Histograms of the unitary evoked potential of the mouse diaphragm show multiple peaks.

1. Two classes of miniature end-plate potentials (m.e.p.p.s) were recorded from diaphragm neuromuscular junctions. Amplitude histograms of both classes had multiple peaks that were integral multiples of the smallest peak (s-m.e.p.p.s). The smaller m.e.p.p.s formed the first three or four peaks of histograms and the number of m.e.p.p.s (skew-m.e.p.p.s) in each peak decreased, forming an over-all skewed distribution. The larger m.e.p.p.s (bell-m.e.p.p.s) formed a more-or-less bell-shaped distribution. The distribution of m.e.p.p.s varied from mainly skew- to mainly bell-m.e.p.p.s. In young adult mice the number of subunits composing the classical m.e.p.p.s varied between ten and fifteen at room temperature; at higher temperatures the range was from three to ten subunits.2. End-plate potentials (e.p.p.s) were reduced with cobalt ions (ca. 4 mm) until most nerve impulses failed to release transmitter. The amplitudes of ;unitary evoked potentials' were of the bell-m.e.p.p. class and histograms show integral multiple peaks that correspond to the peaks in histograms of the bell-m.e.p.p.s.3. The peaks in both m.e.p.p. and unitary e.p.p. histograms remained in the same position throughout the recording period and became more distinct as the sample size increased.4. The variance of the s-m.e.p.p. was estimated from the noise and measurement error and the variance of all peaks in the histograms. Most variance of the first peak (s-m.e.p.p.) was due to noise and measurement error.5. The integral peaks in the m.e.p.p. and ;unitary evoked potential' histograms are predicted with a probability density model based on the estimated variance of the s-m.e.p.p. and the assumption that larger potentials are composed of subunits the size of s-m.e.p.p.s. The data and model support the hypothesis that m.e.p.p.s and unitary potentials are composed of subunits.