This study examines the dynamic relationship between spot and futures prices of agricultural commodities. We first briefly discuss what the non-arbitrage and asset pricing theory has to say about the relationship between spot and futures markets. Next, using recent price data for corn, wheat, and soybeans, we perform Granger causality tests to empirically uncover the direction of information flows between spot and futures prices. Linear as well as nonlinear (nonparametric) causality tests are conducted on both spot and futures returns and their volatility. The results indicate that spot prices are generally discovered in futures markets. In particular, we find that changes in futures prices lead changes in spot prices more often than the reverse. These findings also contribute to the debate on alternative instruments to address excessive volatility in grain markets. Our results support, for example, the viability of implementing a global virtual reserve, recently proposed by von Braun and Torero (2008, 2009), to prevent disproportionate spikes in grain spot prices through signals and, if necessary, market assessment in the exchange of futures.