Paper - Development of the spinal reflex mechanism in human embryos

From Embryology
Embryology - 25 May 2024    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Windle WF. and Fitzgerald JE. Development of the spinal reflex mechanism in human embryos. (1937) J. Comp. Neural. 493-509.

Online Editor  
Mark Hill.jpg
This historic 1937 paper by Windle describes early human neural development.

See the links below for the current notes pages.

Neural Links: ectoderm | neural | neural crest | ventricular | sensory | Stage 22 | gliogenesis | neural fetal | Medicine Lecture - Neural | Lecture - Ectoderm | Lecture - Neural Crest | Lab - Early Neural | neural abnormalities | folic acid | iodine deficiency | Fetal Alcohol Syndrome | neural postnatal | neural examination | Histology | Historic Neural | Category:Neural
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

Development Of The Spinal Reflex Mechanism In Human Embryos

William F. Windle And James E. Fitzgerald

Departments of Anatomy and Obstetrics, Northwestern University Medical School


Aided by grants from The National Research Council and The American Academy of Arts and Sciences.

(Accepted for publication April 12, 1937)


So many details of structure of the human spinal cord are known that one might hesitate to choose this part of the nervous system for further investigation. However, upon examining the literature on anatomic interneuronal connections it soon becomes evident that much remains to be discovered. Specifically, the embryological approach to questions concerning functional reflex mechanisms in man has not received sufficient attention. Several articles dealing with correlated anatomical and physiological observations of reflex pathways in the chick, rat and cat have come from this laboratory (Windle, ’34; Windle and Orr, ’34; Windle and Baxter, ’35). It has been shown that although there are significant differences among these animals, -the fundamental plan of development of the first spinal reflex connections is similar. We propose to describe development of the spinal cord of human embryos between 5 and 8 weeks old and to discuss it in the light of the few physiologic observations available.

Twenty-seven human embryos between 5 and 8 weeks old were prepared by the pyridine silver method. Of this number, fourteen are usable but only eight are well stained. The specimens studied are listed in table 1. Two factors led to failure to obtain a greater number of good results with the staining method so successfully applied to other embryos. Specimens recovered after spontaneous abortion or from ruptured uterine tubes were dead when fixed in the ammoniated alcohol. Even though this fact was not evident from gross examination, the spinal cords show effects of autolysis. A few others were improperly fixed at the time they were removed. In every instance in which an embryo was recovered alive, separated from membranes and placed immediately into fresh 2% amrnoniated alcohol, a good silver stain resulted. A few of the preparations rival the best of our subhuman embryos.

Table 1

Human embryos studied

i DAYS AFTER ‘ "‘§5‘.”‘ 1MSc‘§§i’i§‘fi.LR”.‘c'.1 gggggggg; A’i5‘c“:‘?J£i"5i’i'é’a HOW OBTAINED I

693 ’ 7 14 or 42 5+ l Ectopic

713 10 49 6 Hysterectomy

733 12 43 6+ Ectopic

389 12 “.3 6 + Curettement

698 16 64 1, 7 ‘ Ectopic

731 18 g 47’ E 7 Hysterectomy

725 18 58 7 3 Hysterectomy

766 3 18 55 7 ‘ Therapeutic abortion 712 20 68 7+ Hysterectomy

736 2 1 “Z 7 + Hysterectomy

683 1 23 49 8— T Curettement

738 23 7 8—— ‘ Spontaneous abortion 717 3 23 7 ~ 8—_ Ruptured ectopic

743 5 24 l 66 8 Hysterectomy (local anesthetic)

1 Size is approximate based usually upon measurements taken in reagents during the preparation of the material for sectioning. 2 Number of days after a single intercourse. 3Age is estimated from external morphology.

Structure of the Spinal Cord

Fifth Week

The spinal cord of embryo 693 is exceptionally well stained by the silver reduction method. Its structure resembles that of the 7-mm. cat very closely. A gradient of development from rostral to caudal ends of the cord is clearly indicated in figures 9, 5 and 4; the greatest number of well developed structures are found in the cervical segments.

