In their pursuit of quality, universities have since the 1970s often focused attention on academic development of their faculty. And, ideas about the ways in which such support can best be organized have evolved. In this paper the integrated model of academic development (Hicks, 1999) recently advocated in the literature is scrutinized. We describe how this model has taken shape in the concrete reality of the TA-programs in two universities and then raise questions regarding the implications for support of academic development.The conception of academic development as a multidimensional and distributed activity in which various partners within the university setting have responsibility is now more broadly accepted (Boud, 1999). It is no longer considered only the domain of centralized units with a specific mandate, but the responsibility of every faculty, department and unit (Reid, 2002). Thus, many faculty developers have been struggling in the past decade over the question of the extent to which they should work at the central or local level (Fraser, 2001). Hicks (1999) has proposed an integrated model of academic development, while admitting that such a model might be the most difficult to establish. However, he advocates it as the most robust and most effective, ensuring an ongoing central and local involvement in academic development.The question arises, however, how such 'an holistic approach' (Brew & Boud, 1996: 17) can be shaped concretely. In order to explore this issue, this paper focuses on the TA training developed at the K.U.Leuven (3 programs) and at McGill University (4 programs). Providing some preparation for TA's in order to take up their teaching responsibilities has become a major issue of academic development in many universities (Lowman & Mathie, 1993). Despite contextual differences between the K.U.Leuven and McGill University, the actual programs at both universities show remarkable similarities (with regard to the goals, nature of participation and activities, and duration). Yet, with regard to the ownership of the programs there exists a considerable difference between the Canadian and Belgian approaches. At McGill University the faculties themselves, in some cases even the TA's themselves, have taken the initiative to organize TA programs with the support of the central development unit. At the K.U.Leuven the faculties were encouraged to invest in a TA program by the central educational support unit in association with the TA representatives at the university's educational council. The analysis of our experiences shows that the difference in ownership of and responsibility for the program seems to relate to a distinct view in each university on how an integrated model of academic development might be shaped. Even though the K.U.Leuven central educational support unit very explicitly positions itself as a collaborating partner shaping the training with the local faculty developers (most of whom have had no previous training in higher education teaching and learning issues), the central unit sees itself as the 'driver' of development. At McGill, the central educational development unit only supports local TA programs on the request of the faculties or TA's themselves; however, most of the coordinators responsible for TA preparation took part in the central unit's general program on teaching and learning in higher education.The examples of the TA programs at the K.U.Leuven and McGill University illustrate that realizing an integrated model of academic development is not an obvious, straightforward thing. It appears to be a task provoking challenges as to how, as faculty developers, those in central units and faculties conceive their roles and tasks. Are we in central units prepared to shape our support in 'constructive alignment' (Biggs, 1999) with what all partners engaged define as the desired outcome of academic development? Are we - in order to further such an alignment - prepared to engage in 'learning conversations' (Laurillard, 2002) and to support discipline-based leadership (McAlpine & Weston, 2002)? In order to support academic development within the integrated model it seems necessary to answer these questions. Biggs, J. (1999). Teaching for quality learning at university. UK: Open University Press.Boud, D. (1999). Situating academic development in professional work: using peer learning. International Journal for Academic Development, 4(1), 3-10.Brew, A. & Boud, D. (1996). Preparing for new academic roles: An holistic approach to development. International Journal for Academic Development, 1(2), 17-25.Fraser, K. (2001). Australasian academic developers' conception of the profession. International Journal for Academic Development, 6(1), 54-64.Hicks, O. (1999). Integration of central and departmental development - reflections form Australian universities. International Journal for Academic Development, 4(1), 43-51.Laurillard, D. (2002). Rethinking university teaching. London: Routledge Falmer.Lowman, J. & Mathie, V.A. (1993). What should graduate teaching assistants know about teaching? Teaching of Psychology, 20, 84- 88.McAlpine, L. & Weston, C. (2002). Implementing a more discipline-based approach to educational development: How are we doing after four years? What does it mean for the future? Paper presented at the International Consortium for Educational Development, Perth, Autralia.Reid, A. (2002). Is there an 'ideal' approach for academic development? Paper presented at the International Consortium for Educational Development, Perth, Autralia.