Paper - A presomite human embryo (Shaw) with primitive streak and chorda canal with special reference to the development of the vascular system

From Embryology
Revision as of 09:27, 4 January 2017 by Z8600021 (talk | contribs) (Created page with "{{Header}} {{Ref-GladstoneHamilton1941}} {| class="wikitable mw-collapsible mw-collapsed" ! Online Editor   |- | 90px|left The author suggests an "...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Embryology - 30 Mar 2020    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)

Gladstone RJ. and Hamilton WJ. A presomite human embryo (Shaw) with primitive streak and chorda canal with special reference to the development of the vascular system. (1941) Amer. J Anat. 76(1): 9-44.

Online Editor  
Mark Hill.jpg
The author suggests an "estimated age is 31 days", that would be Carnegie stage 14. The appearance of the upper and lower limb buds are more like a later Carnegie stage 16 embryo occurring in week 6, 37 - 42 days, CRL 8 - 11 mm.
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) 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, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

a presomite human embryo (shaw) with primitive streak and chorda canal with special reference to the development of the vascular system


Department of Anatomy, St Bartholomew’s Hospital Medical College


THE embryo was presented to the Anatomy Department of the Medical College of St Bartholomew’s Hospital, London, by Dr Wilfred Shaw. In appreciation of the gift we have associated the name of Dr Shaw with this embryo. ‘

The embryo shows many features of interest in the development of the angioblastic tissue and blood cells. These features have been described at some length, since they represent an intermediate stage of development which throws considerable light on the relations between angiogenesis and blood formation. In the chorionic mesoderm, including that of the villi, the out- standing feature is the presence in the mesenchyme of angioblastic strands of nucleated protoplasm which, by vacuolation, are becoming transformed into sinuses lined by endothelium and capillary vessels, of which the majority appear to be empty and contain at this stage of development no cellular elements. In the yolk sac haemopoiesis dominates and angiogenesis appears to be secondary. This intermediate stage bridges over a gap in the history of our knowledge of the development of the vascular system in the human subject and confirms the opinion of Streeter (1920), Ingalls (1921), and other authors with regard to the early and advanced stages of angiogenesis in the chorionic mesoderm as com- pared with blood formation, while in the yolk sac the blood islands often appear before the endothelial walls which are subsequently developed around them.

A short description of this embryo was published by Dr Shaw in 1932, and the following account of the clinical history of the case is based on his original article. ’

The embryo was obtained at a subtotal hysterectomy performed by Dr Shaw, on a woman aged 32, for the relief of a right cornual uterine myoma. The uterus was opened immediately after removal. A small prominence, about 1 cm. in diameter, was observed on the posterior wall. This was excised and fixed in Camoy-sublimate solution for 1 hr. and then passed through alcohol, xylol, benzene and wax in the usual way. Dr Shaw cut a complete series of sections of the whole block of tissue in which the embryo was embedded, at 10 p. thickness. The histological condition of the specimen proved to be ex- ceptionally good. and the embrvo was neither diseased nor macerated. 10 R. J. Gladstone and W. J. Hamilton

The woman had had two previous pregnancies. One child had been born at term in 1922, and this was followed by menorrhagia, but the sequence of the menstrual cycle was unaltered; in 1926 the patient had a miscarriage during the fourth month. On 17 February 1930 she came to St Bartholomew’s Hospital complaining of an abdominal tumour and abdominal pain. She had last menstruatedl between 12 and_l7 January 1930. Her menstrual period was one of 28 days; the next period should accordingly have started on 9 February, 8 days before she came to the hospital.

Rising from the pelvis was an abdominal tumour 4 by 3 in. in diameter, which on bimanual‘ examination was located in the right comu of the uterus. The patient was admitted into the hospital, and the operation of subtotal hysterectomy was performed 2 days later, on 19 February. The left ovary was removed and found to contain a corpus luteum.

In discussing the data bearing upon insemination, ovulation and fertili- zation, Dr Shaw ascertained that the probable date of insemination was 24 January. A discrepancy in the history occurred between the patient’s assertion that the February period was due on the 12th and the calculated period 9 February, when the period should have appeared in conformity with her normal period of 28 ‘days. The dating of the specimen, calculated from 19 February, the date on which the operation was performed, is as shown in Table 1.

Table 1 Number of days preceding the Accredited dates of menstruation, insemination, ovulation, operation date, and assumed age of embryo 19 Feb. From onset of last menstruation (12 Jan.)' 38 From assigned date of insemination (24 Jan.) 26 From assumed period of ovulation (25-28 Jan.) 25 to 22 From calculated date of expected menstruation (9 Feb.) 10 From onset of next menstruation expected by patient (12 Feb.) 7 Assumed ‘fertilization age’ of embryo 23 to 20

Assuming that fertilization took place within 48 hr. of ovulation, the age

'of the embryo, according to the data given above, would be from 20 to 23 days.

Presomite embryos belonging to group E of Bryce’s classification (1924) orto groups V or VI of Hertig’s classification (1935), with which this -embryo corresponds, namely, embryos with a completed notochordal process‘(head process of many authors) and chordal canal (archenteric canal of many authors), but before the development of neural folds, are usually thought to be younger than this, e.g. the Heuser (1932) presomite embryo (Carnegie, no. 5960) has an estimated ‘ovulation age’ of 18-19 days, as contrasted with that of the present Shaw embryo of 20-23 days.

