In auto-catalysis one of the _products_ of the reaction acts as the catalyzer. Now Robertson concluded, as a result of his experiments carried out on frogs, that the processes which accompany the excitation of the cells of the neurons are of the nature of catalysis; for he found that they have as one effect the production of an acid; and he also found that acids accelerate such processes which he concludes to be probably of the nature of oxidations. “The chemical phenomena which constitute the activity of a neuron cell,” he says, “seem to us then an _auto-catalytic oxidation_, that is to say, an oxidation in which one of the products of the reaction acts as a catalyzer in the reaction.” It occurred to him then that the physiological correlate of memory might be explained on the principle of auto-catalysis. When, to test this hypothesis, he came to compare the results of certain psychological experiments on memory, made by two different experimenters (Ebbinghaus and Smith), with the law characteristic of auto-catalytic chemical reactions, he found that they corresponded in a surprisingly close way with this law. That is to say, assuming the value of the residua of memory (measured by the number of syllables learnt by heart) to be proportional to the mass of the chemical product of auto-catalysis, we should expect that the increase of the number of syllables or other experiences retained by memory following increase of repetitions would obey the law of catalytic reaction as expressed in the mathematical formula established for the reaction. Now, as a fact, he found that the number of syllables that should be so retained in memory, as calculated theoretically by the formula, corresponded in a remarkable way with the actual number determined by experiment. “The agreement was closer,” the author states, “than that which generally obtained in experiments in chemical dynamics carried out _in vitro_.” Robertson sums up his conclusions as follows: “5th. We have shown that the phenomenon of which the subjective aspect is called ‘memory’ is of a nature indicating that the autocatalyzed chemical reactions form the mechanism conditioning the response of the central nervous system to stimuli. “6th. In admitting that the extent of the trace of memory may be proportionate to the mass of a product of an autocatalyzed chemical reaction unfolding itself in the central nervous system as the result of the application of a stimulus, we have shown that the relation which one theoretically deduces between the mass of memory material and the number of repetitions corresponds to that which has been found by experience.