1. An Early Ovum imbedded in the Decidua (1908)

From Embryology
Embryology - 25 Apr 2019    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Bryce TH. and Teacher JH. Contributions To The Study Of The Early Development And Imbedding Of The Human Ovum 1. An Early Ovum Imbedded In The Decidua. (1908) James Maclehose and Sons. Glasgow.

Online Editor  
Mark Hill.jpg
Note this paper was published in 1908 and our understanding of early development has improved since this historic human study.

See also the later paper by one of the same authors: Teacher JH. On the implantation of the human ovum and the early development of the trophoblast. (1925) J Obst. Gynaecol. 31(2); 166-217.
Modern Notes: Week 2 | Week 3

Stage 6 Links: Week 2 | Implantation | Lecture | Practical | Carnegie Embryos | Category:Carnegie Stage 6 | Next Stage 7
  Historic Papers: 1909 | 1925 | 1937
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)
Title page

1. An Early Ovum Imbedded In The Decidua

By Thomas H. Bryce, M.A., M.D.

Lecturer In Anatomy, University Of Glasgow

And John H. Teacher, M.A, M.D.

Lecturer On Pathological Histology. University Of Glasgow

2. An Early Ovarian Pregnancy

By Thomas H. Bryce; John H. Teacher And John M. Munro Kerr, M.D.

Obstetric Physician To The Maternity Hospital And Gynaecologist To The Western Infirmary, Glasgow

With Ten Plates And Twelve Figures In The Text

Glasgow James Maclehose And Sons

Publishers To The University



The following memoir combines in one publication two separate papers which deal with the processes involved in the imbedding of the human ovum. They have been associated because of the complementary nature of the evidence they afford regarding the histological characters and the activities of the trophoblast. Each breaks new ground, in respect that while the first paper embodies a description of the earliest phase of the human ovum yet recorded, the second deals with the earliest case of ovarian pregnancy hitherto reported.

The fact that the extremely early ovum described in the first paper is a unique specimen presenting features which have not up to the present been observed, necessitated profuse illustration both by coloured plates and photographic figures. This has rendered the production a costly one, and we desire to express our obligations to the Carnegie Trust of the Scottish Universities for giving us a grant towards the expenses of publication.

University of Glasgow,

July 15th 1908.



  1. Paper No I. An Early Ovum imbedded in the Decidua
  2. History of the Specimen
  3. Fixation of the Specimen
  4. Dimensions of the Ovum
  5. General Descriptions of the Sections
    1. The Decidua
    2. The Trophoblast
      1. The Plasmodi-trophoblast
      2. The Cyto-trophoblast
      3. Layer of Large Free Cells
    3. The Contents of the Blastocyst
      1. The Mesoblast
      2. The Embryonic Rudiment
  6. Discussion of Data
    1. Summary of Characters of the Ovum
    2. Comparison with the Ova of Leopold and Peters
    3. The Embryonic Rudiment
    4. The Process of Imbedding
    5. Comparative Data regarding Inbedding
    6. Description of Imbedding of Human Ovum
    7. Analogy with Imbedding of an Embolus of Chorion-epithelioma
    8. Function of the Plasmodium
    9. Fate of Early Plasmodium and Attachment of Ovum
  7. The Age of the Ovum and Relation of Imbedding to Menstruation
    1. Estimation of Age
    2. Comparative Data
    3. Comparison with other Cases
    4. 1._An_Early_Ovum_imbedded_in_the_Decidua_(1908)#Summary_of_data_regarding_Selected_OvaSummary of Data regarding Selected Ova
    5. Table of Selected Ova
    6. Table showing Periods of Fertilization and Imbedding
    7. Relation of the Menstrual Cycle to Oestrus Cycle
    8. Meaning of Menstruation
    9. Table of Menstrual Cycle
    10. Ovulation and Menstruation
    11. Nature of the Decidua


Very early stages of the human ovum are necessarily extremely rare. It is only by fortunate and fortuitous circumstances that an occasional specimen comes into the hands of the investigator. Within recent years a number of young ova have been described, which have considerably extended our knowledge, and have served to show that in certain respects the early stages of development in man differ materially from those in lower mammals. The ovum of Hubert Peters, of which an account was published in 1899, still represents the youngest phase known. A specimen described by Leopold in 1906 is certainly earlier than that of Peters, but no embryonic rudiment was present, and in several other respects it must be considered abnormal. On the other hand, the ova described by Graf v. Spee (1905), Beneke, and Jung amply confirm, though they do not extend, the data provided by Peters' specimen.

Considerable light has been thrown on the problems involved in early human development by recent comparative work, more especially that of Selenka and Keibel on monkeys and apes, and of Hubrecht on Tarsius spectrum. It is now known that the Primates, including Tarsius in that category, form embryologically a group by themselves. All have certain common and peculiar features. There is always present a mesodermic connecting-stalk (Ha/tstiel), through which the vessels of the embryo and chorion are connected without the medium of an allantois ; the yolk sac is very minute and is not coextensive with the blastocyst ; there is a precocious extra-embryonic coelom lined by middle-layer cells, which are present at a very early period before the appearance of the primitive streak or embryonic axis, and therefore before the formation of the dorsal mesoderm of the embryonic body.

There are other features, however, in which the several orders of the Primates differ inter se. In Tarsius the amnion is formed as in the rabbit, dog, etc., by secondary folds, while in monkeys, apes, and man it is already closed in the earliest stages known. The placentation again, in the monkeys (Old and New World), differs from that in the anthropoid apes and man. While the early phases in apes and monkeys, described by Selenka, confirm and explain the corresponding phases in the human subject, none of the stages known reach to the initial stages of the blastocyst, and therefore much is still left for conjecture. The extremely young ovum, which is the subject of the first of the papers in this memoir, represents the earliest stage of any primate form except Tarsius yet recorded, and merits careful and det<iiled description in respect that it pushes back the limits of the unknown in a sensible degree.

The age of young human ova is, of course, from the nature of the case, quite uncertain. It is usually calculated in terms of the conventional rule formulated by Professor His, but the results of the rule as applied to the youngest know^n specimens are unsatisfactory and contradictory. In the present case we are fortunate in possessing very accurate data, and an effort will be made by correlating the facts with those known for other specimens to revise the basis on which the age of early ova is calculated.

Not only do the structural features of the early primate blastocyst remain unknown, but the process of imbedding and the initial phases of placentation are also merely matters of surmise. All the ova described before the appearance of Hubert Peters' monograph were found completely imbedded in decidua, and the hypothesis that the ovum becomes surrounded by a process of circumvallation was generally accepted, though in more recent times the results yielded by comparative embryology had caused some doubt on the matter in the minds of a few observers.^ Several of

  • In William Hunter's Anatomy of the Oracid Uterus 1774, Plate 35, there is figured a complete decidual cast, in which an ovum about the size of a pea lies imbedded. In his diagrams William Hunter clearly indicated that the ovum is at this stage completely surrounded by the decidua, but he expresHed no opinion as to how it Injcomes implanted therein. The theory that the decidua covers the orifices of the Fallopian tubes, and is pushed out by the ovum as it entei-8 the uU.'rus, lias been erroneously attributed to him. (See Historical Introduction to the Catalogue of Um Anatomical ami Pathological Preparations of William Hunter. John H.

'*ie ova, such for instance as that described by Reichert, showed a small area of the decidua capsularis (formerly called reflexa) over the blastocyst, which was of a different nature from the rest of the capsule, and apparently composed of cicatricial tissue. They were, notwithstanding, completely enclosed by organised tissue. In Peters' ovum, however, and also in one described by Graf v. Spec (1905) there is a relatively large area from which decidua is absent, and its place is occupied by a mass of fibrin and blood-clot (the *' Gewebspilz "). The aperture in the wall of the implantation cavity occupied by this mass was considered by Peters, and also by Graf v. Spec, as the point of entrance of the ovum into the substance of the mucosa, but their preparations do not by themselves conclusively demonstrate the actual process by which the ovum is implanted. To prove how this is effected still earlier stages are required. Our young ovum is a further step in the direction of assured knowledge, and as well be seen later necessitates some modification in the interpretation of the "Gewebspilz" completing the capsule in Peters' specimen, while our ovarian ovum, which is the youngest hitherto described implanted in the ovary, throws considerable light on the nature of the imbe^lding process.

In the absence of the early stages in the human subject it is necessary to make use, for the purposes of interpretation, of the data provided by Comparative Embryology, but the remarkable variability in the methods of implantation and in the details of placentation in the diflFerent mammalian orders, speaks for a certain specific character of the embryological processes involved. Caution, therefore, is required in grafting any data derived from the investigation of the conditions in lower mammals on to the facts known for the human ovum, and the more so as the young ovum we have to describe accentuates the very special features of the human blastocyst in its early phases.

The only competent analogy with the higher primate ovum among the lower mammals is to be found among the forms in which there is likewise a decidua capsularis, for instance the hedgehog among the insectivora, and the mice, rats, and guinea-pig among the rodents. It is to be noted that in these forms, as in the Primates, the amnion is closed from the first, and that the blastoderm shows the phenomenon, to a greater or less degree, of "inversion of the germinal layers." Two methods of imbedding, which will be dealt with in greater detail later, have been described in animals with a decidua capsularis. In the hedgehog, mice, and rats, the ovum is said to be received into a recess or fissure of the mucous lining of the uterus ; the epithelium disappears round the blastocyst ; the mucous membrane becomes greatly thickened to form the decidua capsularis ; and the fissure is cut off from the general cavity of the uterus by the fusion of the lips of the decidual swellings from which the epithelium has likewise vanished. In the guinea-pig the observations of Graf V. Spce seem to prove that the ovum, while still in the early blastocyst stage, destroys the epithelium of the surface at the spot where it becomes implanted, by the activity of its ectodermic cells, and then, by a continuance of the process of destruction and solution, imbeds itself in the connective tissue of the mucosa. One or other of these alternatives must apply to the human ovum, and we submit our two communications as a contribution towards the solution of the problem.

In regard to the initial stages of placentation a very large body of data has been accumulated by comparative embryology, and our views as to the deciduate placenta have undergone considerable modification. Apart from the general character of the placentation in the different orders of mammals, debate has centred on the nature of certain layers of cells which separate the foetal from the maternal blood in the placenta. It is unnecessary here to enter on any detailed account of the various and contradictory opinions which have been held on this histological detail, or of the several theories which have been put forward on the subject of the origin of these layers.^

It is now very generally admitted that the evidence afforded by both human and lower mammalian material is in favour of the foetal, i.e. chorionic origin of both layers covering the villi in the human placenta. No doubt appears to exist in the mind of anyone as to the cellular layer, generally known as Langhans* layer, but there is still a lack of decisive proof regarding the plasmodial investment of the villi, or syncytium. Certain authors have maintained that it owes its origin to the maternal tissues — some deriving it from the epithelium either of the surface or of the glands of the decidua, others holding that it represents maternal endothelium spread over the villi in the interlocking of foetal and maternal tissues, which has long- been considered to take place in the development of the placenta.

  • The different hypotheses are fully set forth by Hubert Peters, by Webster {Human Placentation)^ and by Strahl {Hertwt'ffs Uandhuch der EntwiclcduiKjitleJin') ; they have also been dealt with by Teacher in his papers on "Chorion-epithelioma," and are briefly summarised by Bryce in Quain^s Anatomy^ vol. i. 11th ed. 1908.

The idea that uterine epithelium is necessary for the production of the Plasmodium was finally excluded by Catherine Van Tussenbra^k when she demonstrated the existence of a syncytial layer on the villi of a chorionic vesicle imbedded in the ovary, and her observation has been confirmed by several investigators. It is just conceivable, however, that if a fertilized ovum developed in the interior of the GratHan follicle, the follicular epithelium might be responsible for the production of the Plasmodium ; but this cannot be the case if it be proved that the blastocyst may be imbedded in the ovarian stroma outside the corpus luteum. In the second communication embodied in this publication further proof will be provided that such a case may occur.

The theory that the plasmodial layer on the villi is derived from maternal endothelium has become practically untenable in view of the characters of the early ova described in recent years, but it is not quite so certain that it may not owe its origin to the maternal connective tissue in which the ovum is imbedded, altered by the biochemical activities of the ectodermic cells of the blastocyst. In this connection earlier stages in the development of the plasmodium than have been hitherto available in the human subject, are required for the complete demonstration of its foetal origin in the human placenta. Our young uterine ovum, in virtue of its stage of development, and our case of ovarian pregnancy, being a crucial experiment both on the nature of the imbedding process and on the origin of the plasmodium, bring critical evidence to bear on this question.

History Of The Specimen

Figure 1. Portion of Decidual Membrane containing the ovum slightly magnified. The prominent lobule is the actual site of implantation. The indistinct dimple-like mark slightly below the centre of the lobule corresponds to the position of the blastocyst. The darker shading represents congestion, and the dark apot above the dimple is a haemorrhage.