In discussing the relation of ovulation to fertilization, Dr Shaw, after stating that the general belief is that ovulation is limited to the intermenstrual phase of the cycle, mentions that, in his opinion, ‘ovulation is limited to A presomite human embryo 11

between the 13th and 16th days of the menstrual cycle, the first day of the period of bleeding being taken as the first day of the cycle’, and that, ‘in the human subject it is uninfluencedi by such external factors as coitus’ (see also Shaw, 1925).

Further, he stated that, in his view, ‘since it is obvious that fertilization cannot precede ovulation, the union of spermatozoon and ovum must be regarded as being restricted to the latter half of the menstrual cycle, but this view does not imply that insemination before the date of ovulation must be sterile, there being much statistical and clinical evidence that coitus may be fertile during the first half of the menstrual cycle (Siegel, 1916). ‘

Allen et al. (1930), who have recovered, living ova in washings from the uterine tube, and who have studied the young corpora lutea, believe that ovulation takes place on or about the 14th day of the cycle. The recovery of ova is the most convincing method of demonstrating that ovulation has occurred.

The work of Knaus (1934), Venning & Browne (1937), and others confirms the view that ovulation occurs around the 14th day of the cycle.

It is possible that this difference in the degree of development of the Shaw embryo, as compared with others, may be explained by retardation of develop- ment due to the existence of a myoma of the uterus, although, judging from the sound condition of its tissues, the embryo itself does not seem to have been diseased. Other possibilities are: (a) ovulation had occurred at a later date in the cycle than between 25 and 28 January (see Flew, 1941), (b) a delay in fertilization had taken place—in other words, there was an increase in the period between ovulation and fertilization, and (c) the date of fertile in- semination was not 24 January, as the patient admitted. that coitus interruptus had been frequently practised during January and February. '


The chorionic vesicle which is embedded in the compact layer of the decidua is almost completely covered by a decidua capsularis. "It belongs to group E of Bryce’s classification (1924«)——in which a primitive streak and cloacal membrane are differentiated and a complete chorda canal has been formed; in the case of the Shaw embryo the canal appears to be a blind diverticulum. The intrachorionic rudiment (amnio-ectodermic and yolk-sac vesicles) is sur- rounded by the chorion with its villi and has the embryonic plate included within theeopposed walls of the two vesicles of which the yolk sac is the larger. The intrachorionic rudiment is connected with the internal aspect of the wall of the chorionic vesicle by a connecting stalk which consists of two parts, a basal ‘amnio-embryonal’ segment, and a proximal ‘umbilical’ segment con- taining an allanto-enteric diverticulum (Text-fig. 3).

The principal measurements of the specimen are given in Table 2. 12 R. J. Gladstone and W. J. Hamilton

Some of these measurements have been made directly from the sections mounted on the slides, others have been estimated indirectly by counting the number of sections (10 p. thick) in which a particular part appears in the series. Owing to shrinkage and the obliquity of the sections to the longitudinal axis

of the embryonic plate the measurements must be regarded as approximate

(see Text-fig. 1). .In the case of an elliptical disc obliquity of the sections relative to the axis-will cause a diminution in the estimated length of the longitudinal axis, and an increase in the width, which will be proportional to the degree of deviation of the plane of the sections from that of the embryonic disc; the estimation will also be influenced by the difierenceybetween the length of the longitudinal or axial plane of the ellipse and the width of the ellipse.

Table 2 ‘ Approximate measurements ' mm. Maximum horizontal diameter of infiltrated zone around the 15 implantation cavity Depth of infiltrated zone, including the closing plate 5 Chorionic vesicle: maximum external diameter, including villi 11 Chorionic vesicle: vertical external diameter 4-04 Chorionic vesicle: maximum internal diameter . 8 Chorionic vesicle: vertical internal diameter 3 Embryonic rudiment » 1-66 x 1-34 Length of embryonic disc 1~o5 Body stalk 0-61 Amnion 1-35 x l~34 x 0-32 Yolk sac 1-21 x 1-35 x 040 Allanto-enteric diverticulum 0-35 Cloacal membrane 0-09 Definitive _ 0-05 Caudal 0-04 Primitive streak ' 0-43 Preblastoporic segment of disc 0-62 Hensen’s knot 0-04 Notochordal process A 0-17‘


Another well-known circumstance which renders the measurement of embryos difficult is the development of the normal sagittal and transverse curvatures of the embryo, andflthe occasional presence of abnormal twists and tilts: Towards the end of the presomite phase of development the ventral

flexure of the caudal end of the embryo has a marked effect upon the estimate‘

of the length of the primitive streak, and at a. later period when the cephalic flexure is being formed estimation of the length of the embryo by the simple method of enumerating the number of sections is still more profoundly affected. In the specimen under consideration, the ventral bend of the caudal segment of the primitive streak renders any measurement of its length, except upon a model, extremely difficult, since a very considerable extent of its total length is included in two or three sections. This is due to the plane of the sections which pass through the terminal ventrally flexed part of the primitive groove A presomite human embryo 13

being somewhat in the longitudinal axis of the groove, instead of transversely as in its anterior part.

