Paper - The development and structure of the human placenta
|Embryology - 20 Apr 2021 Expand to Translate|
|Google Translate - select your language from the list shown below (this will open a new external page)|
العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt These external translations are automated and may not be accurate. (More? About Translations)
Strachan GI. The development and structure of the human placenta. (1923) J Obstet. and Gynaecol. (1923)
|Historic Disclaimer - information about historic embryology pages|
|Embryology History | Historic Embryology Papers)|
Review - The Development and Structure of the Human Placenta
By Gilbert I. Strachan, M.D. (Glas.), M.R.C.P. (Lond.), F.R.C.S. (Eng. & Edin.),
First Assistant to the Professor of Obstetrics and Gynaecology, the Welsh National School of Medicine; Research Pathologist, the Medical Research Council.
The substance of part of a report, to the Medical Research Council on “The Present Position of our Knowledge Regarding the Human Placenta”
As the placenta is not evolved until after the impregnated ovum is embedded in the uterine decidua it will be proﬁtable, ﬁrst of all, to consider the manner in which the ovum is nourished prior to this embedding.
It is not, at present, known by what means the unfertilized ovum is nourished while still in the ovary : the ﬁne striae seen in the zona pellucida are supposed by some to constitute channels by which nourishment can reach the ovum from the cells of i the corona radiata. But, as the ovum reaches maturity, specialized yolk granules appear round about the central part of the cell and push the nucleus to one side. These granules are highly refractile bodies : they increase in number as the ovum matures and Beneden has given them the name of deutoplasm. It is known that in ova other than human these granules are used up during the earliest stages of segmentation and they are, therefore, to be regarded as a possible source of nourishment elaborated in advance to tide the segmenting ovum over this period.
After fertilization the cells of the corona radiata disappear and are replaced by a sticky layer of albuminous material which is homogeneous and derived partly from the disintegrated cells of the corona radiata (Robinson) and partly from tubal and uterine secretion (Bonnet). “ With regard to its functions, there is little doubt that the degenerating cells of the corona radiata, and later the albumin layer, serve as food for the growing mass of the ovum in the Fallopian tube and uterus ” (Lochhead). In support of this theory Jenkinson has demonstrated the presence of fat and protein substances in this albuminous layer in the mouse while Emrys Roberts has shown that, in the guinea pig, the secretion of the uterine glands contains a protein which probably affords nourishment to the ovum. Bonnet holds that the corona radiata is absorbed by the ectodermal cells of the blastodermic vesicle and is used by the latter as food. After the disappearance of this albuminous envelope the segmenting ovum now lies free in the tube or uterus and has to seek a new source of nourishment.
The result of the cell division, so far, has been to encase the ovum with a layer of ectodermal cells—the extraembryonic ectoderm——while the embryo begins to appear at the spot where an ectodermal cavity——the amniotic .cavity——and an ectodermal cavity—the vitelline sac—are in apposition. The cxtraembryonic ectoderm has a specialized function to perform, namely to ﬁnd nourishment from the surrounding tissues for the rest of the ovum including the embryo; so that aprovision is made for the nutriment of the embryo in actual advance of the formation of the latter. On account of this function Hubrecht gave the name of trophoblast to the extra embryonic ectoderm.
A suitable part of the uterine mucosa—decidua—is selected and into that part the ovum sinks by piercing the surface epithelium, which is dissolved by the histolytic properties of a ferment which is supposed to be secreted by the trophoblastic cells of the ovum. As it sinks into the substance of the stratum compactum this solvent action is exerted on all the surrounding decidual tissue so that the growing ovum comes to be surrounded by a zone of necrotic decidua which is held to constitute still another source of nourishment for it.
So far then, we visualize the ovum as covered with trophoblast whose function is to dissolve the decidua and also to abstract proper nutriment from the surrounding tissues for the various structures, including the embryo, that are beginning to appear in its interior : thus at each stage of its development a source of nourishment is seen to be provided; it has penetrated into the substance of the uterine decidua and the aperture of its entry-through the surface epithelium is scaled up by blood clot (Bryce) or ﬁbrin (Peters).
The above stages which have not yet been observed in the human subject are deduced by analogy from the ova of lower animals. We have avoided the discussion of many attractive questions, e.g., the factors which determine the site of implantation, the time taken in the transit 0.‘ the ovum to uterus, the inﬂuence of the menstrual cycle phases on implantation and many more, remembering that our object is to trace the elaboration of the placenta and not to be side-tracked into other considerations.
It is universally accepted since Peters published the account of his ovum that the embedding in the uterine decidua is by a process of actual penetration so far as the compact layer of the decidua (vide infra), so that the Hunterian and other theories of circumvallation need not be discussed.
The ovum, then, covered by a more or less even coat composed of a layer of trophoblast is lying below the decidual surface and by its presence divides the decidua into decidua vera, basalis and capsularis, these three coming into contact with one another at the margin of the ovum—decidua marginalis. The cavity which separates these deciclual layers, which is usually somewhat larger than the contained ovum, is known as the implantation cavity.
As the ovum penetrates the surface between the mouths of glands (it is too big to enter a gland lumen) its relationship to the glands deserves mention: some glands are destroyed towards the surface of the uterine mucosa and the deeper parts of the lumina of the glands open into the implantation cavity. Again the glands at the periphery are pushed aside by the growing ovum with consequent outward baying of their lumina. As the ovum grows the blood supply to the decidua basalis developes in excess of that to the decidua capsularis; this feature is not marked at present but becomes increasingly evident with growth of the ovum. The blood clot (Bryce and Teacher), or mushroom-shaped closing coagulum (Peters) becomes organized in its basal layers to form Reichert’s scar over the summit of the decidua capsularis—the point of original penetration of the ovum.
It is necessary to take stock of all these features in order that the further stages of development of the ovum, now to be described, may be appreciated. The epiblastic covering now assumes true trophoblastic activities by sprouting out in buds all round the circumference of the ovum—which has been demonstrated in the case of the human ovum, and in this specimen (Bryce—Teacher) it is seen that the trophoblastic cells are arranged in two layers: the superficial layer consists of a plasmodium in which deﬁnite cell walls are not seen and deeply stained nuclei are, scattered irregularly in this plasmoidal mass, in some places scanty and in other places closely congregated. The protoplasm is generally described as being-granular and of a foam-like appearance; Webster and others have shown that this appearance is due to the cutting across at various angles of innumerable strize which penetrate, in‘ a close network, the whole plasmodium. The deep layer is composed of ordinary cells of various sizes and shapes according to the amount of mutual compression exerted, and cell walls are clearly evident with protoplasm and deeply-staining nuclei.
These trophoblastic buds are known as chorionic villi; the super. ﬁcial layer of plasmodium has received several names but the name syncytium, ﬁrst used by Kossmann and Merttens, is now generally applied, while the deep cells at this or later stages are termed Langhans’ layer from the name of the investigator who ﬁrst accurately described them. Prior to this stage the formation of the cavity of the extraembryonic coelom in the common mesoderm in the interior of the ovum has resulted in the trophoblast becoming lined by a layer of mesoblast; this trophoblastic coat from which the buds sprout is known as the chorionic membrane or chorionic plate. The buds or villi, so far, are composed of ectodermal cells and are known as primary villi; soon the mesoderm layer of the chorionic membrane is seen to insinuate itself into the centre of each primary villus to form Tor it a ﬁbroblastic core. The secondary villi are thus formed. The early stage of this penetrative process is well shown in Johnstone’s ovum.
So far the growth of the villi has been uniform all round the ovum; the villi in their centrifugal growth have everywhere invaded the decidua (basalis, marginalis and capsularis). Owing to their proteolytic action these villi destroy whatever tissue they come in contact with and of these the most important structures are the decidual blood vessels, which are eroded by the villi and hence the contained blood escapes between the villi so that in as early a specimen as the Bryce-Teacher or Peter’s ovum the primitive villi are seen to lie bathed in a pool of maternal blood. The next stagethe haemotrophic stage—of development is thus entered upon, and since the mesodermic cores of the villi as yet contain no blood vessels the nourishment of the ovum must be by direct absorption of nutrition from the maternal blood by the chorionic cells. The process of development has, so far, been rapidly sketched; certain points, however, require more detailed consideration before we proceed further.
The trophoblastic epithelium, as. stated, consists of two layers. The inner layer of Langhans’ cells, which at ﬁrst form the major part of the primary villi, by the end of the third \veek have become spread over the mésodermic core in a single or double layer. Here the cells are mostly rounded in shape with relatively large oval or circular nuclei. At the extremities of the villi these cells show a high degree of proliferation, becoming many-layered, polygonal in shape and larger than the others. The ova of Bryce-Teacher and Peters are seen to lie in a cavity in the decidua larger than the ovum itself ; as the villi grow outwards they gain attachment by their tips to the walls of this cavity and, by eroding the blood vessels, so cause the implantation cavity to be occupied by exuded blood. It is, therefore, where the villi are attached to the maternal decidua that the proliferation of the cytotrophoblast of Langhans’ oells occurs. The proliferating cells spread in a direction mainly over the surface of the decidua, but, to some extent also deeply so that in some cases scattered Langhans’ cells are found in the deeper layers of the decidua or even in the muscle of the uterus. These proliferating Langhans’ cells—cell-nodes or cell-columns—may, in some cases, present an appearance very like decidual cells and have frequently been mistaken for the latter; indeed some observers have regarded them as derived from decidua, but the present—day opinion regarding their origin is as stated. The main functions of the Langhans cells would appear to be, ﬁrstly, histolytic on the surrounding tissues, secondly, ﬁxative in order to anchor the ovum, and thirdly, nutritive in abstracting aliment for the ovum and its contents ﬁrst of all from the surrounding tissues and later from the exuded maternal blood.
