View source for Paper - The development and significance of the cell columns in the ventral horn of the cervical and upper thoracic spinal cord of the rabbit (1941)

{{Header}}
{{Ref-Romanes1941a}}
{| class="wikitable mw-collapsible mw-collapsed"
! Online Editor&nbsp;
|-
| [[File:Mark_Hill.jpg|90px|left]] This historic 1941 paper by Romanes describes {{spinal cord}} pathways in the {{rabbit}}.
<br>
Also by this author: {{Ref-Romanes1941b}}

<br>
'''Modern Notes:''' {{spinal cord}} | {{rabbit}}
<br>
{{Neural Links}}
<br>
|}
{{Historic Disclaimer}}
=The Development and Significance of the Cell Columns in the Ventral Horn of the Cervical and Upper Thoracic Spinal Cord of the Rabbit=


By G. J. Romanes

Department of Anatomy, University of Cambridge

==Introduction==
[[File:George John Romanes 1916-2014.jpg|thumb|150px|alt=George John Romanes 1916-2014|link=Embryology History - George Romanes|George John Romanes (1916-2014)]]
Auruovucua great deal of attention has been paid to the arrangement of the cell groups or columns in the spinal cord (Steida, 1870; Obersteiner, 1890; Schiefferdecker, 1874; Waldeyer, 1889; Kreyssig, 1885; Onuf, 1899; Bruce, 1901; v. Gehuchten & de Neef, 1900; Jacobsohn, 1908; Bok, 1928; Angulo y Gonzalez, 1927) and to the relation between these columns and the peripheral nerves or muscle masses (Sano, 1897, 1898; Marinesco, 1898 a; v. Gehuchten & de Buck, 1898; v. Gehuchten & Nelis, 1899; Sass, 1889; Hammond, 1894; de Neef, 1901; Yu-Ch’uan Tsang, 1989), very little is known concerning the development of the cell columns! in the spinal cord and the relation between this development and that of the muscles which the processes of these cells grow out to supply. Frazer (1931) mentions that he can see cell groups already present in the ventral horn of an 8 mm. human embryo cut in transverse section, but he does not trace out the development in any great detail.


The purpose of this communication is to describe the development of the cell columns in the cervical enlargement of the {{rabbit}} and the relation between this and the development of the peripheral musculature and the nerves in the forelimb. 


==Material and Methods==

Under urethane anaesthesia rabbit embryos of known age were removed by laparotomy from pregnant rabbits. These embryos were studied for signs of spontaneous movements and were stimulated with a weak faradic current to determine the age at which the muscles of the forelimb respond by contraction to direct or reflex stimulation. The embryos collected are enumerated in Table 1.

Small embryos were fixed whole in various fixatives including 95 % alcohol and 2% acetic acid for staining with toluidin blue, in absolute alcohol and 1% strong ammonia solution for pyridin-silver staining, and also in 10 % formalin. Larger embryos were injected through the umbilical artery, or the spinal cord was exposed prior to fixation.

Since this paper was completed Angulo y Gonzalez (1940) has published a study on the white rat in which he has found an arrangement strikingly similar to that in the rabbit. 


The spinal cords and dorsal root ganglia were removed intact from the acetic-alcohol fixed embryos, carefully straightened, embedded in paraffin wax, serially sectioned in transverse, coronal and sagittal planes and stained with toluidin blue. Various thicknesses of section were used, the thickest, 20 », being used only where it was not required to study the cell structure. In the case of the smallest embryos the whole of the spinal column was removed with the spinal cord to avoid unnecessary damage. The pyridin-silver preparations were cut transversely at 10 after impregnation in the silver bath for a period of from 5 to 7 days. This increased impregnation was found to give more satisfactory results in the case of rabbit embryos.

The spinal cords of several adult rabbits were fixed by injection with 10 % formalin, removed together with the spinal roots, and imbedded in paraffin wax. The cervical and upper thoracic regions were sectioned in transverse, coronal and sagittal planes and stained with toluidin blue. Owing to the diffuse nature of the cell columns in the adult rabbit the coronal and sagittal sections were not found so useful as in the earlier stages and were superseded by the transverse sections which were much more easily interpreted. For the interpretation of the sections in the coronal or sagittal planes of the younger rabbits the glass plate reconstruction method was used.

{{Romanes1941a table1}}

Table 1 , Crown-rump length of embryos Days from copulation mm. ll 6 12 8 14 11 15 15 16} 19 19 26 21 41 22 56 « 26 80 NOMENCLATURE

The common nomenclature used to describe the cell columns found in the spinal cord of various animals takes into account the position of these cell columns relative to each other in the ventral or anterior horn of the spinal cord. This nomenclature is only of use in cases where the cell columns are few and simple and where names may be used without drawing homologies with similarly named columns, not necessarily of the same function, in other animals. However, both these difficulties have been encountered in studying the rabbit’s spinal cord, and as a result an alternative numerical nomenclature has been used. Similar nomenclatures have already appeared in various ‘publications, and it is stressed here that the numbers used to designate the -various cell columns in this study do not correspond with numbers used by other authors. The ‘columns in this study are numbered in order of their appearance in serial transverse sections traced from the first cervical segment caudally. It should be noted that this nomenclature overcomes the difficulties which arise when a cell column changes its anatomical relationship to the other cell columns in the ventral horn. .

CELL COLUMNS IN THE CERVICAL AND UPPER THORACIC REGION OF THE ADULT RABBIT: 

- Throughout this section frequent reference should be made to Text-fig. 1B.