The ovum was found by Dr. Teacher in a portion of membrane sent to him for examination by Dr. T. Doughis Brown, in a mixture of urine and l)lood-clot. The membrane had been expelled by a young woman who had been married for about two years, but who had not before been pregnant. Dr. Brown recognised a portion of fawn-coloured membrane among the blood-clots, which he considered to be probably a portion of the decidua of pregnancy, as the patient had passed the date on which menstruation was expected by about ten days. By the time the specimen reached Dr. Teacher the haemoglobin had diffused out of the clots, and the membrane was no longer distinguishable from them. In order to differentiate the tissue from the clot, about one-third of its volume of 90% alcohol was added to the fluid, and an hour later the mixture was decanted and fresh 30% alcohol substituted. Two hours later, on examining the material in a white porcelain dish, the membrane was readily identified by its somewhat lighter colour. It was roughly quadrilateral in shape, and somewhat broader at one end, measuring 3*8 cm. in length and 2 cm. in breadth. It represented the greater part, if not the whole of the mucous membrane of one wall, prol)ably the posterior wall, of the uterus. It presented the characters of decidua and was mapped out into areas by shallow furrows. Its outer surface had the characteristic appearances of the detached surface of shed decidua.

The margin was in parts rather thick and rounded, in other parts it faded off into very thin shreds. One of the areas near its centre stood up somewhat prominently from the rest of the membrane. The papilla so produced (Figure 1) at its right margin slightly overhung a deep furrow, but at the ends and on the left side it sloped gradually down to the general level of the surrounding membrane. This elevated area measured roughly 6 mm. by 3 mm. ; it was nearly 3 mm. in thickness, while the rent of the membrane did not exceed 2 mm., and was for the moat part thinner. About the centre of the papilla there was a circular mark like a very shallow dimple, and of lighter colour than the rest of the surface. The adjacent tissue was of deep red colour as if congested, and a still darker patch above the circular mark was regarded as probably due to haemorrliiige into the memlmme. The area thus described is very similar in appearance to the lobule of decidua in which the ovum investigated by Leopold in 1906 was enclosed.

Although it seemed improbable that after twenty hours' immersion in a mixture of blood aud urine the fixation of the preparation could be satisfactory, the membrane was placed in abaolntu alcoliol, this medium having before been found to be the most satisfactory fixative in similar circumstances. After thirty hours the lobule was excised, imbedded in paraffin, and cut by William Price, the laboratory attendant, into a perfect series of sections 7 microns thick, vertically to the surface of the decidua and in the supposed long axis of the uterus.

The probability of finding a valuable ovum being somewhat remote, thirteen sections, at intervals, were cut out of the ribbon, mounted and stained. In three of these Dr. Teacher recognised what seemed to be an ovum of about 1 mm. in diameter, and the remainder of the ribbon containing about 400 sections was then mounted and stained with haemalum and eosin. At a later stage the first series was restained by Weigert's fibrin stain in order to obtain a more accurate difierentiation of the Plasmodium from the decidua.

The following details relative to the history of the case have been supplied by the husband of the lady, himself a man of science and therefore alive to the importance of exact data. They can be implicitly relied upon.

The patient and her husband are both healthy. There has never been any uterine disorder. Menstruation has always been regular, the intervening period averaging twenty-six days with an occasional variation up to two days in either direction, the irregularity being however usually compensated ; if one interval were short the succeeding interval was usually longer by a corresponding amount, and vice versa.

The data may be summarised as follows :

  • September 2nd, 1907. Menstruation began.
  • September 27th, 1907. Menstruation began.
  • October 2nd to 3rd or 3rd to 4th. Coitus.
  • October 19th to 20th. Coitus.
  • October 25th. Menstruation was expected but did not appear. In the succeeding days the patient felt particularly well ; she had no symptoms suggestive of pregnancy except the absence of menstruation.
  • November 3rd to 4th. Coitus.
  • On the morning of the 4th of November a discharge appeared like the commencement of menstruation.
  • On November 5th the bleeding became more profuse than at an ordinary menstrual period, and the clots and membrane were passed towards evening with considerable pain.
  • The patient made an uneventful recovery, and menstruation reappeared twenty-six days after the fourth of November.

Fixation of the Specimen

Notwithstanding the circumstances under which the ovum was obtained the state of preservation of the tissues is wonderfully good. The mixture of blood serum and urine apparently behaved as a neutral fluid, which destroyed the tissues far less than they would have been destroyed by water or even by ordinary normal saline solution. Though we cannot claim that the fixation is perfect, we are satisfied that it is sufliciently reliable for all practical purposes. The nuclei are on the whole well fixed, but the protoplasm is less well preserved and has in parts a slightly macerated appearance. The colouring matter of the blood has been almost entirely dissolved ; the red corpuscles are, therefore, shadowy, but they have for the most part retained their normal shape. There is further a certain amount of granular debris in the blood spaces. The effects of precipitation and maceration are not, however, sufficient to invalidate conclusions regarding the essential facts of the histology of the ovum and of the decidua. The embryonic rudiment is somewhat torn, but it has been possible by reconstruction to make good the defect in this respect. It has generally been assumed that the only hope of getting a young normal ovum, is the chance of an operation or of a suicide. The present case shows that even in the case of an early abortion, an ovum may be cast off* intact in the mucous membrane, and recovered in quite a satisfactory state of preservation. There is no reason for supposing that the mucous membrane was diseased in this case, and it is probable that the abortion was due to mechanical causes, seeing that it ensued immediately after coitus.

Dimensions of the Ovum

The ovum is completely surrounded by decidua and lies in a blood filled space excavated in it. The dimensions of the cavity at their maximum are 1*95 mm. in its longest diameter (parallel or approximately parallel to the vertical axis of the fragment of decidua figured on page 10), '95 mm. in depth, and Vl mm. in its third dimension. The last measurement is arrived at by summing the number of sections 7/^ thick in which the implantation cavity appears. The space enclosing the ovum is thus oval in shape, with its long axis parallel to the surface and about twice the length of its vertical axis, or depth from the surface. The dimensions of the blastocyst itself are rather difficult to determine with exactness owing to its wall being a little folded at several points. In Plate I, Fig. 1 and Plate in, Fig. 3, which represent a section well to one side of the equator of the vesicle, the cavity measures about '36 mm., and it is approximately circuLir. In Plate ii, Fig. 2 and Plate iv, Fig. 4, which is a section cutting the equator of the blastocyst, the internal measurements are '77 by *63 mm. in its two dimensions. The shorter measurement may be taken to represent the maximum size of the cavity, seeing that the blastocyst is obviously compressed and folded at one side ; it is probable that the vesicle is spherical at this stage, because the contour of the unfolded wall is approximately the arc of a sphere.

General Descriptions of the Sections

Plate I, Fig. 1 shows the general relations of parts as revealed by a low power of the microscope. The section cuts the whole length of the decidual lobule described above. It is marlced off above by a fissure which indicates the upper limit of the tubercle seen in Figure 1 ; below it ends in a rounded and slightly overlapping projection which corresponds to the lower edge of the papilla. The free surface of the lobule shows above the mouth of a gland, and over the ovum an irregular depression which is the orifice of the space excavated in the decidua. This does not correspond to the dimple seen on the surface of the lobule ; that is represented by a shallow depression which appears on the surface immediately over the centre of the blastocyst, and is seen in Plate ii, Fig. 2. The blood-vessels of the decidua are greatly dilated, more especially on the deep aspect of the blastocyst, where they form the same "blood cushion" seen in the ova of Peters, Ercole Cova, and Graf v. Spec (1905). The glands are enlarged and irregular. Beneath the blastocyst there is great extravasation of blood and deposit of fibrin, while the upper part of the tubercle is seen to be occupied by a haemorrhage.

The implantation cavity is clearly marked off all round, but more especially at its outer part, by a darker staining band which, as will be seen presently, represents a layer of necrotic tissue. The cavity is occupied by very irregular strands of tissue, the spaces of which are densely packed with blood corpuscles. Above, the space is encroached on and the zone of necrotic tissue broken up by the haemorrhage already alluded to.

Under a higher power (100 d) the blastocyst wall (Plate iii, Fig. 3) is seen to be composed of a lightly staining lamella of protoplasm, in which the cell outlines are indistinctly defined, while the very irregular meshwork occupying the space in the decidua is observed to have plasmodial characters.

In describing the histological details of the several parts, we shall first take the decidua, then the trophoblast, and lastly the embryonic rudiment.

The Decidua

The surface epithelium has completely disappeared. Whether this be the result of maceration, or is a normal feature, it is not possible to ilotermino. The fact that it is also absent from the outlying parts of the decidua, while it is present in the cases of Peters and Leopold, suggests that it is due to superficial maceration.

There is no coagulum or organised thrombus adhering to the surface over the ovum. The mucous membrnno shows all the well known characters of the menstrual decidua. The cells are clear and swollen, the tissue spongy : the vessels aiv givatly dilated; the glands arc enlarged, while their epitbeliuQi is desquamating and their lumen is occupied by extravasated blood. The whole membrane is thickly studded with leucocytes, mainly of the polymorphonuclear variety. All round the ovum the marginal lamella of the decidua is in a state of advanced coagulation necrosis, appearing as a hyaline, darkly-staining, and nearly nuclear-free zone dotted with polymorphonuclear leucocytes. Stained by Gram's method this zone stands out as a deeply staining purple band, forming the wall of the implantation cavity. The plasmodiuni is not everywhere in contact with this layer. Here and there masses of it lie directly against the necrotic tissue, but round the rest of the circumference the two are separated by a space occupied by large mononuclear cells, the nature of which will be discussed below. The necrotic zone is broken up by the haemorrhage in the upper part of the section, and here and there, but in relatively few places, by masses of plasmodium, which are seen spreading outwards along the walls or into the lumina of blood-vessels. The mass of cells seen in Plate ii. Fig. 2 and Plate iii. Fig. 3, occupying the centre of the space between the cyto-trophoblast and the wall of the implantation cavity, corresponds to the layer of cells seen lying elsewhere immediately within the necrotic zone ; the cells have here been displaced inwards by the haemorrhage which has broken into the implantation cavity.

At either extremity of the implantation cavity an interesting condition of the glands is to be observed, from which it would appear that the gland w^alls resist the disintegration longer than the genera] tissue of the mucous membrane. The sections across each pole of the chamber show a gland in section with necrotic tissue and plasmodium on either side of it, as if the process of destruction were extending round the gland wall and isolating it. The implantation cavity is, as already stated, completely closed. There is no large aperture filled with blood clot such as is seen in Peters' ovum, or Graf v. Specs youngest specimen. Over the blastocyst, but distinctly nearer one end of the implantation cavity, is a well marked pocket, '1 mm. in diameter, partially filled with what appears to be thrombus (Plate iii. Fig. 3). Directly continuous with this, and reaching into the implantation cavity, is a spur-like projection of hyaline material only distinguishable histologically from the necrotic, zone of the decidua by being devoid of nuclei.

This depression is (dearly the mouth of the space in the decidua, but there is no direct evidence to show whether it represents the point of entrance of the ovum, i.e. the point where it first began its destructive action on the decidua, or the closed mouth of a fissure into which the ovum had been received. This problem will be discussed in a later section.

The Trophoblast

The term trophoblast, i.e. trophic epiblast, will be used here in the original sense in which it was employed by Hubrecht, to designate that part of the ectoderm which does not share in the upbuilding of the embryo, or of the amnion in the human subject, but only in the attachment and nourishing of the ovum. The trophoblast includes the whole thickness of the wall of the blastocyst, and is differentiated into two parts — the cyto-trophohlast or cell layer, in which the cell outlines are more or less preserved, and the plasmodi-trophoblast or plasmodium, in which they are wholly lost.

The Plasmodi-Trophoblast

The plasmodi-trophoblast forms an extraordinarily extensive spun-out investment for the ovum. It occurs in masses, bands, or threads. It is difficult to differentiate, in places, the fine threads from fibrin filaments. The plasmodial masses are distinguished by the dark, slightly rusty-red tint with which they staiin, forming a sharp contrast with the blue-pink of the immediate wall of the blastocyst, and the red-pink of the necrotic layer of the decidua (Plate in. Fig. 3). The nuclei differ from those of the cyto-trophoblast in respect that they are invariably small and stain darkly. This latter character is due to the finely granular nature of the chromatin network, which in the nuclei of the cyto-trophoblast is more open, loose, and reticular.

  • The terms cytoplast and plasmodiblast were suggested by Van Beneden. The terms used in the text are thtose now employed by Hubrecht. In so far as there consists a certain remnant of doubt regarding the origin of the plasmodium in the human ovum, the use of these terms at this point of our inquiry involves in some sense a /n^tifio principal.

The central strands of plasmodium are arranged round the cytotrophoblast in many places as an apparently laminated formation, with numerous spaces or clefts, which give an appearance of sharp separation of the layers, but between the spaces the cellular layer passes directly into the plasmodial. In some of the isolated masses of plasmodium vacuoles occur, which are either empty or partially filled with granular material. The vacuoles occur in some instances as single spaces, but more frequently they are multiple, and all intermediate stages are seen between masses in which vacuolation is commencing, and the spun-out plasmodial reticulum, the meshes of which are filled with maternal blood corpuscles. Plate v. Fig. 5, shows the characters of the plasmodium as revealed by a higher power of the microscope. To the left and below, a mass of plasmodium is seen lying free in the blood space, and in the early stage of vacuolation, while further to the right, interposed between two portions of decidua, is a larger and more vacuolated portion which is in direct contact with the necrotic zone of the decidua. It lies in the lumen of a greatly dilated capillary which has become directly continuous with the blood space round the ovum by the destruction of its wall. The endothelium of the vessel still persists on the surface of the detached mass of decidua to the left. At first sight it might be inferred that this mass of plasmodium was a portion of the closely adjoining network of the same tissue, but this is not so. When traced through the series of sections it was found to occupy the lumen of the vessel for a considerable distance, and to spring ultimately from the general plasmodial mass much nearer the pole of the blastocyst, where it could be seen entering a gap in the vessel wall.

