The Cord and Spinal Nerves

The cord and spinal nerves are so intimately related that it is difficult to obtain a clear understanding of the one without some knowledge of the other. They will therefore be considered together.

Form of the Cord

The human nerve cord may be used as the basis for the description of the cord, not only because it is of more immediate interest than that of other vertebrates, but also because much more is known about it.

With its ensheathing envelopes removed it is seen to be a slightly flattened tube, having a shallow median furrow running down its dorsal length (Fig. 605), and another wider and deeper groove along the middle of the ventral side. Other less conspicuous longitudinal furrows are also present, giving the whole cord somewhat the appearance of a fluted column.

Diagram of the human cord, dorsal view, showing enlargements and filum terminale

At the anterior end where the cord passes over continuously into the brain, it appears broadened and somewhat oval in cross section, but at the posterior end it tapers rapidly and finally terminates in a non-nervous threadlike prolongation, the filum terminale, into which in early fetal stages the central canal of the cord still extends.

In two regions, at the level of the arms and the legs, the cord becomes swollen to an increased size. That these enlargements are associated with the increased nerve supply of the paired appendages is indicated by the fact that they are absent during embryonic growth before the limbs develop. In flying bats, as would be expected, the anterior far exceeds the posterior enlargement in size, while in leaping kangaroos, which have rudimentary dangling fore legs and powerful hind legs, the reverse is true.

Certain gigantic dinosaurs of the remote past, that were propped up upon a colossal underpinning of pillarlike legs, for example Stegosaurus, actually had a considerably greater diameter in the sacral region of the cord than in the brain itself, which inspired the famous columnist “BLT” of the Chicago Tribune to write:

“Behold the mighty dinosaur
Famous in prehistoric lore
Not only for his power and strength
But for his intellectual length.
You will observe by these remains
The creature had two sets of brains -
One in his head (the usual place),
The other at his spinal base.
Thus he could reason a priori
As well as a posteriori.
No problem bothered him a bit.
He made both head and tail of it.
If something slipt his forward mind
’Twas rescued by the one behind.
And if in error he was caught
He had a saving afterthought.
Thus he could think without congestion
Upon both sides of every question.
Oh, gaze upon this model beast,
Defunct ten million years at least.”

Extent of the Cord

In a typical human adult the nervous part of the cord from the level of the foramen magnum of the skull to the beginning of the non-nervous filum terminale, reaches only about eighteen inches, or to near the level of the first lumbar vertebra. Thus the entire axial central nervous system, including the brain and the functional part of the cord, extends only from the region of the forehead to the “small of the back.” Embryonically as well as phylogenetically the cord is originally practically as long as the backbone itself.

In Ornithorhynchus and a few rodents it reaches as far as the sacrum, but there is a pronounced evolutionary tendency for it to shorten.

Nervous system of a honeybee, larva and adult, compared to show the concentration of ganglia

Among insects the process of cephalization, or shrinkage in the length of the cord, is very striking. The “lower” insects have a primitive nerve chain extending along the entire floor of the body, whereas “higher” forms present a condensed cord, composed of ganglia that have telescoped together into compact masses. During the metamorphosis of an insect, such as a honey bee for example, a ganglionic chain stretches the whole length of the long larval body, but in the adult worker bee, which eventually emerges from the larva, the nerve chain is represented by ganglionic masses that are drawn much more closely together (Fig. 606).

In primates the degree of shortening marks the relative position in animal aristocracy which each form occupies, as the following table shows:

Relative position in animal aristocracy

It is interesting to speculate as to the possible fate of the shrinking nerve cord in the far distant future. Probably the time will never come when the cord will be entirely dispensed with, since there must always remain a structure of sufficient length to furnish a point of departure for the nerves supplying the body.

Spinal Nerves

In man there are typically 31 pairs of spinal nerves, as follows: cervical, 8; thoracic, 12; lumbar, 5; sacral, 5; and caudal, 1. The total number of spinal nerves in vertebrates other than man is naturally subject to wide variation.

