposed causal pathway is that the printing press permitted the ideas of the Reformation to reach a broader audience. I test this hypothesis by analyzing data on the spread of the press and the Reformation at the city level. An econometric analysis which instruments for omitted variable bias suggests that within the Holy Roman Empire, cities within 10 miles of a printing press by 1500 were 57.4 percentage points more likely to be Protestant by 1600. These results are robust, though the effects are weaker, across Western Europe. The analysis also suggests that the early spread of press affected religious choice into the 19th century.