In Plate viii, Fig. 10, an irregular mass of vacuolated plasmodium is seen lying in a bay in the necrotic zone ; it is extending into the decidua in close relation to a vessel which appears in the adjoining sections. In other situations large masses of plasmodium are spread out against the inner face of the necrotic zone — as if anchoring the ovum in the cavity.

The great inequality between the plasmodial mass in the upper part and that in the lower part of the section figured in Plate iii, Fig. 3, is clearly due to the haemorrhage into the upper part of the implantation chamber. The effect of the blood extravasation comes out rather more clearly in Plate ir, Fig. 2, in which the plasmodial meshwork can be observed to be pressed down to some extent on the wall of the blastocyst, while the layer of large cells elsewhere lying immediately within the necrotic zone is displaced inwards. The larger masses and bands have, as might be expected, resisted the pressure, and one process remains applied to the decidua as what appears to be an anchoring strand (Plate iii, Fig. 3). The inner part of the plasraodi-trophoblast has the appearance of irregular branching lamellae laid down round the wall of the blastocyst. In no section, or part of a section, is any portion of the wall of the vesicle left uncovered by a plasmodial layer. Here and there in the formation there are masses of what appears to be coagulum, but it is not impossible that some of these, at any rate, represent portions of necrosed Plasmodium.

The distinctive dusky-red colour of the plasmodium is brought out in Plate V, Fig. 5. At this magnification a granular structure is revealed in the protoplasm which is absent from the cyto-trophoblast, and under a still higher power this appearance is discovered to be due to an alveolar structure in the protoplasm.

The whole appearances presented by the plasmodium lead one to infer that the extraordinary irregularity in the disposition of the layer is due to a process of vacuolation which has broken up the larger solid masses into a sponge-work, and that the trabeculae of this sponge-work have broken down so as to allow the blood shed into the implantation space by the opening of the vessels, to piiss into its meshes.

We thus reach a conception of the origin of the primitive blood lacunae in the trophoblast not unlike that of Peters, but it will be observed that the spaces are ])roduced, in the first instance, entirely in the plasmodi-trophoblast. In this respect our ovum reveals a condition of the human blastocyst hitherto unsuspected ; its walls are at this early stage almost wholly plasmodial, with the exception of a thin germinal layer or mother-zone of cyto-trophoblast forming the immediate wall of the vesicle.

While our specimen speaks for the production of the blood lacunae by the formation of spaces in the trophoblast into which the extravasated blood is shed, it cannot be concluded that they are formed in the first instance in a uniformly thick lamella of plasmodium constituting a tolerably regular wall to the blastocyst. The plasmodi- trophoblast from its very nature is probably highly irregular from a very early stage, and therefore the blood lacunae may in part represent spaces intervening between outgrowing phismodial masses — between what may, in fact, be termed primitive plasmodial villi.

The Cyto-Trophoblast

The cytotrophoblast constitutes a relatively thin lamella which forms the immediate wall of the vesicle. The lamella is sharply differentiated from the plasmodi- trophoblast by its staining reactions. It is tinted, in haemalum and eosin preparations, a delicate blue-pink colour. The cell outlines are nowhere sharply defined, but they can be readily made out in sections in which the blastocyst wall has been tangentially cut. Elsewhere the appearance is rather that of a zone of protoplasm with embedded nuclei. As has been already mentioned spaces occur between the cytotrophoblast and the inner laminated layer of the plasmodium, but between the spaces the two formations are directly continuous. While at these points the cell-layer and plasmodial layer are distinguished in histological characters, the transition from the one to the other is not sharp and defined; the cell-layer changes its tone to pink, and passes uninterruptedly into the dusky-red plasmodial layer. The nuclei are extraordinarily irregular in size, though all show the same loose character of the chromatin reticulum, with one or two chromatin nucleoli (Plate vii. Fig. 7). They differ markedly, as already stated, from the nuclei of the plasmodium, which stain deeply and have a granular appearance. Here and there the innermost nuclei tend to be arranged in a row for a short distance, as if the cells next the cavity were assuming the epithelial disposition which characterises this zone in Peters' ovum. This arrangement of the nuclei is also seen in Leopold's youngest ovum, but in our specimen it is much less definite even than in that extremely early stage.

The nuclei of the cyto-trophoblast are clearly in very active division. Mitotic figures are not numerous, nor are they very well preserved, but their presence indicates that the ovum was in all probability in active growth shortly before being cast oft' The cell characters are best made out in the tangential sections which cut the poles of the blastocyst. Plate VIII, Fig. 11, is a photograph of one pole and Plate vii, Fig. 7, is a drawing of the other. A great many of the cells have either double, triple or even multiple nuclei. This may be observed Ijoth in the drawing and photograph. In the photograph (Plate viii. Fig. 11) the cell outlines are clearly distinguishable ; there is evidence of cell division ; and it may be pointed out that the large dividing cells belong to the innermost layer of the cyto-trophoblast. The absolute continuity of the cyto-trophoblast and plasmodi-trophoblast is brought out clearly in Plate vii, Fig. 7. The a])pearances here, and all round the blastocyst wall, altogether preclude any other conclusion than that the cyto-trophoblast, by active and continuous proliferation of its cells, is gradually diflerentiated at its outer margin into plasmodi-trophoblast. The cyto-trophoblast is, in short, the terminal zone of the trophoblast.

At one or two places the cyto-trophoblast shows minute buds extentlin<^ from its outer aspect. In one situation the bud has taken the form of a narrow cell column, on each side of which, but separated from it bv a s])ace, is a strand of plasniodium. The space does not contain blood, 1 it corresponds to the clefts seen elsewhere between the cell-layer and il'ismodium. These buds, though very rare, we take to indicate a com

miff m proliferation of the cellular layer, which will ultimately lead to m^ n»- the formation of the cellular villi of Peters stage. The grounds on which

kosed will be discussed here; meantime it may be stated opinion is

buds are clearly outgrowths of the cell-layer, and not plasmodial have reverted to a cellular condition.

the cyto-trophol)last show any such columnar

would indicate the i)icscnce of a layer of

part of the wall of the Mastocyst, nor any

rpcont the formative cell mass continuous with

It will be convenient at this point to consider the origin of the large cells which are seen lying free in the blood space within the necrotic layer of the decidua. These cells closely resemble bodies which are certainly cross sections of plasmodial strands, but when traced through the sections it becomes quite certain that they are not continuous with the Plasmodium, but are really isolated cells. Plate vi, Fig. 6, shows them at a point where they form almost a continuous layer for a considerable extent. The necrotic zone of the decidua is depicted. The inner edge of the dead or dying tissue is extremely irregular and is excavated into bays, many of which include one or more of the cells under consideration. Within the layer itself are seen spaces inclosing cells in different phases of degeneration. The nuclei are identical with those of the completel}^ free cells, and the protoplasm stains the same dusky-red in both cases. There are two ways of interpreting these appearances.

  1. The cells within the spaces in the necrotic tissue may be foetal, i.e. trophoblastic, derivatives which, having wandered outwards, are here caught in the act of attacking the necrotic wall of the implantation space and so causing its gradual enlargement.
  2. The cells may belong to the decidua itself, and are being set free from the necrotic zone as it is absorbed and the implantation cavity is enlarged.

The following considerations are in favour of the latter alternative :

  1. The cells cannot be cross sections of plasmodial strands because they cannot be traced back into the general plasmodial meshwork.
  2. The outer shell of trophoblast is entirely plasmodial. No cells which can be definitely identified as embryonic, occur anywhere except in that part of the wall of the blastocyst which we have called the cytotrophoblast.
  3. The cells, both in the character of the nuclei and reactions of the protoplasm, agree with cells further out in the decidua, which are clearly degenerating maternal cells, and are distinguishable from the trophoblast in both these respects. They are quite different from the wandering trophoblast cells of later phases.
  4. The Plasmodium and necrotic zone of the decidua are not everywhere directly in contact with one another; a union between the two occurs only here and there ; elsewhere a space is left between the two containing red blood corpuscles and leucocytes. There is no indication that the wall of necrotic tissue is being absorbed by phagocytosis in the strict sense of the term ; the appearances suggest rather solution by enzymes produced by the trophoblast.

On the whole we are inclined to conclude that these elements are derived from the necrotic zone of the decidua; that they are ifn short decidual cells set free in the process of absorption of the necrotic tissue. It must be admitted however that there is no histological criterion by which it can be absolutely determined whether they are maternal or foetal derivatives.*

  • Since the above was written we have had the opportunity of seeing a demonstration by Graf V. Spee of the early phases of placentation in the guinea-pig, at the meeting of the Anatoniische Ciesellseliaft in Berlin (April 23rd, 1908). Graf V. Spee showed that in the guinea-pig there is a layer of cells of foetal origin outside a plasmodial formation, which, though less extensive, has the same character as the plasmodium in our human ovum, — in short, that there are three layers of the trophoblast. Until his account of his preparations is published we can do no more than point out that Graf V. Spee's researches may possibly necessitate some modification of our interpretation of these elements.

The Contents of the Blastocyst

The Mesoblast

The cavity of the vesicle is occupied by a very delicate cellular reticulun), or loose syncytial tissue which has the characters of mesenchyme. It represents the earliest stage yet observed of the mesoblast. This mesenchymatous tissue shows no signs of cleavage into a parietal and a visceral layer ; it is not yet arranged in a definite and denser layer round the wall of the vesicle, nor are there any processes of it indenting the wall.

The constituent cells of the mesoblast are minute, rounded or stellate elements united together by very delicate protoplasmic threads. They are brought out in the drawing reproduced in Plate ill, Fig. 3, but the network which they form is somewhat obscured by the delineation, in the interests of exact reproduction of what the section shows, of an extremely delicate reticulum, which is clearly an artefact due to precipitation of the fluid basis of the tissue. The coagulation of the tissue has clearly caused some contraction of the mesoblast, which has resulted in its withdrawal from the centre of the vesicle so as to leave a clear space in which the embryonic rudiment is situated. This retraction of the mesoblast has caused some degree of tearing of the embryonic rudiment.

The Embryonic Rudiment

Figure 2. Outline drawings of portions of the Embryonic Rudiment as it appears in thirty successive sections. X 320 diameters. The exact histological characters are shown in Plate VII, Figs. 9 and 10.
Figure 3. Drawing of a reconstruction wax model of the Blastocyst. The outer darker layer is the cyto-trophoblast ; the inner lighter, the mesoblast represented as if solid. The irregular cavity is the retraction cavity in the mesoblast. It contains two vesicles ; the larger is the torn and collapsed amnio-embryonic vesicle, the smaller is the entodermic vesicle.
The model was constructed by building up wax plates to represent what was really a cast of the cavity of the blastocyst, with a space in ita centre containing the vesicles. The model was then cut across so as to present to view the portion of the retraction cavity containing both vesicles. This portion of the model was then cast in plaster of Paris, and an outer case constructed which represents in a diagrammatic fashion the cellular layer ol the blastocyst wall.

In the absence of any spot in the wall of the vesicle which could by any possibility be regarded as embryonic ectoderm, and of any thickening which could represent the inward projecting embryonic knob, we must recognise the embryonic rudiment in two closed vesicles which occupy the central retraction space in the mesoblast; There is a larger vesicle and a smaller. It is unfortunate that the larger is collapsed and considerably torn, while the smaller, though complete, is probably slightly displaced, as it is not directly in contact with the larger sac.

The larger vesicle extends through 24 sections (Figure 2,, 1-24), and therefore measures in this axis '168 mm. The cells forming its walls are small compared with those of the trophoblast, and are cubical rather than columnar, but the protoplasm is frayed, and it is evident that the cell bodies have suffered considerably from defective fixation. The nuclei are rounded and fairly regular, though an occasional flattened nucleus occurs in what appears to represent the roof of the cavity. The vesicle hangs free in the central space, being definitely attached at one point only, where presumably the mesenchymatous tissue was more resistant.

A reconstruction in wax (Figure 3) makes it quite certain that we have to do with a closed but torn and collapsed vesicle with uniform walls; there is no distinction between the cells of the roof and those of the floor. There is no indication of a passage from the cavity towards the surface of the blastocyst.

The smaller sac extends through six sections (Figure 2, 26-31), and measures therefore .042 mm. The cells forming its walls are more flattened than those of the larger vesicle. It is quite certain that though it is of very miuute aize the formiition is a closed sac, and not an accidental groujiing of the mesoblaat cells. It is important to note that while the larger vesicle is attached definitely only at one point, the smaller is closely surrounded by mesoblast strands. These are absent only in one section, and on the side looking towards the position in the central space which is occupied in other seetioua by the larger vesicle.