The outer cell layer—the syncytium—has given rise to much discussion, being one of the most debated structures in histology. Regarding its origin many views have been expressed, and Peters (quoted by Webster) was able to collect 10 different theories in this respect. We may disregard most of these and conﬁne our attention to four :—
(a) Langhans and Strahl held that the syncytium is derived from the surface epithelium of the uterine mucosa which is inverted on the surface of the advancing and penetrating villi. This pre~ supposes that the ovum is implanted on the surface of the uterine mucosa which it does not penetrate, and in view of our knowledge to the contrary this theory must fall to the ground.
(b) ]ansenske’s view was that this layer represented glandular epithelium which had’ been invaginated on the surface of the ingrowing villi. This view is open to the same objection as the previous theory; again, owing to its size, the ovum could not enter a gland lumen, as it is about 0.2 mm. in diameter at the time of implantation (Grosser) while the lumina of the uterine glands is only about 40 to 50 p. in diameter (Frank). Furthermore, in cases of ovarian pregnancy in which no uterine epithelium—surface or glandularis present, well marked syncytium is usually to be seen.
(c) Turner and Waldeyer were of the opinion that the growing villi insinuated themselves into the lumina of the dilated capillaries and so acquired a covering of endothelium which thus formed the outer—the syncytial—layer of the villi. According to this theory the intervillous spaces are greatly dilated capillaries whose walls are invaginated to a varying degree. This view was supported by Pfannenstiel, but was soon disproved by more accurate histological methods.
(d) The universally accepted view at present is that the syncytium is formed from the Langhans’ cells and so is of foetal origin.
The appearances seen in the Bryce-Teacher ovum practically prove this origin. The syncytium is "regarded as derived from the deeper layer of cytotrophoblast but the exact process of formation is unknown; Peters believed that intervening cell walls of the Langhans’ cells break down and so produce the multinucleated appearance, but Teacher regards the process as a continual division of the nuclei of these cells without cell division so that the syncytium represents a very large giant cell. The actual process is not of great moment so long as we know the source.
Function of the Syncytium
The function of the syncytium has been much discussed; Hubrecht, Marchand and Bonnet held that it constituted a reaction on the part of the trophoblast when it comes in contact with maternal structures. But, as Johnstone points out, well developed syncytium present in parts such as the chorionic plate which are not in contact with maternal tissue, so that this theory would require modiﬁcation as the stimulus must be exerted apart from actual contact with, probably mere approximation to, maternal tissues.
Peters regarded syncytial formation as a reaction on the part of the Langhans’ cells to the presence of maternal blood, its formation producing a layer of tissue which prevents coagulation and so allows circulation of the maternal blood.
Bryce, however. quotes van Beneden to show that the syncytium is formed probably before the blood lacunze, but this does not dispose of Peters’ theory, as the syncytial formation may occur in advance of the actual neoessity for it.
Grosser suggests that there are probably two generations of syncytium: the ﬁrst is associated with implantation and the concurrent intense histolysis. Later on this degenerates and the function of breaking down decidua is transferred to the Langhans’ cells. Then, with the formation of the blood lacuna-2 the second generation of syncytium appears, whose function is to extract nourishment from the maternal blood in the lacunae and also to prevent coagulation of that blood. We may take it then that the functions of the syncytium are (a) histolytic, (b) anticoagulative, and (c) alimentary.
The structure of the syncytium is that of a plasmodium spread in irregular thickness over the surface of the chorionic membrane and villi; in some places it is thick and heaped up, elsewhere thin and scanty and, here and there, not to be seen. The nuclei are also irregular in distribution, small and darkly staining. As stated above, the protoplasm is of a granular and foam-like appearance; it very soon shows vacuolation and, as this increases, the syncytium assumes a spongy appearance.
Bryce and Teacher hold that a ferment for the solution of the decidua is produced in these vacuoles, and when this ferment has been expelled the cavities left become ﬁlled with maternal blood, and so the blood lacunae or primary intervillous spaces are formed.
In early specimens the free border of the syncytium is smooth, but after about the second week careful examination will show that it is distinctly striated. This appearance was ﬁrst described by Minot and is seen to consist in the formation of a series of ﬁne striae or cilia running perpendicularly to the free border; at their bases are knobs—blepharoblasts —and Bonnet points out that their presence proves the foetal origin of the syncytium, as cilia are seen always on the superﬁcial aspect of cells. Kastschenko believed that these cilia projected on the surface and by their movement created a stream in the maternal blood in the intervillous spaces. Lenhossek, however, could observe no movements; he called them stationary cilia, and ascribed to them the function of breaking down decidual vessel walls during the burrowing of the ovum. Others again hold that the cilia are merely hollow rods or pores by which nutriment may enter the syncytium or by which a secretion of the syncytium may be expelled into the maternal blood to prepare its constituents for transference to the foetus, while von Spce attributed the appearances to the roughening edge of the protoplasm by a strong ﬁlm of ﬂuid—maternal blood—in the blood lacunae. Finally, these striae have been regarded as artefacts produced by the methods of ﬁxing, but Hofbaucr showed that they are seen best in fresh specimens and so are probably vital structures.
The mesodermal core of the villi, as explained, is an outgrowth of the mesoderm of the chorionie membrane. In the early stages this stroma or matrix is of a very loose myxomatous ﬁbrillated appearance containing spindle and stellate cells and their connecting processes. These connecting processes form a network in the matrix and may represent a series of lymph channels. Later on these cells become more spindle-shaped and more closely packed so that the stroma becomes denser. Blood vessels do not appear in the villi until about the third week and are, at ﬁrst, small and merely endothelial-lined slits, but later become thicker walled and more evident.
The known facts regarding the origin of the intervillous spaces do not render the process of their formation quite clear. Bryce and Teacher’s ideas have been previously mentioned, but Peters supposed that the blood was extruded from the maternal vessels under arterial pressure and so forcibly excavated for itself cavities in the villous trophoblast. The former is the theory generally accepted at present. Frassi was of opinion that the trophoblast cells actively separated from each other, grew apart and joined up again to form cell columns, the spaces between which became the intervillous spaces, while Sfameni and Ercolani considered that the lacunae represented merely dilatations of blood vessels and that the essential feature is the expansion of the uterine wall capillaries caused by the enlargement of the uterus during pregnancy. Two neighbouring capillaries might thus easily come in contact and the wall between them at last become torn. This process would be repeated, and ﬁnally one large blood space would be formed out of a number of fused capillaries. According to these authors the formation of the blood lacunae is the result of purely physical conditions of the uterine musculature. D’Echia is of opinion that the blood lacunae originate from blood capillaries surrounded by syncytium-like elements. The walls of these capillaries wear out and perish, thus allowing the blood" to flow freely into the space surrounded by these elements.
Normally these blood lacunae, which later become the intervillous spaces, are ﬁlled with maternal blood, although Marchand has stated the contrary.
The blood in the spaces does not coagulate, and this would indicate some peculiar function of the trophoblast and especially of its syncytial layer in preventing this. Hofbauer supposes the elaboration of a thin layer of albumose on the surface of the villi for this purpose, but later observers have not conﬁrmed this ﬁnding. At this stage, then, the entire ovum is surrounded by villi which project into a pool of maternal blood which does not coagulate but slowly circulates in the compound intervillous spaces and is, sooner or later, returned to the maternal circulation, no doubt after having parted with its nutritive substances.
It is not proposed to go into detail in describing the structure of the decidua as the main points are known to all. Decidual formation may be deﬁned as the reaction on the part of the uterine mucosa to the conditions of pregnancy.
The most prominent feature of decidual formation is hypertrophy of all the elements of the mucosa; the blood vessels dilate and become tortuous; the glands become hypertrophic, convoluted and dilated, and the epithelium projects here and there into the lumen of the gland as papilliferous elevations with a core of stroma; but the main change occurs in the interstitial tissue where the decidual cells are formed in great abundance.
The decidual cells are large and may be of any size up to 50 /,1. in diameter; in shape they are round, oval or polygonal from mutual pressure. The nuclei. are round and vesicular and stain but slightly with the ordinary reagents.
All the decidual cells are not formed at the same time, so that here and there unaltered interstitial cells are found, and all degrees between this and the fully-formed decidual cell are seen. On this account Marchand in 1904 divided decidual cells into two classes: large and small. Many of the large decidual cells degenerate during pregnancy and are absorbed by leucocytes, but the majority are cast off during or just after birth. The result of this hypertrophic decidual formation is that the uterine mucosa in early pregnancy becomes succulent, hyperaamic, thrown into longitudinal folds and greatly thickened, so that between the second and third month of pregnancy the decidua is more than a quarter of an inch in thickness.
It has been previously indicated that the penetration of the decidua by the ovum divides the former into decidua basalis, capsularis, marginalis and vera. The decidua capsularis at ﬁrst forms a prominence into the uterine cavity covering the growing ovum, but by the end of the third month of pregnancy this abuts on to the decidua lining the opposite ‘wall of the_uterus, the decidua capsularis fusing with theopposed decidua vera and the uterine cavity thus becoming‘ obliterated.
The theories regarding the origin of the decidual cells are considered by Lochhead and may be tabulated thus :—
- Langhans and Hennig held that they were modiﬁed leucocytes, but failed to prove this.