In many specimens, owing to twists around the longitudinal axis and bends in a lateral direction, measurements based on linear reconstructions are also liable to be fallacious. An indication of the amount of error that may occur by simple enumeration of sections, even in a perfectly ellipsoidal or approximately

Text-fig. 1. AOBD is an ellipse representing the outline of an embryonic disc in which it is assumed that the primitive streak and groove occupy the lower half of the major axis of the ellipse, and that the serial sections have been cut at right angles to the major axis. A’0’B’D’ represents an embryonic disc of corresponding size and form the sections of which have been cut at an angle of 30° with the major axis of the ellipse. The maximum length of the disc, if calculated from the number and thickness of the sections which include the major axis, would be shortened by 0-22 mm. and the primitive groove would also be proportionately shortened by 0-11 mm. Y in the interrupted line X Y Z represents the position in which the primitive groove would appear in relation to the edges X, Z of the embryonic disc ina section cut in the position which is indicated in the figure. .

circular embryonic disc, will be gained by a reference to Text-fig. 1, and also the advantage which is gained by an attempt to make a reconstruction model will be appreciated, even though an orientating or basal line may be absent. In the present case, the absence of a base-line was overcome by making use of

a portion of the wall of the chorionic vesicle in the region of the embryonic rudiment. 14 R. J. Gladstone and W. J. Hamilton

The choriomlc vesicle The chorionic vesicle is flattened in such a way that its major axes lie parallel to the membrane lining the posterior surface of the uterine cavity in which the vesicle was embedded. The external surface of the sac is covered throughout its whole extent by villi, and these are more differentiated and

larger at the circumference than at the embryonic pole (future placental site) or at the abembryonic pole (next the lumen of the uterus). '

The epithelial layer of its wall shows a distinct subdivision into cytotro- phoblast and syncytiotrophoblast '(Pl. 3, figs. 7, 8). The extent and relations of the latter to the uterine mucosa will be described in a subsequent publication. The cytotrophoblast has the usual characters of this layer in embryos at this stage of development. .Mitotic figures are occasionally visible in it, and con- nexions with the underlying mesenchyme and angioblastic strands appear to be present; these are discussed in the description of angiogenesis and haemopoiesis.

The chorionic mesoderm consists of a thin stratum of mesenchymal tissue, composed of a network of laminated fibres (Pl. 3, figs. 7, 8), with nuclei at the nodes of the reticulum and enclosing intercellular tissue spaces. Within the mesenchymal tissue there are also strands of angioblastic tissue, ‘many of which contain vacuoles, which by fusion give rise to intracellular spaces; the latter eventually, by further fusion and extension, become the lumina of capillary vessels and sinuses.

The spaces are, therefore, of two types: ( 1) intercellular and (2) intracellular. The intercellular spaces are irregular in form and represent the meshes of the network (Pl. 3, figs. 7, 8). They contain occasional free cells of ailymphoid or phagocytic type. The intracellular spaces at this stage of development usually

contain no formed elements except near the attachment of the connecting '

stalk (Pl. 4, fig. 9). During life they are probably filled by embryonic plasma, secreted by the endothelium. The mesenchymal tissue, on its internal aspect, is limited by a flattened mesothelial stratum which is in relation with the magma reticulare. The mesothelial stratum, \however, is in some places incomplete, and the mesenchymal tissue becomes continuous with delicate syncytial strands of the magma reticulare. The latter contains, in addition to the proto- plasmic strands, groups of cells of both mesodermal and epithelial types. There are also isolated cells showing various stages of degeneration, and some free pycnotic nuclei. The epithelial cells appear to be of both entodermal and ectodermal origin. Some of the formerare found in close relation with the yolk sac and possibly may have originated from a yolk-sac diverticulum or stalk, while the ectodermal cells are found in the vicinity of the body stalk and amnion.

The embryomic disc The embryonic disc is 1-05 mm. in length and 1-84 mm. in breadth. In- cluding the body stalk the length of the whole embryonic rudiment is 1-66 mm.

The primitive streak part of the disc is 0-43-mm. in length and the preblasto- poric segment is 0-62 mm. A presomite human embryo - T 15

The disc is somewhat oval in outline (Text-fig. 2). In the coronal plane it_ is curved with a slight dorsal convexity, but at the margin of the disc the con- vexity is replaced by a slight concavity (Text-fig". 4).

In'the median axis of the disc, in its posterior.part, there is a distinct primitive groove which overlies the primitive streak "(Text-fig. 4). A blastopore is situated at the front part of the primitive groove, and in front of the blaste- pore there is an elevation produced by its anterior lip. The disc in front of this dips away in a. moderately steep slope to the anterior edge (Text-fig. 3).

\ i “"4vN;§!7:i§$ULm'

Text-fig. 2. Wax plate reconstruction of the embryo made at a magnification of 100 diameters and reduced to a magnification of 50 in the reproduction. Viewed from above and in front.

The ectoderm is composed of a pseudo-stratified layer of tall cells (Pl. 2, fig. 4 and‘ P1. 8, fig. 6) which become flattened at the edge of the disc, where they are continuous with the ectodermal cells of the amnion (Text-figs. 4-6); they have a distinct basement membrane except in the regions of Hensen’s knot and the primitive streak. The nuclei of these cells are round or oval and contain a distinct nucleolus.’ Numerous mitotic figures are present in the ectodermal cells of the ‘primitive groove and near its margins. These P“ s- § 8‘ § §. § 5 m §. §

Text-fig. 3. Idealized reconstruction of 9. median sagittal section of the embryo showing the right half of the embryonic A presomite human embryo 17

proliferating ectodermal cells give rise, on each side of the primitive streak, to intra-embryonic mesoderm. Towards the anterior part of the disc the number of nuclear layers in the ectodermal cells becomes fewer—-compare Text-figs. 7-9 with Text-fig. 4 and P1. 3, fig. 6.

