Taken together, our studies indicate that (a) transplants mediate recovery of skilled forelimb movement as well as locomotor activity, (b) combinations of interventions may be required to restore reflex, sensory, and locomotor function to more normal levels after SCI, and (c) that remodeling of particular pathways may contribute to recovery of rather specific aspects of motor function. In conclusion, we suggest that it seems unlikely that any single intervention strategy will be sufficient to ensure regeneration of damaged pathways and recovery of function after SCI. Clearly, work from a number of laboratories indicates that the dogma that mature CNS neurons are inherently incapable of regeneration of axons after injury is no longer tenable. The issue, rather, is to identify and reverse the conditions that limit regeneration after SCI. After SCI, a hierarchy of "intervention-strategies" may be required to restore suprasegmental control leading to recovery of function. The hierarchy may be both temporal and absolute. For example, early interventions (such as the administration of methylprednisolone within hours of the injury) may be required to interrupt the secondary injury cascade and restrict the extent of damage after SCI. At the injury site itself, interventions to minimize the secondary injury effects may be followed by interventions to alter the environment at the site of injury to provide a terrain conducive to axonal elongation. For example, one might envision strategies to downregulate the expression of molecules that limit growth and upregulate the expression of those that support growth. Early after the injury, axotomized neurons may require neurotrophic support either for their survival or to initiate and maintain a cell body response supporting axonal elongation. There may be an absolute hierarchy as well. Particular populations of neurons may have very specific requirements for regenerative growth. For example, the conditions that enhance the regenerative growth of descending motor pathways may differ from those required by ascending sensory systems. One may also want to design strategies to restrict the plasticity of some pathways (e.g., nociceptive) and enhance the growth in other pathways. The demands on the CNS for anatomic reorganization after SCI may be far less formidable than one might at first imagine. If one assumes that recovery of function will require regenerative growth of large numbers of axons over long distances in a point-to-point topographically specific fashion, the idea of recovery of function becomes daunting. On the other hand, it has been shown in many studies and in many areas of the CNS that as little as 10% of a particular pathway can often subserve substantial function. Furthermore, regrowth over relatively short distances can have major functional consequences. For example, relatively modest changes in the level of SCI can have relatively profound effects on the functional consequences of injury. This is particularly true in cervical SCI: an individual with a C5/6 SCI is dramatically more impaired than one with C7/8 injury. One might envision relatively short distance growth across the injury site to re-establish suprasegmental control. Coupled with strategies to enhance the anatomic and functional reorganization of spinal cord circuitry caudal to the level of the injury, even modest long distance growth may have sufficient functional impact. One might imagine the ability to learn to "use" even modest quantities of novel inputs in functionally useful, appropriate ways.