Low fertility soils and in particular low phosphorus (P) availability are a primary constraint to food production in many developing regions including South and Southeast Asia, Latin America, and subSaharan Africa. Legumes are especially sensitive to low P availability since biological nitrogen fixation has a high P requirement. Improved legume productivity in low P soils would improve the availability of highly nutritious grains as well as improve system productivity through increased nitrogen fixation, as well as reduced soil erosion and other agroecosystem benefits.

In recent years significant progress has been achieved in genetic improvement of legumes for better adaptation to low soil P availability, or "P efficiency". Research at Penn State has identified traits conferring P efficiency in common bean (Phaseolus vulgaris), and in collaboration with CIAT has used these traits to identify landraces and bred lines with substantially better P efficiency than the best breeding lines available in the late 1980s. Phosphorus efficient genotypes are now being used in bean breeding programs through CIAT and Bean/Cowpea CRSP networks in Africa and Latin America. In soybean (Glycine max), a McKnight Foundation CCRP project at South China Agricultural University (SCAU) and Penn State has identified promising new sources of P efficiency and has validated that several traits important for P efficiency in common bean are also important in soybean. As a result of this project 8 new genotypes with substantially improved yield on the low P soils of South China are now in the final stages of commercial release. For both common bean and soybean, P-efficient genotypes show substantial yield gains in low P soil compared with conventional genotypes, in some trials doubling yield without additional inputs, thereby demonstrating the potential of this approach.

These genetic resources represent a remarkable opportunity for many developing regions, particularly Subsaharan Africa. Grain legumes with greater P efficiency would produce greater food at low soil fertility and would respond better to fertility inputs than traditional varieties. Legumes are particularly important for food security in many African communities, because of their value in human nutrition, their economic and cultural value, and their agronomic value in enhancing soil fertility through nitrogen fixation. Nutritionally, legumes are rich sources of protein, iron, zinc, and, when consumed as vegetables (e.g. snap beans) provide vitamin A, nutrients that are commonly deficient in many African communities. These nutrients are especially important for growing children, many of whom suffer from anemia and vitamin A deficiency. Recent research also indicates that consumption of grain legumes slows the onset of AIDS in HIV-positive people. Improved legume production would therefore directly address several critical health issues for African communities.

In most African communities, demand exceeds supply for grain legumes, which therefore have high value and are important for household income. Common bean also plays important cultural roles in African communities, many of whom identify themselves as consumers of particular colors and sizes of beans, in contrast to neighboring communities that may have different preferences. Improved legume production would improve rural diets and incomes, and would have beneficial effects on other components of African agroecosystems through enhanced N fixation, reduced soil erosion, and enhanced P cycling. The genetic improvement of legume P efficiency, deployed in consideration of the social and economic context of local cropping systems, has great potential to address a principal constraint to food security in Africa.

A) Mozambique

Mozambique is one of the poorest countries on earth. Of a population of 19 million, 70% are below the poverty line, and 33% are chronically food-insecure. After decades of colonial conflict and civil war, a peace accord was signed in 1992. Since then the country has held three democratic elections and the social situation is improving. Most of the population is engaged in low-input subsistence agriculture and is at risk from periodic drought, poor soil fertility and chronically low yields.

As in many subsistence agricultures, legumes play an important role in Mozambique by providing N inputs through biological nitrogen fixation, by providing a high-quality source of nutrients in otherwise deficient diets, and by providing a high-value crop for marketing locally and for export to South Africa. For example, in November 2004, the wholesale price for common bean in Mozambique was USD 436 per MT, whereas maize was USD 100 per MT. National bean yields average 300 kg/ha, which is only 60% of the world average yield and only 7% of yield potential. Low soil fertility and drought are the primary causes of low yields, as shown by the much higher yields (2.5 MT/ha or more) obtained in INIA trials with fertilizer. Dry beans are being promoted as an export crop by the government. In 2002 Mozambique exported 65,000 MT of beans to RSA, in 2003 this rose to 112,000 MT. Soybean is grown in certain areas for export to RSA, mainly by commercial farmers from Zimbabwe, but is not currently important for direct consumption. The government of Norway recently contracted with Mozambique to buy 50,000 MT of soybeans annually.