In the region of the disc in front of Hensen’s knot the surface of the ectodermal epithelium is irregular and some of the cells appear to be under- going degenerative changes (Pl. 2, fig. 4). Whether this is the result of tension in a rapidly growing area or the result of fixation cannot be determined.

.At the posterior part of the primitive streak there is great proliferation of the cells of the ectoderm. Some of these cells contribute to the formation of the mesoderm of the connecting stalk and pass around the side of the cloacal membrane into the stalk (Pl. 3, fig. 5). '

There is no evidence yet of the limitation of the medullary plate, nor is there as yet any sign of the formation of the medullary groove.

Cloacal membrane

This area is represented by the fusion of ectoderm and entoderm behind the primitive streak. It involves the posterior part of the embryonic disc and the proximal portion of the allanto-enteric diverticulum. The extent of these

two parts is seen in Table 2 and is shown schematically in the graphic recon- .

struction (Text-fig. 3). The ectodermal cells of the embryonic disc and amniotic wall which form the ectodermal part are cuboidal or columnar and have rounded nuclei. The entodermal cells which come into contact with the ecto- derm form a somewhat conical mass; their nuclei are oval in shape (Pl. 3, fig. 5). The mesodermal cells which border the sides of the membrane are separated from the entodermal cells.

Primitive streak and mesoderm

The extent of the primitive streak is given in Table 2 (see also Text-fig. 3). The mesoderm cells pass laterally from the sides of the streak as a layer which divides into two at the edge of the disc (Text-fig. 4). The amount of meso- dermal tissue is much greater at the posterior half of the disc where it forms a compact mass of cells, than in the anterior half where it is only represented by scattered cells and does not form a continuous sheet (compare Text-fig. 4 with Text-figs. 8, 9). Whether any mesoderm arises from the side of the chordal process could not be determined owing to the obliquity of the sections. The entoderm forms a distinct layer below the mesoderm except in the middle line below the primitive streak where it is intimately related to the mesodermal tissue.

Hensen’s knot and chorda canal

There is great proliferation of the ectoderm in the region of Hensen’s knot which is raised above the general level of the ectoderm (Text-fig. 3 and P1. 2, fig. 4). The posterior edge of the knot is sharply delimited from the anterior extremity of the primitive streak. A blastopore leads into a chorda canal; at

Anatomy 76 2 Text-fig. 4. General .view of the embryonic disc at the anterior part of the primitive streak and groove. Blood islands in an early stage of development are seen to the left and right; in the centre, endothelial-lined spaces, some of which contain free blood corpuscles mostly of primitive erythroblast type. Section no. 57, the sections having been numbered from the anterior end of the embryonic disc. x 100. .

Text-fig. 5. General view of the embryonic disc in the. region of Hansen's knot; the section is to one side of the blastopore (compare with Pl. 2, fig. 4). Vascular spaces containing a few fi'ee developing blood corpusoleslare visible in the lower part of the section; they are lined on their distal aspect and sides by endothelioid cells with rounded or lens-like nuclei; on the central side, however, the mesothelial layer is often defective and the wall of the space is formed by entoderm. Section no. 47. x 100.’ A presomite human embryo. . 19

its commencement the latter is a cavity, but farther forward it is only a potential space, and its anterior end appears to be solid (Text-figs. 3, 5). Thelcells in the roof of the canal are tall, columnar, have a clear cytoplasm, and their boundaries are mostly well defined. The nuclei of the cells are deeply stained and are at, or, near, the base of the cells. The cells of the floor of the canal are intimately connected to the underlying entoderm, which~in this region does not appear as a distinct layer (Pl. 2, fig. 4). The chorda canal, as far as can be determined, does not communicate with the yolk sac. Owing to the obliquity of the sections it is not possible to determine accurately the anterior end of the notochordal process, nor is it possible to state whether mesoderm arises from the sides of the notochordal process.

Prochordal plate

In the present specimen, the prochordal plate is irregular in shape, two horns projecting backwards from its posterior edge on each side of the anterior end of the notochordal process. The plate is not sharply delimited from the surrounding entode_rm and it is composed, for ‘the most part, of cuboidal cells in some of which are chromophilic granules (Pl. 3, fig. 6). Dorsal to the plate, in some sections, there is some cell detritus and many chromophilic granules. The plate is separated from the ectoderm, and only a few mesodermal cells are interposed between it and the ectoderm.