The ventral motor nucleus occupies a column in the mantle layer of the basal plate. Within this nucleus, neuroblasts are arranged at right angles to the long axis of the cord. The caudal end of the column contains a great many spindle-shaped cells with new axons growing out but the rostral end is composed principally of more mature unipolar neuroblasts and a few which are forming their first dendrons. This indicates that development began at the rostral end. Axons of most cells of the motor column pass out of the spinal cord in ventral root fascicles and enter into the formation of spinal nerves which pass some distance into the mesenchyme. The nerves of the lumbar regions are shorter than those in the cervical and upper thoracic. A few axons of cells in the motor column emerge in rootlets of the spinal accessory nerve, observed as far caudad as the junction of the seventh and eighth spinal segments (fig. 5). No dorsal root motor fibers were found at any level.

Secondary or connector neuroblasts seem to be confined to the alar plates in the 5-week embryo. They may be seen near the outer margin of the spinal cord and are especially prominent medial to the first few dorsal root fibers which enter the cord. The secondary elements possess no dendrons. In the sacral region, they are less numerous and have shorter axons than farther rostrad. Axons of these cells course ventrad through the motor nucleus toward the floor plate. Most cross in the ventral commissure and turn longitudinally to form the anlage of the ventral spinal funiculus. . The ventral commissure is represented by an occasional fiber in the lower sacral segments and is well formed in the cervical region. Its size varies with the number of secondary neuroblasts of the alar plate. The ventral funiculus is not found below the lumbar region because there the secondary axons end in the commissure. Sections cut coronally reveal that all fibers entering the ventral funiculus at the lumbar level pass rostrad. In similar sections through the cervical flexure, a few descending fibers were encountered but even there at the junction with the brain most of the growing tips of fibers are directed rostrad. We conclude that the ventral funiculus of the 5-week embryo is principally a contralateral ascending fiber tract.

A few silver-stained primary sensory neuroblasts are contained in the spinal ganglia of the cervical, thoracic and lumbar nerves but those of the sacral and coccygeal are almost entirely undeveloped. Nearly all neuroblasts are bipolar cells with distal processes thicker than proximal. Many of the distal processes enter the spinal nerves. Proximal processes enter the spinal nerves. Proximal processes of the cells of the cervical and upper thoracic ganglia form very delicate fascicles which pass to the dorsolateral periphery of the spinal cord. Below the ninth thoracic they fail to reach the cord and in the lumbar region they scarcely leave the confines of the ganglia. A discontinuous dorsal funiculus is formed in the cervical portion.

Although neurofibrillar differentiation must have begun some time before the end of the fifth week, the spinal cord still at this time has a very simple structure. Primary motor and sensory as well as secondary neurons are represented but no functional connections have been made between sensory motor and very few nerve fibers of the brain reach the spinal cord. Apparently, the motor mechanism develops in advance of the sensory.

Sixth Week

Human embryos 713, 733 and 389 are representative of the end of the sixth or beginning of the seventh week. They are not so nicely stained by the silver method as is the 5-week specimen. figures 3 and 6 illustrate two of the specimens. The spinal cord structure resembles that of cat embryos of approximately 10 mm.

The motor cell column has become very large at 6 weeks. Cells at its caudal end are still quite immature but those in the thoracic and especially the cervical segments are sending out many dendrons. Most of these processes spread out at right angles to the long axis of the spinal cord and terminate within the mantle layer but a few are long enough to extend into the marginal zone and are lost among longitudinal nerve fibers there. A separation of the motor column into medial and lateral nuclei is beginning in the brachial region. The spinal accessory nerve arises from nearly all cervical segments.

