Paper - Some observations on myelination in the human central nervous system (1931)

From Embryology
Embryology - 10 Apr 2020    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)

Keene MF. Some observations on myelination in the human central nervous system. (1931) J Anat. 66(1): 1-13. PMID 17104349

Online Editor 
Mark Hill.jpg
This historic 1931 paper by Keene describes neural posterior commissure development.

Also by this author: Keene MF. Some observations on myelination in the human central nervous system. (1931) J Anat. 66(1): 1-13. PMID 17104349 Keene MF. The development and myelination of the posterior longitudinal bundle in the human. (1933) J Anat. 67(4):522 - 535. PMID 17104444

Keene MF. The connexions of the posterior commissure: a study of its development and myelination in the human foetus and young infant, of its phylogenetic development, and of degenerative changes resulting from certain experimental lesions. (1938) J Anat. 72(4): 488-501. PMID 17104719
Modern Notes: myelination | oligodendroglia | gliogenesis | neural

Glia Links: gliogenesis | radial glia | astroglia | oligodendroglia | microglia | Schwann cell | myelination | Category:Glia
Historic Glia Embryology  
1931 CNS Myelination | 1933 Posterior Longitudinal Bundle Myelination | 1938 Posterior Commissure Myelination

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)

Some Observations on Myelination in the Human Central Nervous System

By M. F. Lucas Keene, M.B., B.S. Professor of Anatomy at the London (Royal Free Hospital) {School of Medicine for Women)

E. E. Hewer, D.Sc. Lecturer in Histology and Demonstrator in Physiology at above School


The first appearance of the myelin sheath in fibres of the central nervous system is a subject that has attracted attention for many years, more particularly since Flechsig(1) originated the view that the degree of myelination might be correlated with functional capacity. His original theory stated that myelination was found in projection paths always before association paths, and in peripheral before central paths, and in sensory before motor ones. These statements were not fully confirmed by Monakow, Vogt, Westphal and others (2), and Flechsig subsequently modified his theory but still maintained that with respect to any one projection centre the fibres always myelinated in the same order, namely, first the sensory, then the efferent, and then the association fibres. On the other hand, Vogt(3) states that myelination depends on the number and size of the fibres developed in the path, and is not correlated with the function of the fibres.

Flechsig’s theory has been found to apply in kittens by Tilney and Casamajor (4), who investigated the development of various somatic motor reactions from birth up to 44 weeks, and attempted to correlate these reactions with the myelination found at various ages.

It cannot, however, be a rule of general application that myelination is a criterion of functional activity of the conducting paths. Gonzalez (5) has shown that in white rats myelination does not begin until after birth, and yet in a 19day foetus definite reflexes are possible with an inhibitory control from higher centres. Further work confirming Gonzalez’ view has recently been published by Wenderowic (6).

Quite apart from the functional significance of the appearance of the myelin sheath, the investigation of the myelination present at various stages of development has been used as a means of tracing the course of nerve paths between higher centres and the periphery. No systematic account of this development has yet been published for the human foetus, and the present paper is an attempt to confirm and extend observations of other workers. The collection of suitable material is a difficult matter, and it is realised that the series on which the paper is based is as yet in some ways incomplete, but it is hoped to continue the investigation as material becomes available. Whilst this paper was in course of preparation an account of the myelination present in a 7th month human foetus has been published by Langworthy (7): his observations are in agreement with ours with respect to those fibre groups that are discussed both in his paper and in ours, but he investigated also the myelination in the cerebellum which we have not done.

The origin and formation of myelin has been described by Cornwall (8), Schultze (9), Berberich (10), and others: Cajal(11) regards its formation as due to the activity of the axis cylinder, beginning centrally and extending peripherally, and gradually increasing in thickness.