In the adult rabbit some ten separate cell columns in the ventral horn of the cervical and upper thoracic regions have been discovered; they are as follows:

Columns 1 and 2. These two columns will be considered together because of their close relation throughout the region under consideration. In the first cervical segment columns 1 and 2 appear just caudal to the hypoglossal nucleus at the ventro-medial edge of the ventral horn. Column 1 lies just dorsal and medial to column 2 which is a ventro-medial column throughout the greater part of its length. These two columns retain this arrangement until they reach the upper part of the eighth cervical segment where they fuse into a single column lying in a position intermediate to that taken up by the two separate columns in other regions. Caudal to this, till the upper part of the second thoracic segment is reached, the two columns are irregularly fused. In the upper part of the second thoracic segment they separate; column 1 takes up a ventro-medial position and column 2 an intermediate ventral position. Both these columns are made up of a small number of cells which, in transverse’ section, appear as typical multipolar cells of the ventral horn. It is usual to find not more than four or five cells at the most in each column in any one section, and the columns do not differ significantly in size from one level to another.

Column 3. In the first cervical segment this column is situated just lateral to column 1 and dorsal to column 2. It retains this arrangement throughout the first and second cervical segments. It then passes abruptly in a lateral direction to reach the medial edge of the lateral funiculus in the dorsal part of the ventral horn in the third cervical segment. This abrupt translation is not readily followed in the toluidin blue sections and is only clearly visible in the younger stages where the columns are more compact. However; in pyridinsilver preparations the two parts of the column are seen to be continuous, ‘not only as a cellular band but also by virtue of the atypical course of their processes. Both these parts of the column send their fibres in a dorsal direction to join the spinal accessory nerve. Traced further caudally, the column passes medialwards to fuse with column 5 and takes up a central position in the caudal part of the fourth cervical segment. It is not known whether column 5 extends as far as the sixth cervical segment, in combination with column 8, but fibres can be traced into the spinal accessory nerve from the caudal extremity of this column, denoting the presence of column 38.

Column 4. This column appears in the caudal part of the third cervical segment lying at the lateral edge of the ventral horn between column 2 ventrally and column 8 dorsally. Caudal to the third cervical segment the column remains in the same situation extending into the upper thoracic spinal cord. In the caudal part of the eighth cervical segment it is closely related to and often fused with column 8. In the upper part of the first thoracic segment it.fuses with column 9, mainly due to the fact that column 4 is sweeping medially into the narrower ventral horn of the upper thoracic segments and so impinges on the ventrally situated column 9. Throughout, column 4 is placed ventro-laterally. In the upper cervical region column 4 is of the same size as columns 1 and 2, but, as it sweeps into the laterally expanding ventral horn in the region of the fourth and fifth cervical segments, the number of cells which it contains increases and remains large until the middle of the first thoracic segment where it decreases in size as it sweeps medialwards.

Column 5. In the cephalic part of the fourth cervical segment column 5 is situated centrally in the ventral horn dorso-medial to column 4 and ventromedial to column 3. It extends caudally in this situation to the lower part of the fourth cervical segment where column 8, passing medially, fuses with it. The fused columns pass caudally together as.far as the upper part of the sixth cervical segment where they disappear. It is not possible to say how far column 5 extends caudal to its fusion with column 3, though it is known that eells of column 8 are found throughout the fused part. This column does not show any marked difference in size at various levels, consisting of some five cells in most sections.

Column 6. Column 6 appears in the cephalic part of the fifth cervical segment in the same position as that taken up by column 38 in the third and fourth cervical segments, that is, in the dorso-lateral part of the ventral horn dorsal to column 4. It retains this position throughout the fifth and cephalic parts of the sixth cervical segments, being displaced medially by the appearance of column 7 lateral and dorsal to it. In this new central position it passes caudally to fade out in the lower part of the eighth cervical segment in a position almost directly caudal but somewhat ventral to that taken up by the fused parts of columns 8 and 5. This column increases rapidly in size after its first appearance, reaching a maximum in the sixth cervical segment and decreasing in size again with the appearance of column 7 which seems to be split off from it. Throughout the seventh and upper parts of the eighth cervical segments it remains large, containing more cells than any of the previously described columns. It disappears rapidly in the lower part of the eighth cervical segment. 7

Column 7. This column appears in the caudal part of the sixth cervical segment by fission from the dorso-lateral aspect of column 6. Traced in a caudal direction, it is displaced dorsally by the appearance of column 8 between it and column 4 ventrally. It is continued caudally in this new situation into the caudal part of the first thoracic segment where it ends. This column does not show any great variation in size and contains about five cells when seen in transverse section.

Column 8. Column 8 appears in the eighth cervical segment at the lateral edge of the ventral horn, dorsal to column 4, lateral to column 6, and ventral to column 7. It is continued caudally in this position till it ends in the lower portion of the first thoracic segment. In the lower part of the eighth cervical and upper part of the first thoracic segments it tends to fuse ventrally with column 4, This column forms the most lateral edge of the ventral horn at the point where it extends furthest laterally in the cervical enlargement; it contains some eight to ten cells in transverse section and does not show any great changes in size except at either end where it fades out gradually.

Column 9. This is a short, ventrally situated, column passing from an intermediate ventral position in the ventral horn of the eighth cervical segment to fuse with the medial aspect of column 4 in the caudal part of the first thoracic segment, where this column inclines medially to enter the lateral part of the narrower thoracic ventral horn.

Column 10. This column appears dorsal to column.7 in the lower part of the eighth cervical segment and here constitutes the most dorsally situated cell column of the ventral horn in the cervical enlargement. It extends further caudally than columns 6, 7 or 8, entering into the cephalic two-thirds of the second thoracic segment where it ends in close proximity to the cephalic end of the intermedio-lateral cell column. Column 10, at first small, rapidly enlarges, reaching a maximum in the middle of the first thoracic segment where transverse sections frequently show between five and ten cells.