Connecting‘ stalk

Following the description of Florian (1930), the connecting stalk may be divided into two parts, the amnio-embryonal stalk and an umbilical, or body,

stalk. The latter makes an angle of 80° withpthe embryonic disc. The whole .

connecting stalk is composed of a mass of mesenchymal cells, projecting into which are the allanto-enteric diverticulum and a diverticulum from the dome- caudal part of the amnion (? amniotic duct) (Text-figs. 2, 3). The mesenchymal tissue at the base of the stalk (Pl. 4, fig. 9) consists of a wide-meshed reticulum containing intercellular spaces and enclosing angioblastic strands and de- veloping capillary vessels; in the region of the allanto-enteric diverticulum (Pl. 2, fig. 3) the cells are rounded and have an oval or spherical nucleus. The cells are thus more compactly arranged around the allanto-enteric diverticulum than in the more basal part of the stalk. The allarito-enteric diverticulum arises from the right side of the yolk sac. This is probably due to the fact that the embryo has been twisted upon the stalk as is so commonly found in young embryos. The umbilical stalk is bounded in front by the amniotic ectoderm. There is some irregularity in that part of the amniotic wall which comes into

.contact with the mesoderm of the stalk. Some of the amniotic ectodermal

cells show degenerative changes, and some have become freed into the amnion. The appearance is similar to that described by Florian and, suggests the en- largement of the posterior part of the amniotic cavity at the expense of the

2-2 - 20 y R. J. Gladstone and W. J. Hamilton

stalk tissue. Chromatic particles are present in some of the ectodermal, meso- dermal, and entodermal cells in the anterior part of the stalk (Pl. 2, fig. 3). Well-formed blood vessels and blood islands are present in the stalk; their appearance is described under angiogenesis and blood formation.

Yolk sac and allanto-enteric diverticulum

The measurements of the yolk sac are given in Table 2. There is an infolding of the abembryonic part of the sac (Text-figs. 3, 6), and in our opinion this is an artefact. Further, the sac does not lie symmetri~

.cally below the embryonic disc but more to the right side of it, possibly as

the result of the twisting of the upper part of the embryo (Text-fig. 2).

.“"V’.‘.‘..,k uuwv cowpg.-ug.§.~“.‘ v

D aw” . . ‘we.

' Text-fig. 6. General view of the embryonic disc showing the relations of the thickened entodermal

cells forming the prochordal plate. Typical blood islands are seen in the wall of the yolk sac on the left side of the photograph. Vascular spaces are visible in the lower part of the figure. They contain a few free corpuscles, or are quite empty, vasculogenesis being more advanced in the abembryonjc region than at the circumference. Section no. 31. x 100.

At the anterior end of the embryo there is an infolding from the front and, to a lesser extent, at the sides of the wall of the yolk sac to form what we interpret as the foregut (Text-figs. 7-9).

_ At the posterior end of the embryo there is a. folding of the posterior part of the primitive streak and embryonic part of the cloacal membrane to form the beginning of the hindgut (Text-fig. 3). The allanto-enteric diverticulum A presomite human embryo T 21

Text-fig. 7. Section through the anterior part of the" embryonic rudiment showing amnion, embryonic plate, and wall of the yolk sac. The wide cavity beneath the embryonic plate is believed to be the foregut. The mesoblastic tissue between the embryonic ectoderm andthe entodcrm is here scanty and in the section appears incomplete, this section lying in the pre-_ sumptive region of the buccopharyngeal membrane. The mesenchymal cleft to the left in the figure, and the enclosed space in the corresponding situation on the right, correspond in position to die future pericardial cavity. Apart of the anterior wall of the yolk sac is seen in the lower part of the photograph. Empty spaces lined by endothelial cells are present in the wall of the yolk sac and embryonic plate. Section no. 6. x 100.

Text-fig. 8. The section passes through the anterior boundary of the anterior intestinal--portal and shows the junction of the lateral folds of the yolk sac wall. Blood islands and empty endothelial-‘lined spaces are present in the wall of the yolk sac, andin the floor and roof of the foregut. Section no. 10. x 100. ' 22 R. J. Gladstone and W. J. Hamilton

arises below the level of the hindgut, passes to the right and upwards into the body stalk, and its tip is in close relationship with a detached mass of ento- dermal cells. -

The entodermal layer of the yolk sac is formed of cells the shape and size of which difi'er considerably in different parts of the sac. The entodermal cells below the embryonic disc are flattened, except in the prochordal plate region (Text-fig. 6 and P1. 3, fig. 6) where they are cuboidal, and in the floor of the foregut where they become rounded (Text-fig. 7). In the region of thecloacal membrane and allanto-enteric diverticulum they are again cuboidal. In the abembryonal pole of the yolk sac, where the vascular spaces are well developed,

. 5 “V .-~ $‘.,,§-v(i4s.§.. .~§,_‘ _‘ ~ X...‘

Text-fig. 9. A section just behind the anterior intestinal portal. The lateral infoldings of the yolk sac wall, and the marginal clefts in the mesenchyme of the embryonic plate in the region of the pericardium, are visible. The medullary plate which in Text-figs. 7 and 8 consists of a single layer of flattened eotodermal cells is now considerably thicker and two or three layers of nuclei are visible. Section no. 12. . x 100. '

some of the entodermal cells are columnar and swollen, others contain large vacuoles in their free surfaces, while still others show a ragged edge or are rounded, the vacuoles apparently having discharged their contents into the

yolk-sac cavity (Pl. 1, figs. 1, 2). The entoderm is covered by the mesodermal '

cells. The amount of this tissue present in different areas of the yolk sac is subject to wide variation. Close to the embryo it is composed of flattened cells which form a definite layer, the splanchnic mesothelium, which is intimately related to the underlying entoderm. In other areas the two layers are separated