The material on which this paper is based includes serial sections of brain stem and spinal cord from thirteen foetuses varying in age from 14 weeks to full time, and also from three infants under 1 year old. The ages of the foetuses were determined by taking into account the maternal history, together with the data obtained from careful observation and dissection of the foetus. The foetal specimens include one each of the following ages: 14 weeks, 16 weeks, 20 weeks, 22-24 weeks, 28-25 weeks, 24-26 weeks, 26-28 weeks, 28 weeks (lived 19 days, see below), 836 weeks. In addition there were four full-time specimens, and three infants aged 3, 8 and 9 months respectively. The foetuses dissected varied only within physiological limits, that is, were apparently ‘‘normal.” The tissues have been obtained as fresh as possible, carefully dissected out, and fixed first in formalin and subsequently in Muller’s fluid. Further treatment involved staining by the Weigert-Pal method, the sections being cut in celloidin and mounted serially. In addition to these serial sections nine other specimens of various ages have been investigated by the Bielschowsky and Ranson methods,

General Observations

The appearance of myelin as shown by the typical staining reaction of the Weigert-Pal method is preceded by some change in the nerve fibres that gives rise to a very characteristic brownish grey coloration in sections so stained. This may be termed the “‘pre-myelin” stage: it is of short duration, and is followed by the well-known dark blue staining reaction.

In tracing out the course of the various fibres we have been able to confirm the observation (11) that myelination proceeds centrifugally, and also that in some cases at least a nerve path of several neurones myelinates in neurones in the order of conduction of the impulse. An example of this is to be found in one of the sensory paths: the fibres of the column of Burdach begin to myelinate at about 14 weeks’ development; the second neurone of this path may be the medial fillet which begins to myelinate at about 24 weeks, or the external arcuate fibres which are not myelinated until nearly full time. The thickness of the myelin sheath increases gradually from its first appearance as a thin covering to the axis cylinder: it has been suggested that this gradual increase may be correlated with the increase of growth in the size of the axis cylinder itself.

In the course of this investigation it was found that one specimen did not fall into line with the remainder of the series. The history was one of 28 weeks’ gestation, and in the other specimens of this stage of development no myelination had ever been found in the pyramidal fibres, or in the pons fibres, or in the optic tract: in this particular specimen myelination was present in these three groups of fibres to the extent corresponding to the conditions found about the time of birth (40 weeks’ gestation). Further inquiry into the history revealed the fact that the child, although born at 28 weeks’ gestation, had lived for nearly 3 weeks.

In the following account the main afferent and efferent paths are first considered, then the chief connecting paths, and lastly the cranial nerves. The myelination of the various fibre groups in the thalamic and subthalamic regions is barely touched upon, as it is hoped later to make a further and detailed investigation of this part of the brain.


(1) Posterior root fibres By 14 weeks’ development a very few of the posterior root fibres are finely myelinated: the number of fibres myelinated and the thickness of the sheath increase steadily but more obviously after 24 weeks (fig. 1).

(2) Posterior column fibres

Fine myelination is present in a few fibres of the column of Burdach near the base of the posterior columns by 14 weeks: by 22-24 weeks the number of fibres myelinated has increased considerably, but the sheath is still extremely fine. From the 24th week onward the degree of myelination becomes markedly heavier, and at birth these fibres stand out in the sections as among the most intensely stained.

Fibres in the column of Goll acquire their myelin sheath later, the earliest seen being by 28 weeks: the degree of myelination gradually increases, but even after birth the fibres do not appear so deeply stained as those in the column of Burdach. As is well known the fibres of Goll’s column are smaller in crosssection even in the adult than those of Burdach’s column, and therefore this tract appears less darkly stained of the two (figs. 1 and 8).

The nuclei gracilis and cuneatus are clearly differentiated by 14 weeks, but myelination of the second neurone on this path (the medial fillet) does not appear until about the 24th week of development: this myelination is at first very fine, but is present throughout the whole path up to the thalamus. The staining reaction of the medial fillet increases quickly with age, but from 26 weeks until about 8 months after birth definite non-myelinated bundles, circular in cross-section, can be very clearly seen scattered amongst the myelinated fibres (fig. 11). These bundles can be traced from the origin of the fillet in the medulla up as far as the level of the inferior corpora quadrigemina, and are some of the “aberrant fibres of the pyramidal path” (see below): they cannot be the second neurone on the sensory path of the 9th and 10th cranial nerves that run in the fillet, as these fibres are beginning to myelinate by 14 weeks. ;

The cerebellar connection from the nucleus gracilis and nucleus cuneatus (the external arcuate fibres) is not myelinated at 86 weeks’ development, but the fibres give a faint but definite staining reaction at full time.