Caudal to the end of column 10, there appears the intermedio-lateral cell column which, together with columns 1, 2 and 4, makes up the columnar arrangement of the upper thoracic segments.

==The Cell Columns in Embryos and Foetuses of Various Ages==

Sections of a 6mm. rabbit embryo show that the cervical spinal cord consists of a closed tube made up of a layer of germinal cells thickened in the ventral horn region, but without any sign of differentiation of the cells or any cell columns.

By the time that the 8 mm. stage has been reached a definite ventral horn has been formed containing cells which stain more lightly than the parent germinal cells (Pl. 1, fig. 1). These cells are arranged in a dense mass within which no division into cell columns can be seen in any plane of section. In.the 11 mm. rabbit embryo, however, the first signs of cell groupings are visible and they take the form of an increased density of the undifferentiated neuroblasts into definite longitudinally arranged columns. These columns are difficult to interpret because of the density of the cells in the ventral horn and the undifferentiated nature of the cells forming the columns. In sagittal section (Pl. 2, fig. 3) it is impossible to differentiate the ventral horn cells into particular groups with any degree of certainty. In the coronal sections of the cervical enlargement, however, a laterally placed group, which appears homogeneous in sagittal section, can be differentiated from a centrally placed group (Pl. 3, fig. 2). In this plane of section no further groups are visible and it is in transverse section that the greatest degree of differentiation is demonstrable. Columns 1 and 2 are present but fused with each other throughout the greater part of their length. The lateral part of the ventral horn lodges a large group extending dorsally as far as the dorso-lateral part of the ventral horn (PI. 1, fig. 2). This large column, which is particularly deep in the caudal part of the cervical enlargement, appears to be split in some sections into more than one column. However, since this group shows no clear divisions in sagittal sections, the above appearance is considered to be due to the passage of fibres and blood vessels through it. This is borne out by the inconsistent nature of the division when compared with older embryos. This group probably represents all the cell columns of the lateral part of the ventral horn in an undifferentiated state.


In the cephalic part of the cervical spinal cord, columns 3, 4 and 5 are present ag separate entities, except in the region of fusion between columns 8 and 5. Traced caudally, column 4 fuses with the general lateral mass in the fifth cervical segment. Column 8 deserves special notice because the cells which it contains have reached a greater degree of histological differentiation than the cells of any other group at this stage. These cells are less elongated and contain a more darkly staining and definite Niss] substance in a greater volume of cytoplasm than in the other undifferentiated neuroblasts.

There is some suggestion of a separation of column 6 in the caudal part of the enlargement but this is not so definite as the separation of column 5. It is not possible to differentiate column 9 from the mass of cells forming the most ventral part of the ventral horn in the eighth cervical segment.

By the 15 mm. stage (Pl. 1, fig. 3; Pl. 2, fig. 4) a further degree of differentiation .has been reached, not only in the.cell columns but also in the cells forming the columns, many of which resemble the cells seen in column 8 of the 11 mm. stage. In the first three cervical segments the arrangement of the columns ‘is similar to that in the 11 mm. embryo except that columns 1 and 2 are more readily separable in this region though they fuse again in the fourth cervical segment and are not separable till the first thoracic segment is reached. Columns 8 and 5 are arranged as in the adult. Column 6 is distinct throughout its length and has its adult relations. Column 4 is also distinct throughout the greater part of its length except for occasional fusions with adjacent columns, especially in the eight cervical segment where it is fused with column 9 which is seen at this stage as a thickening of column 4 and only separable from the latter in its most cephalic part. Columns 7 and 8 are irregularly fused but are definitely separated from column 4 ventrally and column 6 medially throughout the greater part of their length. Column 7 is particularly large in the last cervical and first thoracic segments, but column 10 is not visible as a separate entity except in so far as the dorsal part of column 7 appears to extend further Text-fig,


Figs. 1A, B. Diagrams based on serial transverse sections of the spinal cord in 26 and 41 mm, rabbit embryos respectively. 1A, cervical enlargement of the


caudally than in the adult. This appears to be the caudal part of column 10 fused further cephalad with column 7.

In the 19 mm. stage all the columns described in the adult rabbit are visible as separate entities, though considerable fusions occur in various regions between these columns (PI. 1, fig. 4; Pl. 3, fig. 1). These fusions do not appear to be constant and so are not considered worthy of detailed description. Column 7 is now separated from columns 8 and 10, column 9 is separate from column 4, except in its caudal part, columns 1 and 2 show a smaller degree of fusion which extends only from the seventh cervical to the first thoracic segments. 


From this stage no significant changes in the arrangement of the cell columns have been seen and all subsequent changes are in the histological appearance and arrangement of the cells themselves within the columns.

In the 26 mm. stage (Text-fig. 1A; Pl. 1, fig. 5) the cells show as very discrete clumps histologically differentiated from the cells of the other stages and from the indifferent cells of the ventral horn not situated within the cell columns. The columns at this stage are more definite than at any previous or succeeding stage, for though a great number of undifferentiated cells, not appearing in the older stages and especially in the adult spinal cords, still exist between the columns, the density of cell grouping in these columns is very great. In this stage columns 1 and 2 show an even smaller degree of fusion which extends only through the eighth cervical and upper part of the first thoracic segments, and the degree of fusion seen between the various columns of the lateral part of the ventral horn is very much reduced, not exceeding that found in the adult.

