Talk:Paper - Observations on the development of the human suprarenal gland (1927)

From Embryology


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

AnD E. E. HEWER, D.Sc. Lecturer in Histology and Demonstrator in Physiology at above school

‘Tus study of the changes that occur in the minute structure of an organ during its development naturally arouses curiosity as to the nature of any concomitant acquisition of function or changes of function which may occur, and though the main object of this paper is to describe the minute structure of the developing human suprarenal gland as fully and as consecutively as the material at our disposal allows, the scope of the investigation also includes the determination of the earliest appearance of adrenin, the general distribution of lipoid, and the variations that occur in certain cases of abnormal development.

The possible function of those cells constituting the main mass of the human gland during development (the so-called foetal cortex or boundary zone(1)) which disappear during the first year of life offers a most interesting problem, but the very fact that these cells are present mainly during foetal life puts great and very obvious difficulties in the way of any investigation of their physiological significance. We have given considerable attention to the question of the part played by the foetal cortex, but as yet we are unable to offer any definite evidence on this point.

We feel that a solution of this problem may be found in the study of the comparative anatomy and physiology of the foetal ductless glands, but at present the scope of our investigations cannot be further extended.

The evidence presented in this paper is based on our own dissections and observations of some 300 embryos and foetuses ranging from one of 5 mm. to full time. Examination of isolated organs was made also of post-natal material (80 cases) ranging from birth to 8 years. Serial sections were made of some of the earlier embryos comprising one 5 mm., one 10 mm., two 12 mm.}, one 18 mm., one 25 mm. and several 30-40 mm. embryos, and also of glands at various later stages of development.

Owing to the small number of young human embryos amongst our specimens the early development could not be followed very systematically, but

1 One of the 12 mm. embryos was measured after fixation and probably represents a slightly

later stage of development. The other 12 mm. specimen was somewhat damaged, and is not described, ‘The Development of the Human Suprarenal Gland 303

from the 8th week of foetal life to the time when the gland attains its adult characteristics the development of the suprarenal has been followed in detail.

The age in weeks of specimens measuring more than 40 mm. (vertex rump) was computed by taking into consideration all factors bearing on maturity, menstrual history (not always available), length (vertex-rump and vertexheel measurements), weights, ossifications, appearances of hair and subcutaneous fat, general appearances, etc.; and a graph is appended for reference (diagram 1) showing the relation between vertex-rump measurement and the age in weeks of our material.



—- 4400.

— -{300

— 200

- -4100

l I j J i ! l i l 4 8 12 16 20 24 28 32 36 40

WEEKS Diagram 1. Curve showing the relationship of the vertex-rump length of the foetus to the age in weeks,

The following table shows briefly the relationship of the vertex-rump length of the foetus to the age in weeks:

Average vertex-rump

length in mm. Maturity in weeks Number of cases 33 5-6 3 57 10-11 5 98-5 12-14 3 136-5 16 10 178 20 16 215 24 8 252 28 5 284 9

32 340 About full time 50 304 M. F. Lucas Keene and E. E. Hewer


Specimens up to the age of 8 weeks (80-40 mm.) were all cut in serial section without dissection in order that the relations to the surrounding structures might be preserved, and consequently until these have been reconstructed no observations as to size and shape are available; but already at 6 weeks the gland is a clearly defined organ situated on the upper pole of the kidney, and by 8 weeks its maximum cross-section equals that of the kidney. It maintains its relation to the kidney throughout development, its surface at first being smooth, the characteristic fold only appearing after the sixth month of intra-uterine life.

A fresh full time gland when cut open shows a dark central part surrounded by a thin yellowish rim. The central part is largely composed of blood, the organ being extremely vascular at this time.

After birth the gland obviously shrinks, especially in its antero-posterior dimension.


The gland shows a steady increase in weight throughout foetal life, and the following figures give the approximate average weight of the gland in successive months (2):

Weeks Gm. Cases

16-20 0-5 2

21-24 1:0 4

25-28 1-7 11

29-32 2-2 13

33-39 3-0 9 Full time Right, 4-47 65

Left, 4-64 Full time. Weight ratio: Right, 762

Left, 642

MICROSCOPIC STRUCTURE 5 mm. Emsryo (Diagram 1 a).

In a 5mm. embryo a transverse section cut just below the level of the bifurcation of the lung diverticulum shows bilateral coelomic recesses behind the heart. The superficial or lining cells of the posterior wall of these recesses (taken to be pleuro-peritoneal canals) are of a high columnar type, and form the epithelium of the extreme cephalic end of the Wolffian ridges. Since the embryo was not cut absolutely horizontally the right side of the section shows a slightly lower level, and here two ridges are seen, both covered by the high columnar epithelium. This columnar epithelium is supported by bilateral masses of cells very slightly differentiated from the mesenchyme stretching forwards to the gut from the aortae.

A large vein lies posterior to the lateral part of the ridge which is occupied in lower sections by the mesonephric tubules. Neither in this embryo nor in The Development of the Human Suprarenal Gland 305

Columnar epithelium Sympathetic


Diagram 1a. Horizontal section through a 5 mm. human embryo just below the bifurcation of the lung diverticulum.

