Biological control has been successfully adopted in pest control programs to minimize the use of pesticides and reduce environmental and human health risks.

Ten percent of the world's cropped area involves classical biological control. The three major approaches to biological control are importation, augmentation and conservation of natural enemies (DeBach, 1964). Biological control through importation can be used in all cropping systems in developing and industrialized countries (Gurr and Wratten, 2000; van Lenteren, 2006) and has been applied most successfully against exotic invaders. Successful control is most often totally compatible with crop breeding (DeBach, 1964; Thomas and Waage, 1996), and provides economic returns to African farmers of the same magnitude as breeding programs (Raitzer, 2003; Neuenschwander, 2004). In augmentation forms of biological control, natural enemies (predators, parasitoids and pathogens) are mass produced and then released in the field, e.g., the parasitic wasp Trichogramma is used on more than 15 million ha of agricultural crops and forests in many countries (Li, 1994; van Lenteren and Bueno, 2003), as well as in protected cropping (Parrella et al., 1999; van Lenteren, 2000). A wide range of microbial insect pathogens are now in production and in use in OECD and developing countries (Moscardi, 1999; Copping, 2004). For example, the fungus Metarhizium anisopliae var acridum "Green Muscle"® is used to control Desert Locust (Schistocerca gregaria) in Africa (Lomer et al., 2001). Since agents vary in advantages and disadvantages, they must be carefully selected for compatibility with different cropping systems. However, agents are playing an increasing role in IPM (Copping and Menn, 2000). In conservation biological control, the effectiveness of natural enemies is increased through cultural practices (DeBach and Rosen, 1991; Landis et al., 2000) that enhance the efficiency of the exotic or indigenous natural enemies (predators, parasitoids, pathogens).

The economic benefits of biological control can be substantial.

Cultures of the predatory mite, Metaseiulus occidentalis, used in California almond orchards saved growers $59 to $109 ha-1 yr -1 in reduced pesticide use and yield loss (Hoy, 1992). The fight against the cassava mealy bug in Africa has had even greater economic benefits (Neuenschwander, 2004). IITA and CIAT found a natural enemy of the mealy bug in Brazil in the area of origin of the cassava crop. Subsequently, dissemination of this natural enemy in Africa saved million of tonnes of cassava per year and brought total benefits of US$ billions (Zeddies et al., 2001; Raitzer, 2003). Similar benefits for small-scale farmers have accrued from other programs on different crops and against different invaders across Africa (Neuenschwander, 2004).

Weed competition is a significant barrier to yield and profitability in most agroecosystems.

In many developing countries, hand weeding remains the prevailing practice for weed control. On small-scale farms, more than 50% of preharvest labor is is devoted to weed management, including land preparation and in-crop weed control (Ellis-Jones et al., 1993; Akobundu, 1996). Despite these labor investments crop losses to weed competition are nearly universally identified as major production constraints, typically causing yield reductions of 25% in small-scale agriculture (Parker and Fryer, 1975). Delayed weeding is a common problem caused by labor shortages, and reduced labor productivity resulting from diseases such malaria and HIV/ AIDS. Hence, cost-effective low-labor control methods have become increasingly important. In Bangladesh with current methods, one-third of the farmers lose at least 0.5 tonne ha-1 grain to weeds in each of the three lowland rice seasons (Ahmed et al., 2001; Mazid et al., 2001). Even in areas that employ herbicides, yield losses are substantial; in the early 1990s annual losses of US$4 billion were caused by weed competition in the US. For staple cereal and legume crops like maize, sorghum, pearl millet, upland rice in semiarid areas of Africa, the parasitic witchweeds (Striga species) can cause yield losses ranging from 15 to 100% (Boukar et al., 2004). Striga infestation is associated with continuous cultivation and limited returns of plant nutrients to the soil, i.e., conditions typical of small-scale resource poor farms (Riches et al., 2005).

Intensive herbicide use has contributed to improved weed management but there are concerns about sustainable use and environmental quality..

Globally, approximately 1 billion kg of herbicide active ingredients are applied annually in agricultural systems (Aspelin and Grube, 1999). The benefits of judicious herbicide use are broadly recognized. In addition to tillage, prophylactic application of herbicide is the method of choice for managing weeds in industrialized countries and is also widely employed in highly productive agricultural regions in developing countries like Punjab and Haryana States in India. Herbicide use is also becoming more common in small-scale rice/wheat systems in Eastern India and in rice in countries such as Vietnam and Bangladesh where the price of labor is rising faster than crop values (Auld and Menz, 1997; Riches et al., 2005). Substitution of labor by herbicides in Bangladesh reduces weeding costs by 40-50% (Ahmed et al., 2001). Herbicides sold in small quantities are accessible to poor farmers who realize their value; rice herbicide sales have been increasing at 40-50% per year since 2002 (Riches et al., 2005). However, herbicide resistance (currently documented in 313 weed biotypes: www.Weed- Science.org) and environmental contamination are growing problems. Traces of Atrazine and other potential carcinogens are routinely documented in ground and surface water resources in industrialized countries (USGS, 1999), and on