The paper builds on many inputs from ASTPP network members, in particular from presentations at the First ASTPP Workshop in Madrid, June 1996. Special thanks to Dennis Loveridge. The paper was presented in a workshop in Strasbourg, 16-17 December 1996.Science and technology had been professional practices that from their origins had set up mechanism for internal control. Research evaluation could be seen as a system for control of the quality and relevance of the results of research. The most common this control practices had been peer review. When patrons, mainly the State, developed the research funding system formal peer review procedures started to be used and developed as instruments for allocation of funds to research institutions and groups. This extension of the internal control mechanisms of science to the allocation of funds for research had become a widely used instrument, in particular in the realm of basic and fundamental research, and a central element of the legitimisation cycle. The development of the so called strategic R&D programmes show the extension of steering activities, selecting priorities, and direct allocating funds by governments. At the same time the increase relevance of technology and innovation issues in government agendas helped the development and consolidation of specialised S&T policy-making bodies and bureaucracies; new actors in the RTD system, that have different needs of information and knowledge about the S&T dynamics and process, through whom new ideas of S&T (policy) evaluation were introduced. Evaluation is "examining" or "making judgements" (all cognitive process for action include assessments), and policy evaluation could be understood as part of the historical process of development of tools and information systems for public management. But the development of evaluation of S&T policies had evolved mainly from the transformation of the professional control practices of researchers and from the specific forms of management of RTD programmes. The term S&T policy evaluation include activities and practices that usually looks back at the past performance of programmes or policies (or sometimes as they are implemented, as in continuous or real-time evaluation) and they are part of the S&T policy cycle as is traditionally described (design, implementation, monitoring, evaluation, redesign). S&T policy evaluation refers to "retrospective or ongoing examination" of programme performance or impacts. There are other practices such as appraisal that refers to the activities developed at the beginning of a programme or project (what is sometimes called ex-ante evaluation, often related to a selection process for funding or other purposes) or monitoring. In general, evaluation and monitoring can be seen as part of control activities in the policy cycle, with few differences in addition to the more or less "judgement" and to their position in the temporal sequence of the "policy process". The boundaries between those activities are fuzzy and subject to interpretations; depending on the country, the organisational arrangements to carry on those activities diverge in important ways; the use and impacts of evaluations are highly local and contextual; and they are developed in specific national settings and arrangements.