Talk:Paper - The prenatal medullation of the sheep's nervous system (1947)

From Embryology

[ 64 ]


By G. J. ROMANES, Beit Memorial Research Fellow, Anatomy School, Cambridge


Many attempts have been made to correlate the appearance of stainable myelin in the nervous system with the onset of function in developing tracts and nerves (Tilney & Casamajor, 1924; Langworthy, 1926, 1928 a, b, 1980, 1933; Angulo y Gonzalez, 1929). Langworthy (1933) gives a good review of the literature and comes to the conclusion that medullation occurs in different fibre systems in the order of their phylogenetic development, and that normal function and medullation are closely correlated in their time of onset.

The sheep foetus is considered a suitable subject for the study of medullation in the nervous system because, unlike most of the animals which have been investigated, it is born in an advanced state of development and the various stages in the functional growth of the nervous system are well known (Barcroft & Barron, 1936, 19387, 1939, 1942).


Sheep foetuses ranging in age from 53 to 140 days after insemination have been used (Table 1). With the exception of two of these (53- and 68-day-old foetuses) which are aged on the basis of their crown-rump length and weight, all have been obtained by Caesarean section from sheep whose stage of gestation is

Table 1. Sheep foetuses

No. Age in days Weight in g. C.R.L. in mm. — 53 25-5 80-0 _— 63 102 142 299 66 97-2 140 340 78 222-2 180 268 78 229-0 - 192 281 84 348-2 210 331 96 533 250 309 lll 1450 355 682 120 2200 390 689 129 1670 360 98 140 4090 470

accurately known. After fixation by injection with 4% formaldehyde through an umbilical artery the brains were removed, cut into slices not more than 1 cm. in thickness and mordanted in Weigert’s primary mordant (bichromatefluorchrome) for 14 days at 30° C. The pieces were then thoroughly washed in water, dehydrated through absolute alcohol and embedded in either celloidin The prenatal medullation of the sheep’s nervous system 65

or paraffin. The latter method gives good results, and where relatively thin sections (15) are cut these can be affixed to albumenized slides without difficulty by floating on 70 % alcohol, blotting with cigarette paper and drying on the hot plate. Thin sections are infinitely preferable to thick ones for demonstrating early myelin, as the differentiation of the background without destaining the myelin is much simpler.

Sections were then placed for 1 hr. in Weigert’s gliabieze, washed, stained for 4-6 hr. in Kultschitzky’s haematoxylin at 37°C., rewashed, rinsed in gliabieze and differentiated by repeated short immersions in 0-25 % potassium permanganate followed by weak sulphurous acid.

With this method it has been found possible to show the finest myelin sheaths and at the same time obtain adequate differentiation of the background.

Serial sections throughout the whole brain and parts of the spinal cord have been cut in all but the 84-day-old foetus where only the spinal cord had been obtained.


In the following description no attempt has been made to ascertain the degree of medullation of individual tracts at every stage, but only the earliest time at which myelin is seen in each tract.

Table 2 gives a summary of the main observations, showing the earliest age at which myelin is seen. Where a trace (tr.) is marked on the table this means that only a very few fibres have a myelin sheath at this age. A + does not necessarily mean that every fibre is medullated but that the majority have some myelin.

No myelin is present in any part of the nervous system of the 58-day sheep foetus, though by 68 days several elements of the nervous system have begun to acquire myelin sheaths. The most obvious of these is the vestibular nerve whose branches can be traced into the lateral, superior and medial vestibular nuclei. A few fibres pass through the latter to decussate with similar fibres of the opposite side and appear to pass to the contralateral vestibular nuclei. The uncrossed vestibulo-spinal tract contains several medullated fibres, and a small number of homolateral ascending fibres can be traced from the superior vestibular nuclei into the medial longitudinal fasciculus as far as the trochlear nucleus.

