1. To test a previously proposed model for ionic regulation (Weidler and Diecke, 1969), the regulation of sodium ions in the central nervous system of the herbivorous insect Carausius morosus was studied with 22Na. 2. The uptake of 22Na is blocked by 5 mM sodium azide in intact, whole nerve cords (Fig. 2, Table 2). From these findings we concluded that sodium ions are actively absorbed by intact nerve cords. 3. When fat-body sheaths are removed, there is no significant difference between the 22Na uptakes of nerve cords bathed in solution containing 5 mM sodium azide and of those not exposed to azide. Upon comparison of these results with those in Fig. 2 and Table 2, we concluded that the active sodium uptake system is located in the fat-body sheath. 4. The desaturation curve of nerve cords completely saturated with 22Na contains the following three components (Fig. 1): a fast component with a half-time of 1.33 min; an intermediate component with a half-time of 0.39 hr; a slow component with a half-time of approximately 2.5 hr. This three-component system conforms to the proposed model with components probably originating in the fat-body sheath space, the extraneural space and the neural space respectively. 5. The size of the fast component is unaffected by metabolic inhibition (Table 2), and it depends on the external sodium concentration (Table 4); therefore we concluded that this component is outside of the active sodium uptake system. Since the site of active sodium uptake was shown to be in the fat-body sheath (probably in the basement membrane), we concluded that this fast component originates in the fat-body sheath space located outside of the basement membrane. 6. Uptakes of 22Na by intact nerve cords for both the slow and intermediate components are significantly decreased when a metabolic inhibitor is present in the bathing medium (Table 2), but sodium uptake for the intermediate component is increased only slightly and that for the slow component not at all when the external sodium concentration is increased from 15 mM to 180 mM. We concluded from these findings that these two components are located within the boundaries of the active sodium uptake system. 7. The half-time of the intermediate component for intact nerve cords metabolically inhibited during the desaturation period was significantly longer than that of uninhibited nerve cords. This result may be explained by the obliteration of the potential difference across the fat-body sheath by the metabolic inhibitor; we concluded that the intermediate component originates in the extraneural space located between the basement membrane of the fat-body sheath and the neural lamella. 8. Since the origins of the fast and intermediate components have been determined, the only logical anatomical site of origin for the slow component is the neural space (i.e., within the confines of the neural lamella). Results are consistent with this hypothesis. 9. Calculations of mean sodium concentrations from space sizes (determined from photomicrographs) and saturated component sizes were consistent with the model and reported values for similar tissues in other animals. Calculated permeabilities for the fat-body sheath and the neural lamella are consistent with previously published viability studies (see Weidler and Diecke, 1969). We concluded that the present study verifies the significant features of the proposed model with regard to sodium ions.