/7 Dev. Suprarenal


Diagram 2. Horizontal section through a 10 mm. human embryo. 306 M. F. Lucas Keene and E. BE. Hewer

the next (10 mm.) did we see the anlage of the organ in the form of cell buds projecting from the root of the mesentery, as is usually described.

The anlage of the sympathetic system from which is developed the medulla, the third element entering into the formation of the suprarenal, is represented by bilateral masses of rather darkly staining cells situated lateral to the aortae on the anterior aspect of the developing vertebra. ,

10 mm. Empryo (diagram 2).

Unfortunately this embryo is also cut obliquely, the left side of the section representing a higher level than the right side. On the left side is seen the lower part of the left lung; the section also shows the liver, stomach, root of mesentery, and lateral to the root of the dorsal mesogastrium are the Wolffian ridges. Bilateral masses of cells are seen situated medially to the mesonephric bodies, lying in front of the aorta and having a small free surface covered by coelomic epithelium. These masses of cells are the anlage of the foetal suprarenal, and can be readily distinguished from the surrounding tissues. The nuclei of the cells are large, usually round, vesicular, and show at least one well-marked nucleolus; the cell outlines are not easily made out, but the protoplasm is granular and the cells closely packed together. In contrast, the surrounding tissue shows loosely packed cells of smaller size with small nuclei of varying shape, the structure being that of embryonic connective tissue. The covering coelomic epithelium is of the same high columnar type as that covering the rest of the Wolffian ridge: but at this point the cells are very closely packed together, and show an occasional mitosis.

Immediately behind the bilateral masses are the sympathetic chains, consisting of cells with deeply staining nuclei, and nerve fibres are also seen.

12 mm. EmsBryo (diagram 8).

In this embryo definite suprarenal bodies are seen situated on either side of the aorta and pushing forwards into the coelomic cavity medial to the mesonephros and developing genital gland.

The cells of the coelomic epithelium immediately on either side of the root of the mesentery are no longer columnar and are proliferating, forming a cap of cells 3 or 4 layers thick over the developing gland (better seen in an 18 mm. embryo); some of these cells tend to push postero-laterally between the mesonephros and the suprarenal mass. Posteriorly this latter is becoming marked off from the surrounding tissues by a developing connective tissue capsule and the cells are big and becoming faintly eosinophil.

On the medial aspect of the gland migrating masses derived from the . sympathetic are well seen, and form small groups of cells in the root of the mesentery, closely associated with the large blood vessels. These migrating bundles will be termed sympatho-chromaphil cells, 4) from now onwards, and a few are already invading the organ. The Development of the Human Suprarenal Gland 307

__ Epithelial cap Mesonephros eds

t+ Root of Mesentery.

Suprarenal F..17— Blood vessel Gland Migrating Sympatho-chromaphil bundles

Sv “chain

ee $23 — sympathetic

Diagram 4. Horizontal section through the suprarenal region of a 25 mm. human embryo. 308 M. F. Lucas Keene and E. FE. Hewer

25 mm. Empryo (diagram 4).

This embryo exhibits definite paired suprarenal glands. The bulk of the gland consists of foetal cortex or ‘“‘Boundary zone(1)” cells which are large, irregularly shaped and markedly eosinophil, with large nuclei and wellstaining nucleoli. These cells are loosely arranged and the whole mass is very vascular.

The periphery of the gland consists of a thin layer of much smaller cells with darkly staining nuclei: these form the true cortex, namely, that part which will give rise to the entire post-natal cortex.

The migrating sympatho-chromaphil masses are seen in the root of the mesentery, and some bundles have already penetrated well into the gland.

30-40 mm. EMBRYOS.

In embryos of 30-40 mm., that is to say about 8 weeks, the suprarenal glands may be easily recognised with the naked eye. They are seen resting on the upper poles of the kidneys and the two organs are of approximately the same size. The general arrangement of the cells remains unchanged and the whole gland is very vascular, both small and large vessels having extremely thin walls. By now there is a well-marked connective tissue capsule from which delicate strands run into the organ.

The sympatho-chromaphil element is represented by bilateral masses of cells similar to those seen in the sympathetic chains, situated between the aorta and the suprarenal gland. Numerous bundles breaking off from these masses invade the glands chiefly on their medial aspects, and penetrate well into their substance. The cells pass in, in bundles surrounded by connective tissue, penetrating the capsule of the gland where they enter. In addition to this main invasion, bundles of cells pass in with the blood vessels and also penetrate the gland on its lateral aspect.

The invading masses break up to a certain extent after entering and processions of sympatho-chromaphil cells are seen as well as small bundles. Two types of these cells are now recognised, and accompanying nerve fibres are also sometimes seen. The most numerous of the two types of cell are small with a darkly staining nucleus and very little protoplasm; possibly these are neuroblasts because they are frequently found with nerve fibres, and, further, certain of the invading bundles consist of this type of cell only, arranged round nerve fibres and presenting a rosette appearance in section (diagrams 5 and 6). This rosette appearance is that seen in sections of neuroblastomata and affords additional evidence that this type of tumour is derived from embryonic suprarenal medullary cells 6). The appearances were precisely similar to those given in the excellent microphotographs accompanying Wright’s description (17).