Of the other cranial nerves the oculomotor and trochlear contain only a trace of myelin while the motor root of the trigeminal, abducens, facial and hypoglossal contain slightly more—as do the ventral spinal nerve roots in the cervical region. In addition only the reticulo-spinal fibres contain any myelin. These arise both from the pons and medulla as scattered fibres converging on the medial longitudinal fasciculus (Fig. 2 A) where they turn caudally with or without decussation. A well-marked group of these fibres is seen in all the foetuses (Fig. 1 B) lying just caudal and parallel to the fasciculi of the abducens

_ Anatomy 81 5 66

Table 2. Showing the main fibre systems which have acquired a myelin

G. J. Romanes

sheath in foetuses of various ages

tr. =a very few fibres with a fine sheath. + =many fibres with a myelin sheath extending

for at least part of their length.

Nerve or tract

Ventral roots

III, IV, VI, VII, VIII (vest.), XII

V (motor)

Medial longitudinal fasciculus

Vestibulo-spinal, uncrossed

Vestibulo-mesencephalic, uncrossed

Crossed, vestibular

Medullary reticulo-spinal

Pontine reticulo-spinal

Ventral column of spinal cord

Dorsal roots

V sensory mesencephalic and descending


Dorsal spino-cerebellar

Uncinate fasciculus

Cerebellar vermis

Lateral column, spinal cord

VIII descending and cochlear

Crossed vestibulo-spinal

Crossed vestibulo-mesencephalic

Isthmic reticulo-spinal

Solitary tract

Ventral spino-cerebellar

Brachium conjunctivum

Myelin in flocculus

Trapezoid body

Lateral lemniscus

Striae acousticae


Mesencephalic reticular at level of inferior colliculus

Myelin in inferior colliculus

Posterior commissure

Posterior column of spinal cord

Olfactory tract

Optic nerve and tract

Myelin in lateral cerebellar lobes

Brachium pontis

Commissure of the lateral lemniscus

Peduncle superior olive

Intercollicular commissure



Central tegmental fasciculus

Mesencephalic reticular at level of superior colliculus

Stratum profundum of superior colliculus :

Supramamillary decussation


Stria habenularis

Stria terminalis

Fimbria .

Optic radiation

Olfacto-septal fibres

Age in days A

J, ++++44+4+4+4+4+8

PIE L PPL i bid bbe tte e +e t4etettt44 8


co g .

PITT TE ttiids+












_ BJ

tt+tte+ The prenatal medullation of the sheep’s nervous system 67

Table 2 (continued) Age in days


+ tr.?

Nerve or tract 120 129 140

Fasciculus and ansa lenticularis

Fasciculus thalamicus

Dorsal and ventral supraoptic decussation

Medial lemniscus

Fibres in substantia nigra

Efferents from subthalamic nucleus

External capsule

Sigmoid and coronal gyri


Ventral and dorsal external arcuate

Brachium of inferior colliculus

Brachium of superior colliculus

Myelin in stratum superficiale of superior colliculus

Fibres in central grey of mesencephalon yO

Mamillary peduncle

Anterior column of fornix

Mamillo-thalamic tract

Efferents from anterior thalamic nucleus

Efferents from ventral thalamic nucleus

Pyramid to spinal cord

Anterior commissure:

anterior limb posterior limb Pyriform cortex Corpus callosum

I 11 8) I11 8 I 118 lll @

PlIII tl Plt Itt t Plt ttl | bh l ttl


S++ +++ ¢+¢+ +445 +++t4e4 +44 +eteete +44


L || I || [| | I | | [It

| | | | |


hd Ld bh Ltt bh


| | | | | +

t+ ot tte + +44

| | | | | |

t+te + + EEE F Ft FH+ttete +44

SHT+ + + +444 +

bid It | td | Ill Il II || | +

nerve. Together these pontine and medullary reticulo-spinal fibres run caudally just ventral to the medial longitudinal fasciculus forming an ill-defined group which comes to lie dorso-medial to the homolateral vestibulo-spinal tract in the ventral funiculus of the spinal cord. Here it is joined by a few fibres from the lateral reticular formation of the medulla which course ventrally through the lateral funiculus of the spinal cord. In foetuses of 78 days (Fig. 1 A) and older these are obscured by a greater number of fibres which enter the lateral funiculus of the spinal cord from the lower medulla and, lying with the rubrospinal tract, constitute the lateral reticulo-spinal tract. In the 63-day-old foetus a few fibres enter the ventral funiculus of the spinal cord from the ventral reticular formation of the medulla; whether these represent the ventral reticulo-spinal tract or not was not determined.