(3) Direct cerebellar tract

A few very finely myelinated fibres are found at 16 weeks’ development: the myelination increases slowly until about 28 weeks, when the stained sections show that it is considerable in amount. At birth this tract stands out in sections as particularly heavily stained (fig. 2).

The fibres are myelinated throughout their course through the restiform body to the cerebellum.

(4) Gowers’ (antero-lateral ascending) tract

By the 14th week of foetal life a very little myelination is seen in the anterolateral region of the spinal cord, and it is difficult to distinguish between ascending and descending fibres. After the 28th week myelination is more marked, but even at full time the fibre sheaths are not so thick as in the direct cerebellar tract. Fine fibres crossing in the anterior commissure are myelinated at the 14th week: they belong probably to the spinothalamic and spinotectal groups. In the young specimens (14-24 weeks’ development) these crossing fibres are more abundant in the cervical region than lower down the cord (fig. 1). (B) MAIN EFFERENT PATHS

(1) Pyramidal tracts

Some fine myelination is first seen at about 36 weeks’ development, particularly at the higher levels of crusta and pons. At full time the myelination is still very scanty, being more marked in the brain stem and in the cervical region of the spinal cord than lower down. The “ pre-myelin” stage can be well seen in most of the fibres. Even at 9 months after birth the myelination is still very incomplete, and the sheath comparatively thin.

The “aberrant fibres of the pyramidal path” (Dejerine(12)) that run in the medial fillet can be detected as definite bundles (see above) from 26 weeks’ development onwards (fig. 11): they can be traced as unstained circular areas in cross-sections of the brain stem from the level of the lower border of the crura cerebri down to the lower border of the olives, where they appear to join the main pyramidal path. These fibres give the “ pre-myelin” staining reaction by 8 months after birth, and by 9 months they are definitely myelinated although the sheath is thin.

Other “aberrant pyramidal fibres” (also described by Dejerine(2) and referred to in the chart as X fibres), can be seen lying in the lateral part of the substantia nigra, just posterior to the crusta: at lower levels these fibres become more anterior, and join on to the main pyramidal path in the pons. This group of aberrant fibres is not myelinated at birth, but 3 months later shows a very heavy degree of staining. They are frequently shown in diagrams of cross-sections, but are not usually named as part of the pyramidal system.

(2) Rubrospinal tract

A few fibres show fine myelination by 36 weeks’ development, the extent of myelination increasing gradually as in the pyramidal tracts.

(3) Antero-lateral descending tract

It is difficult to distinguish between antero-lateral ascending and descending fibres, but the myelination in this position in the cord gradually increases from 14 weeks’ development onwards. In a 28-weeks’ foetus the olivo-spinal tract can be distinguished in section as a definitely non-myelinated triangular area, and even at 8 months after birth this tract is very incompletely myelinated (fig. 2).

As the fountain decussation of Meynert (tecto-spinal tract) (fig. 18) is finely myelinated by the 24-26th week it is probable that some of these fibres are also myelinated about this time in the spinal cord.

(4) Anterior root fibres

A few fibres are already myelinated in a 14-week foetus, the myelination being somewhat in advance of that of the posterior root fibres. After this date myelination increases rapidly, and is very soon a marked feature of all the preparations (fig. 1).


(1) Ground bundles of the spinal cord

The anterior ground bundle is the first to myelinate, giving a definite staining reaction by 14-16 weeks’ development. The lateral ground bundle begins to myelinate about 22-24 weeks, and the posterior ground bundle about a fortnight later (26-28 weeks) (fig. 1).

