Use-values have been advocated as a tool to compare the value of not just individual species, but also of plant families and forest types to local people, e.g. to identify species or habitats in need of special management or conservation. We estimated use-values in three forest types (upper restinga, lower restinga, tahuampa) on the Amazon flood plain south of Iquitos (Peru), compared two methodologies, identified the most valuable species, and contrasted these valuations with the actual use of forest resources in local villages. A new method for estimating use-values was contrasted with the method of Phillips and Gentry (1993a). Despite philosophical and procedural differences, estimates were highly correlated (r2=0.86). We discuss limitations of both methods and suggest some possible enhancements. The need to discriminate between past, present and potential uses is emphasized.