The importance that IIAM (the national agricultural research organization of Mozambique) places on legumes is shown by the fact that it has a separate Grain Legume program. The principal objective of the Grain Legume program is to develop varieties with better yield and adaptation to regional production environments, and since its inception in 1982, the program has released a number of varieties of peanut, cowpea, and common bean. Another objective of the program is to improve the availability of high quality seed. A large portion of program effort is devoted to genotype evaluation in regional and on-farm trails. Average common bean yields in these trials varied from 200 to 900 kg/ha, while soybean yields were generally larger, from 700 kg/ha to up to 5 MT/ha. The higher yield of soybean may result from several factors, including better root development and thereby better tolerance of low soil fertility and drought, longer growing season, and as a relatively new introduction to the region, freedom from host-specific biotic stresses. As is common in African NARS, there has been no focused effort on abiotic stress tolerance other than general adaptation in regional variety evaluations. Expertise in legume adaptation to infertile soil is limited to one MS level scientist, Magalhaes Miguel, project partner. Our project would therefore address an important yet hitherto neglected constraint.

As part of our project planning Dr. Lynch visited Mozambique in November 2004. Discussions with stakeholders in Mozambique affirmed our project strategy: 1) legumes and especially common beans are an important component of food security in the region, 2) legume yields are strongly limited by poor soil fertility, 3) very little effort is currently devoted to addressing this constraint because of lack of training, infrastructure, and operational support, and 4) institutional support, human resources, and collaborative possibilities exist in the region that would support project goals.

B) South China

China's recent pace of development is rapid but uneven. Currently 46.7% of the population of 1.3 billion is still under the international poverty line. In the Southern province of Guangdong, for example, more than 60% of the 70 million people live in rural areas with an annual per capita GDP less than $1000. As population continues to increase, food security in China will be a primary concern over the next several decades and this should not be neglected in international development.

Grain legumes play a key role in the food security of South China for their unique importance to human nutrition and agricultural sustainability. In South China, demand for legumes far exceeds regional production. In 2004, more than 20 million tons of soybean were imported to China (as contrast to the 15 million tons produced in China), of which approximately 70% were consumed in South China. Therefore, it is of great necessity to increase soybean production in this region.

Among the constraints to soybean production in South China, low fertility, particularly low phosphorus (P) availability, is the primary limiting factor. South China has a humid tropical and subtropical climate, in which warm, moist conditions result in weathered soils (Ultisols, Oxisols) with low nutrient availability. Phosphorus availability is particularly limited in weathered tropical soils because of the cumulative effects of intense leaching and weathering conditions, and also because free Fe and Al oxides bind native and applied P into forms unavailable to plants. Low P availability is especially problematic for legumes, since legume nodules responsible for N fixation have a high P requirement. The overwhelming majority of soils in South China are P deficient. Application of P fertilizer is only a partial solution since P fertilizers are imported and are expensive, and are marginally effective because of immobilization by the soil. Fertilizer costs are the single greatest production cost for farm households.

In the existing project, we have attempted to address this problem by developing improved soybean germplasm with superior root characteristics that enable better adaptation to low P conditions and more efficient utilization of applied fertilizer. After an integrated effort over four years, we are beginning to understand the genetic and physiological basis and agroecological dimensions of efficient P utilization in soybean and its production systems, while at the same time we have developed a number of P efficient soybean genotypes that will be soon released to farmers. These new genotypes show substantial yield gains in low P soil compared with conventional genotypes, indicating a great potential for future application. However, continued efforts are needed to realize the long-term objective of the project in developing new germplasm and technical innovations that may have great impacts on food security both in China and other developing nations, particularly in Africa.

In the current CCRP project, all activities are targeted in South China. In the next phase of the project, we propose to include Africa (specifically Mozambique) as the new target area while maintaining core activities in South China. Specifically, we will continue to breed soybean for P efficiency and other important agronomic traits with the existing materials that we have developed and will test the newly developed genotypes (varieties) under various soil/climatic conditions through National Field Trials. We will optimize the cropping systems (intercropping, rotation, etc) through large-scale experimentation and farmer participatory research. We will create effective and reliable legume (particularly soybean) processing facilities that can be used in households of the rural areas. These research outcomes will not only help to realize the great impacts in South China intended in the existing project but also serve as the technical reserve for generating impact in Africa through training and collaborative research (South-South Cooperation) proposed for the next phase of the project.