(2) Connecting fibres in the grey matter of the cord

Fine myelinated connecting fibres can be seen by 22 weeks’ development, passing between the postero-lateral region of the grey matter and the anterior horn’ cells. These fibres are not myelinated by 16 weeks, and the number myelinated and the thickness of the sheath increase rapidly.

(8) Lissauer’s tract This group of fibres myelinates late: the process was seen to be just begin ning by 36 weeks’ development, but myelination was not constantly found even at full time (fig. 3). (4) Comma tract

The fibres of this tract are not myelinated before those of the posterior columns: in specimens that gave a staining reaction throughout the columns of Goll and Burdach, the comma tract was not identified. ©

(5) Olive and cerebellum connections

Fibres connecting the olives‘and the cerebellum begin to myelinate by 22 weeks: after this myelination proceeds rapidly.

(6) Medial (or posterior) longitudinal bundle By 14 weeks’ development a few of these fibres show fine myelination throughout the entire distribution of the tract from medulla up to mid-brain, and already at this stage there is a myelinated connection between the bundle and the myelinated posterior commissure (figs. 5 and 11).

(7) Dorsal longitudinal bundle (of Schiitz) These fibres begin to myelinate at full time.

(8) Pons fibres A very few fibres are faintly myelinated by 36 weeks’ development, but even at birth the myelination is very scanty (fig. 7). By 8 months after birth, however, the staining reaction is well marked.

(9) Brachia conjunctiva : Myelination begins at about the 26th week of development, and appears to involve all the fibres: it is not possible to distinguish between the afferent and efferent groups. (10) Fasciculus retroflecus of Meynert This bundle (fig. 10) can be distinguished as a collection of non-myelinated fibres in a 14-week foetus: by 22 weeks the bundle is still non-myelinated, but by 24 weeks the myelination is quite distinct in the peripheral fibres. This partial myelination persists for some time, but by 86 weeks the whole bundle is well myelinated and can be traced to the ganglion habenulae.

(11) Cortico-pontine cerebellar tracts Myelination of these paths is comparatively late, being very slight indeed at birth. Nine months later myelination is more advanced, but even at that date in the stained sections through the crusta the central pyramidal fibres appear much darker than do the fibres of these groups. The sections show also that the fronto-pontine path is less heavily myelinated than the temporal and occipital fibres at this date (s year) (fig. 12). Myelination in the Human Central Nervous System 7

' (12) Forniz The fornix begins to myelinate about 8 months after birth.

(18) Ansa lenticularis fibres

There are various groups of fibres that pass across the posterior limb of the internal capsule: some of these begin to myelinate early, before the pyramidal fibres that pass through this region, and are consequently well seen in such specimens (fig. 9). Dejerine(13) distinguishes “radiations strio-luysiennes,”’ which pass from the striate body to the nucleus subthalamicus and then spread out fanwise. These fibres show a “‘ pre-myelin” staining by 24-26 weeks’ development, and a definite myelin reaction by 26-28 weeks. The “striothalamic” fibres lie more laterally than the above in the internal capsule, and they myelinate later, being first seen at 36 weeks.

(14) The bundle of Vicq @ Azyr

This group of fibres is not myelinated at birth, but can be seen in the sections as a definite group of unstaified fibres. By 3 months after birth the tract is well myelinated.

(15) Meynert’s commissure

This band of fibres passes dorsal to the optic chiasma and connects the two lenticular nuclei (Dejerine(14)), and is said by Riese (15) to be the first of the hypothalamic commissures to myelinate. In the 36-weeks’ foetus these fibres were already myelinated, and in one specimen of 28 weeks some finely myelinated fibres were seen which were thought to belong to this commissure although not positively identified. (16) Olfactory nerve

The myelination of these fibres was not investigated.

(17) Optic tract By 36 weeks’ development a few of the optic fibres are myelinated, but the myelination continues to be inconstant and slight up to full time. At full time the connections with the external geniculate bodies and with the optic thalamus are myelinated. In the 28-week foetus that survived for nearly 3 weeks after birth the optic fibres were beginning to myelinate. (For connection with geniculate bodies see fig. 12.)