The second type of cell is larger than the neuroblast and has a vesicular nucleus but is considerably smaller than the cells of the foetal cortex, and The Development of the Human Suprarenal Gland 309

further differs: from these in having a basophil protoplasm. These are the socalled “para-sympathetic cells” described by Zuckerkandl(9) but which he did not observe migrating into the gland. Both types of cell persist throughout foetal life, and are the precursors of cells (other than large ganglion cells) found in the post-natal medulla.

Diagram 5. Human foetus 18 weeks (v.R. 151 mm.). Section of peri-adrenal tissue showing . rosette arrangement of cells.



Diagram 6. Human foetal suprarenal 23 weeks (v.R. 225 mm.). Section showing rosette arrange‘ ment of invading cells within the gland.

10 WEEKs TO FuLL Time.

During the next few weeks of development (10-20) the most striking change occurs in the true cortex. The cells of this layer increase in number (mitoses are seen in an 11-week embryo) and are frequently arranged around a central

_space. This appearance is especially noticeable in 16-week embryos (diagram 7) and the “vesicles” sometimes contain a basophil substance that looks like a coagulated fluid. The vesicular arrangement of the outer. zone becomes. less 310 M. F. Lucas Keene and E. E. Hewer

marked during the later months of foetal life, but is present occasionally even after birth. The bulk of the gland still consists of the large eosinophil cells (“foetal cortex”) which are arranged quite irregularly and are well supplied with capillaries. Often at this time these cells show many vacuoles, but these are not due to the presence of lipoid which normally does not appear until later. A horizontal section cut through the organ at the hilum shows

a Diagram 7. Section through suprarenal of human Diagram 8. Section through suprarenal foetus of 16 weeks (136-5 mm.). a, Vesicular region of human embryo of 12 arrangement of cells of zona glomerulosa. weeks (v.B. 98-5 mm.) showing in filtration by large numbers of cells from adjacent tissue. a, Suprarenal. 6, Invading cells.

that the central part is occupied by large thin-walled spaces filled with blood round which the penetrating bundles are collecting. In a 12-week embryo the immigration was very active, the invading masses consisting for the greater part of the larger cells (diagram 8). This infiltration by cells of the larger type was very marked in this one case only.

No further change is noticed until about the 22nd week, when a wellmarked chromaphil reaction @) is given by the larger of the immigrating cells, The Development of the Human Suprarenal Gland 311

situated both inside and outside the gland, and the nuclei of some of these cells show in addition the typical chromaphil ring reaction.

Throughout the last 10 weeks of foetal life the true cortex increases gradually in thickness, although the foetal cortex constitutes the bulk of the organ until birth. The foetal cortex shows a tendency to congestion, and haemorrhage associated with a progressive degenerative change occurring chiefly in the cells of the deeper layers, the cell boundaries becoming very indefinite and the nuclei acidophil or non-staining. The sections show in addition an increase in the fibrous tissue around the central blood vessels; this is seen by the 26th week and is well marked by the 82nd week.

From the 22nd to the 82nd week the immigration of cells is still well marked, the penetrating bundles consisting at that time almost entirely of the larger type of cell. Nucleated cells, apparently erythroblasts, are frequently seen in capillaries and large sinuses from the 22nd week until full time. It is difficult to account for their presence in such numbers in the suprarenal at this time since already by the 12th week the blood in the heart shows very few erythroblasts.


At full time the suprarenal still consists chiefly of the true and the foetal cortex, but a very thin central zone of medullary cells is seen surrounding large blood sinuses. The outer rim of small polygonal healthy cells, now much thicker, is invading and replacing the inner degenerating foetal cortex (diagram 9). This outer rim gives rise after birth to the whole of the true cortex, and already shows a wellmarked zona glomerulosa and beginning zona fasciculata. The immigration of cells through the cortex is ceasing.

The medulla consists of the two types of cells previously described: (1) small round cells with darkly staining nuclei, the original undifferentiated immigrating cell, (2) larger and

somewhat irregularly shaped cells of varying Diagram 9. Section through full-time size that give a well-marked chromaphil re- _ human suprarenal. a, True cortex action. cells growing in to replace foetal

cortex. 6, Foetal cortex cells de In serial sections of the full-time foetal generating.

gland the medulla is seen to form a central core which links together the blood sinuses near the hilum, and is spread out in a very thin layer around many of these vessels, and stretches a 312 M. F. Lucas Keene and E. E. Hewer

short distance into the peripheral part of the organ. The cells are arranged in irregularly sized clumps containing both types, the larger predominating. In addition, ganglion cells are sometimes seen together with nerve fibres close to the hilum.

The Beilschowsky method of staining shows that the gland at full time is very richly supplied with fibres that stain black by this method. No doubt many of these are nerve fibres, but probably many are collagen (Da Fano); this is borne out by the Mallory method of staining. Hammar(is), making use of this method, states that the gland in the foetus of 16 mm. is already richly innervated.


