Motor giant axons that excite swimming muscles in the jelly-fish Aglantha digitale interface with units of the inner and outer nerve rings in the margin at the base of the bell. External recording electrodes were used to monitor electrical activity at different sites within the nerve ring while events in the motor giant axon were recorded with intracellular micropipettes placed within 100 microns of the synaptic area. In some experiments, 4- to 6-micron-diam patch pipettes were used to record in situ from ion channel clusters at different locations along the axon. 2. Independently propagating calcium and sodium spikes in the motor giant axon were found to arise from different excitatory postsynaptic potentials (EPSPs). Two separate inputs were identified; one EPSP class represented an input from the pacemaker system in the inner nerve ring, whereas another represented an input from the giant axon in the outer nerve ring. EPSPs from the two nerve rings had significantly different time courses and amplitudes. EPSPs from the ring giant axon reached a peak in little more than 1 ms, whereas EPSPs from the pacemaker system reached a maximum in approximately 7 ms. These slower EPSPs may be compound events composed of postsynaptic potentials from multiple synapses excited in series by the passage of the pacemaker neuron signal. 3. The threshold for the production of calcium spikes by the slow EPSPs of the pacemaker system (-51 +/- 2.2 mV, mean +/- SD; n = 5) corresponded well with the voltage at which a net inward "T"-type calcium current first appeared in recordings from axon membrane patches (-55 to -50 mV); the threshold for the initiation of the sodium spike by the fast EPSPs of the ring giant system (-32 +/- 1.2 mV, mean +/- SD; n = 6) corresponded well with the voltage at which a net inward sodium current first appeared (-35 to -30 mV). 4. Inward currents were rarely observed in membrane patches formed using pipettes with tips of < 1 micron OD. Even with 4-micron pipettes, patches of membrane were sometimes obtained with a channel population consisting exclusively of potassium channels; calcium and sodium currents were found in highly discrete areas ("hot spots"). Preliminary findings on the undersurface of the axon, which makes synaptic contact with the myoepithelium, are consistent with a similar distribution. 5. The pathway by which the ring giant excites the motor giant axon is not definitely known. The synaptic delay between the peak of the ring giant action potential (monitored externally) and the initial rise of the fast EPSP (1.64 +/- 0.15 ms, mean +/- SD; n = 21) would allow for transmission at two synapses, because single synaptic delays at neuromuscular junctions in Aglantha are approximately 0.7 ms at 12 degrees C. The mean synaptic delay at the slow EPSP synapse was 0.88 +/- 0.09 (SD) ms (n = 12). 6. The delay between the impulse in the ring giant axon and the subsequent excitation of the motor giant axon may permit the animal to withdraw its tentacles and so lower the drag that would otherwise reduce the effectiveness of any escape swim and might induce tentacle autotomy.
