Multiple sclerosis (MS) is a common neurological disease and a major cause of disability, particularly affecting young adults. It is characterized by patches of damage occurring throughout the brain and spinal cord, with loss of myelin sheaths - the insulating material around nerve fibres that allows normal conduction of nerve impulses - accompanied by loss of cells that make myelin (oligodendrocytes). In addition, we now know that there is damage to nerve cells (neurones) and their fibres (axons) too, and that this occurs both within these discrete patches and in tissue between them. The cause of MS remains unknown, but an autoimmune reaction against oligodendrocytes and myelin is generally assumed to play a major role, and early acute MS lesions almost invariably show prominent inflammation. Efforts to develop cell therapy in MS have long been directed towards directly implanting cells capable of replacing lost oligodendrocytes and regenerating myelin sheaths. Accordingly, the advent of techniques to generate large numbers of oligodendrocytes from embryonic stem cells appeared a significant step towards new stem cell treatments for MS; while the emerging consensus that adult stem cells from, for example, the bone marrow had far less potential to turn into oligodendrocytes was thought to cast doubt on their potential value in this disease. A number of scientific and medical concerns, not least the risk of tumour formation associated with embryonic stem cells, have however, prevented any possible clinical testing of these cells in patients. More recently, increasing understanding of the complexity of tissue damage in MS has emphasized that successful cell therapy may need to achieve far more than simply offering a source of replacement myelin-forming cells. The many and varied reparative properties of bone marrow-derived (mesenchymal) stem cells may well offer new and attractive possibilities for developing cell-based treatments for this difficult and disabling condition.