The following four proposed roles for the acetylcholinesterase (AChE) of the superior cervical ganglion, where the localization of the enzyme is predominantly presynaptic, were studied: (1) limitation, temporally or spatially, of the transmitter action of acetylcholine (ACh) at the postsynaptic site, (2) hydrolysis of liberated ACh to provide an immediate source of choline for uptake and resynthesis to ACh by the terminals of the preganglionic fibers, (3) protection of the presynaptic nerve terminals against endogenously liberated ACh, and (4) prevention of the accumulation and action of ACh liberated during the resting state. Some of the findings reported previously by others which form the basis for the first proposal were confirmed. However, experiments designed to obtain more definitive evidence for this proposal and, particularly, to assess the degree of the intraganglionic spread of ACh were unsuccessful. No evidence was obtained to support the second proposal. Although the rate of failure of transmission produced by repetitive preganglionic stimulation was accelerated markedly by diisopropyl phosphorofluoridate (isoflurophate, DFP), the infusion of choline, over a wide range of concentrations, produced no antagonism of the failure of transmission either before or after DFP. Thus, it is suggested that the failure of transmission under these conditions was due to the accumulation of excessive concentrations of ACh rather than a temporary depletion of transmitter substance. Support for the third proposal was indicated by the findings that: (a) the mean threshold doses, for postganglionic activation, of carbachol were significantly (26-fold) higher in chronically denervated than in normal ganglia, and (b) although the mean threshold doses of ACh were the same in untreated denervated and normal ganglia, after DFP the thresholds were significantly (5-fold) lower in the latter as compared with the former. The mean threshold dose of ACh was 10 times that of carbachol in normal ganglia; however, in the three situations where AChE was inactivated or absent (<i>i.e.,</i> denervated ganglia before and after DFP, and normal ganglia after DFP), the mean threshold doses of ACh were only one-third to one-half those of carbachol. A limited number of similar experiments was conducted with butyrylcholine and tetramethylammonium. It is inferred from these observations that in normally innervated ganglia, threshold doses of carbachol, and possibly of ACh, activate only the presynaptic terminals, causing them in turn to liberate the neurohumoral transmitter, with consequent activation of the postsynaptic site. Hence, it is suggested that the primary role of ganglionic AChE is the limitation of the action of liberated ACh on the presynaptic nerve terminals themselves. Other possible interpretations are also considered. It was observed also that in denervated ganglia the mean threshold doses of both ACh and carbachol were reduced by DFP to the same extent. The possibilities that the butyrocholinesterase of the capsular glial cells, or some nonspecific group at the receptor site, plays a role in the potentiation of these agents by DFP are discussed. The finding that persistent asynchronous postganglionic firing occurred in acutely decentralized resting ganglia following the injection of larger doses of DFP provides support for the last proposal. Postganglionic firing following massive doses of DFP did not occur in chronically denervated ganglia.