In the 41 mm. rabbit embryo (Text-fig. 1B; Pl. 2, fig. 1), the cell columns are very definite but do not appear so compact as in the 26 mm. stage when viewed in transverse section. This decrease in the solidity of the columns increases steadily throughout foetal, and rapidly throughout postnatal, life, till, in the adult rabbit, cell columns are with difficulty separable from each other on account of the scattered nature of their cells (Pl. 2, fig. 2). It seems very probable that this scattering of the cells is conditioned by the growth of the spinal cord which continues late into postnatal development, the postnatal growth of the spinal cord in the rabbit being much greater than in Man, for in a full grown rabbit the spinal cord reaches into the upper part of the sacrum. In this connexion it is to be noted that the area of the grey matter in the eighth cervical segment as seen in transverse section increases rapidly after birth. It is in this stage that the thinning out of the columns is most marked and the growth in length of the spinal cord is most rapid.

In summary, the cell columns of the cervical enlargement of the spinal cord of the rabbit develop very rapidly, reaching their adult arrangement, with the exception of some minor changes, by 16} days after copulation and their most compact stage 2} days later in the 26 mm. rabbit embryo, having commenced their development in the 11 mm. stage, that is on the 14th day after copulation. 


This very rapid development of the columns and the undifferentiated nature of the cells involved make it difficult to determine the exact order of development of the cell columns, but, in general, the columns in the lateral part of the ventral horn appear to develop later than those on its medial aspect out of a common mass of cells whose differentiation passes in a ventro-dorsal direction ; that is, the most ventral columns of the lateral part of the ventral horn develop out of the common mass before the most dorsally situated columns (Pl. 1, figs. 1-4). :

This description of the developing cell columns follows closely that on the white rat which Angulo y Gonzalez (1940) published after this work was completed. It is important to note that this development is very similar in the two types of animals studied, even though the total number of columns present is not. the same.

==Localization==

In order to determine whether any correlation can be found between the cell columns seen in the spinal cord and the peripheral groups of muscles, the following series of experiments was carried out on adult rabbits:

{{Romanes1941a table2}}

Table 2 : Duration of

. experiment
Nerves cut Type of section Position days Median and ulnar Crushing and cutting Wrist 13 Median, ulnar and radial » Elbow 13. Median, ulnar, radial and Simple cutting Axilla 21 musculo-cutaneous Median and ulnar : Violent rupture Wrist 13 Median and ulnar » » 7

All these experiments were performed under aseptic conditions with a view to showing chromatolytic changes in the cells of the various columns in the spinal cord. The animals were killed with an overdose of chloroform and injected with 10% formalin which is a satisfactory fixative for adult rabbit material to be stained with toluidin blue or gallocyanin. The cervical enlargement from the fifth cervical to the second thoracic segments inclusive was removed and sectioned. ,

Study of these sections showed that none of the cells in the cervical region displayed the typical reaction which has beeri designated chromatolysis. Certain of the cells on the operated side showed an arrangement of the Nissl substance which was not visible on the unoperated side, consisting of a grouping of the Nissl substance as a more homogeneous ring around the centrally placed _ nucleus with a clearer zone of cytoplasm between this and the cell membrane. Nissl substance was absent in the dendrites. This was the greatest change observed, and many others which appeared slightly different from the normal were not readily distinguishable from those on the unoperated side. In the case of section of the median and ulnar nerves these changed cells were located in the most caudal and dorsal cell column of the enlargement (column 10); in the case of section at the elbow, cells had also undergone a change in column 7. 



Table 3 shows a similar but more extensive series of experiments carried out on adult mice.

These animals were killed with chloroform and either the spinal cord was removed and fixed in acetic-alcohol or Heidenhain’s Susa, or the whole spinal column was removed and fixed in Jenkin’s fixing decalcifying fluid. The cervical regions, fifth cervical to second or third thoracic segments, were imbedded with the methylbenzoate-celloidin-paraffin technique and serially sectioned. The sections were stained either with toluidin blue or gallocyanin.

There were no cells visible in any of these spinal cords showing definite signs of the chromatolytic reaction and, though there were many which did not appear quite normal, none of them was sufficiently different from those on the uninjured side to allow of a definite analysis of the changed cells. The commonest change took the form of a decreased granularity of the Nissl substance in the cell cytoplasm which thus stained more homogeneously but always with a greater density around the nucleus, though in no case was the clear zone seen in the rabbit visible in any cell in the mouse spinal cord. It is of interest that cells showing these slight changes were present in the most caudal and dorsal cell columns of the cervical enlargement in both animals. These results suggest that chromatolysis is far from being a satisfactory method, at least in the study of the origin of spinal nerves, since, even if some cells are induced to undergo chromatolysis, many others may only show the changes described above or may even remain unchanged and so upset any attempt to determine the central origin of the cut nerve. Several other investigators have found difficulties in producing chromatolysis. Prof. Le Gros Clark (personal communication) has found considerable difficulty in determining the distribution of chromatolysed cells in the lumbo-sacral spinal cord of monkeys after section of the sciatic nerve. De Neef (1901) found that anything but violent rupture was ineffectual in producing chromatolysis, and Marinesco (1898 a) described a series of degrees of chromatolysis depending on the intensity of damage to the nerve cell.


{{Romanes1941a table3}}

Table 3 Position of Duration of

Nerves cut Type of section section experiment
Median and ulnar Simple cutting . Elbow 5 days:

” ” 9” ”

” ” ” 13 ”?