- Overlach and F rommel thought they were modiﬁed glandular cells.
- Ercolani was of opinion that the developing ovum ﬁrst completely destroyed the mucosa, and that subsequently the decidua was formed from the endothelium of the vessel walls.
- Creighton, in 1878, was the ﬁrst.to maintain that they were derived from the interglandular tissue of .the' mucosa which consists of connective tissue of an embryonic t_vp,; and so allows of a rapid transformation of its cellular elements. This view was supported later by Minot, Hart, Gulland, Peters and others, and is the view generally accepted to-day. Since then Hitschrnann and Adler have conﬁrmed these ﬁndings, and have shown that the uterine mucosa is in a state of continual structural" change throughout the whole menstrual cycle.
In the superﬁcial layers the structures mainly affected in the decidual changes are the interglandular tissues which hypertrophy and become almost completely ﬁlled with decidual cells and blood ‘vessels, the lumina of the glands becoming more or less compressed so that this part of the decidua has a firm, compact appearance and is known as the decidua compacta. In the deeper layers the gland lumina are greatly dilated and lined with cubical or ﬂattened epithelium; the interglandular material is relatively scanty in amount and does not show so great a transformation into decidual cells. The deeper part is thus of a more open spongy appearance and is known as the decidua spongiosa. It is only to the deepest strata of the compact layer that the chorionic villi penetrate during implantation of the ovum.
No change is seen in that part of the fundi of the glands situated in themuscular coat of the uterus; it is from these remnants that the whole mucous membrane of the uterus is replaced post partum.
The nature of the decidual formation was for long thought to be a reaction on the part of the mucosa to the presence of an impreg - nated ovum. But as a uterine decidua is formed in cases of ectopic gestation it is clear that some other explanation is necessary. It may be said that decidual formation constitutes a reaction on the part of the uterine mucosa to the condition of pregnancy, but much more than this is not known. The most recent work by Bom and Loéb goes to show that the internal secretion of the corpus luteum is of importance in this respect; according to Loéb this secretion has a sensitizing action on the uterus; he attributes the structural changes of the pre-menstrual and menstrual periods to a reaction to this secretion which, however, in the absence of pregnancy passes off. But in pregnancy when the corpus luteum persists this reaction also persists and increases and represents decidual formation. While, according to this view, decidual formation would appear to be mainly dependent on the internal secretion of the corpus luteum it is evident that mechanical stimulation (the presence of the impregnated ovum) also plays a part, if even a subsidiary part, because if the non-pregnant uterine mucosa be mechanically irritated duringthe time of existence of a corpus luteum, a decidual reaction is produced which, however, soons disappears when the irritant ceases to act. (Loeb). This point will be referred to again.
The functions of the decidua may be summarized as :—
- To synthesize and store glycogen as a supply of carbohydrate for the embryo. This was ﬁrst pointed out by Marchand and was shown by Driesser to occur mainly in the glandular epithelium of the decidua spongiosa. This function, while observed mainly in rodents, is present also in man, but it is exercised only in the early months and ceases presumably when the foetus has advanced in development sufﬁciently to store its own carbohydrate.
- Fat globules have been found at a period of gestation prior to that at which fatty degeneration would be expected, and on this account the decidua has been looked on as an important source of fat for the foetus.
- A provision against a too intense penetration of the ovum; Turner ﬁrst pointed this out, and many subsequent observers have produced facts to support it, especially Webster, Bryce and Teacher. The latter authorities state that decidual formation controls trophoblastic activity until such time as placentation is complete. Young supports this view and regards decidual formation as localizing the haemorrhage and oedema to that part of the mucosa immediately surrounding the embedded ovum. He is of opinion that the physical changes in decidual formation are due entirely to imbibition of ﬂuid by the mucosa.
- A source of nourishment for the ovum. It has been pointed out that prior to trophoblastic formation the ovum ﬁnds alimentary pabulum in the degenerating decidual tissue round it, and Pfannenstiel held that decidual cells are merely degenerate forms of connective tissue cells and are of use only as pabulum to be absorbed by the ovum. While the decidua may be safely regarded as nutritive in the very earliest stages of pregnancy, this function is soon abandoned for the protective when the trophoblast developes.
Vascularization of the Placenta
While, in the early stages, sufﬁcient nutrition for the developing ovum can be obtained from the maternal blood in the intervillous spaces by the chorionic epithelium, by the end of thethird week such a mechanism is insufficient to provide for the needs of the embryo, and a further mechanism has to be elaborated. This consists of a process of vascularization of the chorion and the formation of a connexion between the chorionic and embryonic vessels. This represents the ﬁnal stage of foetal nutrition and the most elaborate. Most authorities, in their descriptions, avoid details regarding the earlier stages of this process; only in embryological works is any attempt made to describe the earliest stages of the process, which is as follows : About the third or fourth week blood vessels are present in the mesoderm of the chorionic membrane and are carried into the villous cores where they proceed to function as the agents of embryonic and foetal assimilation and excretion. At this time the ventral stalk is a very short structure, and the consideration that arises is whether these vessels are formed simultaneously in embryo and chorion and link up with each other to form a complete cardiovascular system, or whether the vessels are formed primarily in connexion with the embryo and grow along the ventral stalk to reach and vascularize the chorion.
The first vessels formed are in the mesoderm on the lower surface of the yolk sac where groups of nucleated corpuscles closely packed together—the blood islands of Pander—appear. The peri apheral cells become the endothelium of the lressel walls while the central mass becomes the primitive-nucleated red blood corpuscles. Cellular processes unite the various islands together and, becoming hollowed, produce a vascular network. This network extends over the sac and joins with a similar system that has been formed in the body stalk and chorion as an extension of the allantoic vessels. Bremer is of opinion that the earliest blood vessels arise separately inthe yolk sac and body stalk by multiple foci of origin and that in the latter situation they take origin in funnel-shaped ingrowths of surface subepithelial tissue. Robinson says: “The first formed villi arecnon-vascular, but by the time the secondary villi have developed the umbilical arteries have grown through the body-stalk (allantoic stalk) into the mesoderm of the chorion, and branches from them enter into the mesodermal cores of the villi, which thus become vascular.” Formed in this way by‘ the fourth week tlte vessels penetrate into the mesodermal cores of the villi and soon extend to the extremities of the villi, a state of affairs which persists until birth.
So far the villi described have consisted of single stems; now these stems branch again and again, and thus the villi become progressively longer with compound branches, and_ the appearance presented is that of an inverted clump of trees. The object of this branching is to increase the area of villous surface available for absorption. As a result of the invasion of the villous stems by the chorionic mesoderm the trophoblast becomes spread over the surface so that on section a central myxomatous mesoblastic core is seen with a small vessel or two in it, covered by a layer or two of Langhans’ cells outside of which is a layer of syncytium usually showing only one layer of nuclei.
Further, as pregnancy advances, structural changes are seen in the villous epithelium. From about the r8th week onward the Langhans’ cells begin to disappear, and with the progressive expansion of the surface of the villi due to arborization the syncytial layer becomes thinned out, sopthat from the zoth week a villus consists of a core of embryonic mesoderm containing blood vessels and covered by a thin layer of syncytium.
Lochhead explains this occurrence by pointing out that, in the earliest weeks of gestation, the various. embryonic organs are not suﬁ-iciently differentiated to be able to carry out their respective functions, and that the Langhans’ cells act for them by selecting carefully the substances to be allowed to enter the foetal circula'tion. Later the organs are able to select for themselves and the Langhans’ cells, therefore, disappear, being of no further use.
The intervillous spaces now become very complex, and it will be appreciated that the maternal blood circulating sluggishly through this space comes in contact with a very large compound chorionic surface so that ample opportunity will be afforded for a complete interchange of substances between the two circulations which, however, are always separated by the foetal vessel‘ walls, mesoderm of the villi and villous epithelium.
Villi are of two kinds: (1) those that extend from the chorionic plate to the decidual attachment, fastening villi or (2) those that hang free in the intervillous blood space—absorbing villi. Again the lateral branches of the fastening villi may hang free in the space or be attached to the decidua.
Where villi are attached to the decidua the chorionic epithelium is not usually reduced to the thin layer, as described, but forms a thick mass of cells and syncytium. at the attached tips of the villi, which mass penetrates into the. decidua. These are known as cell columns, and are, therefore, villous remains into which the mesoderm; has not, so far, penetrated. On this account the attachment of the ovum to the decidua is loose during the ﬁrst six weeks and, as‘ Hitschmann and Lindenthal point out, complete separation of the ovum at this stage is easy._ These cell columns are mainly responsible for the penetration and splitting of the decidua. They are long when 'ﬁrst formed but continually diminish until by the eighth or ninth week they have become invaded by mesoderm and the attachment of the ovum is much ﬁrmer, so that separation now necessitates removal of decidual tissue with the ovum.
Cell islands are likewise massesiof apical trophoblast into which the mesodermal stroma has not penetrated; they differ from the cell columns by lying free in the intervillous spaces, -and therefore cap absorbing villi and not anchoring villi. The cells are1large'and swollen and, especially when cutacross, in section, and. seen apart from the parent villi resemble masses of decidual cells, but true decidual masses are not seen in the intervillous spaces.
Webster maintained that these cell islands may contain decidual tissue, but other observers have not conﬁrmed this. These islands, like the cell columns, disappear. by the tenth week- and are mostly converted into ﬁbrinoid material.
