I. Introduction
In recent years there has been a tremendous increase in the number of
research and development projects involved in agroforestry. This is due to
the recognition that agroforestry has the potential to contribute significantly
to sustainable rural development. Agroforestry is, of course, not new and
has traditionally been practised by farmers in many parts of the world. Only
recently, however, and due to the growing interest within the development
community, has agroforestry become the subject of scientific research. As a
result, there is currrently limited scientific information available to allow
firm recommendations on species, sites, arrangements and management of
trees and other components. Agroforestry extension projects must thus
often work on the unproven hypothesis that the land and farmers will be
better off with agroforestry than without.Agroforestry land use systems, which combine crops, trees and/or livestock,
as well as multiple management objectives by farmers, are characterized
by a high degree of complexity and variability. For this reason, the
development of technology 'packages' through conventional research - a
standard practice in agriculture is probably neither feasible nor desirable for
agroforestry. Many extension and research projects have thus come to
depend on a form of 'action research', in which agroforestry technologies
are developed or adapted as part of the extension process. Such projects try
to monitor the performance of agroforestry technologies introduced onfarm
and their adoption and adaptation by farmers.