Each pair of spinal nerves, with the exception of the cervical ones, takes its name from the vertebra just anterior to its exit. The reason for a reversal of designation in the cervical nerves is that the first pair emerges between the skull and the first cervical vertebra, or atlas, and so, having no vertebra in front of it, takes its name from the vertebra immediately behind it. This method of nomenclature is followed throughout the cervical series, leaving the eighth pair of nerves without any cervical, vertebral godfather standing behind it, and accordingly it is called the eighth pair, although mammals have only seven neck vertebrae.

The next pair of nerves plainly belongs to the trunk, and so begins a new series as the first thoracic pair, taking its name from the vertebra behind which it emerges (Fig. 607).

Diagram showing the relation of the cervical nerves to the cervical vertebrae. Cross section of the central and peripheral nervous systems in man

The size of the spinal nerves is dependent upon the relative area of the body which they supply, the largest in man being the first sacral pair that goes to the legs. The general distribution of every spinal nerve in connection with the region which it supplies is indicated in Figure 608.

The relation between the spinal nerves and the metameric embryonic muscles is very regular and constant, but the rearrangement and migration of these muscles in the adult organism brings about apparent irregularities. A nerve once associated with a muscle, however, remains faithful to it throughout all its subsequent transformations. Since metameric muscles in mammals do not extend to a tail, as in fishes and urodeles, decided differences in the abundance of post-anal spinal nerves appear in the higher vertebrates.

The spinal cord of man, showing plexi, cauda equina, and, on one side, the sympathetic nerve chain

In man particularly, the nerve cord is shortened as already described, so that it does not extend through the entire length of the vertebral column. Nevertheless the emerging spinal nerves maintain their proper intervertebral exits. This necessitates the continuation of the posterior pairs of spinal nerves within the neural canal of the vertebral column for increasingly lengthened distances before they finally emerge, with the result that a brush of spinal nerves, the cauda equina, or the “horse’s tail,” is formed at the end of the cord (Fig. 609). Moreover, while the anterior spinal nerves leave the cord at practically right angles, the angle of departure for posterior pairs becomes more and more acute, until it may be said of the sacral nerves that for some distance they run almost or quite parallel to the filum terminate, that is, to the non-nervous continuation of the cord itself.


The connection between the peripheral nervous apparatus and the central nervous system is effected on either side through the dorsal and ventral roots of the nerves (Fig. 608), which differ not only in structure and function, but also in their origin and manner of development.

Dorsal roots are composed primarily of neurons whose polarity is afferent or centripetal, that is, toward the central nervous system. Ventral roots, on the contrary, are efferent, or centrifugal, carrying messages outward to glandular or muscular effectors. Since these effectors, which are mostly muscular, bring about motion, the ventral roots are commonly designated as “motor roots” (Fig. 610).

Diagram of the relation of spinal nerves to the cord

For the most part dorsal roots are made up of fibers from neurons apparently unipolar, although in reality bipolar, located in the dorsal ganglia, whose origin from the neural crests has already been described. Neurites in the ventral roots, on the other hand, which have no ganglia, take their origin from cell bodies grouped together in the gray matter of the ventral area of the cord.

Among lower vertebrates, dorsal roots in some instances are not always purely sensory but may also contain efferent fibers going to the autonomic system. A few recurrent sensory fibers also may find their way from the dorsal to the ventral root, thus modifying the character of the motor roots.

The spinal nerves of amphioxus show several primitive features. The dorsal root is mixed, including visceral (autonomic) motor fibers as well as the usual complement of sensory elements. No dorsal ganglion is present, the sensory neurons having their cell bodies in the dorsal part of the cord. Further the dorsal and ventral roots, alternating in their attachment to the cord, do not unite into a spinal nerve.

In cyclostomes the dorsal roots are also mixed and alternate with the ventral roots, but, in contrast with amphioxus, the cell bodies of their sensory fibers are located in dorsal ganglia, as in all other vertebrates. In the myxinoids (but not in the lamprey eels) the two roots join to form a united spinal nerve as in all gnathostomes. The combined spinal nerve, therefore, probably represents a secondary evolutionary adaptation.