66-day foetus

In this foetus the dorsal spinal roots in the cervical region contain a little myelin, there is some in all parts of the sensory root of the trigeminal nerve, especially the mesencephalic component which can be traced to the level of the inferior colliculus, and in the glosso-pharyngeal, vagus and accessory nerves.

5-2 68 G. J. Romanes

The lateral funiculus of the cervical and upper thoracic spinal cord contains a few medullated fibres of the dorsal spino-cerebellar tract whose fibres pass to the vermis of the cerebellum (Fig. 2 A). A few fibres of the uncinate fasci

Fig. 1. Drawings.of Weigert-Pal preparations of the medulla and cerebellum of three sheep foetuses. A and B, 78-day foetus, x 7-5; C, 96-day foetus, x 7-5; D, 140-day foetus, x5.

culus arch ventrally from the region of the fastigial nucleus to the vestibular

nuclei which also send fibres into the roof nuclei of the cerebellum. The

’ medial longitudinal fasciculus is slightly medullated, especially in its pontine and medullary extent.

78-day foetus

In younger foetuses medullation is confined to the peripheral nerves and

central connexions of the vestibular apparatus and the reticulo-spinal path ways. In the 78-day foetus there is a moderate amount of myelin in the The prenatal medullation of the sheep’s nervous system 69

tegmentum of the midbrain (Fig. 5 A). A few fibres of the rubro-spinal tract are medullated and can be traced to the cervical spinal cord. In addition, reticulo-spinal fibres arise from the isthmic region and decussate among a very few fibres of the brachium conjunctivum which cannot be traced forwards to the red nucleus, though the medial longitudinal fasciculus is medullated to the ~ level of the posterior commissure in which are one or two fibres with a myelin sheath.

FRNA) ‘\ mares “

Fig. 2. Drawings of sections through the vestibular nerve in sheep foetuses. Stain Weigert-Pal. A, 66-day foetus, x6; B, 78-day foetus, x 7-5; C, 96-day foetus, x 7-5.

All the connexions of the vestibular apparatus are more heavily medullated (Fig. 2B) and include the cerebellar and descending roots. A few crossed vestibulo-spinal fibres are found arising from the nucleus of the latter (Fig. 1 B), and some crossed vestibulo-mesencephalic fibres from the medial nucleus have a myelin sheath. Both cochlear nuclei contain a few medullated fibres, and these can be traced in the striae acousticae and trapezoid body to a finely medullated lateral lemniscus which reaches the inferior colliculus. 70 G. J. Romanes

The solitary tract (Fig. 1B) contains some myelin, and the number of reticulo-spinal fibres from pons and medulla is greatly increased (Figs. 1B and 2 B).

In the cerebellum the flocculus contains some medullated fibres, and the ventral spino-cerebellar tract can be traced over the brachium conjunctivum into the cerebellum (Fig. 4 A).

The lateral funiculus of the spinal cord contains much more myelin on account of the presence of the rubro-spinal, both spino-cerebellar and the lateral reticulo-spinal tracts, and while a few fibres can be traced in the dorsal fasciculi to the gracile and cuneate nuclei, no internal arcuate fibres are present in the medulla and no collaterals from the dorsal funiculi entering the spinal cord show a medullary sheath. The cervical spinal cord contains more myelin than the thoracic or lumbo-sacral regions. The two foetuses of 78 days show no significant difference in their degree of medullation.

84-day foetus

All the funiculi of the spinal cord contain a uniform scattering of medullated fibres without evidence of any localized deficiency. The dorsal funiculi contain the lowest concentration of medullated fibres, and the degree of myelination is uniform throughout the length of the cord with the exception that a few medullated collaterals can be seen entering the dorsal horns from the dorsal funiculi in the upper cervical region.

