of the vertebrates with their most highly developed conductivity, and which will be analyzed in somewhat greater detail. How are we to explain their decrementless conductivity? When we study the decrement of the excitation wave in the series of living substances, before alluded to, we see that this reduces with a progressive increase of irritability. Consequently the extreme irritability of the nerve is a manifestation of its decrementless conductivity. If we study the course of a process of excitation and its conduction in its molecular details, the fact of the decrementless conduction indicates that in excitation, produced by a stimulus, the same number of specific molecules capable of disintegration are broken down in the same manner at every following cross section, as at the point of stimulation; or in other words: an equal amount of energy is set free at every cross section, which, in its turn, acts as stimulus to the next, etc. Such a condition presupposes, however, in an elementary fiber of the nerve, that by the conduction of the wave of excitation from cross section to cross section, all those molecules capable of disintegration are broken down. If it is assumed that the entire number of molecules capable of disintegration do not break down, but only a certain per cent. of the same, then it would not be possible to conceive of a molecular structure of the nerve in which this would take place without decrement of the wave of excitation. With the assumption of a generally homogeneous molecular structure (Figure 23, a) of the elementary fibers it would be entirely incomprehensible how, with the decrementless extension of the excitation, individual molecules capable of breaking down could escape disintegration. If, on the contrary, the molecular structure is not homogeneous it only is possible to explain a conduction, on each cross section of which an equal per cent. of irritable molecules break down, by the hypothesis that the irritable molecules are in their turn ordered in fiber-shaped series (Figure 23, b) within the elementary fiber and are thus protected to a certain degree from one another and from transverse conduction of excitation. This hypothesis would, therefore, only mean that the elementary fiber is not such in reality and would thus transfer the difficulty to the ultimate fiber unit, for which a homogeneous molecular structure would have to be presumed. In short, whatever may be the assumption on which molecular structure of elementary fibers is based, the fact of the decrementless conduction peremptorily demands, from the physical standpoint, that from cross section to cross section the entire number of irritable molecules are broken down. This conclusion is highly important, for it indicates very clearly that the “all or none law” is applicable to the nerve. [Illustration: Fig. 23.]