Farmers’ Bounty: Locating Crop Diversity in the Contemporary World by Stephen B. Brush Yale University Press: 2004. 320 pp. $37.50, £25 Stuart Pimm Biodiversity’s three-part definition — the variety of genes, species and ecosystems — is best appreciated on a beach, with tropical forest in front of you, coral reefs behind and mountains in the distance. On rainy midwinter days, in a city far away, there is always the local grocery. Upmarket ones, the habitat of yuppies, make the point; so do rural markets in poor countries. The former sell coffee from 20 countries, plus broccoli, kale, kohlrabi, cabbage and Brussels sprouts, which are all variants of Brassica oleracea. Rural markets in Central America will have variously coloured maize kernels of assorted sizes and types. In Andean valleys you may find a hundred kinds of potato. This is the biodiversity that Farmers’ Bounty celebrates. It provides insights into ques- tions of distribution, value and survival that apply to biodiversity as a whole. Like species diversity, genetic diversity is concentrated in a few regions. These are known as Vavilov centres after a Soviet plant breeder of the 1920s and 1930s who was executed by Stalin for holding inconvenient scientific views. In Farmers’ Bounty, Stephen Brush wanders bravely into theoretical ecol- ogy to find explanations for how this diver- sity persists. Agricultural experiences are not widely appreciated by theoreticians, yet they offer important insights. Ecologists debate the generality of the idea that heterogeneous environments promote diversity. Mean- while, across the world, farmers discuss the advantages of this or that variety in different environments and years as seeds are sorted, and those retained are bought, sold, ex- changed and mixed for next season’s harvest. In contrast, uniformity was the norm for potato-growers in Ireland in 1845 and US maize-growers in 1972, and the penalty was devastating. Maize-growers in modern Chiapas in Mexico must contend with risks from summer droughts, strong winds, vari- ous soil types and the uncertainties of labour for weeding or applying fertilizer. Growers understand the complexities of the relative advantages in taste, yield and storage of traditional varieties, improved ones and creolized varieties that are the result of decades-long mixing of the two, and decide the proportions of these that they will plant. Simply, crop diversity is the product of a global annual ecological experiment. It is unregulated and informal, changes con- stantly as varieties come and go, and is laden with tradition, language and myth. The participants’ lives depend on their correctly interpreting the experiment’s results. Modern varieties, with their higher yields, better ability to use fertilizer and resistance to specific diseases or stresses such as drought, have been the major challenge to traditional diversity in the past half-century. ‘New’ potatoes are now grown in nearly every Andean valley. Governments encour- age them, as does the grower’s need for a larger surplus. Potato diversity has declined, as might be expected, but traditional pota- toes persist, particularly at higher altitudes. They taste better, especially when boiled or baked, and make better gifts, facts not lost on the clientele at my local grocery. Still, the erosion of variety can be sub- stantial. The Chilean island of Chiloé had some 200 varieties of potatoes in the late 1920s, half that a decade later, and fewer than 40 by 1970. Brush accepts the general trend and the problems it raises, but his frustration over the lack of global documentation is well taken. The putative features of genetic ero- sion are that indigenous varieties have limited geographical distributions, they change little from year to year, and their variety decreases as modern varieties, and fertilizer and pesticide use, increase. Studies certainly suggest these outcomes,but often equivocally. Environment, economy and culture often conspire to conserve diversity for reasons that need broader synthesis and assessment. The familiar solution to retaining diver- sity is the use of seed banks, such as the books and arts NATURE | VOL 430 | 26 AUGUST 2004 | www.nature.com/nature 967 International Rice Research Institute (IRRI) in the Philippines.Brush’s view of such collec- tions is that they are necessary but far from sufficient. The heroine botanist of Jurassic Park enthuses over a tree not seen for 65 million years, then gasps as she notices the sauropod eating it. The film imagines extinct species grown from their DNA, but is silent on how to teach dinosaurs which trees to eat. Similarly, seed banks store DNA but lose the complex relationships between the variety, its symbionts and pathogens, and the traditions of the farmers who knew how to plant, tend, harvest and store it. In situ conservation is essential; so is the selection that goes with it. Women in Rwanda select bean varieties, and in Nepal rice and chick- pea varieties, that perform better than con- ventional crop breeders can manage. If diversity has value, who owns it? In the book’s best chapter, Brush discusses the increasing polarity in answers. Fears of ‘biopiracy’ are widespread in many tropical countries. Stories of how the rich have robbed the poor — and continue to do so — influence research-permit applications even for those of us who travel with only note- books. The Convention on Biological Diver- sity has a strong theme that plant breeders’ rights trump poor farmers’ rights. Yet it is often the poorer country that benefits from exchanges — Vietnam gets almost all of its rice from lines developed from IRRI, com- pared with just a sixth for the United States. Some countries favour outright protec- tion, such as Ethiopia’s ban on the export of coffee plants. Major coffee producers such as Colombia and Costa Rica have a narrow genetic base to their crops; diseases could Growing biodiversity Your local grocery store shows why variety matters. Rich pickings: markets in Kenya have a wide range of potatoes as farmers strive to improve crops. D. GULIN/CORBIS ©2004 Nature Publishing Group