Recently, Staggs et al. (1980) reported that, in crayfish, excitatory postsynaptic potentials (EPSPs) measured at different extracellular calcium concentrations [Ca], are not affected by applied picrotoxin. This finding was interpreted to contradict the observation of Parnas, Rahamimoff, and Sarne (1975) that picrotoxin, by removing a tonic inhibition, leads to an increase of EPSPs. It is important to note that the latter results were obtained at the neuromuscular junction of a crab (Ocypode), while the findings of Staggs et al. (1980) were obtained at crayfish neuromuscular junctions and thus are not seen to "contradict," especially since their conditions were unfavorable for the observation of tonic inhibition. Staggs et al. (1980) also emphasized and gave evidence that in crayfish there is a linear dependence of the amplitude of the EPSP on [Ca],, as reported earlier (Bracho and Orkand, 1970;Ortiz and Bracho, 1972;Zucker, 1974). Since in vertebrate endplates this dependence is very nonlinear (d log EPSP)I(d log [Calo) = 4 (Dodge and Rahamimoff, 1967), and since this nonlinearity is a prerequisite for explaining the observed facilitation by the residual calcium mechanism, this apparent and "suspect" (Atwood, 1977) deviation of EPSP behavior in crayfish should not be left without comment. A recent report by Nickell and Boyarski (1980) showed a dependence of the excitatory postsynaptic current (EPSC) on [Ca];, as well as postsynaptic effects of [Ca], on the EPSC. Figure 1 shows examples of recordings of EPSCs at the neuromuscular junction of the crayfish (Astacus Zeptodactylus) abductor digiti primi at different [Ca],. The synaptic currents were measured at single synaptic spots by means of a patch clamp electrode (Neher, Sakmann, and Steinbach, 1978;Dudel, 1981). Average amplitudes of EPSCs from 19 experiments are plotted in Figure 2 (Fig. 4 from Parnas, Dudel, and Parnas, submitted). In this plot the amplitudes were corrected for postsynaptic effects of low [Ca],, since at 1.7mM [Calo the quantum size is about one-half of that at 13.5mM [Ca], (Dudel, 1981). The slope of the log EPSC to log [Ca], relationship in Figure 2 is 1.6. This slope would have been nearly 2, at low [Ca],, without correction of postsynaptic effects. This agrees with the findings of Nickell and Boyarski (1980), an extensive study of synaptic facilitation (Parnas et al., submitted), and on quantum content (Dudel, 1981) at different [Calo in the crayfish neuromuscular junction. Therefore, there is no reason to assume that the [Ca]o dependence of synaptic transmission, and more importantly, the mechanism of facilitation are basically different in crayfish and vertebrates.The consistent failure of Staggs et al. (1980) to observe nonlinearity may be explained by insufficient access of the low-[Calo solution to the synapses. We have found that careful removal of obstructing fine tissue layers, rapid super-