Bivariate time-series models and related topics of exogeneity and Granger orderings have received much attention from economists over the last decade. Examples in the macroeconomic literature include relationships between money and income (Sims 1972, Feige and Pearce), money and interest rates (Pierce), and wages and prices (Geweke 1975). In agricultural markets, research in this area includes the sorting out of leads and lags between farm, wholesale, and retail prices (Heien, Miller), acreage allotments and acreage supply response (Weaver), and alternative price quotes for eggs (Bessler and Schrader). The methods used in these studies are empirical because they allow the data to specify the dynamic relationship between two variables. Economic theory is used only to suggest the variables to be related. Often theory is ambiguous on the explicit lead-lag relationships between important variables and the variables themselves. In such cases, one can (or must) rely on strictly empirical methods. (Burks discusses the complementary relationship between empirical and theoretical methods in scientific investigations.) Agriculturalists interested in explaining producer behavior and forecasting livestock price and quantity movements have long grappled with problems of leads and lags. Attempts to explain short-run movements in live cattle and hog inventories or slaughter levels using price variables have at times yielded unsatisfactory results. For example, shortrun quantity response to price change was found to be negative for cattle (Nelson and Spreen; Myers, Havlicek, Henderson); however, a positive relation begins to appear with longer lengths of run. The theory of optimal investment under risk explains this type of relationship. That is, treating cattle as both investment goods and as final products will generate the negative short-run response of quantity to price. Such theory, however, will not be explicit about the particular lead-lag relationships to be encountered. Most modern theories of decision making under risk suggest that such lags are dependent upon subjective probabilities of individual economic agents. If one is unwilling to accept the existence of a rational probability (following deFinetti), then he will not be able to specify, a priori, the explicit leads and lags in a particular investment problem. One must look at the data and infer such relationships. This paper examines the lead-lag relationships between livestock prices and various causal variables suggested by theory. It explicitly tests the exogeneity of cattle on feed, cattle slaughter, and income on live cattle prices and the exogeneity of sow farrowings, hog slaughter, and income on live hog prices. Although explicit dynamic multivariate models are not formulated, the paper's results do provide background evidence useful for their ultimate construction (Granger and Newbold 1977, chap. 7).