Nearly all the cranial nerves acquire a myelin sheath early: by 14 weeks’ development some myelinated fibres can be detected in all these nerves (within the brain stem) except in the sensory part of 5 and in the cochlear part of 8. By 24 weeks all the cranial nerves contain myelinated fibres, the motor nerves appearing particularly heavily stained.

Nerve 12

The nucleus (fig. 5) is well differentiated by 14 weeks, the cells being of the large multipolar type and extending throughout the whole of the lower and middle part of the medulla. The myelination is at first fine, but by 26 weeks is extremely heavy, the stained fibres standing out very remarkably in sections

th i ion. rough this region Nerve 11

The medullary nucleus is differentiated and the fibres finely myelinated by 14 weeks’ development. The spinal accessory fibres arising from the cord are myelinated by 26 weeks and possibly before: it is easy to miss these fibres unless complete serial sections are available.

Nerves 10 and 9

The nuclei are differentiated and fine myelination is present in the fibres by 14 weeks’ development. Of the various nuclei the nucleus solitarius (fig. 4) can be the most readily associated with the myelinated fibres in the younger specimens. Nerve 8

(a) Vestibular branch. A few myelinated fibres are present by 14 weeks’ development, but it is not possible to be certain with which of the three nuclei they are connected. The nucleus of Roller’s bundle (descending branch) can be made out with some myelinated fibres. By 26 weeks the myelination of the vestibular fibres is general.

(b) Cochlear branch (fig. 7). The fibres belonging to this nerve are definitely not myelinated by 14 weeks, but by 24 weeks many of the cochlear fibres have acquired a fine myelin sheath: they can be traced to their nuclei. The next neurone (striae acousticae and trapezium fibres) is also myelinated, and the myelination can be traced up in the lateral fillet to the inferior corpora quadrigemina and internal geniculate body. The connection with the superior olive is also myelinated by this date (fig. 6).

Nerve 7

Both efferent and afferent parts of the deep connections are myelinated by 14 weeks, although the sheath is thin and not present on all the fibres.

Nerve 6

The nucleus of large multipolar cells is well differentiated by 14 weeks, and some of the emerging fibres are already faintly myelinated. In the older specimens these fibres stand out in the stained sections as having thick myelin sheaths, similar to those of the 12th and 8rd nerves (fig. 7). ‘

Nerve 5

A few myelinated fibres are found arising from the motor nucleus by 14 weeks, but at this time the sensory fibres are still unmyelinated although the sensory nuclei are differentiated. By 22-24 weeks the sensory fibres are myelinating, and the spinal root in connection with the substantia gelatinosa of Rolando contains finely myelinated fibres in the medulla. By 24 weeks the decussating sensory fibres arising from the main sensory nucleus are well myelinated.



WEEKS GESTATION 10 .15 20 30


& a a 12 8 4 6 6 7 8 9 10 11 12

Posterior root fibres a Burdach’s column

Direct cerebellar tract ff

Gowers’ tract [anterolateral ascending} wenn een Anterior root fibres oo

Anterior ground bundle meoce

Posterior (medial) longitudinal bundle

All “cranial nerves” except cochlear 8, wean sensory 5, 2

Goll’s column

Lateral ground bundle -

Posterior ground bundle —

Connecting fibres in grey matter of cord -=

Olive and cerebellum connections

Fillet (medial)

Fasciculus retroflexus of Meynert

Ansa lenticularis [striae Luysiennes]

Cranial nerves—cochlear 8, sensory 5

Lissauer’s tract --+

Direct & crossed pyramidal tracts

Rubrospinal tract

External arcuate fibres

Pons fibres -

Cortico pontine cerebellar fibres i

Bundle of Schuts

Strio-thalamic fibres -

Commissure of Meynert _

“Cranial nerve” 2

Aberrant pyramidal fibres ‘“‘X”

Bundle of Vicq d’Azyr

Olivo-spinal tract -

Aberrant pyramidal fibres of fillet —_— Fornix - Myelination in the Human Central Nervous System

The mesencephalic root (figs. 10 and 11) of this nerve is beginning to myelinate by 14 weeks, and is closely associated in position with the peculiar large cells that extend up beyond the level of the 8rd nerve nucleus (fig. 8). These cells are usually spoken of as the mesencephalic nucleus of the 5th nerve, and are of several types (Sheinin(16)). They give rise to fibres whose precise nature has been much disputed: they may be afferent or efferent, or possibly comprise both centripetal and centrifugal fibres of the autonomic system (see Winkler (17)).