The model (Figure 3) shows the relative sizes of the two sacs, and their position in regard to one another. The space between the two is obviously considerable, but aa it contained no mesohlast it is probable that there baa l)een a certain amount of accidental displacement. The appearances strongly suggest that the larger vesicle has suffered most in the shrinking of the tissue surrounding it. It seems to have collapsed and drawn away from the smaller vesicle. Two forces must have operated upon it — the centrifugal traction of the shrinking mesoblast, and a centripetal traction due to the precipitation of its fluid contents and contraction of its walls. The result of these opposing forces is the extensive tearing of the walls of the vesicle that has taken place. The other vesicle being very much smaller, and not so directly in the centre of the retractive force, luis suffered much less.

After careful consideration of the sections and model, the conclusion is inevitable that the larger vesicle represents the amnio- embryonic cavity, and the smaller the entodermic vesicle or future yolk sac. The data on which this interpretation is based will be discussed in the next chapter; meantime it may be said that it is in harmony with what we now know of the early primate blastocyst, and with the recent views as to the development of the amnion in several lower mammals. It may also be pointed out that this is the earliest phase of the human embryo yet observed. In Peters' ovum the rudiment consisted likewise of two closed vesicles, but the larger or amuio-embryonic vesicle showed a differentiation of the floor into embryonic ectoderm, and the roof into the amnion. Our ovum shows a still earlier condition of the amnio-embryonic cavity, in which there is as yet no distinction between embryonic and amniotic ectoderm.

Discussion Of Data

From the foregoing description it will be apparent that the present ovum differs in certain important respects from any human ovum hitherto described. It is also in some of its features unlike any ovum with which we are acquainted among the lower mammals.

Summary of Characters of the Ovum

The special characters of our specimen may be summarised as follows :

  1. The blastocyst is completely enclosed in decidua except at one point, where there is a small gap closed by a mass of fibrin and leucocytes. The wide gap closed by the mushroom-like mass of fibrin and blood clot seen in Peters' ovum is entirely absent.
  2. The ovum lies, bathed in blood, in a relatively large implantation chamber with the walls of which it is not united. There is no interlocking or mixing of maternal and foetal tissues. The innermost layer of the decidua lining the cavity is in a state of advanced coagulation necrosis, and this, together with a certain amount of fibrinous deposit, forms a layer of dead material which is practically complete except at one or two points where blood-vessels have been opened up, and at one end where a haemorrhage has broken into the implantation chamber.
  3. The wall of the blastocyst consists of an inner lamella (cytotrophoblast or cell layer) composed of cells rather ill-defined from one another, and continuous externally with an extremely irregular formation which has definitely plasmodial characters (plasmodi-trophoblast). This forms a straggling reticulum, the meshes of which are filled with maternal blood (primitive blood lacunae). There is no protrusion of the cytotrophoblast into the plasmodi-trophoblast strands.
  4. The cavity of the blastocyst is filled by a delicate tissue having the characters of mesenchyme. There is no cleavage of this early mesoblast into a parietal and a visceral layer ; it does not form a distinct lamella round the wall of the cavity ; and there are no protrusions from it representing future mesoblastic villi.
  5. The embryonic rudiment is represented by two eccentrically placed vesicles slung in the mesenchyme by fine protoplasmic threads. They are quite separated from the wall of the blastocyst by mesenchyme, and the cells forming the two sacs have definite and difierent characters, but inter .se show no differentiation. The cells of the larger (amnioembryonic) vesicile are cubical ; those of the smaller (entodermic) vesicle are flattened.

Remarkable as some of the features of this new ovum are, there is no reason to suppose that it is in any way abnormal or pathological. Every one of its characters, as we shall now proceed to show, harmonises admirably with known later stages. It is in no sense contradictory or bizarre. It is not only consistent in itself, but is also consistent both with admitted facts and with inferences founded on these facts.

In the following discussion we shall first establish the position of our ovum relative to the earliest ova hitherto recorded, more especially those of Iieo])old (1906) and of Peters, and consolidate the basis of our interpretation both of the trophoblast and of the embryonic rudiment.

Comparison of the Present Ovum with those of Leopold and of Peters

In Leopold’s ovum nothing like an embryonic rudiment Wtis found. Towards the centre of the implantation cavity is a sac which Leopold regarded as the blast o(*yst. It is a thin walled and irregular vesicle containing tissue similar to the mesenchyme in the present ovum and in that of Peters, but infiltrated with red-blood corpuscles, obviously maternal, and separated more or less from the wall of the vesicle by a s})ace containing l)lood corpuscles (cf Leopold. I DOG, Plate x, Fig. 18). The wall of the blastocyst is described as a thin mantle of ectoderm showing two layers of cells, the inner consisting of rounded or oval elements closely appjied t<3 one another and almost filled by large darkly staining nuclei, the outer of larger granular cell-masses with one or several nuclei, having thus the charactei-s of plasniodium. The cells of the inner layer he distinguishes as Langhans* cells. Here and there the layer thickens into buds of Langhans' cells covered by plasmodium. These buds stretch out into long processes containing cells of both types, but there is never any outgrowth of mesoblast into them. Both cell layers are certainly of ectodermic origin, indeed Leopold regards any separation of them as inconceivable in light of the appearances in his specimen. The processes are attached here and there to the decidua, the terminal cells insinuating themselves among the tissues of the latter. The implantation cavity is a globular space filled with maternal blood in whi(*h the ovum floats freely, being anchored only by the tips of the trophoblastic processes.

There are distinct points of resemblance between our ovum and that of Leopold, but the differences are also strongly marked. The first striking point of contrast, apart from the absence of an embryonic rudiment, is that in our case the trophoblastic processes are entirely plasmodial ; any cells which could bear comparison to Langhans cells are confined to the thick wall of the vesicle, and the anchoring strands of Plasmodium are fewer in number. The undifferentiated condition of the blastocyst wall in our ovum, and the absence of a cellular layer in the trophoblastic processes, lead one to infer that it is probably younger than Leopold's.

While the characters of the blastocyst, the haemorrhage into its interior, and the absence of an embryonic rudiment are abnormal features in Leopold's ovum, on the other hand, in respect of the trophoblast it very possibly represents a normal intermediate stage between the present specimen and Peters' ovum. It is known that in later stages the villi develop practically normally in the absence of an embryo, and doubtless the trophoblast may also do so at this early stage.

The blastocyst wall in Leopold's- ovum, while it has a general likeness to that in the present ovum, is a less definite and a thinner structure, and has a greater resemblance to the layer of cells, with commencing cubical arrangement, covered by endothelium-like syncytium, which constitutes the greater part of the blastocyst wall in Peters' specimen (cf. Leopold, Figure 18, with Peters' Figures 1 and 22). Leopold's ovum further resembles Peters more closely, (1) in having two kinds of cells in the trophoblastic processes, and (2) in the characters of the mesoblast. The principal difierence between the two is that the trophoblastic processes are more numerous and much thicker in Peters' ovum, so that, as compared with Leopold's specimen, the blood lacunae are much reduced in size, and the trophoblast takes the form of a thick layer containing blood spaces. The attachment of the primitive villi to the decidua appears to be very similar in both cases, but the degree of intermingling is much greater in Peters' ovum, and the condition of the decidua is also diflfierent Whereas in Leo})old's ovum (see his Figures 23 and 24) the zone of decidua next the trophobhist is largely in the state of coagulation necrosis, characteristic of the present specimen, in Peters' case the decidua has a much more living appearauite ; there is more mingling of living cellular elements, and the jiinount of necrotic material visible is relatively small. There is evidence of active reaction on the part of the decidua in Peters' case, as shown by the })rescnce of large numbers of polymorphonuclear leucocytes and formation of new vessels; this is less marked in Leopold's case, but the invasion of leucocytes is very striking in the present specimen. All three ova however show a very distinct reaction as far as dilatation of bloodvessels is concerned, and this is greatly exaggerated in Leopold's case by congi'stion whiith is probably due to the mode of death.

In respect therefore of the trophoblast the ovum of Leopold may with some confidence be considered as a stage intermediate between the present ovum and that of Peters' ; but as there was no embryonic rudiment the comparison cannot be carried further.

When the embryonic rudiment in the ])resent case is compared with that in Peters' blastocyst, it is at once apparent, that if our interpretations be correct, this ovum is at a considerably earlier stage of development.

The Embryonic Rudiment

Figure 4. Blastocyst of Tarsius Spectrum. After Hubrecht from Quain's Anatomy 11th Ed. Vol. I. Embryology.
Figure 5. Diagram to illustrate the condition of entypy of the the germinal area. The trophoblast is represented by continuous black lines or masses, there entoderm by interrupted lines, the embryonic ectoderm, and in certain figures the amniotic ectoderm, by epithelial cells. Each figure represents the blastodemic vesicle, a, of the rabbit ; b, of the mole ; c, of the bat ; d, of the mouse or rat ; e, of the guinea-pig ; f, of the kalong (I'teropus edulis).
Figure 6. Hypothetical stage of the human T. H. Bryce from "Quain's Anatomy", 11th Ed. This diagram was designed to show, in one figure, the probable early condition of the trophoblast, and an early phase of the of the cavity of the blastocyst relatively to the size of the amnio-embryonic and entodermic vesicles, ia probably rendered too small. We do not know how soon the trophoblast shell begins to assume the dimensions so much out of proportion, as they appear in our ovum and in Peter's. The trophoblast is shown with a cellular and a plasmodial layer, but it is probable that when the plasmodium is at the stage represented, the embryonic rudlment is not so far differentiated as it appears in the diagram. With these qualifications the figure, which was printed before the present ovum was obtained, is probably not essentially incorrect.

It will be necessary to explain at this point the data on which our interpretiition of the embryonic rudiment is based. The mammalian blastocyst is a hollow vesicle with a knob of cells projecting into its cavity from one point on the wall (Figure 4). The ectoderm of the wall or trophohlast is concerned in the processes of imbedding and placeutation, the inward projection constitutes the embryonic blastema. Whereas in the rabbit the cells forming the embryonic knob become spread out flat at the upper pole of the blastocyst (Figure v, a), and are soon exposed on the surface, by the disappearance of the thinned out trophoblastic covering (Rauber's layer), in another series of mammals the knob remains iuturned and a cavity appears among the cells of the knob. In some cases the roof of this cavity early breaks away {Figure v, h), and the embryonic ectoderm, becoming flattened out, is exposed on the surface just as in the rabbit. In other instances the roof of the cavity persists and forms the definitive amnion (Figure v, e, f). There is very good reason for believing that this is the case in the liuman subject. Owing in all probability to the nature of the processes of imbedding, the embryonic knob or formative-cell mass remains in its original position, and the trophoblast becomes uniformly thickened all over the sphere. The embryonic rudiment lags behind in development, and there is a relatively enormous expansion of the trophoblast shell, which is concerned at first in the excavation of the implantation cavity and then in the nourishment of the embryo.

The entoderm is probably split off, to judge from the stages in Tareius, before the embryonic ectoderm is differentiated from the amniotic. Owing probably to the relatively great expansion of the trophoblast shell, anil tardy difterentiation of the embryonic rudiment, the entoderm does not f^row nuiiid the lihistoeyst wall, but takes the form of a small f\tm"i\ \vAi-\ii, the cavity of which very possibly iippeai-s within a solid mass (if cells. The amnio-embryonic, cavity is excavated in the heart of the embryonic knob {Figure vi), and its floor, necessarily at first concave, and therefore apparently inverted or inturned, becomes the embryonic plate, while the roof becomes the amnion. The mosobhust appears very early and fills the space between the rudiment and the trophoblast. It is uncertain whether the amniotie lamella is from the first separated from the trophoblast. In Figure vi and Figure viii, p. 45, they are represented as remaining connected, because it seemed impossible, without the existence of such a connection, to explain the presence of the amnion duct seen in Beueke'a ovum, or the appearance of an open uninion as seen in Mall's pathological cases. While in man there ia no true reversal ofblastocyst the germinal layers such as seen in mice and luts and the guinea-pig, there is no doubt that the plate of embryonic ectoderm is inturned, and strong probability that the condition is a primary one and not due to a precocious formation of amnion folds.

The views enunciated above find their full justification in this the earliest phase yet known of the human blastocyst, and lead at once to the identification of the larger vesicle as the amnio-embryonic and the smaller as the entodermic sac.2

It is right however to say that a rather different interpretation might possibly be put upon the facts, were the hypothetical stage imagined by Minot2 taken as the initial pluise. He represents the amnio-embryonic cavity as a transverse slit in a thickened portion of the blastocyst wall, (consisting apparently of the formative-cell-mass merged with the trophobliist. Our ovum might represent a phase in which the entoderm had been H(ipanit.ed off, but the amnio-embryonic slit had not yet appeared, and the embryonic ecto<lerm had not been differentiated. The larger sac would in this event be the entodermic sac, and the smaller might be considered a mere accidental grouping of the cells of the mesoblast. There is no doubt however that the smaller sac is closed, and that it is a definite formation. The seccjnd interpretation is further excluded by the absence in any part of the l)lastocyHt wall of any such thickening as represented in Minot's figures. The diagrams here given supply an explanation of the facts at once simpler and more direct.

One of the most puzzling features of early primate ova is the precocious formation of the middle layer cells in the cavity of the blastocyst. The present ovum is not yet early enough to warrant a final judgment on the subject, but two new points emerge. The mesoblast arises at a still earlier stage than it does in Tarsius, and it completely fills the blastocyst. The extra-embryonic coclom seen in Peters' stage has not yet been formed. The probability that the extra-embryonic coelom is produced by splitting in the mesoblast therefore becomes a certainty.

