The adrenal medulla is the ancestral homologue of autonomic ganglia. The innervation of the medulla through a complex of nerves rather than a single trunk allows physiological dissection of synaptic behaviour. In previous experiments, synaptic function within the adrenal medulla of the cat has been studied by testing its response to nervous excitation and by investigating properties such as threshold, fatigue and susceptibility to ganglion-blocking drugs (Marley & Paton, 1961). New data are here presented on the threshold to nerve excitation, spatial and temporal recruitment, fatigue, the effect of drugs on these phenomena and the presence of functional units within the adrenal medulla. The preliminary findings have been communicated to the Physiological Society (Marley, 1961;Marley & Prout, 1963).
METHODSCats were anaesthetized with chloralose (80 mg/kg i.v.) after induction with ethyl chloride and ether. The vagi were cut in the neck and the animal was artificially respired. Carotid arterial blood pressure was recorded with a mercury manometer on a kymograph.Usually the splanchnic nerves of one side were stimulated. They were approached retroperitoneally through a lumbar incision and the adrenolumbar vessels were divided between ligatures. As described elsewhere (Marley & Prout, 1965), there are three or four upper splanchnic nerves to the adrenal medulla. These were identified and cut through and the sympathetic chain was removed from the first to the fourth lumbar sympathetic ganglia. The abdominal viscera including the adrenal gland not under test were removed, and the renal vessels were tied. Heparin (10 mg/kg i.v.) was injected. In several recovery experiments the adrenal glands were chronically denervated under aseptic conditions, the cats being anaesthetized with halothane (Fluothane) and oxygen by a modification of the method for anaesthetizing new-born animals (Marley & Payne, 1962). These cats were re-anaesthetized with chloralose 10-14 days later and prepared as described.The splanchnic nerves were stimulated with platinum electrodes (cathode distal) using supramaximal rectangular (0-5 msec) pulses of 20 V at varying frequencies. The excitation frequency and number of stimuli were counted with a Racal digital frequency meter. For rapid excitation with up to ten stimuli, one stimulator provided the excitation frequency and the other a variable 'gate' during which a Stevens-Arnold 'millisec relay' was closed connecting the first stimulator to the electrodes; the stimuli were counted on an oscilloscope. In experiments to determine the stimulation frequency for optimal sympathin output, as well as those in which cocaine or phenoxybenzamine was injected into the cat, the excitation 31-2