96-day foetus

In this foetus there is a marked increase in the amount of myelin and medullated fibres are present in the forebrain.

Spinal cord. There is a considerable increase in the number of medullated fibres in all funiculi especially the dorsal, and it seems certain that the spinothalamic tracts are medullated by this stage though they were never seen as separate entities. Many collaterals enter the dorsal horns from the dorsal funiculi in all regions of the spinal cord.

Hindbrain. The main advance in this region consists in the appearance of the medial lemniscus (Figs. 1 C, 2 C, 4 B) which can be traced to the level of the trochlear nucleus. In front of this only a few scattered fibres can be found, and it is doubtful whether the lemniscus reaches the thalamus. In addition, the ventral and dorsal external arcuate and olivo-cerebellar fibres are medullated and longitudinal fascicles in the pontine and medullary reticular formation are much more obvious. The medial longitudinal bundle can be traced to the level of the posterior commissure, where its fibres fan out into the reticular substance above the substantia nigra.

The superior cerebellar peduncle (Fig. 4 B) is well medullated and reaches the red nucleus but cannot be traced for any great distance farther. The descending limb of the brachium conjunctivum (Fig. 4 B), which arises in the decussation, receives fibres from the ipsilateral tegmentum apparently The prenatal medullation of the sheep’s nervous system 71

originating dorsal to the substantia nigra, and can be traced caudally to the pons where a few medullated fibres of the middle cerebellar peduncle are present. The lateral lobes of the cerebellum contain- very little myelin.

Midbrain. Fibres of the tecto-spinal and spino-tectal tracts (Fig. 4B) are medullated, and in the isthmus the central tegmental fasciculus, is beginning to be medullated. Its fibres appear to originate mainly from the principal sensory trigeminal nucleus of the same side but could not be traced to the thalamus.

The commissures of the lateral lemniscus and inferior colliculus are medullated.

Lateral to the substantia nigra there is a scattered group of fibres (Fig. 5 B)

which seems to be formed partly from the upward continuation of the medial lemniscus and partly from the stratum profundum of the superior colliculus. These can be traced forwards into the lateral part of the fasciculus thalamicus and also give rise to Meynert’s commissure. This appearance is in accord with the findings of Magoun & Ranson (1942).

Forebrain. The olfactory and optic tracts contain a considerable number of finely medullated fibres. Optic fibres can be traced to the lateral geniculate body and pretectal nucleus. From the former a few medullated fibres pass into the posterior part of the internal capsule but are so scanty that they cannot be followed to the cortex. Both ventral and dorsal supraoptic commissures contain medullated fibres.

The lateral olfactory tract is medullated over its whole extent, but fibres are not found in the pyriform cortex. From the medial olfactory area a few olfacto-septal fibres pass to the nuclei of the septum pellucidum, and a fine scattering of medullated fibres can be traced to this region in the fimbria but no myelin is present in the anterior column of the fornix.

The habenular and terminal striae both contain a little myelin, and a very few medullated fibres are discernible in the habenulo-peduncular tract.

A small number of fibres with fine medullary sheaths arises from the dorsal part of the frontal cortex. These are almost confined to the sigmoid gyrus but a few are present in the coronal gyrus; they enter the internal capsule and pass caudally as far as the putamen of the lentiform nucleus. In. the latter there are a few scattered fibres, and from the globus pallidus a finely medullated bundle forming the fasciculus and ansa lenticularis passes to the subthalamic region. In the subthalamic nucleus there are several myelinated fibres which can be traced to the H, Field of Forel.

111-day foetus

This stage is not associated with such a great increase in the number of medullated tracts, but rather there is an increase in the degree of medullation of those already seen.

The brachia pontis and all parts of the cerebellar hemispheres now contain -a considerable quantity of myelin (Fig. 3 A). In the midbrain (Fig. 5 C) the 72 : G. J. Romanes

brachia of both colliculi are medullated and a few scattered fibres are found in the stratum superficiale of the superior colliculus, with many fibres entering the central grey matter and sweeping ventrally through it towards the oculomotor nucleus and dorsal tegmental decussation. The habenulo-peduncular tract can be followed to the interpeduncular nucleus (Figs. 4 C, 5 C).

