III. On the structure of the so-called apolar, unipolar, and bipolar nerve-cells of the frog

Abstract The author adverts to the opinion generally received with regard to the existence of apolar, unipolar, bipolar, and multipolar nerve-cells, and observes that if cells having such very different relations to the nerve-fibres they are supposed to influence, as apolar, unipolar, and multipolar cells, do actually exist, as many different kinds of action must be admitted. For it is hardly likely that a nerve-cell unconnected with any fibre can affect the fibres at a distance from it in the same way as a cell acts upon fibres which are in structural continuity with it. Neither is it probable that a cell with but one fibre proceeding from it can constitute an organ which acts upon the same principle as the cell from which two or more fibres proceed. If no fibre, or but one fibre proceeds from certain cells, the formation of complete nervous circuits, at least in these instances, is impossible; and if it be admitted that circuits do not exist in every case, a strong argument is advanced against the existence of such complete circuits as a necessary or fundamental condition of a complete nervous apparatus.