The secondary mechanism has made a striking advance in growth. Alar plates contain a large number of neuroblasts lying medial to the dorsal funiculus and a few are found in the basal plates. The axons of these cells form the floor plate commissure. A smaller number of secondary fibers swing laterad into the marginal zone of the same side. These association neurons form the dorsal part of the lateral funiculus, not present at 5 Weeks and more marked at 12 mm. than at 10 mm. The commissural fibers contribute to the large and mainly ascending ventral funiculus. The commissural are relatively more numerous than association neurons in the uppermost spinal segments. The reticular formation of the medulla oblongata is continued into the spinal cord where it occupies a position between lateral and dorsal funiculi. It contains a number of small fiber bundles which become fewer toward the brachial segments. However, one very distinct fascicle (fig. 3) continues caudad into the lower thoracic region. The fibers of this appear to be contributed by neuroblasts of the lower part of the medulla oblongata and the alar plates of the spinal cord; some vagus, accessory and dorsal spinal root afierent fibers run with it but it seems not to be the tractus solitarius proper. This bundle descends. to the lower thoracic region and may form an important _tract by means of which early spinal connections are made by bulbar centers.

The dorsal funiculus is a well—formed continuous bundle of fibers in 6-week embryos. The first little collateral sprouts leave its lateral end and enter the anlage of the dorsal gray matter in specimen no. 389 but the other embryos lack them. They are very short and are best seen in the brachial region. Some of the dorsal root fibers turn into the reticular formation between dorsal funiculus and lateral funiculus; some enter the dorsal end of the lateral funiculus.

The spinal nerves are well formed at 6 weeks and can be followed far toward the periphery. In the arm, the growth tips of brachial plexus fibers can be seen. Some of these ramify beneath the epithelium and others terminate in groups among mesodermal cells of the premuscle masses. No specialized endings exist at this time but the usual enlargements at the ends of new fibers are quite distinct.

All the elements needed for a functional spinal reflex system have been laid down during the sixth week but muscle development and morphologie connections between neural structures have not reached the point at which a flow of impulses from afferent neurons to effector organs is possible. The principal advances beyond the fifth week are the establishment of the secondary association neurons and the formation of a descending tract continuous with the reticular formation of the medulla oblongata. For the most part, the spinal cord motor apparatus remains isolated from the sensory and poorly integrated with structures of the brain.

Seventh Week

Four embryos 16 mm. to 18 mm. long are representative of the seventh week. figure 1 is a photograph of one of these. All were observed alive at the time of operation and were well stained by the silver technique. In structure, these specimens illustrated in figure 7, 10 and 11, resemble cat embryos of about 12 mm. to 14 mm.

Rapid differentiation of the reflex mechanisms takes place at the end of the seventh week. The embryos show all the structures encountered in smaller specimens but increased in number and maturity. The motor apparatus of the basal plates still dominates the picture and the most superficial comparison of 12 mm. and 16 mm. embryos demonstrates a marked enlargement of the neuron group for brachial innervation. The secondary elements likewise are present in great numbers. Commissural neurons are numerous at the rostral end of the spinal cord where they apparently represent the caudal continuation of the great decussation of the medial lemniscus. In the lower cervical region, they are less prominent as will be seen by comparison of figures 10 and 11. The association neurons pass laterad from a large group of cells ventromedial to the dorsal funiculus and build an extensive lateral funiculus into which the brachial motor nucleus bulges. Association fibers actually cut their way through the dorsal part of the motor column (fig. 7) and there come into intimate contact with dendrons of the motor cells. The anlage of the gelatinous substance appears at 16 mm. and is quite devoid of silver stained elements.

Fig.1 Human embryo no. 766; 18 mm. C.R. length before fixation; 7 weeks old. This represents a tage just before the time the first somatic movements should be elieitable.

The progress in growth of the central sensory mechanism is marked by the connections which are being effected with other neural structures. Collaterals are found at the lateral end of the dorsal funiculus but few pass far i11to the mantle layer. As will be seen in figure 7, the thickest and longest fibers lie farthest laterad and some appear to be the main axons of dorsal root neurons rather than branches of these. They curve around the medial side of the gelatinous substance toward the dorsal end of the motor cell column. Some of them form little bundles which turn and course longitudinally for a few sections, gradually shifting laterad into the dorsal end of the lateral funiculns. The state of development of the sensory mechanism is very much like that of cat embryos of 12 mm. which showed no movements and from which we could elicit no reflex responses.