” ” ” 21 ”

» Section with cautery ” 5,

” ” oad 8 ”

” ” 2” 9 ”

” ”? ” 12 ”

” ” ” 21 ”

” Violent rupture » 8 hours

” ory ” 18 ”

” ” ” 28 ”

” ” ” 38 ”

” ” ” 48 ”

” ” ” 3 days

” ” ”? 4 ”

” ” ” 5 9

” ” ” 6 ”

” ” . a’ 9 ”

” » . ” . 12 ”

” ” ” 21 ”

” ” ” 102 ”



However, despite the unsatisfactory nature of these results, in all the rabbits in which section of the median and ulnar nerves at the wrist was carried out the cell changes were confined to one cell column only, column 10, extending from the lower part of the eighth cervical segment to the upper part of the second thoracic segment and situated further dorsally than any other cell column in the enlargement. It seems probable that the cells lying in this column send their axons to the muscles of the hand. This result is borne out by the researches of de Neef (1901) who investigated the cell columns in the cervical and lumbo-sacral enlargements of the rabbit and dog. His description of the columns in the cervical enlargement of the rabbit does not agree in all respects with that given above. However, certain columns are common to the descriptions and one of these is column 10 which de Neef calls column D and shows extending from the eighth cervical to the lower part of the first thoracic segment. By successful chromatolytic experiments de Neef found that this column D contained the cells supplying the hand muscles. Similar findings in regard to column 7 (column C of de Neef) suggest that it contains the cells of origin of the fibres supplying the muscles of the forearm. However, it has been impossible to form any direct opinion on the functional relations of the other cell columns, though on the basis of the findings of de Neef, Bok, v. Gehuchten and other authors it seems likely that the more ventrally situated cell columns supply the more proximal segments of the limb.

It is of great importance to note here that van Gehuchten and de Neef carried out a very comprehensive series of experiments in order to produce chromatolysis. This they were unable to do except by violent rupture of the nerves concerned. This latter technique has been tried here without success, so it seems that chromatolysis as an experimental method has not the value usually ascribed to it, at least in the spinal nerves of the animals described above.

==Development of Cell Processes in the Ventral Nerve Roots of the Cervical Region of the Rabbit==

The first cell processes to develop are very fine, faintly argentophil, protoplasmic strands connecting the ventral region of the spinal cord of the 6 mm. rabbit embryo to the surrounding mesenchyme. It is impossible to follow these fibres further than the most proximal mesenchyme and their junction with another group of fibrils arising from the dorsal root ganglia; beyond this point, although the fibres are not visible, the mesenchymal cells are orientated along. the axis of these protoplasmic strands, In the 8 mm. rabbit embryo a much greater density of staining of the roots is visible, ventral and dorsal roots are fused, and they pass ventro-laterally towards the developing limb bud, stopping abruptly at its proximal end where the various segmental nerves meet to form a plexus. There is a short posterior primary ramus which only penetrates a short distance into the dorsal musculature. At this stage there is no neuromuscular ‘contact between the ventral horn cells and either the developing limb or dorsal musculature.

In the 11 mm. stage the trunks of the nerves have grown peripherally into the limb bud as far as the future wrist joint region and the posterior primary ramus has extended into the dorsal musculature. With a closer study of the nerve fibres in the limb it is found that, though the trunks of the nerves extend far into the limb, only the most proximal muscles contain any such fibres ramifying among them in close relation to the developing myoblasts; thus it appears that a few neuro-muscular units may have been established between the ventral horn cells of the spinal cord and the developing muscles of the trunk and shoulder regions, though neuro-muscular endings are not visible.


In the 15-5 mm. embryo the trunks of the nerves have extended into the distal extremity of the forelimb bud, many more nerve fibres can be found ramifying in the dorsal and shoulder musculature, and a few‘are found leaving the trunks of the nerves to enter the arm and forearm musculature. In this stage therefore a closer relation between the nerves and muscles of the trunk and shoulder regions has been established, and a few new connexions with the arm and forearm musculature have been attained.


In the 19 mm. embryo fine cell processes can be found leaving the trunks of the nerves throughout the limb (PI. 8, figs. 4, 5). These are most marked in the shoulder muscles and least in the muscles of the hand. It seems possible that at this stage neuro-muscular connexions have been established with some muscle fibres of all the segments of the limb and certainly with the dorsal musculature, which in this stage shows reflex contraction on stimulation of the snout, though no neuro-muscular endings have been observed.


The 26 mm. rabbit embryo shows a considerable increase in the number of fine nerve fibres ramifying among the muscles of the forelimb. This increase is especially marked in the muscles of the hand when compared with the previous stage. Definite neuro-muscular endings have appeared in the trunk and proximal limb muscles (PI. 3, fig. 6), and correlated with these is the appearance for the first time of reflex motility in the limb. With the mother under urethane anaesthesia, movements of the trunk, shoulder and arm muscles were seen on opening the amnion of these foetuses. They did not persist, but could be elicited on stimulation of the snout with a weak faradic current. This first appearance of motility is rather later than that described by Pankratz (1989) who found the earliest movements at 163-17 days, but since no intermediate stages have been studied here it seems probable that the earliest movements were just missed, for those of the 26 mm. embryo are quite well developed. 

==Development of The Nissl Substance==

The first cells to differentiate from the parent germinal epithelium appear in the 8 mm. stage. These cells stain lightly, the nucleus being clear and containing a well-formed nucleolus. The two apical spicules of cytoplasm stain a faint uniform colour with toluidin blue but show a faintly granular appearance under the high power.

In the 11 mm. stage two distinct types of cell are seen, the first consisting of those cells developing within the column of the ventral horn, and the second of those lying in the ventral horn but not included in the columns. The latter are similar to the cells of the previous stage but the former stain more darkly with toluidin blue, due to the increased staining properties of the nucleus which shows a dense chromatin network and a darkly staining nuclear membrane. The cytoplasm is similar to that in the cells of the 8 mm. rabbit embryo.