The basal ectoderm is that portion of the troplioblasf which spreads out from the decidual attacliment of the .villi and covers the surface of the decidua. It, therefore-, forms the outer wall of 't«h'e'inte'rvlllous space. It will be seen, then, that the‘interv-ivllous space is lined on every aspect by trophoblast and" is ﬁlled with maternal blood, so that Duval’s deﬁnition of the placenta as “ a maternal haemorrhage encysted by foetal ectodermal elements ” is essentially true. The basal ectoderm may form a single or a Stratiﬁed layer and persists until the end of pregnancy, either as a continuous layer or as groups of oells; it is composed partly of syncytium and partly of Langhans’ cells and, lying on the surface of the decidua, may closely resemble this tissue, so that its presence is often overlooked and it is regarded as decidual tissue. From the latter it is distinguished thus by Frassi: “ The nuclei of these elements are larger and take the stain more deeply and regularly (than decidual cells). With strong magniﬁcations a distinct difference can -be perceived between the endonuclear substance in such cells and that of the decidual cells.” It is to be noted that, whereas transitions in form between these cells and syncytial tissue may be seen, so proving their foetal origin, a similar transition into decidual substance is never found.
It will be seen then, that the decidua forming the lining of the implantation cavity is everywhere invaded by trophoblastic cells so that the ovum becomes ﬁnally ﬁxed in its bed.
Transition zone. This is the name given to that plane of the decidua where foetal and maternal tissues meet, where the penetration of the trophoblast is ﬁnally held up and where the decidua prevents any further penetration by the invader. At ﬁrst the trophoblast stems seem to penetrate almost unchecked, and at this stage decidual formation is imperfect. Later, when the decidua is more fully formed, this invasion becomes resisted, and it was the observation of this fact that led Turner and others to the conclusion that the main function of the decidua was protective; proof of this is found in tubal gestation in which decidual formation is very imperfect in the tube and the chorionic villi penetrate much more deeply than in the uterus. Of this johnstone says: “It is, in truth, the ﬁghting line where the conflict between the maternal cells and the invading trophoderm takes place, and it is strewn with such of the dead of both sides as have not already been carried off the ﬁeld or otherwise disposed of.”
In the Bryce-Teacher ovum the surface of the decidua surrounding the implantation cavity is markedly necrosed A forming a “ hyaline, darkly-staining, and nearly nuclear-free zone." In ]ung’s ovum there was a similarly described layer which was, however, thickly inﬁltrated with red and white blood corpuscles.
In Young’s ovum (1911) the condition is similar, and he traces the changes in the individual cells participating. The cells in the transition zone, especially their cytoplasm, ﬁrst become markedly swollen and stain deeply with eosin, exhibiting a granular appearance, " or, in some cases, a curious character resembling . . . opaque glass.” In places the inner surface of the necrotic zone is tagged and looks as though it was being broken off in fragments into the implantation cavity. Vacuolation is prominent in this degenerating area. In ]ohnstone’s ovum (1914) the border zone consists of oedematous, degenerated tissue. Streaks of ﬁbrinoid material are seen in this area.
Tlie transition or border zone may thus be deﬁned as a zone of degenerated decidual tissue of indeﬁnite extent interposed between the foetal trophoblast on the one hand and the normal decidua .on the other.- In any individual specimen it may be difficult to deﬁne the zone exactly, and certain authors have described what they consider to be its physical characteristics; thus Frassi and Jung state that a leucocytic zone around the ovum is indicative of this area, as the.leucocytes cannot pass beyond the barrier of the trophoblast, but other observers have found such leucocytic. inﬁltration. of the decidua too generalized to be of value in this respect.
The term “ detritus zone,” employed by Bonnet, would appear to be a more appropriate name. He described in this area symplasma formations — areas of coagulation necrosis — of the decidua! cells and enlarged glands in whose lumina are seen secretion, blood and leucocytes.
In this connexion the third layer may be discussed. It has been thought by some to represent a deﬁnite layer of cells between the two layers of trophoblast and the decidua. .Hubrecht gave the name “ deciduo-fracts ” to a layer of giant oells which he described in the decidua of the mouse and hedgehog. They appear in the earliest days of gestation tissue phagocytic, showing absorbed red blood corpuscles and decidual tissue in their interiors. He regarded them as a means of .enlarging the implantation cavity.
It is of little use going into details regarding the various views of the origin of the third layer whether the cells be foetal .(Duval, Sabotta)-—decidual (Kolster, Disse)—or partly foetal and partly decidual (]enkinson)—the important point is that all authors regard them as phagocytes whose function is to erode, absorb and digest the surrounding decidual tissue and so enlarge the implantation cavity. Their life history is short and they soon die and disappear. In the human ovum Bryce and Teacher have. described somewhat similar cells lying within the necrotic layer of the decidua; these cells were partly degenerated and,_ on. account of their staining reactions and inability to trace them to connexion with trophoblast these investigators regard them as maternal in origin.
- Peters and some "other observers have described other cells", but in man; as Lochhead points out, although these cells would appear to be producing a layer of coagulation necrosis-—a symplasma— round the trophoblast, yet they show no signs of ingestion of decidual tissue or blood cells. From this it would appear doubtful whether cells described by Peters are the same as the deciduofracts in the hedgehog and mouse, but even if they should be similar they soon disappear.
The formation of fibrin. can usefully be considered here. Fibrin formation is a degenerative change which occurs in the decidua basalis and capsularis, but not to any great degree in the vera. Attention was first drawn to it by Nitabuch and the subject has since been studied by many others.
Fibrin may be found in four situations:
- as a layer at some distance from the surface of the implantation cavity, most commonly in the basalis butalso often in the decidua capsularis. It was in this area that Nitabuch ﬁrst described it. The time of appearance is uncertain, but Webster has demonstrated its presence as earlyas the fourth week in the basalis and earlier in the capsularis.
- Not so constantly a fibrin layer is found in placentze up to the third month immediately in the wall of the intervillous space. This was first ﬁgured by Rohr and is called Rohr’s stria, or the upper stria, Nitabuch’s being the lower stria.
- Close beneath the chorionic membrane is constantly found in the second half of pregnancy another layer of ﬁbrin——Langhans’ stria. Layers of canalized fibrin are also found here.
- Everywhere, and especially in the intervillous spaces, strize of fibrin are laid down in the second half of pregnancy; they may be seen as small or large ﬂakes attached to one side of a villus, or they mayform-large masses, in the interior of which several more or less degenerated villi are seen. If these masses are small but just visible to the naked eye they are spoken of as ﬁbrin nodes, but if larger they are called white infarctions. They will be referred to again.
Fibrin formation is a sign of degeneration, and it will be seen that degeneration of the placenta is thus started by the fourth week. ‘As pregnancy advances the various layers of ﬁbrin enumerated are laid down to an increasing degree; thus, while preparations are made in advance for the nutrition of the embryo, preparations are likewise made long .in advance for the termination of that nutrition, Regarding the staining properties of fibrin Grosser quotes Hitschmann and Lindenthal that "the typical fibrin reaction is shown only at "its earliest formation, and that later on this is absent; on this account he suggests the alternative name of ﬁbrinoid substance. In the earlier months and before the change has become widespread Nitabuch’s striae are regarded as the boundary between the foetal and maternal tissues; although here and there some decidua] masses may be seen superficial to it, in general this view holds.
Another point is that -leucocytosis in the decidua does not penetrate beyond the stria. Degenerative changes are held to be responsible for the formation of ﬁbrin; but degeneration of what? Langhans and N itabuch held that decidual degeneration, probably as the result of contact with the trophoblast of the villi, was respong sible for the appearance of tl1e Nitabuch striae. ;But much ﬁbrin is formed round the villi and beneath the chorionic membrane, situations in which decidua is absent; ‘so that the trophoblast would appear to be the source in these sites. Again ordinary blood ﬁbrin (maternal), shown by the presenceof red and white corpuscles in the masses, may appear, and thus a third possible source is indicated. This is probably the source of the subchorial (Langhans) ﬁbrin, as the structure here is in layers as ‘though successively deposited, especially as the blood ﬂow would appear to be greatly retarded in the roof of the intervillous space.
In the villi ﬁbrin starts to be deposited partly on the surface where and when the syncytium has almost disappeared, and also between the syncytial covering and the stroma of the villi, pointing to the former being the seat of origin. Langhans’ cells often persist long in these ﬁbrin masses and may stand out prominently as an almost complete row of cells; but ﬁnally they disappear and are, themselves, converted into ﬁbrin. This has been likened to the process by which “ cells, and even whole oell territories become transformed into matrix in some kinds of cartilage ” (Grosser).
With the syncytium transformed into ﬁbrin one of its great fun’ctio'ns—the prevention of coagulation ‘in maternal blood—is lost, and hence the original ﬁbrin mass is added to by ordinary ﬁbrin deposited from blood coagulating on its surface.
By this process a number—it may be a great number—of villi become bound together in ﬁbrinoid material and later the stroma of the villi undergoes hyaline transformation, so that here and there in the dark masses of ﬁbrin lighter staining areas are seen representing the original villi. Later still these become mere ghosts, and ﬁnally are completely transformed into ﬁbrinoid material.
In the chorionic membrane the changes are seen by the fourth month; a partial change, as describedabove, is present by. thesixth month, and the trophoblast in the centreof the placenta is replaced by the ﬁbrin stria. Blood coagulation in layers is thus induced by the factors mentioned and masses of canalized ﬁbrin result.