Diagrammatic section through a vertebra at the level of the roots of a spinal nerve

The dorsal and ventral roots in elasmobranch fishes also emerge from the cord alternately. The dorsal root makes an intervertebral exit, through the intercalary plate, while the ventral root comes out vertebrally through a foramen which penetrates the neural plate. They then join to form a spinal nerve. In higher vertebrates both roots of a spinal nerve emerge intervertebrally in the same transverse plane.

In fishes the dorsal ganglia and the point of union of the roots of the spinal nerves are usually located outside the vertebral column, but in other vertebrates the junction of the roots is effected closer and closer to the cord itself, so that it comes to lie within the neural arch (Fig. 611).


Near the junction of the two roots the spinal nerve divides into four branches, or rami, namely: (1) a thinner, shorter dorsal branch, supplying epaxial sense organs, musculature, and glands; (2) a thicker, longer ventral branch, to similar hypaxial parts; (3) a small meningeal branch, going back into the neural canal to supply the blood vessels of the cord and its membranes; and (4) an autonomic branch, communicating with the autonomic nervous system (Fig, 608).

The important generalization that the branches of spinal nerves are mixed, while their roots are not, was independently established over a century ago by the Frenchman Magendie (1783-1855) and by Sir Charles Bell (1774-1842), an Englishman, who has been characterized by Professor Keith as “an anatomical detective of the highest rank.”


Whenever an extra large area of the body, such as that of the arms or legs, is to receive a nerve supply, several spinal nerves may join forces, forming a plexus. A complication of this kind adds materially to the difficulty in precisely tracing out the actual path and sequence of neuronic lines between the central nervous system and specific end organs. It has been shown, however, that each muscle may be excited by several nerves, while any single spinal nerve may in turn affect several muscles.

The first four cervical nerves in man form the cervical plexus that supplies the neck. The last four, together with the first and usually the second thoracic nerve, make up the brachial plexus of the arm (Fig. 612). Most thoracic nerves do not ordinarily enter into plexus formation. The twelfth thoracic, all five lumbars, and the first three or four sacrals become involved in the large lumbosacral plexus which supplies the leg.

Diagram of the right brachial plexus of man, viewed from the front

In all of these plexuses only ventral rami are concerned, including both sensory and motor fibers.

In various species of long-bodied fishes and urodeles, which do not use their paired appendages for the support of the body, there frequently occurs, anteriorly and posteriorly along the backbone, a shifting of the girdles and their attached appendages (particularly the pelvic girdle), with a consequent variation in the group of spinal nerves that take part in plexus formation. DeBeer records this fact by saying “Limbs may become transposed over the trunk of the animal much as a tune can be transposed over the keys. But it is the same tune and the same limb.”

Pelvic nerves of a snake, showing a simple plexus in the absence of hind legs. The limbless embryo of a snake

There are many variations in the composition of spinal plexuses in vertebrates other than man, in which the legs and arms assume particular Importance. In such instances the anastomoses may even be unlike on the two sides of the same individual. In the enormous brachial plexus of the skate, twenty-five spinal nerves may fuse together, while in certain other elasmobranchs as many as three occipital nerves may join with spinal nerves to form a cervicobrackial plexus. In the long-necked swan the first spinal nerve to join with another in the brachial plexus is the twenty-second.

Certain snakes (Fig. 613) in which no trace of limb-buds appears in the embryo (Fig. 614), possess a lumbosacral plexus, indicating that these highly modified, limbless reptiles were derived from ancestors with legs at some time in the remote past.

Development of the Spinal Cord

With the formation of the medullary tube there are established around the central canal, above and below, roof and floor plates, respectively, which do not increase notably in thickness, and two sides that enlarge enorously until they overshadow both roof and floor (Fig. 615).

In early stages there are two types of cells in the wall of the medullary tube, namely, germinal cells, near the central canal, and spongioblasts, extending from the canal to the outer surface of the cord. Soon several additional types appear, namely, non-nervous ependymal cells; non-nervous neuroglia cells; and neuroblasts, destined to become the important neurons. Germinal cells may give rise to any of these five types with the possible exception of ependymal elements, while spongioblasts differentiate into neuroglia and ependymal cells.

