In the gray substance there is still greater difficulty in tracing the course taken by the various kinds of impulses, and little is known on the subject beyond what is surmised from the proximity of the different parts to the anterior and posterior roots and to the white channels the function of which is known.

Though the attempt to localize the different functions to any anatomical region has not met with success, histology has taught us of the existence of certain groups of cells which, when viewed longitudinally, may be called vesicular columns. Of these, four may be named as distinctively marked. (Fig. 247).

1. The anterior (motor) cellular columns occupy the gray matter seen in sections of the cord as the anterior cornua. They extend throughout the entire length of the cord, the cells being specially numerous where the large motor roots come off. The cells are characterized by their great size and the number of their branches, one of which forms the axis cylinder of one of the motor fibres passing to the roots of the spinal nerves.

2. The posterior cellular columns situated in the gray matter of the posterior cornua are much less obvious than the anterior. The cells are few, small and mostly spindle-shaped. Their processes are not readily traced to the roots of the spinal nerves.

3. The posteromedian cellular column (Clarke) lies on the median side of the posterior gray column, so that it forms the inner part of the posterior cornua near its base. The cells are numerous, but much smaller than those of the anterior vesicular column. Clarke's column is best developed at the junction of the lower dorsal and upper lumbar nerves. It tapers off above and below, and the cells cease to form a continuous column opposite the seventh cervical nerve. But scattered groups of cells in a corresponding position are found throughout all the cervical cord, and seem to link this spinal column with the vagus nucleus in the medulla.

4. The intermedio-lateral cellular columns lie in the lateral concavities seen on section between the anterior and posterior gray cornua. They thus occupy a position between the lateral white column and the central part of the gray matter. They are best marked in the dorsal region, as they seem fused with the cells of the anterior cornua in the lumbar and cervical enlargements.

The facts that cells functionally related are grouped in masses at the points where the spinal nerves arise, and that the various regions of the cord can respond to stimulus when severed from the rest, seem to indicate that a strict homology exists between the spinal centres of vertebrate and the central nervous system in many of the lower animals, which consists of a double chain of ganglia, united together by conducting channels.

We may then suppose the gray matter of the spinal cord to be made up of a series of segments, corresponding in number to the vertebral development, fused together into one continuous organ. These segments may be supposed to receive the afferent impulses from corresponding parts of the body, and send efferent impulses to muscles capable of moving that part, just as the separate ganglia of the invertebrate chain preside over the functions of the corresponding somite of the animal's body.

The various groups of cells in the spinal cord are in more or less direct union with the roots of the nerves and the conducting fibrils of the cord itself, so that they participate in the transmission of the impulses to and from the centres situated in the brain. In the transmission of these impulses the cells seem to have a certain directing and controlling influence which deserves special attention, as it gives us the key to the more complex mechanisms of the higher centres. Although the various powers exerted by the cells of the spinal cord are so intimately associated together as to be practically inseparable, it is found convenient to consider their functions under distinct headings.