The whole gland shrinks owing to the rapid disappearance of the foetal cortex, the cells of which first become swollen, and then lose their nuclei, and


Diagram 10. Section through suprarenal of a child aged 3 months, showing ganglion cells.

are later replaced to a certain extent by temporary fibrous tissue which forms a well-marked zone central to the rapidly developing zona fasciculata. In this fibrous tissue are to be found large isolated eosinophil cells, possibly a degenerative stage of cells of the foetal cortex, and here also basophil granular masses are frequently seen. Up to one year the central fibrous tissue is a marked feature, but can no longer be discerned at three years of age.

The chief change in the true cortex during the first year is the growth of the zona, fasciculata, the cell columns being definitely grouped by the third week, and the zona reticulata being defined by 34 months. It is to be noted that this newly-formed cortex is not nearly so vascular as the foetal structure.

The medulla appears as scattered masses of cells situated principally in the region of the hilum and surrounding the blood sinuses. The thin peripheral portions of the gland show only cortex and the fibrous layer with some central blood vessels. The medullary cells are of irregular shape and of varying sizes, The Development of the Human Suprarenal Gland 313

with darkly staining nucleus and a basophil cytoplasm. The cell boundaries are very indistinct. Nerve fibres are seen between the cells, and ganglion cells are met with in the region of the hilum and sometimes invade the gland (diagram 10). The chromaphil reaction is as a rule not well marked, but as the number of specimens examined was small and removed from infants who had died of various diseases this may not be the normal condition.

Lrporp! (diagrams 11-16).

The distribution of lipoid has been investigated by means of the formalin fixation and Sudan III staining method only, and no attempt has been made to distinguish between the different fatty substances present (6, 16).

oe es | | = EC. \ \

. ae

— ao — B m.m. . 0-1 Re 0-2 ® 0-3 aay ’ 0-4 Ht. 1. 0-5 \ 0-1 . 0-2 0-3 \ \ 0-4 fe 05 \ | \ oe \ De ge ai Diagram 11. Human foetal suprarenal Diagram 12. Human foetal suprarenal 16 weeks (v.R. 136-5 mm.). Section 24 weeks (v.R. 215mm.). Section stained for lipoid. Ca., Capsule. stained for lipoid. Ca., Capsule. T.C., True cortex. B., Central blood F.C., Foetal cortex. B., Central blood vessel. F.C., Foetal cortex. A., Fat- vessel. 7'.C., True cortex. A., Fatcontaining cell migrating centrally. containing cell migrating centrally.

1 A preliminary note on the distribution of lipoid in the gland and also on the earliest appearance of adrenin and of the chromaphil reaction was published in 19242). Since then, with added material, many further observations have been made, and the present findings confirm the previous ones as far as they went. , 314 M. F. Lucas Keene and E. E. Hewer

Lipoid granules are seen in the 16-week embryo, in a few large cells that are migrating through the cortex and settling in the central part of the gland. The migration of these large lipoid-containing cells continues throughout

m.m. 0-1 0-2 0-3 10-4 0-5

Diagram 13. Human foetal suprarenal 32 Diagram 14. Human foetal suprarenal, full

weeks (v.R. 284mm.). Section stained for time. Section stained forlipoid. Ca.,Caplipoid. Ca., Capsule. F.C., Foetal cortex. sule. 7.C., True cortex. F.C., Foetal T.C., True cortex. B., Central blood cortex degenerating. A., Fat-containing vessel. cell migrating centrally. B., Central

blood vessel. (Half only of the crosssection is shown.)

foetal life, and they collect round the central blood vessels in ever-increasing numbers. The nature of these cells is not clear; they may be medullary cells or migrating cortical cells as suggested by Elliot and Armour (1).

The first trace of lipoid in cortical cells is found at 22 weeks when some of The Development of the Human Suprarenal Gland = 315

the cells of the true cortex contain very small granules. By the 24th week the cells of the foetal cortex also contain very minute granules of lipoid, a section under a low magnification showing a diffuse pink staining throughout.

During the later months of foetal life the lipoid increases in amount, especially in the true cortex. As the foetal cortex degenerates its lipoid diminishes in quantity, although certain writers(7) speak of the “fatty degeneration”’ of these cells. At full time the “foetal cortex” contains no lipoid,


q m.m. se m. 0-1 an 0-2 ae 0-3 0-4 se


0-5 :

Diagram 15. Human suprarenal. Five weeks post- Diagram 16. Anencephalic human foetal suprarenal, full natal. Section stained for lipoid. Ca., Capsule. time. Section stained for lipoid. Ca., Capsule. C., T.C., True cortex. D., Remains of degene- Cortex. M., Medulla. B., Central blood vessel. rating foetal cortex. B., Central blood vessel.

whereas the cells of the developing true cortex (which is invading and displacing the foetal cortex) show the lipoid stain in increasing amount especially in the actively growing layers.