' by mesenchymal tissue; in this tissue, and in the spaces enclosed by it, lie

haemocytoblasts and other cells. At the abembryonic pole hit is composed of many layers of cells in which are seen various stages in the formation of endo- thelial-lined spaces and developing blood cells (see under Angiogenesis, p. 29). _ A presomite human embryo 23


_The dimensions of the amnion are 1-35 by 1134 mm., and its height is 0-32 mm. It extends caudally to form a. diverticulum which leads into the connecting stalk (Text-figs. 2, 3). The lining cells of the amnion are flattened and elongated in type except where it comes into contact with the mesoderm of the connecting stalk; here the cells are cuboidal and the surface is irregular. The nuclei of the cells are near the amniotic cavity. Some degenerated ecto-

N dermal cells and cell detritus are found in the amniotic cavity. The free surface

of the amnion, in the greater part of its extent, is covered with rather flattened

. mesodermal cells. In places the mesodermal cells are raised to form small


The early stages of haemopoiesis in the human subject have recently been investigated by Bloom & Bartelmez (1940), who have distinguished the following morphological grades in the development of corpuscles and defined them under the following names:

(1) Haemocytoblasts.

(2) Primitive blood cells.

(3) Lymphoid wandering cells. (4) Primitive erythroblasts. (5) Definitive erythroblasts. (6) Definitive erythrocytes.

Additional types of cells are described by them in both intravascular and extravascular zones of haemapoietic organs or tissues, namely macrophages, giant cells, megakaryocytes, and granular leucocytes.

The characteristics of these cells are defined by Bloom & Bartelmez as‘

indicated below, and as the characteristics agree closely with those which We have observed in the‘ Shaw embryo, we propose to adopt the nomenclature which they have employed, in so far as the types described are encountered at the stage of development which has been reached by the Shaw embryo. In adopting this nomenclature, however, we do not wish it to be inferred that we use the names in any other than a descriptive sense, and also it should be understood that by so doing we do not associate ourselves with certain current theories as to the pluripotentialities of some of the stem cells which are described.

Some of the principal features of the types as defined by Bloom &. Bartelmez are: , (1) H aemocytoblasts. ‘ Spherical, or slightly polygonal cells, often provided with amoeba-like processes ’ which often occur intra- and extravascularly. ‘The cytoplasm is deeply basophil and may contain a few vacuoles.’ ‘The relatively very large nuclei are spherical or slightly indented.’ ‘ The very large and acidophil nucleoli may be single or multiple and are often very angular.’ ‘ The finely granular chromatin is aggregated in larger masses along the nuclear 24 R. J. Gladstone and W. J. Hamilton

membrane in some cells.’ ‘ These cells are identical in appearance not only with the primitive blood cells but also with the large, free, basophil stem cells seen in all haematopoietic foci of both embryonic and adult man.’ .

(2) Primitive blood cells. ‘These are the first free precursors of blood cells and are morphologically identical with haemocytoblasts ’ (which may be found in bloo_d-forming centres in older embryos, foetuses, and adult man).

(3) Lymphoid wandering cells. ‘This name was given by Maximow (1909) to cells of lymphocyte, that is haemocytoblast, type, which he found moving in the mesenchyme.’

(4) Primitive erythroblasts. ‘This is the largest group of cells in the blood-

‘cell forming areas. The youngest ones closely resemble, and are connected by

many transition forms with, the haemocytoblasts. The nuclei become smaller, the nucleoli smaller and often more numerous, and the chromatin particles more prominent as the cells mature. The cytoplasm remains about the same size but loses its basophilia, the colour gradually changing as haemoglobin accumulates within it.’ ‘

(5) Definitive erythroblasts (e.g. thosefound in older embryos (20 mm.) in the wall of the yolk sac, and other situations in foetal and adult life, and which are commonly known as normoblasts). ‘The cytoplasm becomes paler and the chromatin assumes a more regular distribution, with the nucleoli smaller and more numerous; this is often spoken of as the checker board nucleus.’ ‘The nucleus becomes progressively smaller and pycnotic and is finally extruded to produce the mature definitive erythrocyte.’ Proliferation by mitosis occurs in definitive erythroblasts as in primitive erythroblasts.

(6) Definitive erythrocytes. The presence of erythrocytes is first mentioned in an 18-somite human embryo (H.E., H. 1516) in which a few extravascular primitive erythrocytes, singly or in pairs, were noted in the yolk-sac mesoderm. The greater number of the developing blood cells seen at this stage were, however, intravascular haemocytoblasts, and primitive erythroblasts. (In this embryo it is noted that the circulation had commenced and the authors state that the vessels of the villi contain polychromatophil primitive erythroblasts at various stages of maturity and an occasional haemocytoblast; and further, that there wasno haemopoiesis in thefinesenchyme of the villi.)

