Remyelination enables restoration of saltatory conduction and a return of normal function lost during demyelination. Unfortunately, remyelination is often incomplete in the adult human central nervous system (CNS) and this failure of remyelination is one of the main reasons for clinical deficits in demyelinating disease. An understanding of the failure of remyelination in demyelinating diseases such as Multiple Sclerosis depends upon the elucidation of cellular events underlying successful remyelination. Although the potential for remyelination of the adult CNS has been well established, there is still some dispute regarding the origin of the remyelinating cell population. The literature variously reports that remyelinating oligodendrocytes arise from dedifferentiation and/or proliferation of mature oligodendrocytes, or are generated solely from proliferation and differentiation of glial progenitor cells. This review focuses on studies carried out on remyelinating lesions in the adult rat spinal cord produced by injection of antibodies to galactocerebroside plus serum complement that demonstrate: 1) oligodendrocytes which survive within an area of demyelination do not contribute to remyelination, 2) remyelination is carried out by oligodendrocyte progenitor cells, 3) recruitment of oligodendrocyte progenitors to an area of demyelination is a local response, and 4) division of oligodendrocyte progenitors is symmetrical and results in chronic depletion of the oligodendrocyte progenitor population in the normal white matter around an area of remyelination. These results suggest that failure of remyelination may be contributed to by a depletion of oligodendrocyte progenitors especially following repeated episodes of demyelination. Remyelination allows the return of saltatory conduction (Smith et al., 1979) and the functional recovery of demyelination-induced deficits (Jeffery et al., 1997). Findings such as these have encouraged research aimed at enhancing the limited remyelination found in Multiple Sclerosis (MS) lesions, evidenced by a rim of thin myelin sheaths around the edges of a lesion, or, in a minority of acute foci, throughout the entire lesion (Prineas et al., 1989; Raine et al., 1981). It must be said, however, that although remyelination is clearly a prerequisite to sustained functional recovery, other factors such as the state of the inflammatory response and degree of axonal survival within the demyelinated region contribute to the extent of functional recovery that may be possible following therapeutic intervention aimed at halting disease progression. It is not yet clear whether the progression of functional deficits in MS is primarily the result of an increasing load of demyelination, or axon loss, or a combination of the two processes. However, given the increasing recognition that myelin sheaths play a role in protecting axons from degeneration, the success or failure of remyelination has functional consequences for the patient. To understand why remyelination should fail in demyelinating disease and develop strategies to enhance remyelination requires an understanding of the biology of successful remyelination. Firstly, what is the origin of the remyelinating cell population in the adult CNS? Secondly, what are the dynamics of the cellular response of this population during demyelination and remyelination? And thirdly, what are the consequences to the tissue of an episode of demyelination? This review will focus on studies that address these issues, and discuss the implications of the results of these experiments for our understanding of MS and the development of therapeutic interventions aimed at enhancing remyelination.