A sizable literature on quantitative research evaluation procedures has developed, but little attent10n has been focussed on procedures for reviewing or monitoring agricultural research m an external review process. Despite their relative neglect m the literature, external reviews are increasingly relied on by governments and by international donor and lending agencies to assess the performance of research systems. Tlus paper suggests guidelines for external reviews, focussing on their content and procedures, while highlighting key issues Wlth respect to research organizations; discusses the importance of obtaining early agreement among review team members on evaluation criteria and of considenng research system goals, objectives, priorities, organization capacity, management, vertical and horizontal linkages, programme content, impacts, and future needs; examines the issues of ltnkages between national research systems and the international agricultural research centres and the extent to which national systems should adopt a farming systems research approach; and stresses the need for the research system, operatmg in an uncertain world, continually to reassess the long-run demand for particular technologies.