Transverse section through anterior cervical region of spinal cord

Three zones may now be recognized in the wall of the medullary tube, especially in its thick lateral regions, namely: (1) an inner ependymal layer, bordering directly on the central canal; (2) a middle mantle layer, containing many nuclei; and (3) an outer marginal layer, practically devoid of nuclei (Fig. 615).

The spongioblasts, with the ependymal cells, form a supporting framework which extends through all three layers. Although the ependymal cells are concentrated in the inner layer, they send long slender processes to the outer surface of the cord. The spiderweb-like neuroglia cells form a supporting meshwork in the middle and outer layers.

Germinal cells are especially abundant within the sustaining neuroglial network of the mantle layer where they divide repeatedly to produce larger numbers of neuroblasts. The thickening of the mantle layer, due to this rapid cell proliferation, is most marked in two regions on each side of the cord, one dorso-lateral forming the dorsal horn, the other ventro-lateral, the ventral horn (Fig. 616).

Cross section through the fourth thoracic segment of the human spinal cord

Many neuroblasts, when they begin to sprout, send their nerve fibers into the marginal layer. At first these fibers are naked but gradually myelin sheaths are added, probably through the activity of certain neuroglia cells which become temporarily associated with the fibers but do not remain to form a neurolemma. This massing of myelinated fibers in the marginal layer converts it into white matter in contrast with the middle layer, composed chiefly of neuroblasts, cell bodies of neurons and neuroglia, which remains gray matter. Eventually the inner zone becomes reduced to a thin ependymal layer lining the central canal.

Neurons of the ventral gray matter also send out neurites which establish the ventral roots of the spinal nerves and continue into the several branches. Along with the fibers go non-nervous cells which give rise to the two sheaths, myelin and neurolemma. At the same time neurons in the dorsal ganglia are sprouting. 'Their dendrites, accompanied by sheath cells, grow out through the several branches of the spinal nerves to the surface of the body or to the internal organs. Their neurites complete the dorsal roots and extend into the cord where many of them grow forward to the brain by way of the dorsal part of the white matter. These myelinated neurites also have a neurolemmal sheath up to the point where they enter the cord.

Thus the alar plate, the dorsal part of the lateral wall of the cord, is given over to cells and fibers of afferent, or sensory, neurons, concerned with bringing in impressions from without, while the ventral part, or basal plate, becomes headquarters for efferent, or motor, neurons, that convey messages outward to various effector mechanisms.

As the sensory fibers enter the cord those from the outer surface of the body, somatic sensory component, are more medial than those from the internal organs, visceral sensory component (Fig. 617). In like manner the cell bodies of the somatic motor component, which supplies the voluntary muscles, are in the more medial ventral horn while those of the visceral motor component, running chiefly to the autonomic nervous system, are in the lateral column, the more lateral part of the gray matter of the basal plate. Thus the gray matter includes four kinds of neurons, or components, in sequence from dorsal to ventral as follows: somatic sensory, visceral sensory, visceral motor, somatic motor.

Diagrammatic transverse section through the human spinal cord, showing the location of the four types of neurons

In cross section the nerve cord at first has an oval outline, while the central canal appears like a dorso-ventral slit (Fig. 615). With the formation and enlargement of the ventral horns, together with the addition of the surrounding white matter, the cord bulges ventrally on either side of the thin floor plate leaving a median longitudinal groove, the ventral fissure (Fig. 616). The formation of less extensive dorsal bulges results in a shallow dorsal fissure. Further the two sides of the dorsal part of the central canal grow together, giving rise to a dorsal median septum, composed of ependymal cells, which extends from the bottom of the dorsal fissure nearly to the central canal. The septum is separated from the canal by a thin sheet of non-myelinated nerve fibers which grow across between the gray matter regions of the two sides, forming the dorsal commissure. A somewhat similar cross connection between the two sides of the cord is found ventral to the canal, but this ventral commissure consists of a white portion, next to the ventral fissure, in addition to the gray part near the canal.