In regard to the origin of this early mesoblast two remarks suggest themselves. The attachment of the mesenchyme to the amnio-embryonic vesicle more definitely and firmly at one point, may possibly indictate that it is arisinor from the ectoderm and from the same res^ion as it does in Tarsius, i.e. from the future hind end of the embryonic plate, where the primitive streak will afterwards appear. On the other hand it is not impossible that the mesoblast arises contemporaneously with the entoderm from the embryonic knob ; that in short the entoderm, even while coming into existence, is differentiated into an epithelial lamella which constitutes the lining of the yolk sac, and a vascular mesenchyme which is concerned in forming the vessels of the connecting stalks and chorion, and the blood and blood-vessels of the yolk sac. The study of the sections and of the characters of the cells rather inclines one to accept the second alternative, which could without much ditficulty be reconciled with what is known concerning the origin of the vascular mesenchyme in later stages.

1. The views adopted in the text are given at greater length in my article in Quain's Anatomy 11th ed. 1908, vol. I. Kinhryology. They are founded mainly on the works of Graf v. Spee, Van Hen(;den, Hubreeht and Selenka. The chief difliculty which lies in the way of such an interpreUition is the presence of an amnion duct in some lower Primates and in Beneke's ovum and the existence of a ])iuss;ige connecting the amnion with the intervillous space in pathological cases such as desci'ibed hy Mall. Such a condition may well be secondary, and it is so interpreted by Selenka and Keibel, but the appearances might indicate a very precocious formation and closure of amnion folds. In my article in Quain, having never seen any distinct recent statement on Professor Keibel's part regarding the amnion, I was led into the error of implying that lie still held the latter view. In a personal convei-sation with him however, 1 learn that he has long accepted the primary closure of the amnion in man, and in a proof of his forthcoming Xarvu'nta/tfl for man, which he has been gcH>d enough to send to me, I am glad to find that his views are essentially similar to tho.se I have myself been led to adopt. Our case confirms his early jjosition (IHliO) regarding the invei-sion (reversal) of the layei^s in the human embryo; there is no such revt;i*s*il in the strict sense of the term, but in so far as the plate of embryonic ectmlerm is inturned, there is a parallel between the early sUiges in mice, iiits, etc., and those in man, Jis fii-st pointed out, I believe, by (Jraf v. Spee in 1881). It is now known however that this eiirly inturning is a feature of pretty general occurrence in mannnals ; true inveraion (KennbUitterumkit/ir) is brought about by secondary conditions which are not present in the case of the lunnan ovum. T. H. Bryce.
2 . A Laboratory Manual of Embryology 1903.

The Process of Imbedding

Having now discussed the characters presented by the present ovum, and its position in point of development relative to the earliest stages hitherto known, we proceed to consider its relations to the decidua, and the mode of implantation. In the absence of earlier stages, it will however be necessary to consider at greater length than we have yet done the comparative data available on these points. As already explained only those cases among the lower mammals in which there is a decidua capsularis provide a competent analogy.

The hedgehog presents certain resemblances in respect of the gross characters of its placentation to the human subject; there is not only a decidua capsularis, but also a general primary, and a definitive discoidal placenta. For these reasons, among others, Hubrecht undertook the investigation of the development of the placenta, which has led the way towards a clearer understanding of the human placenta.

The ovum of the hedgehog while still a small hollow blastocyst settles down at the bottom of a groove in the uterine mucosa. The epithelium around it disappears and at the same time a decidual swelling occurs in both lips of the groove, which thus enlarge and meet over the ovum. A certain amount of destruction of the connective tissue of the mucosa takes ])lace, bleeding results, an<l a plug of blood clot is formed between the lips of the decidual swellings. The epithelium next disappears from the contracting lips, and they coalesce to complete the inclosure of the ovum.

The ectoderm of the blastocyst proliferates into a relatively thick lamella, which Hubrecht named the “trophoblast." It surrounds the whole blastocyst, and the formative cell mass projects into the interior as a knob of cells. The trophoblast on the one hand rapidly establishes a union with the proliferating decidua, and on the other becomes hollowed out into cavities, into which maternal blood is shed from the capillaries which have been opened up by its destructive activity. At first the trophoblastic shell is complete and the greater part of it becomes penetrated by mesoblastic villi which are vascularised from the yolk sac forming a {)rimitive vitelline placenta, but the definitive placenta is formed by villi which receive their blood-vessels from the allantois.

At a certain stage therefore there is a diffuse placenta, while the definitive placenta is implanted upon the area of decidua adjacent to the uterine wall, i.e. decidua serotina or basalis. As this develops the vitelline placenta applied to the decidua reflexa gradually undergoes atrophy. The main difference between these arrangements and those that prevail in man is that the primitive j)lacenta is supplied in the latter by vessels which develop in the Ilrt/tstiel, or connecting stalk of mesoderm, and correspond to the allantoic vessels. The Haftstiel is, as now well known, a feature peculiar to the Primates, there being further no free allantois.

Accurate data regarding the chronology of imbedding and development in the hedgehog are wanting.

In the mouse the process of imbedding is in some respects very similar to the type seen in the hedgehog. the ovum becomes lodged in a furrow on the side of the slit-like lumen of the uterus farthest from the mesometrium. The e])ithelinm disa})pears in its neighbourhood ; decidual development occurs in the adjacent endometrium, which swells up into a thick cushion all round the ovum. The swelling gradually obliterates the middle region of the lumen, leaving however at first an open passage along the opposite wall of the uterus. The space between the lips of the decidual swellings is at first l)locked by the ectoplacenta, Init later the epithelium disappears — apparently as a result of atrophy due to pressure, the exposed decidual surfaces coalesce, and the capsule is completed. Besides great hyperplasia of the decidua cells, new vessels are formed which are directed towards the ectoplacenta, and when the union of trophoblast and decidua is accomplished, the placenta is developed in the region where the decidua capsularis is finally closed. The part of the decidua, therefore, which is developmentally reflexa plays the part of serotina. Prior to this there is generally some haemorrhage into the implantation cavity, and the shed blood is supposed to be used as pabulum for the growing ovum.

The further development of the placenta does not concern us except in one particular, viz. — the uterine epithelium, as is clear from the work of Burckhard, plays no part in the formation of the placenta ; this is achieved in this case, as in the case of the hedgehog, by the union of foetal ectoderm and decidual connective tissue, and by the opening of maternal blood-vessels into spaces formed in a thickened portion of the trophoblast.

The time relations in the case of the mouse and rat are very accurately known. They will be discussed in the chapter dealing with the age of the present ovum, but it may here be noted that imbedding commences on the fifth day after fertilization, and that the ovum at that period is a small hollow blastocyst, not materially larger than the fertilized ovum before the commencement of segmentation.

In the guinea-pig imbedding seems to occur in a different fashion. According to the researches of Graf v. Spec the ovum, while still in a very early blastocyst phase, loses its zona pellucida, and actively attacks the endometrium. It destroys a small area of the epithelium, and passing through the gap so produced, continues its destructive action on the connective tissue of the mucous membrane until a cavity is formed sufficient for its complete inclosure. The opening in the epithelium is blocked for a time by the cells of the future ectoplacental pole of the ovum, the epithelial and foetal cells remaining in contact at this point. Nec^rosis, followed by solution of the dead cells, now overtakes a (considerable area of the endometrium, with the result that a considerable space {Imi)lantatioushof) is formed round the ovum. This is filled with fluid and is lined by more or less necrotic and dissolving tissue. Outside this zone, active proliferation takes place leading to the formation of a decidua which completely surrounds the blastocyst. The placenta is formed from the ectoplacenta, which becomes drawn down, owing to changes in the uterus, towards the middle of the lumen ; maternal bloodvessels are developed in this position, and the final result is that, in spite of the initial difference in the manner of implantation, the relations of placenta to uterus are very similar to those in the mouse. It is sufficiently clear that in this case also the uterine epithelium takes no part in the formation of the placental tissues.

The imbedding of the guinea-pig ovum is eff'ected on the seventh day after coitus ; segmentation is complete, but the size of the ovum is not appreciably increiused. When implantation hius taken place the ovum grows and develops with great rapidity.

In all these cases fertilization apparently occurs in the upper part of the Fallopian tube ; segmentation is completed during the transit of the ovum to the uterus, which it reaches but little increased in size and still enclosed in the zona pellucida. This envelope now rapidly disappears, and implantation of the ovum takes place forthwith. In the mouse and guinea pig the endometrium does not show the slightest trace of alteration from the non-pregnant state until the epithelium has disappeared, and the ovum commences to attack the deeper tissues ; decidual development then rapidly follows. Whether this is the same in the hedgehog is not quite clear. In all these cases the ovum becomes completely enclosed, and after a period in which it lies free in the implantation cavity, it obtains attachment to the maternal connective tissue by means of the development of a special part of its ectoderm.

While the present ovum is the youngest yet described it is not yet sufficiently early to demonstrate the actual mode of implantation in man, but it brings us nearer to a comprehension of the process in several respects.

In regard to the antecedent phenomena there is no reason for believing that they are different from those in other forms, and we may consider that the ovum reaches the uterus as a small blastocyst still enclosed in the zona pellucida. The cavity of the human uterus is of course vastly larger, relatively to the ovum, than that of the hedgehog, mouse or guinea-pig, but it is narrow and slit-like. It differs from that of the mouse in having, as far as the corpus uteri is concerned, a perfectly smooth surface, free from crypts or furrows, with the exception of the mouths of the glands, which are too minute in an undilated state, to admit an ovum. The idea that the ovum becomes imbedded in a gland is apparently not now held by any one. Our ovum is imbedded near the centre of a smooth, rounded elevation, similar to, but higher than, many other small areas lying between the furrows, and it appears highly probable that the elevations and furrows alike have arisen jis the result of decidual proliferation. We have no evidence that this development takes place prior to the imbedding of the ovum. Indeed, from what occurs in the mouse and guinea-pig, we might from analogy infer that the human ovum may imbed itself in practically unaltered endometrium. This question will be considered again in the chapter on the age of the ovum, and the relations of imbedding to menstruation.

When we consider the appearances of this elevation, the shape of the uterine lumen and the characters of the mouth of the implantation cavity in our specimen, we must conclude that in the initial stages the process of imbedding in the case of the human ovum is in all probability similar to that which Graf v. Spee describes as occurring in the guinea-pig rather than to that which occurs in the mouse or hedgehog, though it is not impossible that the ovum settles into one of the slight hollows which are present on the surface of the normal endometrium. This opinion is of course no new one ; it has been advocated by Peters, Graf v. Spee, Leopold and many others, but our case enables us to amplify the conception in some particulars.

We would describe the imbedding of the human ovum as follows : The ovum having attained the stage of an early blastocyst, and measuring about '2 mm. in diameter {i.e. approximately the size of the mature oocyte), comes to rest in a slight depression, but neither a crypt nor a fissure, in the endometrium, destroys the surface epithelium, and continuing its destructive activity passes into a space in the decidua which has thus been produced. Necrosis followed by solution (digestion) of a considerable mass of the endometrium follows, resulting in the formation of an implantation cavity. So far we agree with the views expressed by Graf V. Spee regarding the human ovum in 1905. Changes leading to the production of decidua begin immediately after the solution of the epithelium, and the elevation is formed which is the characteristic resting place of all the four earliest ova at present known. The mouth of the implantation chamber is probably blocked by a mass of blood clot, the cavity having meantime been filled by blood shed from the opened-up maternal capillaries. The present specimen shows a narrow orifice only •1 mm. in diameter, and the sealing substance is rather thrombus-like material than blood clot. The relatively- wide gap ('8 mm.) in the surface of the mucosa closed by blood clot and fibrin, which characterises the ova of Peters, Jung, Stolper and Graf v. Spee, is here entirely absent. If we have proved our thesis that the present ovum is the earliest yet recorded, we must conclude that this gap is not the portal of entrance properly so called — it is much too large — but is a secondary formation, being produced by the subsequent destructive activity of the trophoblast threatening to destroy the roof of the implantation chamber. The smaller extent of the fibrine cap in Leopold's ovum, and the fact that it is in two portions, woakl, if our view be correct, fiml a re;iily expljmation.

The ovum now rapidly differentiating develops a thick trophoblast all round the blastocyst as iu the hedgehog, and not at one point aa in the other species (see Figure vi, p. 33), The ovum is at first free in the implantation cavity. The trophoblast from a very early stage,


P.r., poiiil of entrance: iryl., cy to- trophoblast ; /U., plmniodi-trophoUast ; II.:., necrotic xane of ticciilua; gl., gland; cap.. ca|)illBry ; pi.', muwea of cauiiolHling plBsmodium invorliny inpinkrica. The cavity of the blastocyst is completely filled by ineaolilwt, and iinbed<l«rt therein are the nmniD-embryoDic Ulit Gubxleraiii! veBicles. The ntittinil proportions of the several pttrta have boon strictly observod, its ill the bat (Van Bcneden), ahows a cellular layer (cyto-trophoblast) and a plasmodial layer (pla-smodi- trophoblast). the plasmodium throws out bud.'^ which stretch towards tbe walls of the deciduid chamber, and it is continually being added to by active proliferation in the cellular layer. In the first place the plasmodial masses exert mainly a destructive action ; this results in thy production of a relatively large implantation cavity such ;ts seen in our specimen (Figure vn). The destructieui of the decidua is necessarily associated with the destruction of vessel walls and the opening up of glands. Haemorrhage occurs into the cavity, but the blood does not coagulate. It serves to nourish the ovum, and after a time it begins to circulate among the trophoblastic processes. Up to this stage the ovum has not become attached to the decidua. It now becomes fixed, first ])y anchoring strands of phismodium and later by the development of j>rimitive cellular villi. Our. specimen shows the ve;y initial stages of attachment.

