The emergence and establishment of the world's agricultural systems can be seen in retrospect as a very gradual process in which, through the Holocene, humanity became more and more dependent for food on fewer and fewer species of crops and domestic animals – to the extent that now just three cereal crops, rice, wheat, and maize, provide most of the energy humans derive from plant foods. Much of this dramatic reduction in diet breadth occurred in the twentieth century, and it conceals a 12,000-year history of plant and animal domestications, dispersals, adoptions, and exchanges revealed by biological, archaeological, and ethnohistorical evidence. In recent years new genetic, bioarchaeological, and dating techniques have increasingly been used in investigations of the beginnings of agriculture; in particular there has been an upsurge of research on plant and animal domestication (Zeder et al. 2006). At the same time much new archaeological evidence has been acquired of early crops and domestic animals in regions previously largely disregarded in the search for agricultural origins, such as parts of South Asia, tropical Africa, eastern North America and lowland South America east of the Andes (e.g., Piperno and Pearsall 1998, Neumann 2003, Fuller 2006, Smith 2006a,b).

It is now clear that the processes by which systems of agricultural production eventually became established in almost all cultivable areas were more varied and complex than previously assumed, and there are still great gaps in our knowledge of the diverse trajectories that led to the emergence, development and worldwide spread of agriculture. One approach to a better understanding of those trajectories is to examine how many so-called hunter–gatherers enhanced their food supplies by intervening in the life cycles of the plants and animals on which they depended.