Fig. 3. Drawings of sections through the vestibular nerve in sheep foetuses. Stain Weigert-Pal. A, 111-day foetus, x6; B, 140-day foetus, x 5.

_ The mamillary peduncle, the fornix and the mamillo-thalamic tract can all be traced to the mamillary body though the fornix is very poorly medullated.

Fibres of the brachia conjunctiva can now be traced upwards through the red nucleus into the subthalamic region, and both medial lemniscus and spinothalamic fasciculi extend to the thalamus. The prenatal medullation of the sheep’s nervous system 73

Fig. 4. Drawings of sections through the decussation of the trochlear nerve in sheep foetuses. Stain Weigert-Pal. A, 78-day foetus, x 7-5; B, 96-day foetus, x 7-5; C, 111-day foetus, x6 3 D, 140-day foetus, x5. 14 G. J. Romanes

Fig. 5. Drawings of sections through the oculomotor nucleus in sheep foetuses. A, 78-day foetus, x 7-5; B, 96-day foetus, x 7-5; C, 111-day foetus, x6; D, 140-day foetus, x5. The prenatal medullation of the sheep’s nervous system 75

Most of the thalamic nuclei, with the exception of the pulvinar and dorsal part of the medial nuclear group, contain a few scattered fibres, the anterior nuclei receiving the mamillo-thalamic fasciculus and giving a well-marked group of efferents into the internal capsule (Fig. 6 C), and while the central tegmental fasciculus (Figs. 4 C, 5 C) can be followed to the nucleus reuniens its cephalic part is poorly medullated, though at 140 days (Fig. 5 D, 7 A) this portion is more heavily medullated than the rest.


Fig. 6. Drawings of sections through four levels of the forebrain from a sheep foetus of 111 days. Stain Weigert-Pal, x2.

Efferent fibres are present from the ventral, lateral, anterior, lateral geniculate and medial geniculate nuclei of the thalamus.

The number of medullated cortical connexions is increased though the fibres are not numerous; they are present in the cingulate, coronal, lateral, mid-suprasylvian, posterior suprasylvian, anterior and posterior ectosylvian gyri (Fig. 6).

Efferent cortical fibres can be traced through the corpus striatum into-the cerebral peduncle. The great majority of these contain no myelin caudal to 76 G. J. Romanes

the subthalamic nucleus, but a few (Figs. 5 C, 6 C), whose nature is not determined, pass caudally to the level of the isthmus and appear to end in the ventral part of the tegmentum of the midbrain. There is no myelin in any of the pontine pyramidal fasciculi.

There are no medullated associational fibres in the cortex, and the corpus callosum is devoid of myelin (Fig. 6).


Fig.-7. Drawings of sections through four levels of the forebrain from a sheep foetus of 140 deys. Stain Weigert-Pal, x2.

120-day foetus

The central nervous system of this foetus shows no obvious increase in the number of medullated tracts. More myelin is present in the cortex, and a few fibres are found in all regions except the pyriform lobe, the orbital gyri and the extremity of the occipital pole. The prenatal medullation of the sheep’s nervous system 77

Scantily medullated pyramidal fibres can be traced through the pons and a few extend as far caudally as the cervical spinal cord.

All thalamic peduncles contain more myelin and intrathalamic fascicles are more obvious than in the previous foetus, but there is still little myelin in the dorsal part of the medial nuclear group and pulvinar. A few short associational fibres are present in the cortex, but the corpus callosum is devoid of myelin.

The anterior limb of the anterior commissure contains medullated fibres, and though no commissural fibres with a myelin sheath are present in the. posterior limb a few derived from the fascicles of the internal capsule run in its lateral part. ,

180- and 140-day foetuses

In both of these foetuses all the thalamic connexions are medullated (Fig. 7), ‘and in addition there are medullated fibres in the posterior limb of the anterior commissure (Fig. 7 C) and in the corpus callosum. All parts of the cortex contain several medullated fibres though these are still not numerous in the orbital gyri, the pyriform lobe or the occipital pole; in all regions there are fibres showing the beaded appearance characteristic of early medullation.