Thus the nervous elements of the two sides of the cord are almost completely separated from one another, dorsally by a fissure and septum, ventrally by a fissure. Only the commissures act as nervous bridges between the two sides.

Internal Architecture of the Adult Cord

In cross sections of the fully formed cord the gray matter somewhat resembles a butterfly or a capital letter H with a hole (the central canal) through the center of the crossbar (the gray commissures). On each side the projecting horns of the gray matter, the upper and lower parts of one upright of the letter H, separate the white matter into three large masses, the dorsal, lateral, and ventral funiculi. These funiculi are further differentiated into parallel tracts, or fasciculi, each of which is a group of fibers of common function and with their cell bodies in the same general location. These fibers arise from cell bodies located (1) in the dorsal ganglia; (2) in the cord itself; and (3) in the brain. Together they constitute path-ways of communication between more or less distant parts of the entire body.

The determination of the facts now known concerning these tracts has been a difficult though fascinating work, engaging many different investigators over a long period of time. Nevertheless our knowledge in these matters is all too limited. In tracing the course of the path-ways over which the countless messages of communication go, workers have employed four outstanding methods, namely, Wallerian degeneration, study of pathological changes, observation of the order of myelinization, and electrical stimulation.

As long ago as 1852 Waller found out that when a fiber is cut the part of the individual fiber separated from the nucleus degenerates. This serves to indicate in which direction the nucleus of a nerve cell is situated. Animals whose nerve fibers have been allowed to degenerate following a particular type of cut are observed for motor paralysis, loss of sensation, or both, in any region of the body. These experiments reveal the nature of the fibers involved, the location of their cell bodies and the region or regions of the body which they supply. This type of research is aided by methods of differential staining which render fibers and sheaths visible.

Pathological conditions, especially in man, furnish similar information. Cadavers of individuals who have suffered either loss of sensation or motor paralysis during life may, upon autopsy, reveal what group of cell bodies or fibers was concerned with the symptoms.

Another method (Flechsig’s) of determining the course and direction of fiber groups is dependent upon the fact that the nerve fibers, which are at first naked, do not all acquire their medullary sheaths at the same time. Because most of the fibers belonging to any one tract tend to myelinize concurrently, it is possible to recognize different pathways by this means.

The fourth method, employed considerably in recent years, involves the use of electrical stimulation. Minute electrodes are used to stimulate a limited group of nerve cells in anaesthetized animals. After tabulation of the effects produced in numerous experiments of essentially one type, it is possible to determine at least some of the connections of the stimulated areas. By this means it has been possible to obtain detailed information concerning many complicated pathways, especially those in which only a few or scattered fibers are involved and therefore other methods do not succeed.

Thus the details of the invisible individual highways of the nervous system have been pieced together accumulatively, until now quite a complete picture of the whole complicated system of nervous traffic lanes is available. As would be expected, the various paths in the Great White Way of the nerve cord are much more definite and have been more thoroughly studied in higher groups than in lower vertebrates. Representative tracts, as found in man, may be grouped as follows:

I. Ascending (Sensory) Tracts
1. Dorsal Columns
  Fasciculus gracilis (Tract of Goll)
  Fasciculus cuneatus (Tract of Burdach)
2. Spinocerebellar System
  Dorsal spinocerebellar tract Ventral spinocerebellar tract
3. Spinothalamic System
  Lateral spinothalamic tract Ventral spinothalamic tract
II. Descending {Motor) Tracts
  Fasciculus corticospinalis lateralis (Crossed pyramidal tract)
  Fasciculus corticospinalis ventralis (Direct pyramidal tract)
III. Association Tracts
  Fasciculus proprius dorsalis (Dorsal ground bundle)
  Fasciculus proprius lateralis (Lateral ground bundle)
  Fasciculus proprius ventralis (Ventral ground bundle)

Ascending Sensory Tracts

The fibers which give rise to the so-called dorsal columns are neurites coming from cell bodies located in the dorsal ganglia of the spinal nerves, outside of the cord. As they come into the cord by way of the dorsal root of the spinal nerve, they branch and go both anteriorly and posteriorly, with the longer branch toward the head, and the shorter branch away from the head. Both the longer ascending craniad fibers and the shorter descending caudad fibers send off collaterals that enter the gray matter, where they form synapses with either intermediate neurons or motor neurons. Many of the cranially directed fibers reach without relay as far as the brain itself.