The cell islands mentioned above are an important early source of ﬁbrin, and it ‘is thought that they may give a. stimulus for the formation of the larger masses. In the centre of some of the larger ﬁbrin masses liquefaction may occur, and so placental cysts ‘are produced; they are normal structures, usually‘ very small, and present in mature placentae, showing on section or through the amniotic surface. Such cysts have been studied by Giese, who found a lining of ﬂattened endothelium-like cells, derived from trophoblast, surrounded by masses of ﬁbrin in which ghosts of chorionic villi are seen.
In infarct-formation it is important to remember that the ﬁrst event is ﬁbrin-formation round the villi; degeneration is secondary to this as the villi are thus cut off from their source of nourishment —the maternal blood in the intervillous space. The idea of Ackermann and sortie others that degeneration of the villi is the primary occurrence is wrong; the‘ villi are nourished by the maternal blood and the villous (foetal) circulation is entirely for the purpose of eﬁecting blood interchange just as is the pulmonary circulation in the adult.
A great deal of discussion has raged round the origin of these so-called placental infarctions. Their presence has long been noticed in a large proportion of placentae; indeed Williams, investigating 500 consecutive placentae, found naked-eye infarcts in 63 per cent., while minute ﬁbrin nodes were present in every case.
Eden, Williams and Ackermann are of opinion that the primary factor is an endarteritic process in the arteries of the villi; by this the nutrition of the peripheral villi is interfered with allowing ﬁbrin to be deposited, ﬁrst just beneath the syncytium, and secondly on the surface. With this deposition occurring at various foci it is easy to imagine that numbers of villi may become glued together in masses of ﬁbrin and undergo ﬁnal degeneration. Looked at in this way the term “ infarction ” is justiﬁed. Williams and Eden insist that this is a normal condition in the placenta and indicates a senile change in the organ. From this it would appear, however, that the nutrition of the villi is through the chorionic vessels; a little consideration will show that such is not the case, and Young in 1914 convincingly marshalled the facts to prove that it is the maternal blood in the intervillous spaces that nourishes the villi.
Young is of opinion that the white infarctions are the end results of red infarctions caused by the retro-placental haemorrhages. Williams admits that one type of infarction, characterized by vascular dilatation and clumping of the villi, may be due to such a mechanism, but while the writer has evidence to show that a number of placental haemorrhages do end as white infarcts it cannot reason. ably be claimed that such is the usual method of production in view of the frequency of these ﬁbrinoid deposits. Clemenz, in 1922, could not detect the endarteritis described by Williams and others, and concluded that the main factor is a clotting of maternal blood in the intervillous spaces which is initiated by ﬁbrin deposits upon villi with local epithelial defects, and is aided by the slowness of the blood stream in the placental circulation. This would appear to be the most reasonable explanation as it meets all the facts of the case. Most investigators will agree with the dissatisfaction of Clemenz with the term “ infarction.” He suggests in its place the better name of “ white necrosis of the placenta.”
Talbot in 1921 put forward the view that infarct formation was a sign of some hidden sepsis, but his paper is unconvincing.
As we know that the villi are nourished by intervillous blood it is obvious that the cause of so widespread degeneration must be looked for in someinterference with the maternal circulation,and on this account it seems that the views of Clemenz are the most satisfactory. Usually the presence of these necrotic patches makes no difference to mother or child, showing that placental tissue is formed in an amount beyond the actual requirements of the foetus; but when the change is very extensive it is probable that foetal death may be caused by ablation of a sufficiently large area of placenta.
In this connection we may discuss the subchorial closing ring. This is a plate of cells of varying thickness which forms. a rim at the placental margin. It is not always present and is often incomplete; it lies between the ﬁbrous tissue of the chorionic membrane and the overlying ﬁbrinoid material (Langhans), laterally it passes insensibly into the epithelial layers of the chorion laeve. It is composed of cells which, towards the centre of the chorionic membrane, form a single layer, and peripherally often several layers; it represents one of the situations in which the Langhans’ layer has persisted until the end of pregnancy.
Winkler described a continuous layer of cells lying beneath the connective tissue of the chorionic membrane; he called this the closing plate and supposed that it was derived from the decidua.
Pfannenstiel supposed" that this was produced by the undermining of the decidua marginalis by the marginal villi. Much discussion has centred round the origin of this plate into which it is not our purpose to go; suﬁice it to say that the modern conception is that the closing ring is formed as described above by trophoblastic cells, and that any supposed derivation from decidua is due to a wrong interpretation of the appearance of the cells. A "number of text-books describe and ﬁgure this structure as a separate process of decidua—the subchorionic decidua-—but this is opposed to the modern conception of the intervillous space being limited in every direction by foetal trophoblast.
Formation of the Placenta
So far the chorionic villi have been described as entirely surrounding ‘the ovum——a condition well shown in the BryceTeacher, ‘Peters and Leopold ova-—bathed in maternal’ blood 630 Journal of Obstetrics and Gynaecology
in the lacunae. Between the villi and‘ decidua basalis the attachment is intimate, but between the villi and decidua capsularis it is never so intimate, and penetration and attachment of the villi is not so deﬁnite as in the case of the decidua basalis. As pregnancy goes on the ovum grows and protrudes more and more into -the uterine cavity and the decidua capsularis becomes stretched more and more over the surface of the ovum; also, on account of the stretching, the blood supply to the decidua capsularis and to the related intervillous spaces is interfered with and increasingly limited. From both these causes the nutrition of the villi in relation to the capsularis is interfered with and they begin to atrophy. In a four weeks’ ovum a spot bare of villi has been seen at the summit of the decidua capsularis (decidual pole) (Pfannenstiel), and even at the second week early signs of atrophy have been noted. To compensate for this loss of nutritional surface the villi connected with the decidua basalis continue to branch and to grow in length while becoming correspondingly thinner; by these means a relatively increasing surface area is exposed to the maternal blood in the intervillous spaces and so an increased blood exchange occurs between the two circulations.
Thus with the villi atrophying at one region andundergoing hypertrophy in another the chorion ceases to present the generalized shaggy appearance of the early ovum, and becomes smo0ther—chorion lzeve—all over except where it is_related to the decidua basalis——chorion frondosum.’ Microscopically, degeneration is very evident in the chorion laeve. The epithelium of the villi disappears and hyaline degeneration occurs in their stroma, and between these hyaline masses the remains of cells and leucocy.tes are seen. At the decidual pole even these hyaline vi-lli vanish, but they persist in the neighbourhood of the placenta. '-The epithelium of the chorionic membrane is, however, usually .rec‘og-nizable even at maturity, and extending to it is a zone of detritus containing the remains of the villi, the capsularis and the compact layer of the apposed decidua vera in which also some degeneration has occurred. The process is a gradual one and is" established by about the ninth week, being ﬁrst seen towards the capsular pole of the chorion; it is completed by the twelfth week when the hypertrophied villi are now limited to a circular area on the chorion related to the decidua basalis—the discoid placental area. This area is about one-fourth of the area of the uterine wall, and this ratio is maintained more or less up to the end of pregnancy.
Several considerations arise out of these facts. What are the factors which arrange the disposition of the contents of the ovum to that the body stalk, the future umbilical cord which connects the "embryo is in practically every case with that part of the chorion which is related tothe decidua basalis? Again, in the event of the usual mechanism becoming disarranged and the umbilical stalk becoming ﬁxed in relation to the decidua capsularis with the consequent formation of the placenta there, would the. insufﬁciency of blood Supply then lead to death of the ovum and early abortion? Or would a sufficient blood supply be determined in these circumstances to tide the ovum over until it had gained an attachment to the decidua vera and opposite uterine wall where it might develope ? Further, if such a state of affairs would cause the ovum to perish would it not be possiblefor a certain proportion of early abortions of unknown cause to be due to this? To these questions there is at present no answer, and future work can alone solve the riddle.
Fate of the Decidua
The progressive and degenerative changes already described in the decidua occur mainly in the basalis. Fibrinoid formation, for instance, does not occur in the decidua vera, ‘and it would appear from this that ﬁbrin is‘ deposited only where foetal and maternal elements meet. In general terms it may he said that after the third month, as pregnancy advances, the decidua degenerates until at term it constitutes only a thinand irregular membrane easily cast off.
The decidua vera is of the same general structure as previously described showing a superﬁcial compact and ,a deep spongy layer. After the fourth month the decidual cells of the compact layer lose their polygonal or rounded appearance, "become smaller and more spindle-shaped and arr'anged'in'a direction parallel with the surface of the uterus. ' In the spongy layer the epithelial papillae disappear and the gland cavities now lose their irregular and open appearance and become low and broader in shape. This, ‘as well as the ﬂattening of the cells of the compact layer, is due to the progressive enlargement and stretching of the uterine wall. The epithelial cells lining the glands, at ﬁrst columnar, become cubical and later ﬂattened like endothelium, and these ﬁnally are seen to be absent in places. The glands thus come to resemble small elongated clefts like empty venous or lymphatic spaces. These changes cease in the deepest layers where muscle ﬁbres are seen; here the glands retain a more or less normal appearance and form the elements of regeneration of the entire mucosa.
The surface epithelium soon becomes ﬂattened with the expansion of the uterine cavity, and the cilia are lost; according to Marchand fatty degeneration occurs, but this is denied by Grosser, who says “ the fatty degeneration "of the decidua parietalis, which was formerly regarded as the rule, occurs at most only in exceptional cir'cumstances."~ By the end of the third month,—i.~'e., at the time of the obliteration of the decidual space by the fusion of the decidua capsularis and vera—the epithelium has degenerated and disappeared.