We may here digress for a moment to point out a very suggestive analogy with the normal ovum in respect of its influence on adjacent maternal tissues, presented by an embolus of chorioU'Cpithelioma in its development into a secondary tumour.^ The embolus, usually somewhat larger than the imbedding ovum, lodges in the fork of a blood-vessel, the wall of which soon shows degenerative changes identical in character with those seen in the decidua around the present ovum. These occur prior to the invasion of the tissues by the tumour cells. The injured blood-vessel dilates into a more or less globular aneurismal cavity, and there may be considerable growth of the embolus in its interior before invasion begins. The blood in the neighbourhood of the embolus does not coagulate, until secondary changes, which need not be discussed here, bring about that result. After a time the tumour elements invade the maternal tissues and the embolus becomes attached. At this stage appearances may be found very similar to those around the margin of Peters' ovum.

The formation of the aneurism with its necrotic wall, round the embolus finds a parallel in the behaviour of the decidua round the human ovum. We may assume that, as in the case of the guinea-pig ovum, there is primarily destruction of a fairly wide zone of tissue, which draws away from the ovum partly, at least, on account of the swelling which accompanies coagulation necrosis. The shed blood must further stretch the soft decidua just as the wall of the vessel is stretched into the wall of an aneurism, and the necrotic zone of the decidua is removed, like the necrotic wall of the aneurismal sac, before attachment can take place. Also, as in the case of the embolus, the necrosis is at first progressive, but soon, reaction taking place in the surrounding tissue, it becomes more capable of resistance, and consequently of effecting a union with the trophoblastic processes, as will be described later.

  • 0n the Development and Natural Healing of Secondary Tumours of Chorion-epithelioma," by John H. Teacher, Jour. Path, ami Davt. vol. xii. ixirt iv. 1908.

The appearances of the necrotic zone of the decidua and of the Plasmodium in our ovum all suggest that the process of destruction is not one of erosion by direct cellular activity or phagocytosis, but a sort of digestion or solution due to the action of extra- cellular substances probably of the nature of enzymes. Further, it appears probable that the vacuolated and spun-out condition of the plasmodium is due to the formation in the vacuoles of a digestive fluid, and that when this is shed by the rupture of the vacuoles its place is taken by maternal blood. In this way all the appearances of the plasmodium are readily explained. Possibly the secretion also contains an enzyme which prevents coagulation of the blood ; but it is also not impossible that the surface of the plasmodium may behave towards the blood like an endothelium, thus preventing coagulation. The fact that coagulation readily occurs in the neighbourhood of the placenta, or tumour, when a considerable extravasation has taken place, is rather against the idea that there is an active coagulation-hindering enzyme.

The plasmodial formation characteristic of the present ovum has hitherto been described only in connection with a young ovum recorded by Stolper. The ovum which measured 2*5 x 2.2 mm. in diameter, was obtained by curetting undertaken on account of menorrhagia simulating abortion, although no menstrual period had been omitted. It presented a close resemblance to Peters' ovum, and was fairly well fixed. On one side of the ovum the villi were long and rather more advanced in appearance than in Peters' case, but on the other side there was an extraordinary development of spun-out plasmodium, w^hich was not attached to the adjacent decidua, and was not accompanied by a corresponding development of the Langhans* layer. Much of this plasmodium was degenerated or even necrotic, and Stolper states that *' es kann nicht physiologisch sein." In the light of our ovum, the occurrence of such a condition of the plasmodium in Stolper's case, probably represents the abnormal persistence of what is normal in an earlier stage.

It will be recollected that in Lcopok's ovum both layers of the trophoblast are represented in the irregular villi, and that in Peters' specimen the epithelial villi are broader bands composed of Langhans' cells and covered by an endothelium-like layer of syncytium bounding the blood lacunae. Our specimen seems to represent a phase hitherto unknown, in which the ovum is largely unattached and the villi are wholly plasmodial. Leopold's ovum is a stage in which attachment is proceeding and the plasmodial villi are becoming replaced by cyto-tro})hoblastic processes or villi, while Peters' ovum represents a stage in which attachment is taking place, and the necrotic zone having largely disappeared, there is a mingling of foetal and maternal tissues ; in short, formation of a placenta has begun. In the later stages destruction of decidua no doubt continues, but fiir less actively.

With regard to the manner in which the purely plasmodial m^isses become replaced by processes containing both cyto-trophoblast tmd plasmoditrophoblast, three possible explanations present themselves :

  1. The trophoblast may from the beginning be plasmodial and the cell-layer may arise by differentiation within the j^lasmodium. Such a possibility has been suggested, among others, by one of us (Dr. Teacher in his studies on Chorion-epithelioma), but it is very difficult to reconcile with the characters of the spun-out and vacuolated plasmodium in the present ovum.
  2. In terms of the view expressed by Leopold and already referreil to, the broad processes seen in Peters' ovum may arise by the fi»rmation of buils of the cyto-trophoblast which grow outwards into the Tiift>3iKxlial ijitrands, much in the same way as the mesodermic processes £:r V :rTo the priinitive epithelial villi. Much of the plasmodium would "TtiBs >%f r^TUvxxI to the endothelium-like lining of the blood lacunae.

Tii^ ,'\vlAUiition also presents difficulties in view of the characters of -:t! ^u"'- 'ii;i«>m»xlium. It is difficult to picture the process imagined, and -t^.fi».«Tr- \v »w ruoliue^l to fall Kiek on a third possibility, viz. : '^^ti rhf ••vt».>" trophoblast grows out as cellular masses and •«uiniT>) -x'l ^ 'lUtt-h into tho plasmodial strands, but into the spaces

The fm-ther changes could lie conceived as follows : the greater part of the eurly ])hismo(liimi disappears after being spun out into the fine threads seen in many parts of our sections, while the marginal cells of the py to- trophoblastic columns continue to form new plasmodium. On the other hand, the plasraodium may in part pereist, the strands arranging themselves over the cytoblast eohimna as an eudothelium-Iike layer, while the outlying parts reniiiin as the irregular masses invading the dcridiia. It appears probable that the extensive plasmodium is in great part a temporary furmation provided for the early enlargement of the implantation cavity. The attachment of the ovum is effected when the columns of cyto-trophoblast reach the decidua at points from which the necrotic layer has been removed, and become fixed by the terminal cells insinuating themselves iimoug the elements of the decidua. The accompanying figure (Figure viii) presents in a diagrammatic form an ovum at a stage intermediate between our ovum and that of Peters.


The later stages of attachment and of the formation of the placenta do not here concern us, but it must be pointed out that our specimen establishes a difference, though perhaps it may be one of degree not of kind, between the processes involved in the imbedding of the ovum, and those which characterise the commencement of placentation. A distinction must therefore henceforth be drawn between the initial stages and phases of implantiition or imbedding and those of attachment or placentation.

The Age of the Ovum and Relation of Imbedding to Menstruation

Thk following chapter is of the nature of an appendix to the descriptive part of our memoir. It deals with the age of our ovum and certain theoretical considerations regarding the decidua which arise in connection with that problem. In attempting to place our specimen, in point of age, among other recorded cases we were met with so many difficulties and contradictions that it seemed worth while to ascertain whether, by applying the precise data of our own case to the other records, we could arrive at a more consistent chronological sequence than afforded by His' rule. We have therefore reviewed the literature from this point of view, and shall now give the result of our enquiries.

In estimating the age of the human ovum or embryo, it is customary to make use of the convention of His, and to rec^kon, iis the approximate age of the ovum, the interval between the date on which the first omitted period should have commenced and the termination of the pregnancy. This is fairly satisfactory from the embryological point of view when dealing with embryos of the sizes commonly found, but it conflicts not only with some facts of clinical experience, but also with the data of comparative embryology, and it fails altogether when applied to many of the very young ova which are on record.

For the calculation of the age of the present ovum we have quite precise data regarding coition. The coitus of 2nd October mny be definitely rejected as a possible factor in the case. The data [u-ovided by several of the cases which are summarised in this section make it certain that fertilization must have taken place after, and was presumably effected by, the coitus of 19th October. Any other conclusion would be so incoiisisteut with the character of the ovum itself and would involve so many gratuitous and unnecessary assumptions, that the case has the practical value of ascertained pregnancy after a single coitus. As the abortion l)egan on November 4th and was completed on November 5th the total interval between coitus and expulsion is IGi days.

While the time occupied by the spermatozoa in traversing the genital passages is not known in the human subject, the observation of BirchHirschfeld ^ proves that they reach the Fallopian tubes in large numl)ers within 24 hours, and in the rabbit the distance from the vai^ina to the upper parts of the tu])e is known to be traversed in from a I hour to 2 hours (jMilncs-Marshall). It is therefore unnecessary to allow for an interval of more than 24 hours between coitus and fertilization. In those animals in which the phenomena have been actually observed, such as the mouse and the rabbit, fertilization occurs forthwith, and although the jmssibility of its being delayed is not denied in view of other facts of com])arative embryology, there is no valid reason for supposing that it may not occur in the human su])ject immediately after the spermatozoa reach the upper third of the Fallopian tube, where it may be assumed that they normally encounter the ovum.

There are no very reliable data regarding the length of time that spermatozoa may survive in the genital passages. They have been observed in a living state after nine days, but it by no means follows that they retain fertilizing power so long. The comjiarative infre(|uency of fertile insemination during continuous (tohabitation, seems to point to there being some special circumstances connected with a successful impregnation. Were spermatozoa to retain for long their power of fertilization, no ovum should escape fertilization. It seems reasonable to assume, on the analogy of the lower mammals, that the favourable conditions are the occurrence of insemination and ovulation simultaneously or at no great distance of time. The shortness of the oestrus period in lower mammals may perhaps be looked upon as a means of attaining this end. If allowance must be made for the possible survival of spermatozoa for long ]>eriods, any attempt to establish a sequen(!e of stages is futile. We proceed therefore on the assumption that in the majority of instances insemination and ovulation nearly coincide, and for the sake of argument allow 24 hours for fertilization to occur.

  • Cited from Tioopold, P^'r7^'? luid KM^ T-.oipzi<(, lf^97.

If we now deduct from the total interval of 16i days 24 hours for the period between insemination and fertilization, there remain 15^ days.

The favourable state of {^reservation of the ovum, as already shown, precludes the possibility of its having been retained in the uterus for any length of time, and this is consistent with the fact that not more than 36 hours elapsed between the first symptoms of abortion and its completion. If it be assumed that development ceased soon aftei- the commencement of abortion, at least 24 hours must be deducted at this end of the scale, and we have left a period of 14^ days as the probable maximum age of the ovum.

If the ao^e be calculated from the data regarding menstruation it works out thus : — Menstruation should have occurred on October 25th, or 28 days after the date of the former period. Between this and the first symptoms of abortion 10 days intervened, making the minimum possible age of the ovum 10 days. If on the other hand the intermenstrual interval had equalised itself to the usual average of 26 days, the allow^ance on this occasion, seeing that the previous interval had been 25, should have been 27 days. On this basis the minimum possible age would come out as 11 days, or, according to the convention of His, 12 days. Now we know that degenerative changes commence some days before the appetirance of the menstrual discharge. It therefore seems necessary to allow some time, probably as much as 3 days, during which the fertilized ovum exercises some influence over the uterus which results in the arrest of the destructive changes. This would bring the probable minimum age of the ovum up to 13 days. Taking all the factors into account, the absolute limits of age may be stated as from 12 to 15 days, but, as will be seen later when the data regarding other recorded ova are considered, a computation of from 13 to 14 days is probably correct. Moreover, it will be found to agree very w^ell with such facts of comparative embryology as are applicable.

The chronology of the early stages of development are most accurately known in the •inase of the white ni«>a^. Kat thev tutve also been ascertained with eonsidemble exai:tne?ss in. the •^a.^et^ *>f the guinea-pig, nibbit, pig and doer. The nature of the att.urhmrnt of the ovum to the uterus and its course of develpment are :> •iLrfen^tit in the pi? and dog from what they are in man. that ct>m[<iri^>n U im^••:'«?i^iWe. ex^-ept with regard to one point whirh will h-r taken [:iter. The relatively short gesUitioii perio<l of the three Dxlent>. an«l the in«tivi*la:dity in the behaviour of the ova of difft-rent >jH.vies, «s>mpel us t«> iL-xen-i^e «?autiMn in applying the facts of their development to the elu':-i*{ati«-'n of human embryology.

