Increasing cotton oil production in developing countries : prospects from new technical practices in cotton growing

Although ranking among the major food oils in the world, cotton oil is basically a byproduct of cotton fiber production. Change in cotton oil production depends mainly upon change in seedcotton production, even if better oil processing from seedcotton would have positive impact as well. However, area devoted to seedcotton production in the world, including in developing countries to some extent, has been stagnating for decades while yield has also come to stagnate since mid-1980s. Increase in cotton oil production would derive mainly from seedcotton yield improvement which might, in some cases, induce positive change in cotton area. The issue of modifying the current seedcotton yield trends demands to clarify reasons of yield stagnation being observed. Reasons differ among countries. In economically developed countries, yield is stagnating at a somewhat high level, existing high performing technical packages are fully applied and the challenge being set is to carry out new packages to enhance the yield potential. In developing countries where smallholder production dominates, reasons of yield stagnation at a low level are different so that new research outputs adapted to developed countries would not automatically be relevant in less favored countries. It is now fully recognized that many technical practices being recommended to smallholders in developing countries are not relevant as regard these farmers' liquidity constraint and risk adversion while they are submitted to climatic hazards in addition to impediments in input and credit provision. Research work must then begin with taking into account these contraints in order to carry out new technical packages less demanding in cash expenses, more cost effective, providing better protection against climatic hazards and cotton pests. As in many developing countries, cotton is grown under rain fed conditions, tolerance to occasional drought is of major importance. CIRAD has contributed to accertain that the above research objectives are not unrealistic. CIRAD has implemented the approach called direct sowing over vegetative cover during the last decade, either in commercial farming or smallholding farming. This paper states the principles of the approach, exposes the results obtained in Brazil and Madagascar and suggests the assumed phenomena that reverse a negative trend of soil degradation leading to yield unstability and profitability decrease into a positive process of soil improvement and higher input efficiency to ensure better profitability Finally, relevance e of experimenting the approach in Mozambique is discussed.