From this stage onwards the nucleus gradually loses its strongly staining properties and the cytoplasm stains with increasing intensity until, in the 19mm. embryo, the nucleus is again quite clear and the cytoplasm stains densely and is finely granular.

In the 26 mm. embryo true Nissl granules appear for the first time around the periphery of the cells situated in the columns of the ventral horn (Pl. 3, fig. 3). These consist of large aggregations of the finely granular substance, diffused throughout the cytoplasm in the earlier stages. Around the nucleus there is still a large quantity of the diffuse substance though the cytoplasm stains less darkly here than at the periphery. The further development of the Nissl bodies consists of the decrease in density of staining of the perinuclear cytoplasm and the gradual extension of these bodies into this region of the cytoplasm where they appear in the 80 mm. rabbit foetus, though it is not till 11 days after birth that all traces of the fine granular deposit are lost, and the Nissl bodies are free in a clear cytoplasm.

In regard to the arrangement of the Nissl granules it is interesting to note that they are present to a greater degree in the more ventral cell columns than in those more dorsally situated (Pl. 3, fig. 3, cf. a, b). This may be correlated with the greater degree of activity in the proximal rather than the distal segments of the limb at this stage, for no movements attributable to forearm or hand musculature are visible in the 26 mm. embryo.

==Discussion==

There are four distinct stages in the development of the cells in the ventral horn of the spinal cord and the groupings which they form:

# A stage of differentiation of the cells in the ventral horn from the germinal epithelium. This consists of a decrease in their density of staining and the appearance of a bipolar shape. In this stage the first cell processes make their appearance. These are at first faintly argentophil protoplasmic strands extending a short distance into the surrounding mesenchyme, but later they stain more strongly and are traceable to the base of the limb bud, and towards the dorsal musculature in the form of a rudimentary posterior primary ramus. No cell groupings are visible, but the cells are orientated with their long axes directed towards the point of exit of the ventral root.
# This stage is marked by the appearance of cell groupings or columns (PI. 1, fig. 2) and begins with the entry of the cell processes into the limb and dorsal musculature, but as yet no functional or anatomical connexion has been established between the muscles and nerves. The cell columns form two distinct groups. The first is ventro-medial and is found throughout the greater part of the spinal cord; the second, which is lateral in position and extends dorsally for a variable distance, is only present in that part of the cord supplying the limb musculature, and represents the mass of cells out of which the various cell columns of the lateral part of the ventral horn will develop. The cells are similar to those of the last stage but those in the columns tend to show a more darkly staining nucleus and cytoplasm, a considerable quantity of Nissl substance being freely scattered in the cytoplasm without any sign of Nissl granules.
# This stage is marked by the splitting of the lateral cell mass into subsidiary columns (PI. 1, figs. 2-4) commencing with the separation of its most ventral part and terminating with the final formation of a separate columnar form in its most dorsal cells. Peripherally the cell processes enter into close relation with the various muscles of the limb and trunk. This process proceeds proximo-distally, and as each segment of the limb is progressively innervated a further cell column separates from the main lateral mass of cells in the ventral horn of the spinal cord, in such a way that at the end of this stage all the cell columns present in the adult are visible. Though it has been impossible to show any definite nerve terminals on the muscles at this early stage, a functional connexion between muscle and nerve must be present at least in the trunk by the 19 mm. stage when slight reflex activity of these muscles can be obtained on stimulation of the snout. The cells of. this stage are very similar in shape to those of the preceding stage though now: the cytoplasm stains a denser homogeneous blue and the nuclei are clear.
# During this last stage definite reflex and spontaneous movements which appear first in the proximal part of the limb were observed, but there was no appreciable change in the arrangement of the cell columns. The Nissl substance within the cells shows an interesting change, being bunched together in blocks or granules around the periphery of the cells. These granules.are at first most marked in the ventrally situated cell column of the lateral group (Pl. 3, fig. 3) which is presumably the column associated with the innervation of the proximal part of the limb, the first part to become active. Several authors (Flesch, 1884; Hodge, 1892; Mann, 1894; Nissl, 1896; van Durme, 1901; Carlson, 1902; Dolley, 1909; Malone, 1913; Kocher, 1916; Cowdry, 1924; Hopkins, 1924; Clark, 1926) have described differences in the arrangement of the Niss] substance in nerve cells after various fixatives and degrees of activity at the time of fixation. It is therefore interesting to note that the Nissl granules appear at the time when the cells first show signs of functional activity.


Superimposed upon this general arrangement of the development of the cell columns is a tendency for their more cephalic members to appear first, thus showing a general cephalo-caudal differentiation. The last column to appear, therefore, is the most caudal and dorsal of the lateral group.

Correlated with the anatomical changes just described are three different functional periods in the development of the cells of the ventral horn:

# A period during which the cells are sending out their processes towards the muscles which they are to supply. In this stage all the cells, whatever their destination may be, are functionally alike in their growth processes and so might be expected to form a single undifferentiated aggregation.
# A period during which the cells consolidate their connexions with the peripheral muscle fibres. Such a period must of necessity affect the proximal neuro-muscular connexions first and the distal connexions last, but in each segment of the limb all connexions occur almost simultaneously. ‘So, for example, while one group of cells is connecting with the proximal muscles of the limb, others passing to more distal parts are as yet still in stage one. Such an arrangement produces two groups of functionally and histologically different cells within the original cell column. It is this differenee which is regarded as the factor in the splitting of the main cell mass into successive subsidiary groups, each related to a more distal segment of the limb in order of appearance.
# A period of functional activity of the nerve cells which leads to active contraction of the muscle fibres. This change results in a histological rearrangement of the Nissl substance (whose adult arrangement is not reached until postnatal activity has been in progress for some time) and the appearance of definite neuro-muscular endings, but there is no change in the arrangement of the cell columns which are, apart from their greater density, adult in form at the end of stage two.