Nerve 4

In the 14-week foetus the fibres of this nerve are just beginning to myelinate. At a slightly later stage myelinated fibres can be easily traced passing from the nucleus of this nerve to enter the posterior longitudinal bundle.

Nerve 3

Myelination is present in a few fibres at 14 weeks, and by 26 weeks most of the fibres are heavily myelinated.


The dates of myelination of the important groups of fibres have been summarised in the accompanying chart. It will be seen that there are four definite periods (between 12 weeks’ gestation and a year after birth) at which the myelination process appears to receive an impetus. These periods are as follows: (i) About 14 weeks’ gestation. (ii) 22-24 weeks’ gestation. (iii) Just before full time. (iv) About 8 months after birth.

In addition, in the tracts that begin to myelinate at about 14 weeks the process is markedly accelerated at about 24 weeks. In this connection it is interesting to note that the “quickening movements” of the foetus are felt about the 5th month, that the weight ratio of the brain remains practically constant from the 24th week onwards (18), that the brain grows most rapidly during the second half of pregnancy (19), and that the majority of the fissures and lobes of the brain are differentiated between the 24th and 28th weeks (19). There may also be some correlation between the marked increase of myelination at this time and the appearance of lipoid in the suprarenal cortex of the © foetus (20): the lipoid first appears in traces by 22 weeks’ development, and then increases rapidly during the later months of foetal life.

On referring to the chart it will be seen that myelination occurs first in sensory paths, together with the motor cranial and spinal nerve roots, and the medial longitudinal bundle. It is remarkable that the sensory fibres of the trigeminal nerve and the cochlear fibres of the auditory nerve are later in myelinating than any other nerve roots. Of the association paths, the medial longitudinal bundle and the anterior ground bundle of the cord are the earliest to myelinate. Myelination in the Human Central Nervous System 11

Myelination in the column of Goll is considerably later than that in the column of Burdach. The significance to be attached to the acquisition of the myelin sheath is a debated point, and so it is difficult to explain this delay. Since the rate of growth of the hinder parts of the body lags behind that of the remainder, it is tentatively suggested that there may be some relationship between the rate of growth and the myelination of afferent fibres.

By 28 weeks’ development, at which stage the foetus is viable, all the important sensory paths are myelinated, with the exception of such connecting paths as Lissauer’s bundle in the spinal cord, and the external arcuate fibres from the nuclei gracilis and cuneatus to the cerebellum. At this age, the long motor paths are not myelinated, but as the association paths in both brain stem and spinal cord and also the motor nerves are all myelinated the nervous activities must be purely reflex, with no central control of cord function.

At birth the long motor paths are beginning to myelinate: the whole of the cortico-pontine-cerebellar path is also myelinating, providing the path for cerebral control of cerebellar responses. In addition, the external arcuate fibres are myelinating, also the bundle of Schiitz and Lissauer’s tract. Thus all the important connecting paths are myelinated at birth.

The tracts that are not even beginning to myelinate at this time are the olivo-spinal tract and the aberrant pyramidal fibres: of these the aberrant pyramidal fibres X myelinate about 3 months after birth, and the other two groups begin to myelinate together at about 8 months. These aberrant fibres of the pyramidal path are usually missed in stained sections in which both fillet and pyramidal fibres are fully myelinated, although they can be picked out by the small size of their fibres: in this series they show extremely well, because they do not myelinate until so very much later than the fillet fibres. Dejerine (12) has traced their course in Marchi stained preparations of a case of pyramidal degeneration, and our series confirms his description.