The pyramidal tract is much more strongly medullated in the 140- than in the 180-day foetus and can be traced into the upper cervical spinal cord (Figs. 1-3).


The present observations on a single series of sheep foetuses cannot be considered exhaustive as there may be considerable variation in the degree of medullation at a given age, not only between different breeds of sheep but also in the same breed. Such a variation is present in the literature dealing with the development of myelin in the human foetus and infant. Table 3 shows the range of variation in the results given in two papers dealing with the first appearance of myelin in man (Lucas. Keene & Hewer, 1931; Langworthy, 1983). Some of these variations are very marked involving tracts at all levels , of the nervous system, and, though it is not quite clear from Langworthy’s (1933) paper how he arrived at a table giving the medullation at ten ages after - describing five different ages, it is obvious from his text that his table refers to the first appearance of myelin.

Though the above variations may be the result of differences in material, technical procedures or the level at which some of the tracts are studied, they are too great to allow any conclusions to be drawn about medullation in one foetus and functional development in another.

Despite these differences in man and the fact that the sheep foetuses used in this study have been collected in a random fashion over a period of years, each shows a distinct increase in the degree of medullation over the previous member of the series.. Also the two sheep foetuses (268 and 340), 78 days old, show an exactly similar degree of medullation though born in different years. Thus this limited series of observations can be taken to give only a general 78 G. J. Romanes

survey of the development of stainable myelin in the sheep, and an attempt may be made to correlate this with the development of behaviour only because the material for this paper is part of that used by Barcroft & Barron in their physiological experiments.

Table 3. Comparing the results obtained by two authors on the development of myelin in man

Age of medullation in weeks >— Difference as Lucas Keene & Langworthy Difference % of smaller

Fibre system Hewer (1931) (1933) in weeks observation Olivo-cerebellar 22 : 4* 22 100 Ventral spino-cerebellar 14 28 14 100 Tractus cuneatus 14 28 14 100 Tecto-spinal 24-26 4* 18 70 Fornix 32* 8* 24 50 Posterior spinal roots and 14 20 6 43

cranial nerves , Dorsal spino-cerebellar 16 20 4 25 Cortico-spinal 36 4* 8 22 Cortico-pontine Birth 8* 8 20 Habenulo-peduncular 23-24 28 4 16 Medial lemniscus 24 28 4 16 Ventral roots 13-14 16 2 16 Pallido-subthalamic 26-28 32 4 14 Rubro-spinal 36 32 4 11 Mamillo-thalamic 12* 8* 4 7 V (sensory) and VIII (cochlear) 22-24 24 0 0 Optic tract 36 - 36 0 0

  • — weeks after birth.

In general, the order of medullation of the various elements in the nervous system of the sheep agrees with those found in other mammals (Tilney & Casamajor, 1924; Langworthy, 1928 a, b, 1980, 1988; Lucas Keene & Hewer, 1981, 1933), but the degree of medullation is much more advanced at birth. This precocity of foetal medullation has disclosed the fact that not only do some nerve fibres, such as the ventral spinal roots, become medullated long after there is physiological evidence of their ability to function (Barcroft &

- Barron, 1942), but also that the optic tract begins to be medullated at least 50 days prior to the exposure of the eye to light.

Unfortunately, it is impossible from the physiological findings to determine exactly which tracts in the nervous system begin to function at a given time, but it has been possible to associate particular regions of the brain with certain phases of development by transecting the brain and spinal cord of older foetuses at various levels and comparing the activity which results with that seen during development (Barcroft & Barron, 1942).

Their results, in conjunction with Table 2, show clearly that activity precedes medullation by a considerable period in all levels of the brain. The onset of inhibition described by Barcroft & Barron (1942) corresponds in time with the first appearance of stainable myelin in the spinal cord, medulla and pons. But by their experiments these authors have demonstrated that when the The prenatal medullation of the sheep’s ‘nervous system 79

midbrain is severed from the pons at this developmental stage, the pontine and medullary segments of the brain demonstrate no evidence of inhibition though they alone contain myelin. Thus the development of myelin is not responsible for the inhibition, the operation merely causing the foetus to revert to a less advanced stage in its functional development without showing any evidence that the presence of myelin in regions caudal to the level of section has altered their functional capabilities. Even when the spinal cord is severed in the cervical region at a time when it contains a moderate number of medullated fibres (70 days), the activity of the trunk and limbs caudal to the section reverts to that found in normal foetuses 836-40 days old when no myelin is present.

