Female peasants' participation in migration can best be understood, it is suggested, through an analysis of the role played by gender ideologies and practice in peasant household livelihood strategies.1 As gender acts as a major organisational principle in peasant household maintenance, it assigns labour and responsibilities over the agricultural cycle and the household developmental cycle. Individuals defined largely in terms of their gender roles, take on tasks and duties over their lifetimes, which are governed by a general cultural pattern. When this pattern is adhered to, it structures migration for men and women. In a minority of cases however, migration is a function of strategies, whether individual or household, which stretch these patterns and challenge traditional female roles. Female migration in the Andes, as in Latin America generally, is a neglected field for research. Women in Latin America make up the majority of rural-urban migrants yet the full implications for this for the peasant economy has not been explored. Previous research on female migration has tended to focus on the migrant women's urban experience. For example, work on female migrants in a number of Latin American cities has described the incorporation of women into the urban economy (both in the informal and formal sectors) (Bunster and Chaney, 1984; Smith, 1971,1976; Young, 1978; Arizpe, 1980; Lowder, 1973); their demographic characteristics (Bunster and Chaney, 1984; Harkness, 1973; Smith, 1970; Macisco, 1972); and their social mobility (Harkness, 1973; Smith, 1973; Whiteford, 1978). Such an urban focus is misleading for a number of reasons. Firstly, the movement of women is seen as the amorphous group of individuals, a group defined solely in terms of gender, whereas women's social roles are differentiated according to class, cultural identity, age and marital status. All of these factors influence migration and labour market participation. Secondly, past work has completely neglected the position of women in the rural areas, from which the women migrate. Beyond the studies of male migrant's selectivity (Browning and Feindt, 1969; Balan et al., 1973) and a few studies which mention in passing the women left in the rural areas as their husbands migrate (e.g. Bourque and Warren, 1981; Laite, 1985), no systematic work has been carried out on the rural context for female migration. Furthermore, the implicit assumption in these studies is of an individual