Sympathetic preganglionic neurons (SPN) receive direct input from many levels of the neuroaxis, including the cervical, thoracic, and lumbosacral spinal cord, brainstem, hypothalamus, and even the cerebral cortex. In animals with intact spinal cords, nearly all tonic excitatory drive to SPNs descends from supraspinal systems. In these animals, tonic inhibitory input to SPN from segmental and propriospinal neurons is more important than excitatory input from these sources. Indeed, spinal excitation of SPN is inhibited both by descending pathways originating from supraspinal systems and from intraspinal systems. Thus, tonic and reflex-elicited sympathetic activity after spinal cord injury (SCI) is thought to result from disinhibition of excitatory spinal interneurons (IN) and from loss of descending excitation of spinal inhibitory IN. Where are the segmental and propriospinal IN that affect SPN located? Both retrograde transynaptic tracing and neurophysiological recordings indicate that the longitudinal distribution of excitatory neurons affecting renal sympathetic nerve activity (RSNA) is similar to the distribution of the SPN that generate that activity; the maximum concentration of both SPN and their associated IN is between T10 and T12 (1,2,3). On the other hand, propriospinal neurons projecting from rostral cervical spinal cord to caudal thoracic cord appear to be exclusively inhibitory (4). Both sympathoexcitatory and sympathoinhibitory propriospinal projections from lumbosacral spinal cord to caudal thoracic spinal cord have been reported. Surprisingly, these ascending propriospinal pathways, important as they are in autonomic dysfunction after SCI, are only now being examined in detail. Although previous studies have confirmed direct projections from the major sympathoexcitatory site in the rostral ventrolateral medulla (RVLM) to SPN, the degree to which the RVLM’s excitatory effects are mediated by these direct projections, or by projections to excitatory IN, has only been determined recently. We found that the incidence of retrogradely-identified renal sympathetic IN closely apposed by RVLM projections was only one-fifth the incidence of SPN closely apposed by these projections (5). These data confirm our electrophysiological observations that in rats with intact spinal cords, many SPN but few spinal IN exhibit activity correlated with RSNA (6). In recent years, a major concern of this laboratory has been the possibility that treatments currently under development for spinal cord injury may disrupt the function of the complex (and still incompletely-characterized) interactions between segmental, propriospinal, and supraspinal sympathetic systems. Yet in developing treatments, few laboratories assess autonomic function and dysfunction other than bladder control and the severity of autonomic dysreflexia. In a recent study, we used as our model the abundant sprouting of the corticospinal tract (CST) rostral to a chronic spinal cord injury (7). This model is relevant because encouragement of sprouting of lesioned descending pathways is being investigated in experimental animals as a promising treatment for SCI. We hypothesized that the synapses of sprouting axons, if they impinged on SPN, might amplify the modest increase in RSNA elicited when we microstimulated the thoracic CST. We identified “renal” SPN and IN by retrograde transport of pseudorabies virus from the kidney. We identified CST axons by anterograde transport from the cortex. Six times as many labeled SPN and three times as many labeled IN were closely apposed by CST axons one month after a chronic lumbar lesion than in unlesioned rats. Nevertheless, responses in RSNA to stimulation of the thoracic CST just caudal to an acute spinal transection were unaffected in the lesioned, sprouting rats. Apparently, the new synapses on sympathetic neurons were not sufficiently powerful, dense, or numerous enough to affect CST control of RSNA. These results are encouraging. They show that an increase in contact between a sprouting descending pathway and spinal sympathetic neurons need not lead to dysfunction. Nevertheless, we suggest that it would be prudent to conduct similar experiments routinely to assess the autonomic effects of treatment-induced sprouting and regeneration after SCI. In summary, appropriate levels of sympathetic activity are achieved through a balance of excitatory and inhibitory input from both spinal and supraspinal systems. Spinal cord injury disrupts that balance, resulting in dysfunction. Treatments for SCI may result in similar, or even more serious, imbalance. Therefore, assessment of autonomic function should be included in the development of treatments for SCI.