Decidua basalis. This has not the compact and spongy layers so well marked as the decidua vera. The impregnated ovum penetrates to the depths of the compact layer of the decidua or to the junction of that and the spongy layer; so that the only part, if any, of the compact layer in the decidua basalis is a very thin layer of tissue deep to the level of penetration. This point appears to be forgotten in a number of the published descriptions of the decidua basalis.
What remains of the compact layer forms the decidua] or basal plate which closes the intervillous space towards the uterine sid.As in the case of the vera the decidua! cells become more fusiform towards the end of pregnancy. Here are seen the ﬁbrin strize already described, also the remains of the trophoblast which originally formed the basal ectoderm and the cell columns of the anchoring villi. The trophoblastic remains may be either syncytial masses or large mononuclear cells similar to those found in connexion with ﬁbrin formation. The point is that, especially in the later stages, these cells may be found beyond the Nitabuch ﬁbrin striae (see discussion re border zone). The terminations of the anchoring villi have, later on, lost their epithelial—trophoblastic covering, and their tips in many cases end by dipping into ﬁbrinoid masses while degeneration in varying degrees is seen in the stroma.
The decidual pillars are elevations of the compact layer of the decidua basalis which have escaped the eroding action of the early trophoblast; they extend to a varying degree into the intervillous spaces and towards the chorionic membrane. In some sections they appear to be separate from the basal plate, and on this account Leopold applied the name “decidual islands” to them; but in serial sections they can be traced directly to the basal plate.
These pillars, or placental septa, in later months divide the placenta into cotyledons, an appearance not found in early placentze. The septa extend only partly towards the chorionic membrane. In the mature placenta the decidual cells in these septa have almost always degenerated and sometimes have disappeared entirely, so that only an open meshwork is seen in which are patches of ﬁbrinoid material; into these the anchoring villi penetrate so that the appearance presented may be very confusing. It is on this account that much discussion has arisen as to the origin of these pillars; Whitridge Williams, for instance, says that the decidua] conception of their origin is erroneous: “. . . . most of them represent masses of trophoblast into which the chorionic connective tissue has not grown, and which therefore have not developed into typical villi.” While the matter cannot be con sidered as settled the present.inclination is to regard them as decidual.
The spongy layer is of similar appearance to that of the decidua vera, but is not so thick in this situation, and gland spaces are not so prominent on account of the degeneration of the walls of some of them which have been ﬁlled with blood in the early stages of development; between the gland spaces are seen occasional detached fragments of syncytium—syncytial giant ce1ls—which have been detached and have penetrated deeply.
The decidua capsularis early shows degenerative changes. The aperture of penetration becomes closed either by inward growth of the margins or by organization of the closing coagulum after the protruding portion is cast off.
In F rassi’s ovum the decidua capsularis is a smooth even layer and, as would be expected, glands are present at the margins but not towards the pole. Streaks of ﬁbrin have started to appear and are most evident at the summit. But no trace is found of the implantation opening in the later stages of development of the placenta; in this area a tissue rich in ﬁbrin and poor in cells is found which is called Reichert’s scar; this denotes either the complete organic closure of the capsularis or, more probably, the ﬁrst stages of degeneration. As pregnancy advances the decidua capsularis becomes more and more stretched, exsanguinated and correspondingly degenerated, and by the end of the third month its remains fuse with the decidua vera covering the remainder of the uterine cavity. Some hold that traces of the decidua capsularis can be found up to term, while others maintain that it entirely disappears and that the atrophied remains of the villi of the chorion laeve, which have come into apposition with the decidua vera, have been erroneously regarded as decidua capsularis. Grosser shows extensive atrophic changes in the decidua capsularis over the region of the internal os where there can be no contact with the decidua vera and argues from this that complete atrophy occurs.
It is a point almost impossible of proof either way as the decidua capsularis consists entirely of a compact layer which will become apposed to the superﬁcial surface of the compact layer of the decidua vera; further both these layers are degenerated, stretched and atrophied in varying degrees and with the surface and glandular epithelium lost their junction would be impossible to trace.
Decidual vascularization is effected by the ultimate branches of the uterine arteries. While these penetrate and feed both the decidua vera and basalis the supply, and therefore the number of vessels and the degree of their dilatation, is much greater in the latter than in the former.
It has previously been pointed out that in decidual formation in general the arterioles become enlarged, dilated and convoluted. In passing through the spongy layers of the decidua basalis these characters are seen, but when the vessels reach the basal plate the muscular coat becomes lost and they then become merely endothelial-lined dilated blood channels as was ﬁrst pointed out by Webster; according to some observers even the endothelium is lost when the ﬁbrin layers are reached, so that it is an easy matter for these dilated vascular channels to open into the intervillous spaces.
The supplying vessels are said to.open mostly in the region of the placental septa, while the draining vessels take origin towards the middle of the cotyledon. The mouths of these supplying and draining vessels are easily seen in sections and are ﬁgured by all authors; they are funnel-shaped and open obliquely on the surface, and in some cases are blocked by syncytial masses. Veit first demonstrated this and showed how such chorionic elements were continually entering the circulation and could be demonstrated in the blood strata at distant parts. He based his theory of toxzemia of pregnancy on this fact, and, later still, Abderhalden founded his blood test for pregnancy" on it.
The chorio decidual vessels, first described by Ruge, are not to be confused with the vessels just described. They are relatively large and may be visible macroscopically. They pass out from the anchoring villi where the epithelium is lost at the apex andeithez’ terminate in the basal plate or enter the stem of a villus ascending from the basal plate. The epithelium and some of the stroma of the villi have degenerated, but the vessels have persisted and have become engorged and dilated.
Ruge believed that the presence" of these vessels indicated a vascularization of the decidua by foetal vessels. It is to be emphasited, hmtve*ve*r', that this does not indicate any anastomosis between the fwtal and maternal circulutxion; such. does not, ‘in. any circumstances, exist.
The Microscopic Structure of the Amnion
The microscopic structure of the amnion has been described by several observers, especially Webster. At the fourth week of gestation the lining ectoderm cells are ﬂattened and ‘almost like an endothelium; the lines of division between the cells are usually indistinct, and here and there absent, so that a syncytium‘-like formation is produced. The nuclei are round or oval and may be in more than one row. The mesoblast varies‘ in thickness; its outer layer is called the mesothelium and resembles the epiblast in appearance; the inner layer next the amniotic epiblast is ﬁbrilliatcd and contains but few cells, and is thought to be formed from the mesothelium.
Fig. 1. High power photomicrograph of Johnstone’s ovum (30 days) to show 9. branching villus covered with two layers of trophoblastic epithelium, and with a ﬁbrooellular core of embryonic mesoderm which has retracted slightly during preparation. (By kind permission of Dr. R. W. Johnstone.)
Fig. 2. High power photomicrograph (untouched) of a. villus from a. four weeks ovum. The syncytial covering‘ the underlying layer of Laughing’ cells and the mesodermal core are all seen. In the lower part is seen a section through a eyncytial bud—a. syncytial giant cell. Fig. 8. Striated border of the syncytium. On, the lower edge several adherent blood corpuscles may be seen. (This illustration is kindly lent by Dr. R. W. Johnstone.)
Fig. 4. Low power photmnicrogreph oi the villi of a six week: ovum. The double layer of covering trophoblaet, the cellular core of mesoblaet. which now contains blood vessels, and some eyncytial giant. cells are shown. Fig. 5. Photomicrograph of a. full term placenta. The trophoblast is now reduced to a. single layer of eyncytium : well-formed blood vessels are present in a stem villi: while some maternal blood corpuscles are present in the intervillous space.
Fig. 6. The decidua. vera. compacts at the eighth week. The decidual cells are polygonal and closely packed, while the cell outlines and villi are vague. ‘zw +4.9-yum -‘r. c
Fig. 7. The decidua. spongiosa in a two months ovum. The glandular spaces are greatly dilated and are starting to be drawn out. Papilliform projections into the gland lumina are Seen
Fig. 8. Photomicrograph of a white “ infarction” of the mature placenta. The villi are seen as lighter areas in varying stages of degeneration in the darker mass of fibrinoid material which glues them together. In the lower part of the ﬁeld several unaltered villi are seen.
As pregnancy advances the epithelial cells increase in number and may form more than one layer; they become of cubical or even, in some places, columnar appearance in the latter half of pregnancy, and as gestation proceeds the cells become mainly columnar with the nuclei distant from the basement membrane. The lines of separation between the adjacent cells remain dim. The edges of the cells are irregular and the adjacent projections blend so as to form bridges between the oells, thus producing a “ prickle ” appearance.
Stomata between the cells have been described, especially by Hiiter and Winckler, but Lange believes that such “ stomata ” are merely lacunae formed by the breaking down of degenerating cells, and Bondi could not ﬁnd any stomata.
Apart from such stornata it is quite possible that minute openings may exist between adjacent cells for the passage of ﬂuids. Lonnberg, in 1901, found fat granules of frequent occurrence in the epithelial cells; granules appear in the cells after the third month, and Bondi found such granules at term. They stain with neutral red, but he could not assign a function to them.
Now and again, especially at the placental margin, the epithelium becomes many-layered in places with corniﬁcation and desquamation of the superﬁcial layers forming the amniotic villi or caruncles described by Ahlfeld.
