Early epidemiological studies identified bovine spongiform encephalopathy as a feed-borne infection associated with infected meat-and-bone meal in animal feed. The infection may have derived from scrapie in sheep, a spontaneous genetic mutation in cattle, or a transmissible spongiform encephalopathy in another mammalian species. Experimental work on the risk of transmission has necessarily tried to identify risk materials and their infectivity levels, the nature and size of species barriers, the infectious dose, the route of infection, the strain of the agent and the genotype of the animals at risk. The identification of levels of infection in cattle tissues has aided the removal of risk materials from the human and animal food chains. Maternally associated transmission is unlikely to maintain an outbreak, but the offspring of clinical cases appear to be at greater risk when the rate of food-borne exposure is high. Studies of embryo transfer have not shown infection to be transmitted by this means. While the control measures appeared to be straightforward, compliance has at times proved difficult to enforce and quantify. This has necessitated more extensive prohibitions, aggressive enforcement and thorough auditing of compliance levels.