The failure of the adult human central nervous system (CNS) to repair following injury has significant clinical consequences. Lesions caused by trauma, vascular defects or chronic inflammation can result in long-standing damage and considerable functional impairment for which there are no effective remedies at present. A major goal of neuroscience research is, therefore, to devise strategies for effective repair following CNS damage. One of the most important of these strategies is transplantation. The goal of this work is to transplant cells into the damaged brain either to replace tissue directly or to stimulate the ability of the CNS to repair itself. Promising initial results have emerged with transplantation for the very focal defect of Parkinson's disease, using fetal dopaminergic neurons placed directly into the denervated striatum. However, to repair widespread lesions it seems likely that the transplanted cells will have to be altered so as to enhance their potential to initiate or facilitate repair. In this review, we emphasize the importance of understanding the developmental biology of the system in question before attempting manipulation of cells prior to transplantation. This point of view stems from observations that mechanisms used during development are often reexpressed in those systems that repair effectively. It follows that manipulating cells to be transplanted so as to re-express molecules present in development may enhance repair in those areas where repair is normally minimal.