The mesothelium degenerates, this process starting in the second month and being well advanced by the sixth. The loose tissue formed by the mesothelium and lying between it and the epiblast increases in amount, but contains few cells, and in the later months of pregnancy it is of a mucoid appearance. The connection between this mesoblast and the chorionic mesoblast is by ﬁne threads only. A
There are no blood vessels in the amnion, and this may explain the lack of organic union between that membrane and the chorion.
The Fixation of the Placenta
The ﬁxation of the placenta is by the attachment of the anchoring villi. When the ovum is tiny the tendency to separation is slight on account of its light weight, but it has been shown that up to the end of the second month of gestation this attachment is not ﬁrm; after that, however, the villi penetrate further and as the covering epithelium is lost in many, the stroma of the villi comes into contact with the decidua so that the junction is more ﬁrm. Again by the end of the third month the expanding ovum reaches to the opposite uterine wall and there gains an attachment by the fusion of the decidua capsularis and vera while the chorionic villi atrophy except in the part apposed to the decidua basalis. Thus a fresh attachment being formed for the ovum a certain amount of strain is taken off the placental attachment and so the likelihood of separation is further diminished. Further, amniotic ﬂuid becomes secreted from the earliest weeks and increases rapidly in amount, so that the placenta is ﬁnally kept in position not only by the grip of its own tentacles but also by the intrauterine ﬂuid pressure, which is augmented by the painless uterine contractions of pregnancy, pressing it against the uterine wall.
During the second stage of labour the uterine contractions and retraction may cause some disproportion in size between the placenta and its uterine site, but this disproportion is small in amount as the uterus is still occupied by the foetus and the major part of the liquor amnii; in the third stage, however, these factors no longer exist to assist in maintaining the placenta in position, the uterus is free to retract strongly, the disproportion between the placenta and its site becomes greatly increased, and as the placenta and membranes cannot adapt themselves to this reduction they are separated and then expelled.
Frankl has recently stated his opinion that the placenta is more adaptable to variations in the area of its attachment than is generally supposed. His view is that the separating agent is haemorrhage from the engorged decidual vessels and that uterine contractions act as detrusors only after the placenta has been thus separated. The placenta is usually attached to the decidua, but in some few instances the condition of placenta accreta or increta is found, in which partial or complete atrophy of the uterine mucosa, due to various pathological states, is present and in which, owing to defective development of the decidua the villi penetrate to the uterine muscle and cause damage or partial destruction. Dietrich, in 1922, collected 19 recorded cases of this condition, and advised hysterectomy, as normal or complete manual separation is impos sible owing to the penetrative attachment.
The stinmlativc agent to ch-orionic attachment is believed to exist in the corpus luteum of the ovary. This view was originally put forward by Pnenant in 1898, and later by Ludwig Fraenkel in 1901 who concluded that the corpus luteum is a ductless gland which in the human female is renewed every month; its function is to control the nutrition of the uterus from puberty until the menopause, to prepare the endometrium for the embedding and maintenance of the ovum, and to maintain the raised nutrition of the uterus during early gestation. Fraenkel supported his theory by experimental double oophorectomy on rabbits, performed from one to six days after coitus, the period of gestation being 30 days. Abortion followed in all cases but not in the controls, in which only one ovary was removed. Similar experiments on dogs and rats were reported by Marshall and Jolly in 1905 with proper controls, and similar results were obtained, but in this series it was noticed that abortion occurred only when both ovaries were removed early in pregnancy. “ In other cases (in the rat) in which the ovaries were removed at periods varying from the sixth day until near the end of pregnancy, the young were produced normally at full time.”
Blair Bell and Hick, in 1909, performed a similar series of experiments on rabbits and does with controls, and obtained results similar to those quoted, but Bell is of opinion that a comparison can hardly be made in this respect between rabbits and the human subject. On the other hand a number of cases have from time [0 time been reported of double otiphorectomy performed early in pregnancy in the human subject with no interruption of gestation, and while in some of these, no doubt, a portion of ovarian tissue had been left this cannot be claimed for all.
Lochhead says : “ It must be pointed out, however, that there is no evidence that the corpus luteum governs the ﬁxation of the embryo in any other than the indirect sense implied in the supposition that the secretion elaborated by that organ acts as a stimulus which excites the uterine mucosa to undergo the necessary hypertrophy. In this general sense’, also, it is probably true that the luteal secretion (or, at any rate, the secretion of the ovary) assists in nourishing the embryo during the ﬁrst stages of pregnancy, since there is every reason for concluding that it helps to maintain the raised nutrition of the uterus.” The corpus luteum therefore is not necessary as an aid to ﬁxation in the later months. Physiologically the corpus luteum undergoes atrophy at this period, and it has been suggested that the atrophic changes which the decidua basalis and vera undergo in later pregnancy are due to the withdrawal of this secretion. Loeb showed tl‘at decidua] formation was due, at least partly, to mechanical stimulation of the mucosa sensitized by the corpus luteum. By introducing foreign bodies such as glass rods into the uterine cavity localized overgrowths of decidua—deciduomata—were produced.
The present opinion, therefore, is to regard the blood-borne hormone of the corpus luteum as a sensitizing agent to the uterine mucosa to facilitate the embedding of the ovum and to produce the necessary hyperaemia to allow uterine hypertrophy and growth of the ovum. 638 Journal of Obstetrics and Gynmcology
The Circular Sinus
The circular sinus is a structure prominently described in most text books. It is a blood sinus incompletely encircling the periphery of the placenta. Well marked in some specimens, in others no trace of it is to be found, and in the majority of cases it is but indistinct. It is situated in the angle between the placental margin and the chorion laeve. It is not a formed vessel but an irregular space of varying diameter formed in the marginal dilated portion of the intervillous spaces, the inner and upper walls being formed by the ﬁbrinous cohesion of adjacent clumps of villi, while the ﬂoor is formed by the basal plate of the outermost portion of the decidua basalis. From the outer wall of the sinus veins take origin, the ruptured mouths of which can usually be seen, while through the openings the villi on the inner wall are visible.
Some observers, however, e.g., Friolet, have described an endothelial lining to the sinus, and have deduced from this that the above description is incorrect and that the sinus is a trulyformed blood vessel, but Grosser holds that such endoth-elial-like cells are really trophoblastic epithelium stretched out and correspondingly altered in appearance.
From the above description it will be seen that the main structural changes which occur in the placenta in its later period are degenerative. The villi, at ﬁrst, large and covered with a complete double layer of surface epithelium with an open myxomato.us stroma containing blood vessels, which are thin-walled even in the villous stems, become progressively smaller in section due to continual branching; the layer of Langhans’ cells disappears, and later the syncytium also becomes incomplete so that areas are left here and there on the surface of the villi, denuded of epithelial covering, on which ﬁbrin is deposited in increasing ﬂakes from the coagulating blood in the intervillous spaces. “ Infarcts,” visible to the naked eye, or microscopic masses of ﬁbrin gluing adjacent masses of villi together, are thus produced.
As the stem villi increase in size and the arborizations multiply an increasing blood supply is called for, so that in the stem villi the arteries become of an adult appearance with crenated tunica intima, thick media and a whorled externa. Some observers have described endarteritis which, if present, would further add to the appearance of degeneracy.
The decidua is at first a thick vascular membrane in which well formed decidual cells are plentiful; from the third month degenerative changes appear. Fibrin stria: are deposited here and there, the tips of the anchoring villi invade and erode it, and, finally, at term the membrane is greatly atrophied and ﬁbrotic and forms only a thin and imperfect layer on the uterine surface of the placenta.
In post-maturity one would expect to see these degenerative changes more evident even than usual, and this expectation is fulﬁlled according to Ballantyne and Browne. These authors describe the changes of post-maturity in the placenta as consisting of non-vascularity, many of the small villi having no vessels and all of them possessing fewer than is normal at full time. More or less obliteration of the Iumina of the vessels by endarteritis was constant. Here and there areas of calciﬁcation were present. There was no enlargement of the villi and so the organ as a whole was not enlarged. “ The changes were not pathognomonic, but revealed in a higher degree what is normally found in the mature placenta; it was a matter of degree.”
The amount of blood in the intervillous spaces varies; in a placenta ﬁxed in situ the spaces are seen to be full of red blood corpuscles, but in an expelled placenta they are relatively few in number. it is supposed that the uterine contractions during labour expel most of the intervillous blood into the uterine veins before the placenta is shed; also, as Young points out, intervillous blood does not coagulate, so that, if the placenta be placed in ﬁxing ﬂuid for some days before sections are cut, the greater part of the blood in the intervillous spaces becomes washed out. Fibrin deposits, of course, are retained in the placenta.
It is important to remember these points in the microscopical examination of a section of the placenta lest these physiological senile changes may be interpreted as evidence of placental disease; this probability is heightened by the fact that owing to its spongy structure the placenta is by no means an easy organ to cut and stain satisfactorily. Again, our knowledge of placental pathology is, at present, so imperfect that even the best histologists are at times divided in opinion as to the interpretation of the appearances seen on examination of placental sections.
The presence of fat and glycogen in the villous epithelium has not been described; it is proposed to reserve this for discussion when the physiology of the organ is considered.
The Expelled Placenta
Plane of separation. It has been noted that the deeper layers of the decidua are of an almost cavernous structure, and this is generally regarded as a preparation for facilitating the separation of the ovum through this layer at birth. This was ﬁrst pointed out by Langhans and supported by Barbour, but Webster, going on the fact that the amount of decidua present on the surface of the placenta is small, concluded that the plane of separation is mostly through the deeper parts of the compact layer or through the junction between the spongy and compact layers; but he agreed with Priestley and Leopold that in manual separation the line of cleavage would be through the spongy layer.
