This paper summarizes the early results of the two systemwide off-peak free-fare transit experiments being conducted in Trenton, New Jersey, and Denver, Colorado. These experiments, which are sponsored by the Urban Mass Transportation Administration (UMTA) under its Service and Methods Demonstration Program, are the first free-fare programs of such size and comprehensiveness. The demonstrations have already provided a number of interesting, if still tentative, conclusions. The first major conclusion is that, while free fare induces large and sustainable ridership gains (19 percent in Trenton and 34 percent in Denver), the general aggregate behavior of the population in making their modal choices is not significantly different from what it would be with any other absolute change of an equal amount. The price elasticity of demand for transit implied by the Trenton results was -0.42, which is virtually identical to the transit industry's experience. Saturday evening youth riders and walk trips made up fully 7 percent of the ridership in the free-fare system in Trenton. The demonstration appears to have reduced the peak-load capacity requirements in Trenton's transit system and caused a dramatic shift from the peak to the off-peak. The most surprising finding was that complaints of rowdiness, vandalism, and other incidents increased at both sites to such an extent that some groups called for the abandonment of the experiments.