Thus a neurite entering the cord through a dorsal root makes numerous central connections many of which are through its large number of collaterals (Fig. 618). Some collaterals terminate in the ventral gray matter where they synapse with motor neurons which run out through the ventral roots to muscles or to the autonomic nervous system. Other collaterals, ending in the dorsal gray matter, synapse there with intermediate neurons, the cell bodies of which are in this part of the gray matter. The neurites of these intermediate neurons usually run into the white matter where they may enter one of several tracts, either ascending or association. Still other collaterals run through the dorsal commissure to end in the dorsal gray matter of the opposite side of the cord.

The various central connections, within the cord, of sensory neurons

The fasciculus gracilis and fasciculus cuneatus, which make up most of the dorsal funiculus, are composed mainly of ascending branches of the neurites of sensory neurons (Fig. 619). Fibers from sacral, lumbar, and lower thoracic spinal ganglia form the fasciculus gracilis, while those from anterior thoracic and cervical ganglia constitute the cuneatus. In the cervical region these two tracts are separated from one another by the dorsal intermediate septum. The courses which these and other tracts take upon entering the brain will be considered later. Although some of the short descending branches of the sensory neurons may also run in these two tracts, most of them take other pathways.

Some of the intermediate neurons with which the collaterals synapse, as mentioned above, run into the spinocerebellar tracts, located chiefly near the surface of the cord in the lateral funiculus. The dorsal spinocerebellar tract arises from cell bodies located in the dorso-median gray matter (nucleus dorsalis, sometimes known as Clarke’s column) of the same side of the cord. The cell bodies of the fibers making up the ventral spinocerebellar tract are found in the dorsal and intermediate gray matter, of the opposite side of the cord as well as the same side.

The principal tracts of the cervical region of the human spinal cord

Other intermediate neurons, with which collaterals join in the dorsal gray matter, run through the ventral white commissure to enter the spinothalamic tracts, which are therefore crossed tracts, being composed of fibers which have their cell bodies on the side opposite that in which they run. The ventral spinothalamic tract is at the surface in the ventral funiculus, while the lateral spinothalamic tract lies just medial to the ventral spinocerebellar tract.

The primary sensory neurons associated with the lateral spinothalamic tract differ in two important respects from those connected with the other tracts. First of all, they are entirely unmyelinated. In addition, they do not send branches up and down the cord but, instead, enter the gray matter almost at once to synapse with neurons of the second order whose fibers then run into the white matter where they enter the lateral spinothalamic tract.

We may further observe that of these six ascending tracts the fasciculi gracilis and cuneatus have their cell bodies in spinal ganglia, while the other four have theirs in dorsal gray matter. Also, the spinothalamic tracts and some fibers of the ventral spinocerebellar are “crossed,” while the other three have their cell bodies on the same side as that in which they run.

These tracts also differ in function. Impulses from “muscular-sense” organs in muscles, joints, and tendons and those from sense organs of touch in the skin are carried by sensory fibers which run at least a short distance in the fasciculi gracilis and cuneatus. These impulses are carried to the brain through three different paths, namely: (1) the original fibers of these two fasciculi; (2) the spinocerebellar tracts; and (3) the ventral spinothalamic tract. In the case of “muscular-sense,” most of the impulses going to conscious levels are carried to the brain in the original fibers, while those going to the subconscious areas of the cerebellum are transferred to the spinocerebellar tracts. Tactile impulses may also be carried long distances in the original fibers, but most of them are transferred to the ventral spinothalamic tract before reaching the brain. The pathways for sensations of pain, heat, and cold go at once into the gray matter, then across to the opposite side, and forward in the lateral spinothalamic tract.