It is interesting to note that the late date of myelination of two of the descending paths—namely the olivo-spinal tract (an efferent path from the cerebellum) and the aberrant pyramidal fibres in the fillet (an efferent path from the motor cerebral cortex)—corresponds to the time at which the baby begins to maintain its equilibrium and to perform purposive, co-ordinate movements such as crawling and standing. The baby can carry out certain co-ordinate movements at a considerably earlier date, but since these particular activities become possible at about the same time as the olivo-spinal and aberrant pyramidal fibres myelinate, these may be the special nerve paths concerned, assuming that there is some relation between myelination and function.

At the beginning of this paper it was mentioned that one of the specimens was obtained from a child born prematurely at 7 months and surviving for nearly 3 weeks, and that in this case the myelination found corresponded in certain respects to that of a full-time foetus. This acceleration of the myelination process may be associated with and due to the premature assumption of post-natal conditions, although it is not possible to make out a definite correlation from one case: if this is so, it would lend further support to the view that myelination and functional activity are related.


  1. In the course of human development there are four stages at which the process of myelination receives an impetus: (i) About 14 weeks’ gestation. (ii) About 22-24 weeks’ gestation. (iii) Just before full time. (iv) About 8 months after birth.
  2. Of these periods the second (22-24 weeks) is the one at which the myelination process is most active.
  3. Dates of myelination are given for the more important of the fibre paths, and for the cranial nerves.

Our grateful thanks are due to Dr Cruickshank, to Dr Masterman, and to others who have kindly supplied us with material and with histories of the specimens, and also to Dr J. G. Greenfield for access to his library. We are indebted to Mr V. Willmott (Royal Free Hospital) for the excellent microphotographs.

This research is part of a general enquiry into foetal development, the expenses of which are being met in part by a grant from the Thomas Smythe Hughes Medical Research Fund, and a grant to one of us (E. E. H.) from the Medical Research Council.


(1) Frecuste, P. (1895). Neurol. Centralbl. vol. xtv, pp. 1118, 1177.

(2) Casa, Ramon (1911). Histologie du Systéme Nerveuz, vol. 1, p. 866.

(3) Voat (1908). Neurol. Centralbl. vol. xxvu, p. 137.

(4) Tuney, F. and Casamagor, L. (1924). Arch. Neurol. and Psychiat. vol. xu, 1, pp. 1-66.

(5) GonzauEz, A. W. A. (1929). J. Comp. Neurol. vol. xivm, p. 459.

(6) WEnDEROWIC, E. (1930). Arch. f. Psych. u. Nervenkrank. vol. xc, pp. 789-804.

(7) Lanewortay, O. R. (1930). Contrib. to Embryol. No. 120. (Johns Hopkins University.)

(8) Cornwatt, L. H. (1927). Arch. Neurol. and Psychiat. vol. xvi, p. 240.

(9) ScuuLtzz, W. H. (1909). Ergebn. d. allgem. Pathol. vol. xm, 2, p. 253.

(10) BErBeEricn, J. (1926). Klin. Wochenschr. vol. v, i, p. 211.

(11) CasaL, Ramon (1928). Degeneration and Regeneration of the Nervous System, vol. 1, p. 202.

(12) Drserine, J. (1901). Anatomie des Centres Nerveuz, vol. 0, pp. 543, 578.

(13) —— (1901). Anatomie des Centres Nerveuz, vol. 1, p. 327.

(14) —— (1901). Anatomie des Centres Nerveux, vol. 0, p. 409.

(15) Rimsz, W. (1925). J. fiir Psychol. und Neurologie, Bd, xxx1, pp. 81-122.

(16) SHErmin, J. J. (1930). J. Comp. Neurol. vol. t, 1, p. 109.

(17) Winxter, C. (1921). Anatomie du Systéme Nerveux, vol. 1, pp. 1-100.

(18) Medical Research Council (1924). Special Report Series, No. 86, p. 80.

(19) Fetpman, W. M. (1920). Principles of Antenatal and Postnatal Child Physiology, pp. 20911.

(20) Keng, Lucas and Hewer, E. E. (1927). J. Anat. vol. uxt, 3, pp. 301-24.