After birth the lipoid staining corresponds to the development of the new cortex, so that at about three months a cross-section of the gland shows a wellmarked lipoid reaction in the zona glomerulosa and fasciculata, and also in certain cells of the medullary region. Separating these two areas is a narrow band showing no fat, which corresponds to the region of the degenerating remains of the foetal cortex.




It is generally accepted that there is some relation between the chromaphil reaction in cells and the presence of the active principle of the suprarenal medulla known as adrenin (13). In this investigation the chromaphil reaction was tested by two methods:

(1) Fixation of fresh material in formo-acetic-Miiller.

Anatomy LXxI 21 316 M. F. Lucas Keene and E. FE. Hewer

(2) Ogata’s method (8) of fixation in 5 per cent. K,Cr,O, followed by the addition of formalin.

The presence of adrenin was detected by making a 10 per cent. extract of the fresh gland in either 0-9 per cent. NaCl or Ringer’s solution, and testing by some or all of the following reactions in frogs!:

(1) Vaso-constrictor effect on peripheral circulation by perfusion.

(2) Accelerator and tonic effect on heart by perfusion.

(3) Dilator effect on isolated eye.

(4) Secretor effect on cutaneous glands.

Chromaphil reaction. It was found that the chromaphil reaction was never obtained before 22 weeks of foetal life, but after this age the reaction was fairly constant. Failure to obtain the characteristic colour may be due in some cases to the material not being absolutely fresh.

Adrenin, Extract of the gland of a 12-week embryo (one case) contained a trace of adrenin (vaso-constrictor test only). At 16 weeks (three cases) the active principle was definitely present. At 18 weeks the extracts gave a more marked adrenin response to the tests used, and from this age onwards adrenin was found present in considerable amount.


While examining the material some notes were made on the developing sympathetic elements, and are recorded here as confirming, for the greater part, the descriptions given by other workers (9, 17).

The sympathetic anlage is present in the 5mm. embryo and consists of bilateral masses of cells lying in front of the developing vertebrae, and lateral to the aortae in sections showing these structures. The masses are differentiated from surrounding tissue by the darker staining of the nuclei of the cells.

In the 10 mm. embryo these masses consisting of one type of cell are seen in the same position, are better differentiated from the adjacent tissue, and contain nerve fibres. .

In the 8-week embryos (30-40 mm.) sections show the developing ganglia to be composed chiefly of cells with small, darkly staining nuclei, and in addition a few larger faintly basophil cells with big faintly staining nuclei. A 12-week embryo shows these two types of cells, and moreover they are seen to be the same sort of cells as those that are invading the suprarenal gland.

The rosette arrangement already mentioned in connection with the masses migrating into the suprarenal (17) was seen in sections of a chain removed from a 16-week embryo.

In the 22-week embryo the nerve fibres are seen much more clearly, and the cells with large nuclei are definitely nerve cells of the sympathetic ganglia, and some show the distinctive surrounding nucleated sheath. As development

1 N.B. We are indebted to Professor F. Ransom, M.D., and to Miss E. Scarborough, M.Sc., M.B., B.S., for the carrying out of these tests, The Development of the Human Suprarenal Gland = 317

proceeds these cells gradually acquire more protoplasm, until at full time they have the characteristic appearance of the fully developed ganglion cell.

Paraganglia. The nerve cells of the semilunar ganglia acquire the adult type more quickly than those of the sympathetic chain, showing typical structure by 28 weeks. Neither in the sympathetic chain nor in the outlying ganglia has the chromaphil reaction been observed before full time (sections of material for the most part fixed in formo-acetic-Miller and stained with haematoxylin and Biebrich scarlet).

Zuckerkandl bodies. In the early embryo masses of cells derived from the sympathetic anlage grow forwards into the root of the mesentery together with nerve fibres; small bundles break off from these masses and invade the suprarenal gland, others form the semilunar ganglia, whilst yet others go to form the organs of Zuckerkandl.

The pre- and para-aortic tissue of a 16-week embryo from the region of the suprarenal glands is seen on section to contain the semilunar ganglia, collections of haemo-lymphoid tissue, and other encapsuled masses (the Zuckerkandl anlage) consisting of cells arranged in groups. These cells have large faintly staining nuclei, and although the cell boundaries are rather ill defined the cells are arranged in groups and are well supplied with capillaries. In embryos of 28 weeks these future Zuckerkandl masses are much larger, and the cells are irregular in shape and show a certain amount of fibrous tissue between the groups. At full time these bodies have the appearances described by Zuckerkandl(9). Under the high power of magnification the cells of the Zuckerkandl bodies at full time resemble the larger cells described in the suprarenal medulla, and like them show the chromaphil reaction in both protoplasm and nucleus. Fibrous tissue is increasing and the body presents the appearance of progressive atrophy, which becomes much more marked in specimens taken from a four-months-old infant.


(1) Anencephalic. The organs of six specimens were examined, five fulltime and one aged 24-26 weeks. Unless the specimen is hardened and fixed by injection of formo-saline through the umbilical vein it is difficult on account of its small size to isolate the suprarenal and clear it completely of the surrounding abundant fatty tissue. Thus only in one case of the six were the suprarenal glands weighed. The right gland weighed 0-35 gm., the left 0-27 gm. (average weight normal gland of the same age 4-78 gm.), giving weight ratios of 5280 and 6844 respectively (average weight ratios 748 and 679).