Muscles of the neck, back and shoulder have begun to develop in the 16 mm. embryos, and there is a little differentiation in the distal half of the arm. In this respect the specimens are not much more advanced than embryo (‘IX of Bardeen a11d Lewis (’01). The motor nerves break up into fine branches as they enter the muscles and individual fibers tend to alter their course to H111 parallel to the newly formed muscle fibers. As seen in figure 2, they end upon these in little swellings. Sometimes branching nerve fibers terminate in several bulbs resembling grape—like endings. It is quite possible that should impulses course over motor fibers they would be transferred to muscles at these primitive motor endings. Many of the peripheral nerve branches course very near the integumental epithelium and individual sensory fibers are often found terminating; there as free endings with or without growth cones. Thus we find receptors and effectors of a primitive type present before the central reflex connections have been completed. 6

Eighth Week

We have six embryos between 7 and 8 weeks old which were stained by the silver method and two are illustrated in figures 8, 12 and 13. The two smallest (nos. 712 and 736) have reached a stage of development comparable to eat embryos of 14 to 16 mm. long.; they are representative of the first half of the eighth week. Two of the 23-mm. specimens (nos. 738 and 717) were apparently dead at the time they Were obtained and early autolytic changes can be observed. The remaining two were observed alive but no movements were seen even though one of them was removed under a local anesthetic. The end of the eighth week marks the change from embryo to small fetus.

The spinal cord is larger at'8 weeks than at the end of the seventh mainly because the secondary mechanism has grown so rapidly. The longitudinal tracts contain many more fibers at this period than before. But the most significant advance has been made in the development of connections between the sensory fibers of the dorsal funiculus and the centers forming in the mantle layer. Collaterals are very numerous at 20 mm. (fig. 8) and run far into the anlage of the gray matter. At 23 mm. (fig. 12) some of the longest of these nerve fibers can be followed under the microscope almost as far ventrad as the motor cell groups. The greatest development of the sensory collaterals is found in the lower cervical spinal cord segments. The fibers are much fewer and shorter in the thoracic and lumbariregions as will be seen by comparing figures 12 and 13.

Skeletal muscle development is widespread and numerous motor nerve fibers can be seen among the muscle fibers. The terminations of growing nerves are usually little bulbs which lie upon the muscle fibers. Many more of these are found at 8 weeks than at 7. Superficial nerves ramify beneath the integumental epithelium in many places. Thus, all elements necessary for simple spinal reflexes have been laid down.


A few studies of living human embryos have been made under conditions which may have been favorable for observing early movements. Those of the eighth week are arranged according to size of specimens in table 2. In our opinion, absence of motility does not mean that responses could not have been obtained under proper conditions. When a general anesthetic is used it is to be expected that movements may be absent. Furthermore, early behavior occurs phasically and the chances are good that some observations will be made in quiescent periods. Too much stress cannot be given the fact that surgical manipulations in exposing and removing the uterus set up a state of anoxemia which leads to depression of the embryo even though no general anesthetic is administered. This may explain why we failed to find movements in our specimen removed under local anesthetic; several minutes elapsed between clamping uterine blood vessels and opening the uterus. It has been shown (VVin(lle and Barcroft, ’37) that anoxemia affects finer movements, such as those of the limbs, very rapidly.

Table 2

Human embryos 7 to 8 weeks old examined for motility

EMBRYO . . _ ' , SIZE, C.R. IL‘ ESTI AN}JSTI{E'l‘IC RESULTS LENGTH J6 I ‘ .Avertin and ' N 0 movements 16 fitzgerald - ' N 0 movements ]8 l fitzgerald - ' ' No movements 18 fitzgerald ‘ I ' _ N 0 movenients 18 fitzgerald - ' ' ‘ No movements sterilization 20 fitzgerald Hysterotomy No movements 20‘ % Hooker 19_51 Hysterotomy N o movements 20 Yanase 1’ ' ‘. Forelimb movements occurred abortion spontaneously ('1?) after free— ing from membranes 21 fitzgerald Hysterotomy N 0 movements 22 Hooker Spinal novocaine Hysterotoiny No movements 24 fitzgerald Local y ' V No movements when touched ‘ with scalpel 22” Strassmann . Ruptured tubal Spontaneous movements of

pregnancy limbs seen through intact

l amnion 25 Spinal ysterotomy Colltralateral trunk bending l novocaine after‘ stimulating the neck and morphine ‘ Twins.

