were offered for recruiting women farmers, the use of leadership training so that women might learn to express their problems, and techniques to increase participation in credit and soil conservation programs. It was noted that there are a variety of household types and that women both as wives and as heads of households need to be targeted.
Prior to the circular, some male extensionists included women farmers in their programs, but in general, the inclusion of women was neither consistent nor reflective of women's contributions and needs in an area. Most male extensionists believed that women farmers were to be contacted by female extension workers and that rural women only should study home economics. The Circular legitimized and mandated that male extensionists work with women farmers in the smallholder sector.
These techniques might not work in every corner of the world, but the general argument that says that men cannot work with women farmers needs to be reexamined. New methods and techniques have to be devised that are feasible and that consider cultural traditions. In addition, female extensionists must not be left out. The number being trained in agricultural subjects must be increased. In the case here, female extensionists subsequently were taught how to cultivate and innoculate soybeans. Incentives and promotions are also necessary for them. In addition, the curriculum of the home economics courses for rural women must be modified to include materials that are directly related to women's productive roles.
CONCLUSIONS
Those involved in FSR/E must be prepared to keep their eyes open at all stages of the work. In the prediagnostic stage they must consider primary and secondary sources that detail the sexual division of labor and the changing roles of various household members. Sometimes this type of information is available, other times it must be collected in the field. An example of a focus on women's roles in crop production was prepared for a commodity oriented project. The Bean/Cowpea Collaborative Research Support Program, funded by USAID, specifically developed resource guides for a number of African and Latin American countries. These guides detail women's roles in agriculture and the ways that breeders and other production scientists can make use of this information (Fergusen and Home, 1985). Where this type of information is not readily available, researchers may have to disaggregate agricultural data sets. in order to ascertain gender differences or to collect their own data from local women and men farmers.
In the practice of FSR/E, researchers must confront extension workers with the need to include in their surveys a diversity of farmers in terms of resources, household and family types, and to consider people at different points of their life cycle in the diagnostic stage. Strict instructions need to be given to extension personnel for them to include in their surveys: a) low as well as high resource farmers, b) women farmers with both low and high resources, and c) women as both household heads and wives. The sexual division of labor and differential management strategies will have to be discerned. If questions on the allocation of labor and resources, problems and needs, remuneration, and investments cannot be answered for different categories of farmers, the work is incomplete. If only men provide the answers about women, the data are most likely biased.