There is therefore no evidence that the onset of medullation produces any marked change in the physiological response of the foetus, and the appearance of sustained movements (Barcroft & Barron, 1942) precedes the development of myelin. In contrast to Langworthy’s (1926) findings but in agreement with those of Windle (1929), decerebrate rigidity appears in development at a time when the midbrain contains practically no myelin and the rubro-spinal apparatus is quite devoid of it.

These facts are in favour of the view that function precedes medullation and is independent of it for a considerable time. On the other hand, the optic nerve in the sheep begins to be medullated while the foetus has still 50 days to remain in utero. Here medullation can be conditioned neither by the onset of normal function in this nervous pathway nor by association with the ability to open the eyes (Langworthy, 1933), for the eyelids are still tightly fused 15 days later and only begin to separate 24 days after medullation in the optic nerve has begun. It may be that the development of myelin in the optic system is initiated by factors quite different from those found elsewhere in the nervous system, but it is clear that the transmission of impulses resulting from light reaching the retina is not a factor in the sheep or in man where the optic nerve also begins to medullate before birth (Lucas Keene & Hewer, 1931; Langworthy, 1938).

The conclusion seems warranted that medullation can occur in the absence of normal function and, though this does not help to clarify the mechanism, that the developing nervous system prepares itself for the demands which are to be put upon it at birth. It is obvious that a sheep is born with a nervous mechanism more highly developed than that of a newborn rabbit, yet each has as little knowledge of the world outside as the other. This inherent development in the absence of function has been demonstrated by Harrison (1904), who has shown that development of functional capabilities in the amphibian nervous system is not retarded by immersion in chloretone which removes from the larva the ability to mould its nervous system by its activity. Similarly, Goodman (1932) states that placing newborn rabbits in the dark for 6 months does not impair medullation of the optic nerve and, though Held (1928) believes that exposure to light hastens this process, it is clear that the absence of function does not prevent or decrease medullation in the optic system. 80 G. J. Romanes

The connexion between thickness of the myelin sheath, diameter of axon and speed of conduction is well known in the adult nervous system, but the part played by the myelin is not understood. The present study produces evidence that in some cases nerve-fibre systems may function before they develop a medullary sheath, while others become medullated before they are called upon to function. Medullation forms one of the last phases in maturation of the nerve fibre, and it seems likely that tracts once medullated are capable of functioning, though it does not follow that where there is no myelin there is no function. This lack of direct correlation between medullation and function agrees with Angulo y Gonzalez (1929) and Windle (1929) but disagrees with Tilney & Casamajor (1924). The reason for this may be that in studying the early postnatal development of animals such as kittens the process is so rapid that the delay which may exist between function and eclipsed, but in the paper by Tilney & Casamajor the prenatal behavioural activity is not studied.


1. The order of medullation is described in many of the tracts in the nervous system of the foetal sheep.

2. Medullation starts at the 63rd day of gestation with the vestibular apparatus, the lower motor neurones and the reticulo-spinal tracts. By 78 days the midbrain connexions have begun to acquire medullary sheaths, and the first myelin appears in the forebrain at 96 days.

8. A comparison of the time of medullation with the functional development described by Barcroft & Barron shows that function in most regions of the . brain is established 3-4 weeks before medullation begins. In contrast the optic nerve begins to be medullated 50 days before full term. The relation between the onset of medullation and function is not clear, but it is certain that normal function can occur in the absence of myelin and, in the optic tract at least, medullation begins in the absence of normal function.

4. The possibility is discussed that medullation is an inherent developmental characteristic which is independent of function but may be accelerated by it.