In section the full time glands show the feature described by Elliott and Armour (1), that is to say that traces only of foetal cortex are present in the specimens. On account of this absence of the foetal cortex the whole organ is thin. The true cortex is, however, thicker than in the normal full time foetus, and shows a well-marked zona glomerulosa and fasciculata, and in fact has the appearance of a 12-month post-natal gland, except that there is no sign

21—2 318 M. F. Lucas Keene and E. E. Hewer

of the fibrous layer that is present at that time between the cortex and the medulla (diagram 17). The cells of the medulla show a well-marked chromaphil reaction, and the medulla appears normal except that it is rather more abundant than in the normal full time organ: this observation was also made by Meyer (20).

The distribution of lipoid is similar to that seen in the one 12-month postnatal gland examined for this substance, that is to say, all the cortex contains lipoid, and the reaction is especially well marked in the zona glomerulosa and _ in the deeper part of the zona fasciculata. It is however remarkable that the medulla shows no cells giving the lipoid reaction, a marked difference from the normal.

ee ee ei eee eo'e,2) ee Sai} a

<7 ee hy: . SF es Pao i

Diagram 17. Htuman anencephalic foetus (full time). Section through suprarenal showing (a) well-developed true cortex, (6) well-developed medulla.

W-PiLcRin. CEL

The gland of the foetus aetat, 24-26 weeks is extremely small but consists of both true and foetal cortex. The cells of the latter are degenerating, and the true cortex is actively proliferating. Penetrating bundles are seen passing through to the central part of the gland. In this case anencephaly is associated with complete spina bifida.

The extract made from the gland of a 28-week anencephalic foetus not examined histologically contained adrenin as shown by all the four tests previously mentioned.

(2) Monster. A monster of about full time showing abnormal development of the external genitalia and neither anal nor genital apertures, was found on dissection to have an enormously distended uterus and no rectum; the ovaries were cystic. Further, there was complete absence of the thyroid gland, although serial sections of the neck showed parathyroids to be present. There was evidence that the thyroid gland had completely degenerated. In this case the suprarenal gland showed a relatively narrow and very inactive true cortex (diagram 18).

(3) Osteogenesis imperfecta. In a case of osteogenesis imperfecta the suprarenal gland was found normal in every respect, including the lipoid reaction. The Development of the Human Suprarenal Gland 319

(4) Cerebro-meningocoele. (a) A 32-week foetus showing the frontal lobes joined in the mid-line, but having well-developed sulci and gyri, and with most of the cerebellum in the hernia, presented a normal suprarenal gland.

(b) A 84-week foetus in which most of the brain, in a very degenerated condition, lay in the hernia, had the facies of an anencephalic, and the supra


Diagram 18. Suprarenal of monster II. Note scarcity and inactivity of true cortex (a). Compare with diagram 9.

renal gland showed a marked diminution in the thickness of the “foetal cortex.” This gland had the appearance of that of a 3-month infant, except for the absence of the fibrous layer normally found at that time between cortex and medulla.

(5) Hydrocephaly and spina bifida. A case of marked hydrocephaly and spina bifida showed normal development of the suprarenal gland. 320 M. F. Lucas Keene and HE. E. Hewer

The Suprarenal Gland in certain Pathological conditions.

Condition * Age Condition of gland (a) Mongol (1) 3 months post-natal Normal (6) Status lymphaticus (1) 4 months post-natal Normal (c) Erysipelas (1) 5 weeks post-natal Cloudy swelling and cells looking unwell (d) Tubercular meningitis and miliary 9 months post-natal Normally developed,showtubercular peritonitis (1) ing caseous foci (e) Meningitis, secondary to osteo- 3 months post-natal Normally developed, cells myelitis (1) looking unwell ForTaL (f) Pulmonary tuberculosis of mother 21 weeks Great excess of fat present 1 in the foetal cortex Pulmonary tuberculosis of mother 24 weeks Ditto. and father (1) (g) Syphilis of mother (3) 20-26 weeks Excess of fat in the foetal cortex (kh) Syphilis of mother (5) Full time No abnormal amount of fat

(Foetal Wasserman reactions not done.)

N.B. It has not been our aim to collect pathological material, but the cases quoted had to be classified from their history as “‘not normal,” and a few notes were made as to the conditions found.