Journal of Anatomy, Vol. LX VI, Part 1

myelinated fibres appear black, as do also the capillaries.

Plate I

1. T.S. Cervical region of cord. x 12. Foetus of 24-26 weeks. The myelination in the column of Burdach is very marked, and clearly heavier than that in the column of Goll. In addition, anterior and posterior root fibres are well myelinated. The myelination in the ground bundles shows particularly well owing to the absence of myelination in the adjacent tracts. Myelinated fibres crossing in the anterior commissure should be noted.

2. T.S. Thoracic region of cord. x 9. Full-time foetus. The direct cerebellar tract, posterior columns and the ground bundles are intensely stained. The lack of myelination is very obvious in the direct and crossed pyramidal, rubro-spinal and olivo-spinal tracts.

3. T.S. Thoracic region of cord. x 13. 36-week foetus. This-section of low thoracic cord shows differences in staining in the two posterior columns, that of Burdach being more intensely stained than that of Goll. A very faint degree of myelination can be observed in Lissauer’s tract; and the direct cerebellar tract is at this age well myelinated, though the staining exhibited is not so intense as in fig. 2 in full-time specimen. The absence of staining in the crossed pyramidal and rubro-spinal tracts is well seen.

4. T.S. of medulla at mid-olivary level. x 9. Foetus of 28 weeks. (Lived 19 days and myelination further advanced than is usual in foetus of this age.) The myelination of sensory and connecting fibres is well shown, particularly in the solitary bundle, medial longitudinal bundle and fillet. The lateral part of the olive is devoid of myelinated fibres, whereas the medial and dorsal portions (older parts) have well myelinated connections.

Plate II

5. T.S. Medulla. x 30. Full-time foetus. The section shows the origin of the 12th cranial nerve from its nucleus.

6. T.S. Lower border of pons. x 9. Foetus of 26-28 weeks. The staining of the section is somewhat patchy, but the myelinated connection of the cochlear fibres of the 8th nerve with the superior olive can be well seen.

Plate III

7. T.S. Lower border of pons. x 6. Full-time foetus. The 6th, 7th and 8th cranial nerves can be seen. When compared with fig. 6 it is clear that the myelination of the trapezium fibres is further advanced in this older specimen.

8. T.S. showing nuclei of 5th cranial nerve. x 25. Full-time foetus. The large rounded cells of the mesencephalic nucleus are in sharp contrast with the multipolar cells of the motor nucleus. The lining epithelium of the fourth ventricle is shown at the left of the picture, and the mesencephalic root of the 5th nerve can be seen passing dorsally and lateral to the large cells.

9. Section through subthalamic region. x 8. 28-week foetus. (Lived 19 days.) The myelinated fibres passing from the nucleus subthalamicus across the internal capsule to the striate body are well shown.

Plate IV

10. T.S. mid-brain. x 8. 28-week foetus. (Lived 19 days.) The fasciculus retroflexus of Meynert is seen as a partially myelinated bundle of fibres in the red nucleus.

11. T.S. At upper border of pons. x 11. Full-time foetus. In the posterior part of the section is seen the posterior longitudinal bundle medially, with the mesencephalic nucleus of the 5th nerve, the brachium conjunctivum and the lateral fillet laterally. The crossing of the brachia conjunctiva can be seen in the mid line, and still more anteriorly the medial fillet. Particularly noticeable are the unstained “aberrant pyramidal fibres” among the stained fillet fibres.

Plate V

12. Section cut obliquely through the red nucleus. x 34. Child of 8 months. The optic fibres are seen passing to the external geniculate body, and also sweeping round to the thalamus. The myelination of the pyramidal fibres in the crusta is still incomplete.

13. Section through mid-brain. x 11. Full-time foetus. This section shows the fountain decussation of Meynert and the 3rd cranial nerve.

Cite this page: Hill, M.A. (2020, April 10) Embryology Paper - Some observations on myelination in the human central nervous system (1931). Retrieved from

What Links Here?
© Dr Mark Hill 2020, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G[Category:Glia]]