The expelled placenta consists from within outwards of an inner layer of amnion lying on the chorionic membrane from which project the uncountable host of villi now in various stages of degeneration, with fibrinoid deposits lying between them; between the villi is maternal blood, and attached to the outer or uterine surface is a layer, usually thin but of varying thickness, of the compact and spongy layer of the decidua basalis.
The appearance of the expelled placenta is well known, and it is not proposed to enter into a minute description which can be found in any good obstetrical text book.
Certain points, however, require discussion. As regards the weight, this is usually between 1 lb. and 1} lb. (453—566 grms.), but it varies within limits, and normal placentae up to 6oogrms. are often found. Of more importance than the weight alone is the ratio of the weight of the placenta to that of the foetus — the weight ratio - which is usually quoted as 1 to 6, although Holland found the weight ratio in 94 fresh and 64 macerated foetuses to be 1 to 8. Sfameni (quoted by Luciani) estimated the normal weight of the placenta as 408 grms. without the cord and membranes, and the usual weight ratio under these conditions is I to 7.78 and not 1 to 6 as usually quoted. This is in agreement with Holland’s ﬁndings as he always separated the membranes before weighing the placenta.
The weight ratio varies normally with the age of gestation, the younger the foetus the larger relatively the placenta, and hence the weight ratio will be reduced in premature cases.
Dividing foetuses into three classes (a) those under 1,800 grms., (b) those between 1,800 and 3,000 grms., and (c) those over 3,000 grms., Holland fou-nd in fresh foetuses that the weight ratio was 1 to 6, 1 to 8 and 8 respectively, while in macerated foetuses the weight ratios were respectively 6, 8 and lo, and in 34 syphilitic placenta: the weight ratio was 5 and 7 in the ﬁrst two classes, while no cases came into the third class. This subject will be returned to when discussing pathological conditions of the placenta in a future review.
The 1:lacental-umbil-ical-‘vessels radiate out from the insertion of the cord in convoluted branches which subdivide, and their ultimate terminations are seen extending to the extremities of the villi. It is important to remember that nerve endings cannot be found in the walls (Sf these vessels, therefore their calibre must be controlled by some agency other than innervation. Schmitt, in l922, published his experiments on this subject and found that the placental vessels dilated in response to amyl nitrite and warmth and contracted to histamine, pituitary extract and barium chloride, but showed little reaction to adrenalin even in 51 to 1000 concentration. The vessel tone, however, was increased in response to oxygen, and from these experiments it is suggested that the calibre of the placental vessels is regulated by the amount of oxygenation of the foetal blood, that when oxygen is plentiful the vessels contract and in conditions of anoxzemia they dilate. The mechanism of calibre control of the placental vessels is thus unique in the body.
This resume will indicate to the reader the large number of problems connected with the growth and structure of the placenta that still await solution, and it is in the hope that the necessary research will be stimulated that the review has been compiled.
1. Ahlfeld. Arch. '1‘. Gyn(ilcol., Bd. xiii, s. 241.
2. Ballantyne, J. W., and F. J. Browne. Jaum. Obatet. and Gymzrol. Brit. E'mp., Summer 1922.
3. Bell, W. Blair. “The Sex Complex.” London, 2nd edit., 1920.
4. Bell, W. Blair, and Hick. Brit. Med. Jaurn., 1909, i, 655.
5. Bondi. Ze-ntralb. f. 6lyna'Icol., 1905.
6. Bonnett. Anat. Arm, xiii, 1897.
7. Bonnett. Manataechr. f. Geburtsh. u. G'ymi]col., xviii, 1903.
8. Bremer, J. L. Amer. Jaum. Anat., 1914.
9. Bryce, T. H., in “ Quain’e Anatomy,” vol. i, Embryology. London, 1908.
10. Bryce, T. H., and J. H. Teacher. “ The Early Embedding and Development of the Human Ovum.” Glasgow, 1908.
11. Clemenz. Zeitschr. f. Gcbiirtuli. u. Glynéc.-01., 1922, 758.
12. Creighton. Jmu--n. Anat. amt Phy.s:'o'l., xii, 1878.
13. D’E'1-chia. Morzacsschr. f. Geburtah. u. (lg/ndkol., 1921, iv, 65.
14. Dietrich. Zeitechr. f. Geburtshr. u. Gy-ndkoI., 1922, 579.
15. Eden, T. W. Journ. of Pafhol. and Bar-teriol. London, 1897.
16. Ercolani. Mam. dcllvlcad. di Bologna, 1876.
17. Frankl, 0. 31179., Gyna-col. and 0batct.., xxxii, 450.
18. Fraenkel, L. Arch. f. G'_i/mlI:ol., 1902, lxviii, 438.
19. Frank, R. T. “Gynecological and Obstetrical Pathology.” Appleton, 1922.
20. Frassi. Arch. /. Milcroslc. Anat., 1907 and 1908.
21. Frommel. Zeitnchr. f. Gcburtsh. u. Gyndlwl, xxxvi.
22. Grosser, 0., Keibel and Mall. “Manualpf Human Embryology,” vol. i. London, Lippincot, 1910. Various references.
23. Hennig. Quoted by Lochhead.
24. Hitschmann and Adler. Mori_atssch~r, f. Gelmrtsli. u. Gy~ndIcol., 1908.
25. Hitechmann and Lindeuthal. Ze~n!.ralbl. f. GyndIcol., 1902 and 1903.
26. Holland, Eardley. “The Causation of Fetal Death.” H.M. Stationery Office, London, 1922.
27. Hubrecht. “ Plecent-ation of the Shrew.” Quart. Jvmrn. of Ms‘:-r. Sr-ii-m-M, 1894. 642
28. Jenkinson. “ Observations on the Physiology and Histology of the Placenta. in the Mouse." Tijd. N edcrl. I)ie.rk., v. 11.
29. Johnstone, R. W. “ Contributions to the Study of the Early Human Ovum."
30. 31. 32. 33. 34. 35. 36. 37.
38. 39. 40. 41. 42. 43. 44. 4 5.
46. 47. 48. 49. 50. . Spee, Gmf v. Arch. f. Anat. 1:. Phys., Anat. Abt.., 1883.
Juum. Obstet. and Gyntrcol, Brit. l9'mp., 1914, 231.
Langhans. Arch. f. Anat. u. I"7L'_I/8., Anat. Abh, 1877.
Langhana. Beitr. z. 6'eImtrsln'. u. G'ym§lcol., v, 1901.
Lochhead, J ., in Marshall's “Physiology of Reproduction,” 2nd edit., London, 1922.
Loeb, L. Journ. A-mm‘. Mod. Assam-., 1908 and 1909.
Loeb, L. Su~rg., 0;:/ruecal. and 0bstet., xxv, 1917.
Luciani. “ Human Physiology,” vol. v, London, 1921.
Marchand. Ara.-h. f. Gy/m2kal., lxxii, 1904.
Marshall and Jolly. “ The Ovary as an 0.1-gen of Internal Secretion.” Phil.
T1-an8., B, 1905. Nitabuch. Dissertation, Beme, 1887.
Overlach. Arch. f. Miler. Amct., xxxv, 1885.
Peters. “ Uber die Einbettuug des Menschlichen Eies.” Leipzig and Wien, 1899.
Pfannensteil. Hamlb. (ler Geburtslar, Wiesbaden, 1903, i.
Prenant. Rev. Gén. des Sciences, 1898.
Roberts, E. Emrys. Jam-n. Aunt. and l’71»;:/3-iol., 1910, xliv, 192.
Robinson, Arthur. Hunterian Lectures. Journ. Anat. and PM/aiol., xxxviii, 1904.
Robinson, Arthur. Cunningharrfs Anatomy (Section of Embryology), London. 1922. l ' Bohr. Virchow’s Archives, 1889.
Ruge. Zeitsclw. /. Geburtsh. u. 0'1:/1161901., 1898, xxxix, 550.
Sabotta. Arch. Miler. Anat., 1903. .
Schmitt, W. Zentmlbl. f-. G’;:/n(ikol., 1922.
Selenka. Biol, Zent1albl., 1898.
Strahl. E’~rgeb. J. Anat. u. Entw-ickl., 1899,
Talbot, J. E. Surg., G1/nwcdl. and Obstet., 1921, xxxii, 552.
. Turner, Sir William.‘ “Lectures on the Comparative Anatomy of the Placenta."
Edinburgh, 1876. Van Benedin. Anat. Anzeiger, 1876.
Veit. “Die Vex-schleppung der Chorionzotten.” Wiesbaden, ‘1905.
Webster, J. Clarence. “ Human Placentation.” Chicago, 1901.
Williams, Whitridge. Amer. Journ, of 0bate¢., 1909.
. Williams, Whitridge. “Obstetrics,” 4th Eclit., Appleton, 1920.
Young, J. “Reproduction in the Human Female." Edinburgh, 1911.
Young J. Proc. Roy. Soc. Med. (Sect. Obstet. and Gynaacot), 1914, iv. 291
Cite this page: Hill, M.A. (2021, April 20) Embryology Paper - The development and structure of the human placenta. Retrieved from https://embryology.med.unsw.edu.au/embryology/index.php/Paper_-_The_development_and_structure_of_the_human_placenta
- © Dr Mark Hill 2021, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G