This paper examines state agricultural total factor productivity (TFP) data, 1950-1982, for evidence of convergence, i.e., TFP growth rates of the future are inversely related to the TFP level at the starting data. After finding evidence of convergence, the paper examines the contributions of public and private R&D to convergence and presents implications for a more efficient organization of public agricultural research. For example, we find that increasing a states own investment in public agricultural research reduces the rate of TFP convergence but larger public investments in surrounding areas that potentially spillin increase the rate of convergence. Also, the results imply that the average rate of convergence in our best fitting model is about 10 percent per year. The finding of strong positive interstate spillover effects of public agricultural research suggests incentives should be considered for stronger cooperation across states on agricultural research funding and new political jurisdictions for financing public agricultural research.