The first step essential in the search for a cure of human spinal cord injury (SCI) is to appreciate the complexity of the disorder. In this regard, it is not only the loss of ambulation but the sensory and autonomic changes that are equally important in recovery. In addition, there are the serious social emotional psychological and lifestyle effects of SCI which should also be taken into account. It is also true that no two SCI lesions are alike as each is the result of a SCI unique to that individual. Clinically of utmost importance is the segmental level of injury and whether it is complete, incomplete or discomplete (loss of all neurological functions below the injury but with physiological or anatomical continuity of Central nervous system tracts across the lesion). We are not concerned here with primary and secondary prevention or methods designed to limit the severity of the lesion after the event, important as they are, but with the requirements for a cure. Clearly, the greater the number of nerve fibers that can be preserved in the acute stage, the better will be the end result. Our focus at present is on the end-stage lesion with the aim of showing that a cure for SCI will depend upon establishing functionally useful central axonal regeneration and reestablishing physiological reconnections. Existing experimental methods are based on stimulating axonal regeneration by neutralizing inhibitory factors, adding positive trophisms and creating a permissive environment. Better results are obtained by bridging the gap with grafts of peripheral nerves or transplants of Schwann cells and genetically engineered fibroblasts. Recently, the potential for stem cells to enhance this process has created great interest. This is because of the ability of pluripotential cells to differentiate into neural tissue. A cure based on the physiopathology of SCI requires pyramidal, extrapyramidal, sensory, cerebellar and autonomic pathways to be regenerated with their appropriate neurotransmitters restored and reflexes integrated physiologically and in synchrony. In human SCI, there is a very long distance anatomically for axonal regrowth to occur in order to reach their relevant nuclei. This is because of continuing Wallerian degeneration. It also presumes that the target neurons are intact and that there has been no transneuronal degeneration above or below the lesion. Alternatively, in place of regenerated long axons, a multisynaptic pathway may be constructed from stem cells that have developed into neurons. Whether such a pathway would restore useful neurological functions is unknown. At present, the transplant and grafting research teams are exploring these possibilities in experimental animals. Moderate success in gaining axonal regeneration has been reported; however, it must be appreciated that the human lesion differs considerably from that of the experimental animal. In order to be successful, the neuropathology and neurophysiology of human SCI must be taken into account. The purpose of this review is to place the req...