A few words of explanation are perhaps necessary here in order to dis- tinguish the exact meaning which is implied by the use of the qualifying adjectives ‘primitive ’. and ‘definitive’; thus the primitive blood cells are stated to be ‘morphologically identical with haemocytoblasts’—the difference here lies in the circumstance that the primitive blood cells are the first haemo- cytoblasts to appear whereas the name haemocytoblast is a more general term than primitive blood cell, and is applicable to all developing blood cells having the special characters of haemocytoblasts (e.g. basophil cytoplasm and amoeboid processes) wherever and whenever these cells are recognizable. Thus

_they may be present in mesenchymal connective tissue, lymphoid tissues, or

the medulla of adult bone; whereas the use of the term primitive blood cell is A presomite human embryo A 25

restricted to the parent cell of the early embryonic stage when blood is com- mencing to be formed on the yolk sacror vascular area. The terms primitive and definitive, when used in descriptive embryology, usually denote re- spectively the first or undifferentiated and the final stage of development of any particular cell _or organ. When, however, it is implied that successive independent generations of developing cells follow one another, there is apt to be confusion unless this is clearly stated. I Our observations on angiogenesis and haemopoiesis in the Shaw embryo will be described under the following headings: (1) The chorionic mesoderm and villi. (2) The connecting stalk. (3) The yolk sac.

Chorionic mesoderm and villi

This presents the usual appearance of embryonic mesenchyme at the pre- somite stage of development, namely, a loose reticular tissue, formed by a delicate syncytial network of feebly stained protoplasmic strands and laminae, which enclose irregular intercellular spaces. Basophil nuclei of varying size and shape which occasionally are seen to be undergoing mitosis are present at the nodes of the network. The loose reticular stratum is limited, on its inner side, by a thin membranous layer of flattened cellular elements. This is incomplete in certain places where the chorionic mesenchyme becomes con- tinuous with protoplasmic strands of the magma reticulare. In those places where the membrane is well developed it forms a definite mesothelium which resembles the splanchnic mesothelium covering the basal part of the body stalk and the yolk sac, but the cellular elements of the chorionic membrane are more delicate and flattened than in the yolk-sac mesothelium. Branching strands of angioblastic tissue are present throughout the Whole extent of the chorionic mesenchyme and the central mesodermal cores of the villi. They are readily distinguished by their diffuse pale pink colour and finely granular cytoplasm. It is difficult to determine the exact relations of the strands but they appear to be connected with (1) the surrounding mesenchymal tissue, (2) the chorionic mesothelium, and (3) the cytotrophoblast. Vacuoles are frequently visible in the cytoplasm and, by fusion of these with one another, intracellular spaces or lumina are formed in the larger and more differentiated strands which are becoming transformed into capillary vessels and vascular sinuses bounded by an endothelial wall. .

The spaces in the mesenchyme are thus of two kinds: (1) the irregular, intercellular spaces or meshes of the reticulum, and (2) vascular spaces lined by endothelium.

The vascular sinuses and developing capillary vessels are mostly empty (Pl. 3, figs. 7, 8); very occasionally, however, various stages in the ‘rounding up’ of the primarily flattened cell elements of the wall of a vascular sinus are seen, until the stage is reached of a round bud with a constricted neck pro-I 26 R. J. Gladstone and W. J. Hamilton

jecting into the lumen of the sinus. Within the lumen of the sinus (Pl. 4, fig. 9), and in the Vicinity of these buds, free cells having all the characters of the endothelial cell elements forming the buds are frequently visible. These liberated cells arising from the endothelium vary in size, but are on the .whole smaller than the erythroblasts in the blood islands and vascular spaces of the yolk sac. Occasionally the free cells are seen to be undergoing mitosis; on the other hand, some show degenerative changes.

The vascular spaces and developing capillaries are seen in the chorionic mesenchyme and central cores of the villi throughout the whole extent of the vesicle, but spaces and capillaries containing free endothelial cells appear more frequently in the chorionic mesenchyme in the vicinity of the body stalk. The transformation of theangioblast by canalization into sinuses and vessels is more advanced on the placental side of the chorionic vesicle, and at the equator where the villi are largest, than at the abembryonic pole. ‘

Free spherical cells, or groups of such cells, ‘of mesothelial type are also seen

in the magma reticulare, and more particularly in the neighbourhood of the I

body stalk.

Fibroblasts and small round cells of a lymphoid type are also occasionally seen in the intercellular spaces of the chorionic mesenchyme; the nuclei of some of the fibroblasts show mitosis. I

The appearance of the angioblastic strands and vessels in the mesenchyme forming the central core of the chorionic villi is similar to that in the parietal mesenchyme, though, on the whole, the development is less advanced than it is in that layer. The development of angioblastic tissue and vessels in the villi is taking place throughout the whole extent of the vesicle but, as in the parietal mesenchyme, the largest and most differentiated vessels are found in villi near the attachment of the connecting stalk. As in the parietal mesenchyme, the vessels arise by vacuolation of the protoplasmic strands and the fusion of vacuoles to form an intracellular lumen which contains, as a rule, no formed elements. The appearance of one of these vessels and the relation of its wall to the surrounding mesenchyme are well seen in the transverse section shown in P1. 3, fig. »8, in which it will be noted that the upper wall of the vessel, as seen in the photograph, is formed by angioblast, whereas the lower is com- pleted by a delicate film or membrane which is continuous with both the angioblast above and with the surrounding mesenchyme. The vessels of a villus are often continuous at the base of the villus with a vessel in the parietal layer of the chorion, sometimes by a T-shaped junction suggesting outward growth into the villus of a sprouting bud from a previously formed vessel in the mesenchyme. Other strands, however, appear to be formed independently by transformation of the mesenchyme, although it is possible, as claimed by Hertig (1935), since the strands often appear to be continuous with the tropho— blastic cells, that they may have arisen at an earlier stage by delamination from the cytotrophoblast, but the end-product, namely, the endothelium lining a blood sinus or capillary vessel, is so similar -to that which occurs in the A presomite human embryo 27

"formation of the blood vessels in the yolk-sac mesoblast and other mesodermal

tissues, that we are inclined to think that the vessels arise by the transformation or differentiation of mesodermal cells. When once the angioblastic tissue has become differentiated it reproduces itself by budding out processes which spread by proliferation of endothelial cells and extension, as shown by Clark & Clark (1932 and 1937) and also Clark (1909) in their accounts of observations upon the growth and extension of new blood and lymphatic capillaries from pre-existing capillaries in the liver, rabbit ear, and in amphibian and mammalian material respectively.

