Agricultural research, as with many other governmental services, can be performed efficiently at the state level but produces benefits that accrue to a broader area than just the originating state or region. Results from basic research, for example, would be unrestricted by geographic boundaries. Even applied research which is designed to solve specific problems encountered in a particular state may result in spi1lovers--geographically external benefits--to other areas. For example, some research results can readily be applied over wide geographic areas while other results need only additional adaptive research before they are suitable for other areas. The idea that the benefits of agricultural research are not realized solely by the state or region providing the research expenditures is not a new one. Several researchers have analyzed the interregional diffusion of a particular technology (Peterson and Hayami, 1977, pp. 524-526). In the study of hybrid corn diffusion, Griliches (1957) found that differences among regions in adoption rates were dependent on such factors as the size and density of commodity production and profitability of the new technology. Despite the widespread concern over the diffusion of a particular technology, the external benefits of agricultural research have not received much attention from economists working in the general area of research evaluation and planning. Attempts at measuring the contribution of agricultural research to agricultural production have often utilized a production function for a commodity or agricultural sector as a whole in such a manner that research was included as a separate variable (Peterson and Hayami, 1977, pp. 520-521). The majority of studies which have included research as a separate variable in a production function have been aimed at the national level rather than the regional or state levels. Griliches' (1964) work was one of the first publications in the area and Evenson's (1967) work was really important because it revealed the nature of the lag between the research input and increased output. The production function approach provides an estimate of the marginal product of agricultural research which is particularly useful in guiding decisions about allocation of resources to agricultural research. Studies directed at the state or regional level confront a major problem not encountered in a national analysis: interregional spillovers of the benefits from agricultural research results. This problem has been termed pervasiveness, indicating the tendency for research results generated in one region to be incorporated into farm production functions in other regions (Evenson, 1971, p. 173). Latimer and Paar1berg (1965) and Evenson (1971) recognized the pervasiveness problem. Latimer and Paarlberg were unable to find a statistically significant relationship between research expenditures within the state and agricultural output. They attributed these findings to the pervasive nature of agricultural research results (Latimer and Paarlberg, p. 239). Evenson included a variable which measured the intensity of commodity research in an attempt to control for the pervasiveness of research (1971, p. 177). If research results were completely pervasive, Evenson argued, this variable would dominate the state research variable. The variable was statistically significant indicating that the interregional transfer of agricultural research results should be taken into account in cross-sectional analyses. The existence of spillover benefits has a bearing on the allocation of research funds both within and between states. One important problem is to determine the appropriate balance between federal and state government in financing agricultural research. More specifically, what portion of the research expenditures should be financed by the federal government? The federal government initially served as a catalyst in developing the institutional framework to conduct agricultural research. The Morrill Land Grant College Act of 1862 and the Hatch Agricultural Experiment Station Act of 1887 reflect the emergence of a dual federal-state approach to agricultural research (Peterson and Fitzharris, 1977, pp. 72-73). Under these acts, each state received funds for a college of agricultural and mechanical arts and for an agricultural experiment station. This institutional framework is still a dominant force in agricultural research. Federal funds are allocated by a formula which is based largely on a state's rural and farm population (Peterson and Hayami, 1977, p. 522). Assuming that this system of finance was appropriate when it was first devised, is it still equitable after almost a century? This paper deals with the effects of spillovers of agricultural research benefits on the financing of research by federal and state governments. It considers conceptual problems of financing government services which produce spillovers and proposes a model to align a region's investment in agricultural research with social benefits by compensating for spillovers with funds from the federal government. Interregional spillovers of the benefits from agricultural research results are empirically measured in order to determine the appropriate balance between federal and state funding of agricultural research.