Central to the postulated relationship between the brain and the immune system has been evidence for the direct neural innervation of primary organs of the immune system. It has been reported previously that the thymus gland in rats and mice receives a substantial innervation from the "retrofacial" nucleus of the brain stem and ventral horn cells of the upper cervical spinal cord. Based on the proximity of the thymus to thoracic viscera and neck musculature known to receive motor fibers from these same areas of the brain stem and spinal cord, we examined the possibility that retrogradely labeled cells in the brain stem and spinal cord following injections of tracers into the thymus are due to spread of tracer into the esophagus and neck musculature. Small injections (0.5-2.0 microliter) of wheatgerm agglutinin-horseradish peroxidase (WGA-HRP) were made into the thymus, the esophagus, and the longus colli muscle of rats or mice. Also, the effects of a unilateral cervical vagotomy on cholinesterase activity in the thymus were examined. Finally, the source of the sympathetic supply to the thymus and the presence of catecholamine and cholinesterasic fibers in the thymus was reassessed. Injections of WGA-HRP into the thymus produced little or no labeling in the brain stem and spinal cord. In contrast, control injections into the esophageal wall resulted in numerous intensely labeled cells in the compact formation of the nucleus ambiguus, irrespective of the rostral-caudal level of the esophageal injection. Similarly, tracer injections into the longus colli muscle resulted in numerous intensely labeled cells in the ventral horn of the upper cervical spinal cord. Unilateral vagotomy did not alter cholinesterase activity in the thymus even though it was largely depleted in the ipsilateral nucleus ambiguus. The histochemical studies verified a major sympathetic innervation of the thymus gland. In keeping with this result, in animals in which no labeled cells were observed in the brain stem or spinal cord following thymus injection, labeled cells were, however, observed in the sympathetic chains from the superior cervical ganglia caudal to the T3 ganglia. In summary, all labeled cells in the brain stem and cervical spinal cord observed following tracer injections into the thymus can be accounted for by spread of the tracer into surrounding structures, leading to spurious labeling.(ABSTRACT TRUNCATED AT 400 WORDS)