In addition to examining our own incomplete series of human embryos, we have had an opportunity also of studying preparations of embryos of 10 mm., 12 mm., 15 mm., 16 mm., and 25 mm. kindly lent us by Professor J. E. Frazer. In connection with the origin of the true and the foetal cortex our preparations are so strongly suggestive to us of an origin different from that usually accepted that we wish to put forward this view for consideration even though we are aware that there are many gaps in the evidence we present for its support. Our preparations suggested to us that the cap of proliferating coelomic epithelial cells seen in the 12 mm. embryo may represent the anlage of the true cortex. These cells appear to be pushing postero-laterally between the mesonephros and the suprarenal mass, which was seen to be differentiated in the 10 mm. embryo, and which would therefore represent the foetal cortex constituting the main mass of the gland throughout foetal life. In the 12 mm. embryo this suprarenal mass is limited posteriorly by a delicate connective tissue capsule, and on the medial aspect there is a procession of sympathochromaphil bundles, but no capsule. Zuckerkandl’s diagram of the suprarenal of a 15 mm. embryo shows the aforesaid cap of cells separated from the main mass by a large vessel. This diagram, however, shows no differentiation between the cells forming the cap and those lying in the mesentery. Our sections which are cut less than 5 thick show that these two are clearly The Development of the Human Suprarenal Gland 321

differentiated, and further that the cells of the epithelial cap are very similar to those in the genital ridge, but differ from the epithelium over the mesonephros.

The two parts of which the foetal suprarenal is composed—the true cortex which persists post-natally and the foetal cortex which atrophies after birth— are generally considered to have a common origin, differentiating from the cells forming the original mass seen in the 10 mm. embryo, but we suggest that this — original mass forms only the foetal cortex, the true cortex arising from the cap of cells seen at a later stage. If these cells do form the true cortex they must push their way round between the connective tissue capsule and the main mass. This actual encircling of the suprarenal mass has not been traced but preparations of the 12 and the 18 mm. embryos are suggestive of the initiation of such a process.

Further, in support of our view of a different origin of the true and the foetal cortex is the following observation on the staining reactions of the cells. The suprarenal anlage is eosinophil even in the 10 mm. embryo and the cells of the foetal cortex are eosinophil throughout their persistence, whereas the cells of the true cortex differentiate after the appearance of the eosinophil anlage and never during foetal life give an acidophil reaction. We think it unlikely that small basophil cells, .such as form the true cortex pre-natally, should differentiate from a mass of large eosinophil cells, and although many examples could be quoted of cells originally showing a basophil reaction subsequently becoming eosinophil we cannot recall any examples of the reverse process, although this has been described by de Beer @1).

The specimens lent us by Professor J. E. Frazer in our opinion supported this view of the possible origin of the true cortex except in the case of the 16 mm. embryo. The latter certainly showed some cells pushing round between the mesonephros and the suprarenal from the coelomic surface, but they were slightly flattened cells and appeared not unlike enlarged capsule cells. This 16 mm. embryo taken by itself would not have suggested the view put forward above.

With regard to the medulla of the gland our evidence agrees with the observations of others, namely, that it is derived from the sympathetic anlage. Its further development, however, as seen in our specimens does not entirely agree with the description given by Zuckerkand] (9). He describes the invading masses as composed of one type of cell only, namely, the ‘“‘sympathoblasts,”’, (smaller darkly staining type), except occasionally in foetuses of 50-60 mm. in which he observes a marked invasion of the gland by “‘chromaffinoblasts.” We also noted this occurrence in one case aged 12 weeks. In addition, many of the invading bundles seen in our material contained even in embryos of about 8 weeks both types of cell, the smaller type being by far the more numerous until after the seventh month. This invasion by two types is also noted by other observers. There was no evidence that the migrating sympathochromaphil masses, when inside the gland, gave rise to ganglion cells as has 322 M. F. Lucas Keene and HE. EH. Hewer

been stated (4); but ganglion cells were seen being carried into the gland with nerve fibres.

Hammar (18) has recently published an account of the human foetal suprarenal from the second month onwards, especially from the point of view of its functional activity. Comparing the foregoing evidence with his results it is found that he also notes that the time of invasion of sympatho-chromaphil masses occurs at about the 15 mm. stage. He finds ‘“‘ chromaffin cells” in an embryo of 90 mm., whereas we first obtained a chromaphil reaction at 22 weeks. (In this connection it is interesting to note that various observers mention the appearance of this reaction at stages of development ranging from 9 mm. to 7 months.) With respect to the lipoid, Hammar and other workers have found evidence of the presence of lipoid from 25 mm. onwards. We have been unable to investigate this point in embryos younger than 16 weeks, but in these (three cases) there was no trace of lipoid in the cortex.

The observations made as to the early atrophy of the foetal cortex in cases of anencephaly agree with those of Elliott and Armour, and later workers (20), but the suggestion that there is possibly some connection between the foetal cortex of the suprarenal and the development of the cerebral hemispheres is not supported by a specimen very kindly shown us by Professor J. E. Frazer. In this anencephalous embryo of 25 mm. the suprarenal is well developed and consists of the usual mass of foetal cortex and the narrow rim of developing true cortex. Moreover, our sections of the anencephalous suprarenal (24-26 weeks) showed a considerable amount of degenerating foetal cortex, though the gland as a whole was extremely small. It is interesting in this connection to note that in the case of cerebro-meningocoele (4 (b)) in which the brain is mostly degenerated the foetal cortex of the suprarenal gland was very much diminished, whereas in the other case (4 (a)) in which the cerebellum only was involved in the hernia the suprarenal gland was normal for its age. (See also Cosmettatos (10) on coexistence of malformations in brain and adrenals.) That the acephalous state is constantly associated with certain abnormal appearances of the suprarenal gland is indisputable, but the nature of the relationship is obscure.