In the mouse the rate of development i> known vvry ai^urately up to the 10th ilav. an»l the iv:>uh> of S»->f>'>na. Bun-khanl, Jenkinsoii, and Melissin's show no i^reat difference in n^sw^-t of rhronolouv- Fertilization occurs verv >ix»n after roitu<: bv the euil of the first 12 hours the ova are fur the mi»>t i»;irt in the twoMvlk**! ^tair-- of seirmentation ; and by the end of 4> hours they t-iaisist »f fn>m l»» i*> 24 bhistomeres. Development then proceeils more rapi«lly. and the differentiation into an outer layer and an inner trpuij* of k\\U U»iv»mes ap{»arent. The zona pellucida is still complete. Bv the eml oi the onl dav the ova have rearheil the uterus, the zona j>ellui:ida is lH\iriiiuinir to disap|.K?ar. and the morula has become a hoUow bhistocyst. The ovum now elouirates, and up to the beginning of the 5tli «lay is engageil in the piwesc^ of inversion of the germinal layers. On the 4th day im]>hintation evMumenres hy the removal of the uterine epithelium in the neighbourht^'Hl of the ovum, and fixation occurs on the 6th clay at the earliest, thouirh not until the 8th day does the ectoplacenta take form. At this |K>int, however, the lagging behind of the embryo relatively to the tmphobhist in the human ovum produces a state of affairs so different that further eom|^Kirisi>n l)ecomes impossible.

Duval indicates the 7th ilay as the jKM-iod of attachment in the rabbit, but the ovum in this case is not ci>mj)letely imluHWed, and the development of the blastocyst is ijuite ilifi'erent from that in the human subject. On the other hand, the similarity of the process of imbedding in the guinea-pig to what is theoretically probable in the human subject makes it a more favourable subject for comparison.

Von Spee figures numerous ova of the guinea-pig either free in the uterus or in process of imbedding. None of those are earlier than the beginning of the 7th day after coitus, and nearly all are in the early blastocyst stage. There is no marked change up to the end of the 7th day. At 7 days 13i hours the ovum is a hollow vesicle of considerable size, completely imbedded, and shows an enormous advance compared with ova 24 hours younger. This may be explained l)y sup})osing that whereas the growth of the ovum is relatively slight before implantation, development proceeds very quickly after imbedding is complete.

It seems improbable that the human ovum could be imbedded earlier than that of the guinea-pig, and we therefore agree with the opinion expressed by (Jraf von 8pce that imbedding in the human subject cannot take place sooner than 7 days after fertilization. By this time in all probability segmentation is complete, but the ovum has hardly increased in size, still retaining the dimensions of the unfertilized oocyte, Le, about '2 of a millimetre in diameter.

The rate of development after imbedding is more rapid in the rabbit than in the mouse, and much more rapid than in the guinea-pig. These variations are probably, in j)art at any rate, to be explained l)y the different methods of implantation. Development seems to be retarded in the early stages in forms with a, decidua capsularis, and in the human ovum the retardation of the evolution of the embryonic rudiment compared to the trophoblast is a very striking feature. If, therefore, seven days be arbitrarily allowed for segmentation and the early phases of implantation, another seven days does not seem too much to allow for the growth and development of the ovum up to the stage j)resented by our case, when we consider the rate of growth in the mouse and guinea-pig, and allow for the much longer gestation period in the human subject.

In comparing the present ovum with other human ova, it seems advisable, in order to restrict the range of error as far as possible, to consider only those which may be estimated as at most a week older than it is; therefore no ovum larger than Graf v Spee's "Gle" is included in our series. Further, only those are considered w^hich have a reasonably complete and precise history, and have been obtained under favourable circumstances, or in such condition as precludes retention in utero for more than a day or two after the cessation of development. A number of otherwise valuable young ova have therefore been omitted, namely those of Ahlfeld, Beigel and Lowe, Breuss, van Heukelom, Hitschmann and Lindenthal, Hoffmeier, Keibel (1890), KoUmann, Leopold (1906), Marchand, Rauscher, v. Spee (1905), Stolper, and Wharton Jones. The ovum No. 11 of Mall is also omitted as being pathological. There remain eleven recorded ova which meet our requirements, namely those of Peters*, Jung, Merttens, Beneke, v. Spee (v. H.), Leopold (No. 1), Reichert, Rossi Doria, Eternod, Frassi, and v. Spee (Gle). The data regarding these eleven cases will now be briefly summarised and tabulated.

For purposes of comparison in the histories and table, the age of the present ovum is taken as 13-14 days. In assessing the relative age of other young ova, the total size of the ovum, the size of the blastocyst, and where it is known the size and state of development of the embryo are all taken into account. It is probable that ova vary slightly in the rate of development of the embryo, and still more in the rate of growth of the trophoblast. Consideral^le variation in the rate of development is known to occur in certain of the lower mammals ; for example, Keibel found in the pig difterences which he regarded as equal to from 24 to 48 hours' growth, at such an early date as the 14th day of pregnancy. The computations must therefore be regarded as merely approximate ; but as the total period considered is only 20 days, and the difference between the youngest and the oldest only 7 days, they may be regarded as fairly correct.

Summary of data regarding Selected Ova

I. Ovum of Peters

Dimensions — external, about 24x1.8 mm.; blastocyst, internal, r6x 8x*9 mm.; embryo, '19 mm. The specimen was obtained at sectio a few hours after death. The fixation is very good, and the details of its structure and imbedding have been very well worked out. The patient who had borne one child menstruated for the last time on 1st September, 1895. Towards the end of September, besides other symptoms, vomiting occurred. On the menstrual discharge failing to appear, she committed suicide on the 1st October by swallowing caustic potash. No definite data as to coitus could be obtained. Although this ovum is considerably more advanced in develo})ment than our ovum, it can scarcely be regarded as more than 15 days old, in view of the facts relatincr to Merttens' ovum, and the very ojreat advance which occurs in the succeeding 7 days. Peters reckoned the age as 3 to 4 days, but this is obviously a great understatement. Estimated age, \Z\ to 14 J days.

II. Ovum of Jung

This ovum, measuring 2*5 x 2*2 x I mm. in diameter, was obtained by curetting on account of leucorrhoea with retroflexion and prolapse of the uterus, from a multipara aged 28. Menstruation was irregular, occurring at intervals of from 5-6 weeks, and the last completed period began 4 J weeks prior to the operation. The last coitus occurred 4 days before the operation. The specimen is extremely well preserved, showing numerous mitotic figures. It was fixed while still warm in 70% alcohol. It presents a very close resemblance to that of Peters, but there are more definite villi. The case has not yet been published in full detail. This ovum probably comes next in point of age among other known ova to that of Peters, with the possible exception of Graf v. Spee's ovum of 1905. Estimated age, 14^-151 days.

III. Ovum of Merttens

Dimensions — external, 4x3 mm. ; internal, 3x2 mm. It was found in curettings sent for examination to Langhans laboratory under suspicion of malignant disease — the presence of an ovum not being suspected. Only a few sections were mounted, 4 of which contained an ovum. The four sections were of practically the same dimensions, and were therefore regarded as representing the greatest diameter of the blastocyst. From the general characters of the ovum, it is distinctly, but not greatly, older than that of Peters. The villi, which averaged '364 mm. in length, are somewhat thick columns of trophoblast with short mesoblastic cores containing no blood-vessels. The curetting was performed 21' days after the onset of the last menstruation, which lasted 5 days, leaving a clear period of 16 days, and this in all probability represents the maximum age of the ovum. Merttens regarded it as much younger than 16 days. It could hardly be less than 14^ days when compared with the present ovum or more than 15^ days when its history is considered.

IV. Ovum of Beneke

The cavity of the ovum measures 4*2x2'2xr2 mm. and the embryo appears to be rather younger than that of ovum '*v. Herff" of V. Spee. It was obtained by curetting on 30th March, 1903. The last menstruation extended from 5th to 10th March. There was no cohabitation after 22nd March. the period of 8 days l)etween this and the operation is however unimportant, since by comparison with the present ovum the minimum age could scarcely be less than 1 5 days. The period of twenty days from the end of menstruation to the date of operation represents the maximum possible. Probably sixteen to seventeen days would be a fairly accurate estimate.

V. Ovum "v. Herff" of Graf v. Spee

This specimen was obtained in a decidual cast expelled by abortion 2 days after the onset of influenza, and, as we have ascertained from the author, exactly five weeks from the end of the last menstrual period. Allowing 5 days for the period, the abortion occurred 40 days after the date of hist menstruation and 12 days from the expected menstruation. The implantation cavity measured 7 x 55 mm. — the external diameter of ovum 6x4*5 mm. — bhistocyst 4 mm. — embryonic rudiment •37 mm. There were definite but very short villi. The state of preservation is extremely good. The age of this ovum may be estimated as from 17 to 18 days.

VI. Ovum No. 1 of Leopold

The blastocyst measures externally 6x65 mm., internally 4x3.7 mm. It was found in a uterus which had been excised on account of carcinoma of the cervix. The patient was a married woman, Vpara, and her periods latterly had been somewhat irregular. The last extended from 14th to 19th August, 1887. On 20th August coitus occurred for the last time, and the operation was performed on the 29th August. Leopold regarded the ovum as of the 8th day on account of the history, but it is clearly much older. It is possible that the supposed menstruation was a haemorrhage due to the presence of the tumour, but it is also possible that it represented menstruation occurring during pregnancy. The embryo in this case was not recognised, but from the size of the blastocj^st and the .state of the villi, the age may ])e reckoned as about the same as that of ovum v. H." of V. Spee, namely, 17 to 18 days.

VII. Ovum of Reichert

This ovum, the histological details of which are not very exactly known owing to the <late (1878) at which it was described, measures internally 5*5 x 3*3 mm. It was ol)tained at a sectio upon the body of a young married woman who had not previously been pregnant. Her husband left Berlin " in the first days of November," and on 7th November the expected menstruation did not appear. During the night of 21st to 22nd November she died suddenly, her husband still being absent. Reckoning from the date of the missed period, and allowing an additional three days for the arrest of the menstrual ])rocess, the age of this ovum would be 17 days. The dimensions l)eing very similar to those of the two preceding ova, it should probably be regarded as from 17 to 18 days.

VIII. Ovum of Rossi Doria

The ovum measures internallv G x 5 mm., and the decidual chamber 9x8 mm. The patient was 30 years of age, had been 8 years married, and had 5 children. The last menstruation began on 30th January, and lasted between 3 and 4 days, as was usual. On the 23rd February, after dancing at a ball, she observed a few drops of blood, which she regarded as the return of menstruation almost a week too soon. This increased during 2 days into definite haemorrhage, and there were slight pains. Lat^r she suffered very severe pains and returned for treatment on **7th February*' (surely a misprint for 27tli February?). A quantity of clots and a complete decidual cast were removed from the vagina. The ovum formed a projection like a pea. The decidua capsularis appears to have been ruptured, and the embryo is not figured. The preservation of the ovum is remarkably good. There are no useful data as to coitus. The interval between the cessation of menstruation and the first symptoms of abortion is al>out 21 days. Allowing 24 hours for the occurrence of fertilization, and iissuming that development ceased soon after the appearance of symptoms, an age of 20 days is possible, and by comparison with Etornod's ovum, it might be stated as 18 to 19 days.

IX. Ovum of Etemod

This ovum, from the point of view of chronology, is one of the most important in the literature — as we have here the absolute maximum time occupied by development to a stixge almost identical with that of ovum **Gle" of v. Spee. It is excellently preserved, and has been most carefully describe*! and figured. It was obtained from a woman ** worthy of belief" ('^digne de confiance"), w^ho, through force of circumstances, had coitus with her husband only on one occasion, namely, the night of 6th-7th November, 1891. Menstruation did not appear as expected on November 22nd. On November 28th she aborted. According to Eternod this ** allows, reckoning upon possible delay of fertilization and of days prei)aratory to abortion, of the deduction that the specimen was of the end of the 2nd or beginning of the 3rd week." The ovum was of flattened disc shape, measuring with the villi 10x8'2x6 mm. The villi measured from 1*2 to 2 mm. in length, and the embryo IS mm. in length by from •18 to '22 mm. in width. It is in much the same stage of development as ovum *'Gle" of von Spee, but is slightly smaller in all respects. On the principles which have been applied to the present ovum, we have to deduct from the total period of 21 complete days and part of the 22nd day, about 24 hours for the occurrence of fertilization and rather more than one day for the abortion, leaving a probable maximum of 19 days. The age may therefore be stated its 18 to 19 days.

X. Ovum of Frassi

The ovum which is described as of loss tliaii 10 days was obtained by hysterectomy on account of persistent meuorrhagia due to mt»tritis. Tiie patient was 40 yours of age and had had 11 children — the last 2\ years previously. Menstruation occurred regularly every 2S days. Her first period was believed to have occurred 2 weeks before the operation, but she afterwards confessed that that period had been omitted ; she had con(*ealcd the fact for fear the operation would be put off if it was known that was pregnant. She had suttered greatly in her other pregnancies. The uterus was fixed whole in :V ; formalin, and the preservation of the ovum is very good. The implantation cavity is wholly surrounded by decidua, and there is no ** cicatrix.'* There are villi all over the circumference, measuring from '5 to TO mm. in length, but they are best developed round the sides, and least over the apex of the vesicle. Blood-vessels arc seen in the base of the cord, but there are none in the villi. The implantation chamber measures 13x5 mm., and the cavity of the ovum i)'4x3'2 mm. The embryonic rudiment measures l*17x"G mm., and it is regarded by Frassi is slightly younger than that of the ovum "Gle" of Spee. The ovum is very similar to that of Eternod, and may be considered to be of the same age, namely, 18 to II) days.

