This study is an evaluation of the relation between farm land values and the returns to farm land. The primary motivation for this study has been to explain the observed relation between farm land values and returns which has in some years been contradictory to the suggested theoretical relation between land values and returns. The primary objective of this study has been to examine the importance of growth of the returns to farm land in explaining the historical levels of farm land and its relationship with the returns to farm land. Data were collected on the values of cash rented farms and estimates of gross cash rents paid on those for 24 states for the period 1929-1980. Twelve states were selected for analysis. The 12 states were Ohio, Indiana, Illinois, Iowa, Missouri, Pennsylvania, Michigan, South Dakota, Minnesota, Wisconsin, North Dakota, and Oklahoma;The theoretical land value model was developed from the present value approach for determining values. The explanatory variables identified from this approach are the returns to farm land, the capitalization rate, the growth rate of the returns to farm land, the inflation rate, inflation rate, and taxes. Because of lack of data, taxes are not included in the empirical models;The empirical models to determine the association between the explanatory variables and farm land values were estimated by regression analysis. Two basic sets of regression were estimated; one using farm land value per acre as the dependent variable and one using the ratio of gross cash rent to farm land value as the dependent variable. The results of the empirical models indicate that gross cash rents are strongly related to farm land values. In general, the growth rate of gross cash rents did not perform as expected. The coefficient on nominal and real growth rates was significantly different from zero but did not have the correct sign. The coefficients on the lagged growth rates were found to be not significantly different from zero. The general rise in the price level did not appear to have a significant impact on the nominal or real value of farm land. The capitalization factor was found to be statistically significant in estimating the rent-to-value ratio;Granger causality tests were also applied to the data. Overall, the results did not support the hypothesis that gross cash rents "caused," in the Granger sense, farm land values.