With regard to the localization of function within the cell columns of the spinal cord, direct evidence concerning the connexions of the cells is lacking owing to the unsatisfactory nature of the chromatolysis reaction in the cells of origin of spinal nerves, but certain facts concerning the development of the cell columns throw light on their function:

# The cell columns are developed at a time when no functional activity in the form of muscle movements has been noted, so it seems certain that they cannot be considered as the result of the relation of the cells to peripherally functioning groups of muscles.
# Several cell columns have already appeared before all the muscles of the limb have been innervated, thus suggesting that each cell column does not supply all the muscles of the same functional activity throughout the limb.


Combined with the fact that one cell column only has been affected in lesions denervating the muscles of the hand in the rabbit, and that further spread into a second column has been noted on denervation of the forearm and hand, the two points mentioned above bear out the findings of de Neef (1901), Bok (1928) and other authors, localizing segments of the limbs in specific cell columns.

The cell columns described above cannot be considered as aggregations of cells formed as a result of the similarity of function of the muscles which they supply, as has been suggested by Goering (1928) in the white rat, but must be considered as related to specific segments of the limb, both embryologically and functionally. This will mean that within any one cell column there exist side by side cells supplying flexors, extensors and rotators of the segment innervated by that column. It seems likély that as these cells differ more functionally from each other the cell columns will not tend to become more definite by closer and closer aggregation of their component parts, as would probably be the case if all the cells were exactly similar in function, but the cell columns might even tend to become less distinct and broken up into several parts. Such a process of de-differentiation of the cell columns has been shown to take place, starting with the beginning of activity in the foetus and receiving added impetus during postnatal development,. with the result that cell columns in the spinal cord of adult rabbits and in all other mature animals yet studied are made up of very diffuse cell groupings which are difficult to determine in comparison with the same columns in the late embryonic stages (cf. Pl. 1, fig. 5; Pl. 2, fig. 2). This de-differentiation may not be an active process and may be related to the growth of the intercellular substance of the grey matter in the spinal cord, but it shows that there is no functional tendency for the cells within the columns to aggregate.


If the interpretation of the mechanism of development of the cell columns given above is correct, then their configuration will be dependent upon the arrangement of the segments of the limb supplied. Thus mammals showing a similar arrangement of the limb might be expected to show a like arrangement of the cell columns, though variations in the number of cells within the columns would be expected in relation to the increased or decreased amount of muscle supplied (Shorey, 1909; Hamburger, 1934). That such a similarity exists, despite functional differences, is evident from a comparison of the cell columns in the cervical and lumbo-sacral enlargements of the spinal cord of man (v. Gehuchten & de Neef, 1900, Jacobsohn, 1908, Romanes, 1941) which are practically the same, the only marked difference lying in the relative size of the various columns. That this similarity can be extended to other mammals with like limbs seems certain from the findings of de Neef and other workers, and the preliminary researches of the author into the comparative anatomy of these cell columns in a variety of mammals. That the converse of this is true is well known and it is only necessary to mention two examples here. The first is the condition of the spinal cord in the porpoise (Hepburn & Waterston, 1904) where reduction of the limbs and cell columns is present, and the seconds reported by Yu-Ch’uan Tsang (1939), is the further differentiation in a polydactylous strain of mice of the most caudal and dorsal cell colurnn of the lumbar enlargement which supplies the foot.

Thus the cell columns of the spinal cord undergo changes in their development which reflect and coincide with the proximo-distal differentiation of the various parts of the limb musculature and its associated nerves.

==Summary==

The adult anatomy and development of the motor cell columns in the ventral horn of the cervical enlargement in the rabbit are described in relation to the developing peripheral structures. Four chief stages have been noted:

# A thickening in the ventral part of the germinal epithelium of the neural tube to form a ventral horn.
# The outgrowth of processes from the ventral horn cells, the latter grouping themselves together to form longitudinal aggregations.
# The secondary splitting of these aggregations into subsidiary columns of cells coinciding with their connexion to the muscles which they supply, but prior to reflex activity.
# The appearance of coordinated neuro-muscular activity leading to contraction of the peripheral muscles. In this stage true Nissl granules appear in the cytoplasm of the ventral horn cells.


The establishment. of neuro-muscular connexions, which are distinct only in the last stage, proceeds peripherally from segment to segment along the limb, a new column appearing as each segment is innervated. Thus the columns are embryologically related to segments of the limb, and there is evidence in the adult to suggest that such a functional relation exists, though it is not complete owing to the unsatisfactory nature of the chromatolytic reaction.


I wish to express my thanks to Prof. Harris for his help and encouragement throughout this investigation.


==References==

ANGULO Y GonzALEz, A. W. (1927). J. comp. Neurol. 48, 115.

—— (1940). J. comp. Neurol. 78, 469.

Bok, 8. T. (1928). W. v. Méllendorf’s Handbuch der Mikroskopischen Anatomie des Menschen, 4/1, 478.

Bruce, A. (1901). A Topographical Atlas of the Spinal Cord. London.

Cartson, A. J. (1902). Amer. J. Anat. 2, 341.

Cuark, 8. L. (1926). J. comp. Neurol. 41, 423.

Cownkry, E. V. (1924). General Cytology. Chicago. Dotiey, D. H. (1909). Amer. J. Physiol. 25, 151. , .

v. Durmg, P. (1901). Névraze, 2, 115.

Fiescu (1884). Quoted from v. Gehuchten, Systéme Nerveux. Louvain.