XI. Ovum "Gle" of Graf v. Spee

This was a perfectly fresh aborted ovum in a complete decidual cast, expelled 5 weeks after the frrnutHtflon of the last menstruation. The preservation of the specimen is excellent, numerous karyokinetic figures being recognisable. The patient had been iieallhy, menstruating regularly every 4 weeks, and the abortion was regarded as the return of the period a little after its usual date. The internal dimensions of tlu* chorion are 10 X 8'5 X 6*5 mm., and the embryo measures l\^)4 mm. The stage of development of the embryo is slightly more advanced than that of KtiM'nocl, and the age may be stated as 19-20 days.

If the foregoing summary of the facts regarding these eleven ova be analysed, it will be found that the most precise data are supplied by the cases of Merttens, Beneke, Rossi Doria, Eternod, and Reichert. In the case of Merttens 16 days, in that of Beneke 20 days intervened between the end of last menstruation and operation ; in the case of Rossi Doria 24 days ela})sed between the beginning of the last period and the commencement of symptoms of abortion ; and in Eteruod's case the ovum was expelled 21 days after a single coitus. Eternod's case supplies us with a fairly precise upper limit, on our method of estimating, and at the same time justifies us in excluding the possibility that in the first three cases fertilization occurred before meustruation. The case of Merttens may be taken as defining the maximum possible age of an ovum slightly older than that of Peters. The circumstan(.'es of Reichert's case were very similar to those of our own ; fertilization in all probability occurred shortly before an expected period which was arrested.

By utilising these data we can arrange the ova in a series in which dimensions and ages gradually increase. The results of the rule of His applied to these ova are brought out in the column in the table under the head of '* days elapsed from the omitted period." It will be seen at once how contradii'tory the results are in the cases to which it can be a])plied, and it will be noticed that it is not applicable at all to five cases out of the tw^elve. This is of course due to the admitted error of three weeks under the rule. In dealing witli these young ova it seems therefore necessary to proceed on the basis of some such *' normentafel'* as here given, and we submit our table as a tentative in that direction. That it is absolutely correct is not pretended, but that it is fairly correct in a relative sense is probable, because it is consistent in itself and with the data of comparative embryology.[1]

Having now obtained a chronological secjuence in which the age is calculated from fertilization, a second table was constructed, to see at what periods of the month fertilization and imbedding would have taken place in these twelve ova, on the basis of the first table. The termination of pregnancy being known, the date of fertilization was arrived at by subtracting the age of the ovum in each case. As a definite figure was required the higher figure was arbitrarily selected in each instance, allowance being made for the various circumstances under which the ova were obtained. The date of imbedding was fixed by assuming that seven days ehipse between fertihzation and imphintation of the ovum in the endometrium.

The second table is the complement of the first, and it shows that, if the ages of the ova be correct, fertilization may occur at any time during the intermenstrual interval, and that imbedding may take place either in the period of quiescence or in the period during which, without the occurrence of pregnancy, the premenstrual and menstrual changes would have been progressing.

Table of Selected Ova

Bryce1908 table01.jpg

Table I. Chronological Table of twelve recorded early pregnancies. the table is constructed on principles explained in the text. Fertiliziition is assumed to be effected about 24 hours after insemination, and 24 to 48 hours is allowed for the completion of abortion. The leading data are supplied by the histories of Nos. 1, 4, 5, 8, 9, 10, and the position of the remainder is adjusted according to their dimensions and state of development. The ages, according to the convention of His, are shown in the column headed "Days Elapsed from Omitted Period."

Table showing Periods of Fertilization and Imbedding

This result of our seriation of these early cases is not consistent with the older views regarding menstruation and its I'clations to imbedding, for it carries with it the conclusions that the menstrual decidua is not a jjreparation for the reception of an ovum (in the old sense of the words) ; that menstruation is not an abortion of an unfertilized ovum ; and that ovulation does not necessarily coincide with menstruation. The conclusion is not inconsistent, however, with the recent views on menstruation as developed by Heape, and supj)orted by a considerable numl)er of comparative researches. It is this consideration which gives weight and interest to our argument.

It is now generally admitted that the menstrual cycle in man and monkeys is homologous with the oestrus cycle of the lower mammals. The oestrus cycle is divided by lL'ap(» into pro-oestnnn, oestrus and dioestrum, and this division has been confirmed for many manjmals by his own researches and those of F. 11. A. Marshall. During })ro-oestrum the generative organs of the female show signs of special activity, such as swelling of the vulva, coloration or flushing of the surroundings, and a discharge of l)lood or mucus from the vagina This is immediately followed by the ** oestrus," or period of desire," during which alone the female is capable of impregnation and will receive the male. If pregnancy does not occur, oestrus, after a brief space in which desire subsides (metoestrum), is succeeded l)y a period of quiescence or dioestrum, which lantB till |)ro-oe8trum again sets in. In polyoestrous mammals several cycles of this kind may follow one another. Menstruation in the human feniale is homologous with pro-oestrum, as first pointed out by Heape. Though there is no fixed ** period of desire ' there is an indication that a vestige of this persists, in the fact that a phase of more pronounced oestrus commonly suf^ceeds the cessation of menstruation.

The accompanying table, which is borrowed from Heape's papers, but has be(».n slightly modified and adapted to our present purpose, will cx|)lain at a glance the general relation of the oestrus cycle of a polyoestrous mammal to the menstrual cycle.

Table of Menstrual Cycle

Bryce1908 table02.jpg

Table II. Showing the relation of the dates of fertilisation and imbedding to the menstrual cycle calculated from the data given in Table 1. The higher figure in the age column is arbitrarily chosen in each instance, and allowance is made for the special circumstances of each case. The Roman numerals indicate the first days of two successive menstrual periods.

Bryce1908 table03.jpg

Table III. Similar to No. 2 illustrating the division of the menstrual cycle of the human female and its homology to the oestrus cycle of a oestrous mammal unaffected by pregnancy, modified from Heape. The chronological division of the cycle is founded on that of Milnes-Marshall, 1893. The complete cycle is shown extending from the 10th day of one lunar month to the 10th day of the next.

It is quite clear from the results of the comparative method that the signiticance of menstruation does not lie in the mere periodic growth and sul)se(|uent destruction of the mucous membrane, but in the cycle as a whole. Tlui csseiKie of the process is not the preparation of a menstrual decidua, but the formation of a new endometrium ; in other words, the growth and swelling which precedes "discharge," does not represent a preparation for the ovum ; it is merely a phase in that reconstitution of the new endometrium which is the real preparation for the reception of the ovum. In this case of most lower mammals, the generative organs lie dormant through a large part of the year, and when the breeding season approiiches, the endometrium undergoes in the pro-oestrum a sj»eciea of regeneration resulting in the development of a new surface on which the ovum may implant itself. In the case of the human female there is no longer this regular breeding period, but desire and the possibility of impregnation occur at irregular periods all the year round. Instead therefore of an annual (or in some mammals a seasonal) renewal of the endometrium, and the preparation of a new surface on which the ovum may be engrafted, we have in the human subject a mostly regeneration and preparation of the endometrium for the same purpose.

Much of the difficulty regarding the decidua disappeared when it was shown that it had no part in the formation of the placenta. The ?? performed by the decidua in the imbedding of the ovum require the connective tissue elements shall be in a condition of special activity. The cellular "primitive" structure of the eudometric stroma permits that special and ready reaction which constitutes the conservative function of the decidua, and by a recurrence of the menstrual cycle the mucous membrane is maintained in the necessary condition of immaturity.

Accepting this interpretation of menstruation our Table II no longer presents the difficulties which it presents to the ohler views regarding the process. In the lower mammals ovulation takes place apparently during oestrus or in late pro-oestrum, and in most animals fertilization of the ova occurs forthwith. It is quite possible that in the human subject considerable delay may occur between insemination and ovuhition on the one hand, and between insemination and fertilization, or the actual union of the male and female element on the other, as there is no fixed period corresponding to heat." But it seems reasonable to suppose that the most favourable condition for successful impregnation lies in the occurrence of insemination and ovulation simultaneously, or at no great distance of time, and the occurrence of fertilization immediately the two elements meet, just as in the lower animals.

Looking now at Table II, it is apparent on these premisses that fertilization of the twelve ova concerned must have taken place at wide intervals during the month, and comparing it with Table III it will be noticed that the first three must have been fertilized while repair was possibly not yet complete, and the next three clearly during the period of quiescence. Von Spec's two ova may, and the last four certainly must, have been fertilized in the period of swelling, if we assume that the sw^elling continues in spite of the occurrence of fertilization.

Most probably also ovulation corresponded more or less closely with fertilization, and occurred at intervals throughout the month, perhaps not even excepting the latter days of the actual period of menstrual discharge. A further inference is that ovulation and menstruation may, but do not necessarily, coincide. This is supported by the researches of Bland-Sutton on menstruation of Macaque monkeys and baboons, and of Heape on menstruation in Semnopithecus entellus and Macacus rhesus. They have shown that in these animals menstruation and ovulation do not necessarily coincide. Further, it is supported by the observations made by gynaecological surgeons in the course of operation, viz., that apparently ripe or recently ruptured Graafian follicles are found in the ovary at any point of the intermenstrual period, and that, on the other hand, there is frequently no trace of either ripe or recently ruptured follicles immediately before or immediately after the menstrual period. It would therefore appear possible that the disappearance of a fixed oestrus in the human female has been associated with scveran(*e of ovulation from oestrus, but it is probable that the most fiivoural)le circumstances for successful impregnation, are ovulation about the period of menstruation, and insemination shortly before or after the period of discharge.

On the other hand it is absolutely certain that if the table be correct, imbedding occurs quite independently of the menstrual growth, and at any period in the cycle save the destructive phase.

In the menstruation of monkeys, according to Heapc, ** the stage of growth" is characterised by, firstly, swelling of the endometrium, proliferation of the inter-glandular tissue elements, congestion, and later an increase in the number of the leucocvtes visible in the blood-vessels. In the first stage of degeneration all those phenomena become more pronounced, and extravasation of blood takes place into the tissues with migration of leucocytes. In other words, the phases represent the early stages of that reaction of the tissues in response to stimulus which is called inflammation. In the later stages the degenerative changes in the tissue elements, the haemorrhage, and the disintegration of the membrane which follows, constitute a form of ulceration of the endometrium which can scarcely be characterised as other than severe. The discharge contains red-blood corpuscles, masses of uterine stroma, fragments of uterine epithelium, squamous cells from the vagina, leucocytes and blood-clot in somewhat varying amount ; in short, the elements are the same as are found in a human menstrual discharge, where the reaction has been intense, and accompanied by the shedding of membrane.

The swollen endometrium is described by Heape as menstrual decidua, following the usual custom of obstetricians and pathologists. The decidua of pregnancy, so far as we can judge from the limited amount of material belonging to the early stages which we possess, is very similar in structure to menstrual decidua; in fact many authorities deny the possibility of distinguishing between them, and doubtless the processes concerned in their formation are similar. But although the processes are similar, it does not follow that their purpose is identical. As already pointed out, it is apparent from the tables that the ovum can imbed itself during the period of quiescence of the endometrium, and the case of ovarian pregnanc)'^ described in the second part of this memoir demonstrates that the presence of decidua is not necessary to iml>edding. In the mouse and guinea-pig there is no change in the endometrium until it is attacked by the ovum. In the human subject, no doubt, in certain instances, menstruation is arrested and degeneration is checked before imbedding takes place ; but it is not essential that actual development of the endometrium into decidua should precede its invasion by the ovum. Indeed, the analogy of the guinea-pig and other mammals makes it probable that the development rapidly takes place immediately after the invasion, and that it may in the first instance be of local occurrence only. This apparently is what occurred in the case of Peters, where a definite decidual reaction is said to be limited to the immediate neighbourhood of the ovum. On the other hand, should an ovum be fertilized after the premenstrual changes have set in, there is no reason to suppose that they would offer any obstacle to its implantation ; indeed, as the early changes in the formation of the menstrual swelling are identical with those of decidua formation, it might be favourable to the process. We have no evidence, however, to show whether fertilization may not completely arrest the whole menstrual changes.

Menstruation on the premisses detailed above is thus a cyclical process which provides for the maintenance of the endometrium in a suitalile condition for producing the decidua of pregnancy, but the ovum is not dependent upon the development of a menstrual decidua for a suitable nidus in which to become imbedded. The uterus is capable of developing a decidua of conservation whenever its integrity is threatened by the urowinff ovum.

  1. The table does not affect the ordinary obstetric reckoning, the basis of which is statistical.

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)


Bryce, T.H. & Teacher, J.H. (1908). An Early Ovum Imbedded In The Decidua. Glasgow: James Maclehose and Sons.

Bryce, T.H., Teacher, J.H. & Munro Kerr, J.M. (1908). An Early Ovarian Pregnancy. Glasgow: James Maclehose and Sons.

Cite this page: Hill, M.A. (2019, April 25) Embryology 1. An Early Ovum imbedded in the Decidua (1908). Retrieved from https://embryology.med.unsw.edu.au/embryology/index.php/1._An_Early_Ovum_imbedded_in_the_Decidua_(1908)

What Links Here?
© Dr Mark Hill 2019, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G