‘ This measurement taken after fixation.

The smallest specimen in which movements were seen was that described by Yanase (’07 b). The length, 20 mm., and description of the embryo agree fairly well with the fact that menstruation ceased 68 days before. It was obtained at artificial abortion and there is no reason to believe that a general anesthetic was used. Its heart continued to beat for 20 minutes, indicating that it must have been i11 reasonably good condition when first obtained. Freed from its membranes in Ringer ’s solution at body temperature, six or seven movements of the arm were observed; no other activities were reported. That Yanase was well qualified to describe early behavior cannot be doubted. His earlier studies in peristaltic movements of the embryonic gut of guinea pigs (Yanase, ’07 a) must have stimulated his powers of observation.

In a discussion of embryonic motility, Strassmann (’03) held that movements of skeletal muscle begin very early and can be seen with the naked eye, especially in extra-uterine pregnancies. To illustrate this point, he described an instance in which he saw slow movements of the extremities through the intact amnion of a fetus about 8 weeks old (22 mm. after fixation). The amniotic sac bulged from a laceration in the uterine tube, but the chorion was fastened tightly to the wall of the tube and projected only with a few tufts. This is the only case recorded in which observations of a human embryo were made with placental circulation intact. True, the patient was probably anesthetized (no mention of that fact was made) a11d one may consequently question whether the conditions were any better than they were in other studies; but one cannot doubt that Strassmann saw movements which he thought to be active. Continuing his generalization, he said “die ersten Bewegungen sind ein Anhebung der oberen Extremitaten, um eine Atmung in Gang zu bringen.” It is quite probable that this writer had had other opportunities to observe fetal behavior.

Hooker (’36) studied a motile embryo of about 8 weeks (25 mm.). When the neck was touched the body and limb-girdle muscles of the same and opposite sides contracted. This is clear proof that the 8-week embryo is capable of a small amount of reflex activity even after a state of anoxemia has been set up. The one point in common brought out by all three investigators under discussion is contractility of muscles moving the arms. This is confirmed by our own studies in the cat, rat and sheep (Windle et al., ’34, ’35, ’37; Bareroft, Barron and Wlndle, ’36) which indicate that forelimb movement is one of the very first behavioral responses.

From our present study of neural structures, it is clear that the nervous and muscular systems have reached a state of development at 20 mm. in which the anatomic basis for function is present. In fact, it is possible that reflex movements can occur a little earlier than this in some human embryos. Sensory endings and primitive motor terminals are formed at 16 mm. and adequate spinal reflex connections are made between 16 and 20 mm. It may be significant that Yanase saw movement of the arm only. We find that the peak of growth of the spinal reflex mechanism is located at the level of the brachial plexus. The human nervous system at the end of the eighth week is the equivalent of that of rat, cat and sheep embryos from which reflexes of tlie neck, trunk and forelimbs were elicited and in which some integration of elemental movements had taken place.

Hooker (’36) reported some histologic observations made by Hogg who found “a very undeveloped state of the dorsal funiculus containing the afferent fibers” and “a few ventral commissural fibers from a nucleus situated in the midregion of the gray matter” of a 22—mm. embryo. This implies a less mature condition than our observations in several specimens of about this size indicate. There are embryos larger and smaller than the average for any particular developmental stage. Perhaps Hooker’s specimen belongs in the former group.


Neurofibrillar differentiation begins before the end of the fifth week in the human spinal cord. Primary motor, primary sensory and secondary neurons are present but functional reflex connections have not been made.

All the elements needed for a functional spinal reflex system are laid down by the end of the sixth week. However, muscle development and morphologic connections between neural structures are still incomplete. Tracts descending from the reticular formation forecast the foundation of an integration mechanism.

The seventh week is a period of rapid growth and expansion of elements already represented. Peripheral motor and sensory endings of a primitive type are forming. The first collaterals of the dorsal (sensory) funiculus are making their way into the region from which spring the association neurons. Human embryos of 7 weeks resemble those of other species which are not quite ready to execute spinal reflex movements.