Descending Motor Tracts

The cortico-spinal system, on the other hand, runs from the brain to the cord, constituting a motor transmission highway that reaches its highest expression in mammals. Arising from cell bodies in the cortex of the cerebrum in the anterior part of the brain, it grows down by means of neurites into the cord, eventually reaching the ventral horn of the gray area to synapse with the neurons there whose fibers extend outward through the ventral roots of the spinal nerve to the effector mechanisms.

Not all of the cortico-spinal fibers follow the same course. Most of them, upon reaching the posterior part of the brain (medulla), cross over from the side of their cortical origin to enter the cord on the opposite side, forming the crossed pyramidal tracts, which are bordered on the outside by the spinocerebellar bundles of the lateral funiculi (Fig. 619). These tracts together form “a great motor strand which brings the spinal motor apparatus under the control of the will” (Cunningham). Other fibers, fewer in number, continue from cortex to cord on the same side of their origin, forming the direct pyramidal tracts, that come to lie on either side in the ventral funiculus close to the ventral fissure. These latter tracts are found in only a few mammals including the anthropoid apes and man. Just before the fibers composing them terminate, they also cross over, through the white commissure of the cord, and synapse with motor neurons of the side opposite from that of their origin, so that finally all neurons of the cortico-spinal system make connections with effector neurons on the side of the body opposite to that occupied by the cell bodies from which they originate in the brain.

Other lesser descending bundles run from different parts of the brain to the cord where they synapse with motor neurons. These tracts aid in the complicated regulation of the effectors of the body.

Association Ground Bundles

Earlier we described simple intrasegmental reflex arcs, one segment and involving one sensory, one motor, and usually one association neuron. We have also shown how intersegmental reflex arcs may extend over segments in addition to the one through which the sensory neuron entered. In the latter type neurites of the sensory neurons ran in the dorsal columns giving off collaterals at various levels. These collaterals in turn went into the gray matter to carry impulses to motor neurons either directly or through an association neuron.

Another type of intersegmental reflex arc is that in which impulses are transmitted from one sensory neuron to motor neurons in a number of segments through an intermediate neuron. A collateral from a sensory neuron enters the gray matter where it synapses with the short dendrite of an intermediate cell. The neurite of this cell runs into one of the white ground bundles lying immediately around the gray area. These bundles, or fasciculi propni, are found in all three funiculi where they form an almost continuous layer around the gray matter (Fig. 619). A neurite entering one of these bundles splits into ascending and descending branches each of which runs for only a few segments (Fig. 620). Along their courses these fibers give off collaterals which return to the gray matter where they synapse with motor neurons. In this manner reflex responses are spread over several segments through association neurons.

Stereogram of the spinal cord showing a diffuse reflex of one side of the body

As in the case of intrasegmental reflex arcs and the intersegmental type in which the different levels are reached through sensory neurites, this type involving proper fasciculi is not limited to one side of the cord. Some of the intermediate fibers cross in the ventral white commissure to send impulses out through motor neurons of the side of the body opposite that to which the sensory neuron belonged. In these various ways the two sides of the body and the various segments are interconnected so that a response is by no means limited to a localized area. Under ordinary circumstances, however, the response will not be diffuse as in the case of nerve-net transmission. Instead the impulses pass along appropriate pathways as determined by synaptic connections within the cord.

It is evident from the above description that motor neurons may synapse with fibers coming from a variety of places. In fact a single motor neuron with its much-branched set of dendrites may have impulses funneled to it through several primary sensory neurons as well as by fibers from fasciculi proprii, corticospinal tracts, and other tracts which space has prevented us from discussing. As Sherrington has so aptly said, a motor neuron is “the final common path” of many reflex arcs.

Also it is evident that the cell bodies of the fibers which make up the tracts of the spinal cord have three different locations: (1) spinal ganglia, giving rise to the fasciculus gracilis and fasciculus cuneatus, as well as many collaterals which enter the gray matter of the cord; (2) gray matter of the cord, giving rise to the three fasciculi proprii (dorsalis, lateralis, and ventralis), the dorsal and ventral spinocerebellar tracts, and the lateral and ventral spinothalamic tracts; and (3) the brain, giving rise to the crossed and direct pyramidal tracts.

