acquire the energy requisite to discharge the muscles to which the motor-fibre carries them. In vertebrate animals, sensory nerves are branches of neurones of which the cell-bodies lie in cranial or spinal ganglia. They resemble the ganglion-cells of the leech in as much as they are unipolar; both branches, the one which collects impulses from sense-organs, and the one which distributes them to the spinal cord, come off from the cell in a common trunk which afterwards divides, although the unipolar condition of the cell of the spinal ganglion is not primitive, but acquired. In the earliest stages of its growth the cell is bipolar. Its two ends subsequently grow together for a certain distance, the common portion being the vertical limb of the =T= (_cf._ Fig. 21, which shows the growth of a granule of the cerebellum). The body of the cell contains a network not unlike the network of the leech. It is probably related to what may be termed the charge of the neurone, the development of a suitable degree of force in the impulses which pass through it. The neuro-fibrillæ of a large nerve-cell, such as a motor cell of the spinal cord, are exceedingly slender (Fig. 22). They branch and reunite. A certain number gather towards the axon; but the majority pass through the cell from one dendrite to another, or from one branch of a dendrite to another branch. It is very tempting to suppose that neuro-fibrillæ are connected with conduction. When first discovered they were regarded as conducting strands; but it is evident that they are not comparable with telephone wires or other isolated or separate conductors. There are good reasons for regarding dendrites as collecting processes, taking up impulses from the end-twigs of the nerves which branch in the grey matter around them, passing them through the cell-body into the axon. The continuation of neuro-fibrillæ from dendrite to dendrite seems to be irreconcilable with the hypothesis that they are disposed in the lines of conduction.