I am indebted to Sir Joseph Barcroft who supplied the material for this study and the figures necessary for computing the age of the two youngest foetuses, which were obtained through the kindness of Dr D. V. Davies, of the Anatomy School, Cambridge. ,


ac anterior commissure cic commissure of inferior colliculus

be brachium conjunctivum cll commissure of lateral lemniscus

bed descending limb of brachium con- cm mamillary body . junctivum cn cochlear nucleus

bic _ brachium of inferior colliculus cp cerebral peduncle

bp _ brachium pontis ctf central tegmental fasciculus

c cerebellar root of vestibular nerve dc ___ dorsal spino-cerebellar tract

ce corpus callosum .dvn descending vestibular nucleus The prenatal medullation of the sheep’s nervous system 81

f fornix rs rubro-spinal tract tg facial genu rz medial reticulo-spinal tract and defn facial nucleus cussation g glossopharyngeal nerve 8 solitary tract hp __habenulo-peduncular tract 8 superior colliculus te inferior colliculus 8e sulcus ectolateralis ig interpeduncular ganglion 8g sulcus cinguli to inferior olive al sulcus lateralis tv trochlear nerve . 80 superior olive

lateral geniculate . sr sulcus rhinalis u lateral lemniscus . 88 sulcus splenialis lr lateral reticulo-spinal tract ssu sulcus suprasylvianus lun lateral vestibular nucleus st spino-tectal tract m motor root of trigeminal nerve sv sensory root of trigeminal nerve ml medial lemniscus . syn _ superior vestibular nucleus mf medial longitudinal fasciculus sy sulcus sylvianus mp mamillary peduncle t stria terminalis mv mesencephalic root of trigeminal nerve tb trapezoid body moun medial vestibular nucleus ts tecto-spinal tract na nucleus ambiguus uf uncinate fasciculus nd _ nucleus dentatus v vestibular nerve nl nucleus of lateral lemniscus ve ventral spino-cerebellar tract 0 oculomotor nerve or nucleus vd _— descending trigeminal root or optic radiation vw abducens nerve ot optic tract vit__— facial nerve Dp pons vm _ vestibulo-mesencephalic tract pe posterior commissure v8 uncrossed vestibulo-spinal tract pm pyramid vin ventral tegmental nucleus r reticular fibres ‘ ve crossed vestibulo-spinal rb restiform body x crossing vestibular fibres rn __ red nucleus


ANGULO y GonzaLEz, A. W. (1929). J. comp. Neurol. 48, 459. Barorort, J. & Barron, D. H. (1936). J. Physiol. 88, 56.

Baxrcrort, J. & Barron, D. H. (1937). J. Physiol. 91, 329.

Baxrcrort, J. & Barron, D. H. (1939). J. comp. Neurol. 70, 477. Barorort, J. & Barron, D. H. (1942). J. comp. Neurol. 77, 431. Goopman, L. (1932). Amer. J. Physiol. 100, 46.

Harrison, R. G. (1904). Amer. J. Anat. 3, 197.

HEtp (1928). Quoted from Mann, I. C., The Development of the Human Eye. Cambridge. Lanawortsy, O. R. (1926). Contr. Embryol. Carneg. Instn, 17, 125. Lanawortay, O. R. (1928a). Contr. Embryol. Carneg. Instn, 20, 127. Lanewortsy, O. R. (19286). J. comp. Neurol. 46, 201.

Lanawortay, O. R. (1930). Contr. Embryol. Carneg. Instn, 21, 37. Lanewortsy, O. R. (1933). Contr. Embryol. Carneg. Instn, 24, 1. Lucas Krenz, M. F. & Hewer, E. E. (1931). J. Anat., Lond., 66, 1. Lucas Krenz, M. F. & Hewesr, E. E. (1933). J. Anat., Lond., 67, 522. Macoon, H. W. & Ranson, M. (1942). J. comp. Neurol. 76, 435. Trey, F. & Casamagor, L. (1924). Arch. Neurol. Psychiat. 12, 1. Writ, W. F. (1929). J. comp. Neurol. 48, 227.

Anatomy 81 6