Diagram showing by curves the area, at several levels of the cord

The pattern of the gray and white matter, as well as the actual size of the whole cord, varies greatly at different levels in any individual. Since the white is the transmission highway between the cord and the brain, it naturally becomes cumulatively larger the nearer it approaches to the brain itself. Relative increase in the white matter at different levels within the cord of mammals is explained as machinery that makes possible more reliance on the brain in regulating the behavior of the animal.

The relative amount in square millimeters of the white and gray matter throughout the human cord is shown by the curves plotted in Figure 621.

Comparative Anatomy of the Cord

The cord of amphioxus resembles that of the vertebrates in having an inner region of cell bodies outside of which are massed the fibers. The entire cord is made up of gray matter, however, as no medullary sheaths are laid down in this animal. Some of the cell bodies of sensory neurons are located in the cord, while others are distributed along the dorsal roots, never being grouped into compact spinal ganglia. The central canal is a narrow slit with only a thin roof-plate, as in vertebrate embryos (Fig. 622).

Cross section through the spinal cord of amphioxus, showing location of cell bodies

Cross sections of the broad, flat cord of cyclostomes show clearly the three zones which are typical of vertebrates (Fig. 623). The nearly circular central canal is surrounded by large ependymal cells. Then there is a middle zone of nerve cell bodies outside of which is a peripheral layer of fibers, non-medullated as in the case of amphioxus. The cell bodies are not confined to the vicinity of the central canal but are distributed in a broad band that extends into the lateral parts of the cord. Although the cell bodies of some sensory neurons are located in the cord most of them are in dorsal root ganglia as in higher vertebrates. The neurites of these sensory cells run short distances in the dorsal part of the cord giving off collaterals into the gray matter. In general the cord is for local reflexes but there are a few very large motor fibers of Muller which run from the brain to the posterior end of the cord.

The cartilaginous fishes show definite advances toward the plan of higher vertebrates. For the first time myelinization of the fibers occurs, forming true white matter. Further the gray matter shows an arrangement into dorsal and ventral horns, although the dorsal horns are usually combined into a single broad region so that the gray matter is shaped like an inverted T rather than like an H as in mammals (Fig. 624). Instead of being arranged in compact dorsal funiculi, the sensory neurites form small bundles scattered through the dorsal gray matter. Also these sensory fibers are usually short, not extending into the brain region as they do in mammals. Compared with cyclostomes there has been an increase in the number of tracts set up between brain and cord. There are ascending fiber groups, probably corresponding to the dorsal spinocerebellar tract, as well as several descending tracts from the medulla and mid-brain.

Cross section of the spinal cord of a cartilaginous fish

The amphibians are the first to have brachial and lumbar enlargements, which make their appearance as the result of the development of the legs. As large dorsal funiculi are laid down, the gray matter assumes the butterfly shape. Various tracts have been described including the dorsal spinocerebellars and the probable forerunners of the spinothalamies, but the brain still exerts only limited influence upon spinal reflexes.

Reptiles have well developed brachial and lumbar enlargements except in the limbless snakes. In turtles the cord is very thin in the thoracic region, a condition probably due to the almost complete absence of thoracic musculature. Reptiles are the first vertebrates in which the dorsal funiculi carry great numbers of fibers to the brain and therefore increase in size as they go anteriorly.

The mammalian cord, unlike that of reptiles, birds, and most anamnia, never extends the entire length of the vertebral column. The great development of the dorsal funiculi, which began in reptiles, becomes progressively greater in passing from the lower to the higher mammals. The corticospinal tracts, from the cerebral cortex, are peculiar to mammals. Only the crossed tract has been described for most mammals; but the ventral, uncrossed tract is said to occur in rodents, cetaceans, and some ungulates, in addition to the primates. Although the crossed tract is ordinarily in the lateral funiculus, as it is in man, it is found in the dorsal funiculus in monotremes, marsupials, rodents, and ungulates.