Case II, the monster with the brain normally developed but with abnormalities of the genitalia, is certainly suggestive of some association between the true cortex of the suprarenal and the development of the genitalia (15), for in this case the true cortex was relatively narrow and inactive although the rest of the gland was normal. It must be remembered in this case, however, that another gross abnormality was present, namely, absence of the thyroid gland.

The question of the connection between adrenals and the sex organs has been fully discussed by Glynn (11, 14): Krabbe (12), however, regards any coexisting malformations of these organs as due to the close proximity of the derivative cells. Recently Rolleston(19) has given a classified description of cases of primary adrenal tumours, in many of which associated reproductive abnormalities were present. The Development of the Human Suprarenal Gland = 323


1, At 5mm. the embryo shows the sympathetic anlage.

2. At 10mm. the cortical anlage of the suprenal is evident.

3. At 25 mm. the suprarenal gland is defined and consists of true cortex, foetal cortex and immigrating sympatho-chromaphil bundles.

4, The immigration of the sympatho-chromaphil masses begins at about 12 mm. development, is most active between the 12th and 22nd weeks, and ceases about full time.

5. The foetal cortex degenerates progressively during the last 10 weeks of foetal life, although still forming the bulk of the gland at birth; it is replaced by a downgrowth of true cortex, the cells at full time already arranging themselves in columns (zona fasciculata).

6. The following changes occur during the first year: the foetal cortex has disappeared by the 12th month; the zona fasciculata is defined by the 3rd week; the zona reticulata is defined by the 14th week; the medulla gradually increases in amount.

7. Lipoid appears normally in foetal and true cortex at 24 weeks.

8. Adrenin appears at 12 weeks; the chromaphil reaction appears at 22 weeks.

We offer our sincere thanks to Lady Barrett, Miss Mabel Ramsay, Dr E. W. G. Masterman, Dr A. L. Baly, Miss Mouillin, Dr John. Adams, Dr A. Wylie and to all others who have kindly sent us material. We are still anxious to increase our collection of embryos under 12 weeks.

We are indebted to Mr W. Pilgrim for some of the diagrams which he drew for us at very short notice.

This research is a 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, a grant to one of us (M. F. L. K.) from the British Medical Association, and a grant to one of us (E. E. H.) from the Medical Research Council.


(1) Extiorr and Armour (1911). “Development of cortex in human suprarenal.” Journ. Path. and Bact. vol. xv, p. 481.

(2) Keene and Hewer (1923). “Studies in foetal development.” Journ. Obstet. and Gyn. vol. xxx, pt 3, p. 345.

(3) SwaLE Vincent (1922). Internal secretion and the ductless glands. p. 123.

(4) Coopsr (1925). Histology of the more important human endocrine organs at various ages, p. 34.

(5) Murr (1924). Textbook of Pathology, p. 210.

(6) Papers by CuaurFraRD, LarocHE, GricavT and others, 1918 onwards, quoted by SHARPEYScuaFer in The Endocrine Organs, 1924, vol. 1, p. 106.

(7) SHarpry-ScHarer (1924). The Endocrine Organs. Vol. 1, p. 92. 324 M. F. Lucas Keene and FE. E. Hewer

(8) Oaata (1923). In Medical Science Abstracts, p. 155.

(9) ZucKkERKANDL (1912). In Human Embryology, by Keibel and Mall, vol. 11, p. 170. (10) Cosmurratos (1920). Gréce méd. vol. xxu, p. 17. Quoted in Endocrinology, 1921, p. 240. (11) Guynn and Hewetson (1913). Journ. Path. and Bact. vol. xvi, p. 81.

(12) KrasBeE (1924). Rev. frang. dendocrin, vol. x1, p. 103. (13) Oxtver and ScHarer (1895). Journ. Physiol. vol. xvi, p. 230. (14) Guiynn (1912). Quart. Journ. Med. vol. v, p. 157. —— (1921). Journ. Obstet. and Gyn. vol. xxvu, p. 23. (15) Buttock and Sequeira (1905). Trans. Pathol. Soc. vol. ivi, p. 189. (16) Cramer (1920). Brit. Journ. Exp. Pathol. vol. 1, p. 184. —— (1920). Journ. Physiol. (Proceedings), vol. Liv, p. ii. (17) Wricut (1910). Journ. of Exp. Med. vol. xu, p. 556. (18) Hammar (1925). Upsala Lékareforenings forhandlingar Ny Foljd, Bd. xxx, pp. 5, 6. (19) RotiEston (1925). West London Med. Journ. vol. xxx, p. 3. (20) Meyer (1921). Deutsch. Arch. Klin. Med. Bd. cxxxvu, S. 225. (21) DE BrsErR (1926). Anatomy, Histology and Development of the Pituitary, pp. 10-15. (22) KeErnr and HEwEr (1924). Lancet, p. 111.