A reconsideration of the Poisson hypothesis for transmitter release at the crayfish neuromuscular junction.

1. Transmitter release from excitor nerve terminals in the crayfish opener muscle was studied with both intra- and extracellular electrodes.2. The amplitude distribution of quanta recorded extracellularly from a single release site did not fit the compound Poisson hypothesis in two thirds of the data sets.3. Some data appeared to be non-Poisson because specific quantal multiples were released; the non-Poisson nature of release for other sets could be due to variable invasion of the axonal spike into the terminal region.4. In some muscle fibres, the number of quanta released at one ending to any given impulse in a stimulus train was correlated with the number of quanta released at other endings on the same fibre. This suggests that such endings may be supplied by nerve twigs arising from a common branch point with a very low safety factor for transmission.5. No consistent relationship was found between the amplitudes of the intra- and extracellular spontaneous potentials recorded from the same muscle fibre.