Frazer, J. E. (1931). Manual of Embryology. London.

v. Genucuten, A. & DE Buck (1898). J. Neurol., Bruxz. Quoted from Systéme Nerveux.

v. Grnucuten, A. & pz Nzsr, C. (1900). Névrave, 1, 201. ‘ 

v. Genucuten, A. & Nexis (1899). J. Neurol., Bru. Quoted from Systéme Nerveux. 

Gornina, J. H. (1928). J. comp. Neurol. 46, 125. 

Hampouneer, V. (1934). J. exp. Zool. 68, 449. 

Hammonp, G. M. (1894). N.Y. med. J. 59, 15. Quoted from Angulo (1927). 

Hepsurn, D. & Warrrston, D. (1904). J. Anat., Lond., 38, 105. 

Hopes, C. F. (1892). J. Morphol. 7, 95. 

Hoprgins, R. E. (1924). Anat. Rec. 28, 157. 

Jacopsoun, L. (1908). Ueber die Kerne des menschlichen Riickenmarks. Berlin. Kreyssie (1885). Virchows Arch. 102, 286. 

Kocusr, R. A. (1916). J. comp. Neurol. 26, 341. Matong, E. F. (1913). Amer. J. Anat., 15, 121. Mann, G. (1894). J. Anat. Physiol. 29, 100. 

Marrnesco, G. (1898a). Rev. Neurol. 6, 463. —— (1898b). Pr. méd. p. 201. Both quoted from Systéme Nerveux. 

DE Neer, C. (1901). Névraze, 2, 71. Nissz, F. (1896). Z. Psychiat. 1147. 

OBERSTEINER, H. (1890). The Anatomy of the Central Nervous Organs, trans. Alex. Hill. 

Onvr, B. (1899). J. nerv. ment. Dis. Panxratz, D. S. (1939). Anat. Rec. 73, Suppl. p. 72. 

Romangs, G. J. (1941). J. Anat., Lond., 75, 145. Sano, F. (1897). Premier Congrés International de Neurol., Bruzx., p. 37. (1898). Ann. Soc. Méd.-Chir. Anvers. Both quoted from de Neef, C. (1901). 

Sass, A. (1889). Virchows Arch. 116, 243. 

ScHIEFFERDECKER, P. (1874). Arch. Mikr. Anat. 10, 471. Snorey, M. L. (1909). J. exp. Zool. 7, 25. 

Srerpa, L. (1870). Studien ueber das Centralnervensystem der Wirbelthiere. Quoted from de Neef, C. (1901). 

WALDEYER, W. (1889). Das Gorilla Biickenmark. Berlin. Yu-Cw’uan Tsane (1939). J. comp. Neurol. 70, 1.



==Explanation of Plates 1-3==

===Plate 1===

Fig. 1. Transverse section through the ventral horn of the spinal cord of an 8 mm. rabbit embryo in the 8th cervical segment. Stain, toluidin blue. x 400.

Fig. 2. Transverse section through the ventral horn of the spinal cord of an 11 mm. rabbit embryo in the 8th cervical segment. Stain, toluidin blue. x 400.

Fig. 3. Transverse section through the ventral horn of the spinal cord of a 15 mm. rabbit embryo in the 8th cervical segment. Stain, toluidin blue. x 320.

Fig. 4. Transverse section through the 8th cervical segment of the spinal cord in a 19 mm. rabbit embryo. Stain, toluidin blue. x 100.

Fig. 5. Transverse section through the 8th cervical segment of the spinal cord in a 26 mm. rabbit embryo. Stain, toluidin blue. x 60.

===Plate 2===

Fig. 1. Transverse section through the 8th cervical segment of the spinal cord in a 41 mm. rabbit foetus. Stain, toluidin blue. x60.

Fig. 2. Transverse section through three cell columns in the ventral horn of the spinal cord of an adult rabbit. These are examples of the densest cell columns seen in the cervical enlargement at this stage. Stain, toluidin blue. x 120.

Fig. 3. Sagittal section through the lateral part of the ventral horn of an 11 mm. rabbit embryo -in the 8th cervical segment. Cf. Pl. 1, fig. 2. Stain, toluidin blue. x 400.

Fig. 4. Sagittal section throygh the lateral part of the ventral horn of a 15 mm. rabbit embryo in the 8th cervical segment. Cf. Pl. 1, fig. 3. Stain, toluidin blue. x 400.

===Plate 3===

Fig. 1. Sagittal section through the lateral part of the ventral horn of a 19 mm. rabbit embryo in the region of the 8th cervical and Ist thoracic segments. Cf. Pl. 1, fig. 4. Stain, toluidin blue. x 100.

Fig. 2. Coronal section through the upper part of the cervical enlargement in an 11 mm. rabbit embryo. The right half viewed from above is shown. Stain, toluidin blue. x 240.

Fig. 3. Sagittal section through the lateral part of the ventral horn of the cervical region in a 26 mm. rabbit embryo. A, ventral cell column; B, dorsal cell columns. Stain, toluidin blue. x600. : .

Fig. 4. Section through the biceps of a 19 mm. rabbit embryo. Stain, pyridin-silver.. x625.

Fig. 5. Longitudinal section through the subscapularis muscle of a 19 mm. rabbit embryo. Stain, pyridin-silver. x 625.

Fig. 6. Longitudinal section through the subscapularis muscle of a 26 mm. rabbit embryo to show neuro-muscular endings. Stain, pyridin-silver. x425. 


{{Footer}}
[[Category:Historic Embryology]][[Category:1940's]]
[[Category:Spinal Cord]][[Category:Neural]][[Category:Rabbit]]
[[Category:Draft]]