A Belousov-Zhabotinsky solution in a marginally stable but excitable steady state can propagate trigger waves of oxidation that move undamped through the solution. Such a wave can be initiated by a positive square wave pulse of 0.8 V or more applied to a silver electrode previously biased negatively with respect to a platinum electrode in the same solution. The critical pulse duration necessary to initiate a wave can be measured within a few tenths of a millisecond and has been determined as a function of composition of the solution and of other relevant parameters. Passage of a trigger wave can be followed by recording potentials between platinum electrodes in its path. Because velocity of propagation at any point is a positive function of time since the last previous wave passed that point, three or more successive waves separated by no more than a few minutes will tend to approach more uniform separations as they propagate. If a pulse is applied to an electrode less than 50s after the solution underwent an induced oxidative excursion (lOE), the incipient wave may be “localized”for up to 2 min before it leaves the electrode to propagate as a trigger wave; pulses applied 50s or more after an IOE generate waves without delay. Repetitive pulses at a frequency of about 0.8 min−1 or less each generate a wave, but it is not possible to generate waves at a frequency greater than this no matter how frequently pulses are applied. Dissolved oxygen increases the concentration of bromide ion in the solution apparently by affecting the stoichiometry of oxidation by fcrriin of a malonic-bromomalonic acid mixture; this increased bromide makes the system less sensitive to trigger wave initiation. An agar gel impregnated with silver nitrate solution can also initiate trigger waves. If a pair of silver electrodes is connected by a wire, a trigger wave impinging on the first can be discontinuously)-advanced so that future propagation is as though the wave had continued in the same direction but had been at the second electrode at the time it actually reached the first. All of these observations can be interpreted by a chemical mechanism previously developed to explain other features of this remarkable system. The behavior reported here exhibits several features similar to those observed during transmission of signals by a nervous system, and it may suggest useful analogies to ncurophysiologists. © 1979, American Chemical Society. All rights reserved.