Agricultural Geography as the Geography of Food P. J. Atkins, Durham University Area, Vol. 20, No. 3 (Sep., 1988), pp. 281-283 We should be grateful to Bowler and Ilbery (1987) for their timely reassessment of agricultural geography, which has recently seemed becalmed in a Sargasso Sea of production functions and tedious typologies. They rightly profess a progressive and more coherent view of the sub-discipline which emphasises the links between agriculture and the wider society and economy. Just how radical is their redefinition, however? Their revised agenda can be seen as state-of-the-artism rather than a true demolition and reconstruction job. It is hardly surprising that agriculture qua production should be receiving less attention in this country. With only about 2 per cent of the employed UK work force and the same contribution to GDP, it is a sector of declining relative significance. According to Burns (1979) over 50 per cent of the value of food on the supermarket shelves is added after it has passed through the farm gate, and clearly it would be irresponsible of geographers to neglect the downstream portions of processing and supply networks. I therefore strongly support the call by Bowler and Ilbery for a study of supply systems. I wish to go further, however, and argue for the metamorphosis of agricultural geography into a geography of food along the following lines. First, with certain notable exceptions, there has been an unfortunate neglect by agricultural geographers of the rural development process in poor countries. Agricultural geography has been dominated by concern for farmers and farming in the 'north ',with the rural ' south 'left to development specialists. In his recent textbook, for instance, Ilbery (1985) explicitly excludes the developing world from his project. This imbalance is difficult to justify. About 87 per cent of the world's farmers and agricultural workers live in countries with an annual per capita income of less than $1,600, where the ultra poor may spend up to 80 per cent of their disposable income on food. Here the production and consumption of food dominate people's lives and livelihoods. Traditional agricultural geography can play a valuable part in our comprehension of both the ecological setting and the socio-economic circumstances of production in less developed areas, but understanding the onward connections into regional and national food systems needs a broader vision. Recent work has increased our knowledge of the place of agricultural marketing in the process of underdevelopment through the investigation of exchange relations (Harriss, 1984), of famine by examining entitlements to food in crisis situations (Sen 1981), and of food problems and policies at various scales (Grigg 1985; Tarrant 1980), but the holistic reconstruction and exploration of food systems (production, processing, transport, marketing, consumption) has so far attracted less geographical attention. This is understandable perhaps in view of the size and complexity of the task. Secondly, a related and complementary theme is the role of food systems in our own historical development process. Anthropologists and historians (Fenton and Kisban 1986; Rotberg and Rabb 1985) seem to have staked out the only claims so far to this important field, but historical geographers should be aware of the crucial place of food in the evolving