1. Transmission of a single nerve impulse has been investigated at the nerve-electroplaque junction of Torpedo marmorata in the presence of 4-aminopyridine (4-AP), a drug which powerfully potentiates evoked transmitter release.2. Three methodological approaches were used conjointly. These were (i) electrophysiological recording of the compound electroplaque potential (e.p.p.), (ii) radiochemical measurement of evoked acetylcholine (ACh) release and (iii) analysis of the content of ACh and ATP in the tissue at brief time intervals during the course of the e.p.p. and soon after. The last was achieved by using a stimulator coupled to a rapid tissue freezer.3. In the response to a single stimulus, 4-AP enhanced in a dose-dependent manner the size of the e.p.p., increasing the duration much more than the amplitude. At 10(-4) M-4-AP, this resulted in the generation of a characteristic ;giant e.p.p.' whose area (in V x ms) was approximately 120 times greater than that of a normal e.p.p.4. The giant e.p.p. consisted of an initial peak, lasting for some 100 ms, a late rebound at about 300 ms, and finished between 500 and 1000 ms after the stimulus. Temperature changes greatly affected the shape of the giant e.p.p., modifying particularly the amplitude and time course of the late rebound.5. The amount of ACh released in response to a single stimulus was measured radiochemically and was found to greatly increase in the presence of 4-AP, explaining the potentiation of the e.p.p. With 4-AP concentrations ranging from 10(-6) M to 10(-4) M, the augmentation of ACh release showed a close correlation with increase of the e.p.p. area.6. The large potentiation of evoked transmitter release occurred in spite of a reduction of ACh stores. After treatment with 10(-4) M-4-AP, the total ACh content was reduced by 30-40% in the absence of any electrical stimulation. The reduction affected to a similar extent the vesicular and extravesicular compartments of ACh. This was accompanied by a general increase in the resting rate of ACh turnover.7. Synaptic vesicles were isolated from small fragments of electric organ, rapidly frozen with our device. Compartmental analysis was carried out by labelling the transmitter pools with a radioactive precursor and it was confirmed that vesicular ACh has a relatively low metabolic rate, whereas free ACh (most probably cytoplasmic ACh) turns over more rapidly. The same finding was obtained after treatment with 4-AP, but the starting levels of ACh and the yield of synaptic vesicles were lower.8. The total ACh content was measured at 30 and 100 ms intervals during the course of the giant e.p.p., and soon after. We found characteristic and significant changes which were (i) an initial fall of total ACh occurring within 100-150 ms, (ii) a transient ACh increase which occurred later and seemed to correspond to the late rebound of the giant e.p.p. and (iii) a steady 20% lowering of total ACh, observed from the end of the giant e.p.p. and lasting for more than 1 s.9. The ATP content of the tissue, during and after the giant e.p.p., followed a time course which was remarkably similar to that of total ACh. A significant ATP/ACh relationship was found in most experiments separately, and in the pooled results with a higher degree of significance.10. Vesicular ACh did not exhibit any significant change during and after the giant e.p.p. Neither the transient initial variations of total ACh nor its later lowering were reflected in similar changes of vesicular ACh. It was therefore the extravesicular pool of ACh which was concerned in the characteristic pattern of changes of total ACh.11. Compartmental analysis of transmitter stores was performed during the course of transmission, after labelling ACh in the tissue with a radioactive precursor. It was found that no detectable transfer of ACh occurred from cytoplasm to vesicles, either during the giant e.p.p., or within the following second.12. The following conclusions were reached. The effect of 4-AP is to cause a very strong and long-lasting potentiation of ACh release, resulting in a giant and complex electrical discharge. Transmitter release under these conditions was not only due to sudden liberation of the preformed, available ACh but also to a marked contribution of new ACh made during the giant e.p.p. These changes in ACh content were very significant and took place exclusively in the extravesicular pool of transmitter.