The first spinal reflex arcs are completed during the eighth week. Other investigators have reported human embryonic movements in three favorable instances at or before the end of the eighth Week. Their observations can be correlated with the present morphologic study.

Literature Cited

BARCROFT, J., D. H. BARRON AND W. F. WINDLE 1936 Some observations on genesis of somatic movements in sheep embryos. J. =Physiol.;‘vol. 87, pp. 73-78.

BARDEEN, (7. R., AND W. H. LEWIS 1901 Development of the limbs, body-wall and back in man. Am. J. Anat., vol. 1, pp. 1-35.

HOOKER, D. 1936 Early fetal activity i11 mammals. Yale J. Biol. and Med., vol. 8, pp. 579-602.

STRASSMANN, P. 1903 Das Leben vor (let Geburt. Samml. klin. Vortr. 19001903, Grynak., no. 132, S. 947-968.

WIN]:-LE, W. F. 1934 Correlation between the development of local reflexes and reflex arcs in the spinal cord of cat embryos. J. Comp. Neur., vol. 59, pp. 487-505. 1937 On the nature of the first forelimb movements of mammalian embryos. Proc. Soc. Exp. Biol. and Med., vol. 36, pp. 640-642.

WINDLE, W. F., AND J. BARCROFT 1937 Some factors governing the initiation of respiration in the chick. (Manuscript unpublished.)

WINDLE, W. F., AND R. E. BAXTER. 1935 Development of reflex mechanisms in the spinal cord. of albino rat embryos. J. Comp. Neur., vol. 63, pp. 189209.

WINDLE, W. F., W. L. MINEAR, M. F. AUSTIN AND D. W. ORR 1935 The origin and early development of somatic behavior in the albino rat. Physiol-. Zool., vol. 8, pp. 156-185.

WINDLE, W. F., AND D. W. ORR 1934 The development of behavior in chick embryos: spinal cord structure correlated with early omatic motility. J. Comp. Neur., vol. 60, pp. 287-307.

WINDLE, W. F., D. W. ORR AND W. L. MINEAR. 1934 The origin and development of reflexes in the cat duri11g the third fetal week. Physiol. Zool., vol. 7, pp. 600-617. 506 W. F. WINDLE AND J. E. fiTZGERALD

YANASE, J. 1907 a Beitréigc zur Physiologie der peristaltischen Bewegungen des embryonalen Darmes. Arch. f. d. ges. Physiol., Bd. 117, S. 345-383.

1907 b Beitriige zur Physiologic rler peristaltischen Bewegungen (lcsembryonalen Darmes. II. Mitt. Arch. f. d. ges. Physiol., Bd. 119, S. 451-464.


Plate 1

2 Nerve fibers ending on myoblasts in the shoulder region of 7-week-old human embryo no. 725. Pyridine silver stain. X 450.

3 Spinal cord at the fifth cervical segment of 6-week-old human embryo no. 733. The tract descending from the reticular formation is illustrated (d.t.). Pyridine silver stain. X 250.

4 to 8 Camera lucida drawings of the spinal cord of human embryos stained by the pyridine silver method. figures 4 and 5 illustrate. the sacral and seventh cervical segments of 5-week-old embryo 110. 693. figures 6, 7 and 8 illustrate the seventh cervical at 6 weeks (110. 713), at 7 weeks (no. 698) and in the eighth week (no. 712). x 70.

Plate 2

9 to 13 Photomicrographs of sections of the spinal cord of human embryos stained by the pyridine silver method. X 75. figure 9 illustrates the first cervical level of 5—week—o1(1 embryo no. 731. figures 10 and 11, the first and seventh cervical segments of 8-week-old embryo no. 731. figures 12 and 13, the fifth cervical and fifth thoracic levels of 8~week—0ld embryo no. 683.

Cite this page: Hill, M.A. (2024, May 25) Embryology Paper - Development of the spinal reflex mechanism in human embryos. Retrieved from

What Links Here?
© Dr Mark Hill 2024, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G