Connecting stalk

This region must be considered as consisting of two segments: (1) the basal part at the chorionic end of the stalk, where the supporting tissue consists of a loose mesenchymal reticulum continuous with the parietal chorionic mesen- chyme, and (2) the proximal, or embryonic, end—-the umbilical stalk—which encloses the allanto-enteric diverticulum and in which the supporting tissue

consists of closely packed, rounded, embryonic cells continuous with the yolk- .

sac mesoblast. In the former (Text-fig. 10) a large capillary vessel is seen which extends into the parietal mesenchyme. Its endothelial wall shows bud- like projections into the lumen, and some free rounded cells lying in the lumen. These cells are smaller than the haemocytoblasts and early erythro- blasts found in the blood islands in the wall of the yolk sac (cf. Pl. 4, fig. 9, with P1. 4, figs. 10, 11), and they differ considerably in appearance, having compact deeply stained spherical nuclei surrounded by a relatively small amount of cytoplasm. In contrast with these isolated cells, definitely arising by a process of budding from a preformed endothelial wall, are the cells found in the region of the allanto-enteric diverticulum, which lie in the mesoblast in close relation with the entodermal wall of the diverticulum. These cells closely resemble those of the blood islands of the yolk sac, with which they are strictly homologous, and with the blood-vascular system of which they will later become continuous, when an endothelial wall has been differentiated from the mesoblast sur- rounding them, and the two systems become joined. A connexion is also formed, at a later stage, of the plexus of vessels which gives rise to the umbilical arteries and the umbilical vein with the independently formed vessels in the chorionic segment of the stalk. The time at which this occurs appears to be variable; thus, in Ingalls’s human embryo at the beginning of segmentation (three somites) the author (1921) states that ‘the distal portion of the stalk is -occupied by a few small blood vessels through which a narrow connexion is set_ up with the vessels of the chorion’.

In contrast with this we find in Stieve’s 13}-day human embryo Hugo (1926) —total length of embryonic rudiment 0-75 mm. and length of embryonal shield 0-57 mm.—the development of solid blood islands and blood vessels in the stalk unusually far advanced considering the early stage of development of this embryo. Stieve regards the mesoderm which contains the blood islands as 28 l R. J. Gladstone and W. J. Hamilton

arising from the ‘sickle node’ and, therefore, axial in origin. He also believes that the cells forming the islands in the stalk, like those of the islands in the yolk sac, may originate, ‘in the human subject, from both entoderm and

Text-fig. 10. The photograph shows the junction of the connecting stalk with the chorionic mesoderm, andthe advanced stage of development of the vessels in the stalk, adjacent parietal mesenchyme, and mesenchymal tissue of the In the cavity of the chorionic vesicle isolated groups of degenerate mesenchymal cells and free nuclei are visible. The con- nexions of the villi with the decidua, and of the intervillous spaces with the basal sinus, are also clearly shown. Section no. 144. x l0Q.

mesoblast, as in other species of mammalian embryos. The mesenchyme at the chorionic end of the stalk, and the chorionic mesenchyme, which contain blood vessels but no islands, are regarded by him as ‘morula mesoderm’, which A presomite human embryo 29

difference, we believe, may explain the difference in type of angiogenesis in the two parts of the stalk. It is not until the vessels are fully formed in both parts and are connected that, when the heart commences to beat, blood flows from the umbilical arteries into the chorionic or placental vessels.

Yolk sac

There are three types of blood island on the surface of the yolk sac :'

(1) Haemocytoblasts and early primitive erythroblasts not enclosed inlan endothelial covering.

(2) Solid masses of haemocytoblasts enclosed in an endothelial wall.

(3) Endothelial-lined cysts or vessels containing a few free haemocytoblasts and other cell types.

Text-fig. 11. The photograph shows the early formation of intercellular and intracellular spaces. Some of the intercellular spaces are lined by entoderm (upper layer in the photograph) on one side and lay mesenchymal tissue on the other. Some haemocytoblasts and primitive erythroblasts (cells with clearer cytoplasm) are lying free in the spaces. Section no. 46.

x 560. ‘

Intermediate stages between these three types of island are frequent, and intercellular spaces not lined by endothelium are also present in the mesoblastic tissue between the splanchnic mesothelium and the yolk-sac entoderm (Text- fig. 11).

Angiogenesis and haemopoiesis are most advanced at the distal or abem- bryonic pole of the sac (Pl. 1, fig. 1) where, presumably, the blood islands first commenced to be formed in the zone corresponding to the vascular area of lower types of Mammalia; these two processes are less advanced in the circum- ferential zone near the margin of the embryonic disc where the earlier stages 30 R. J. Gladstone and W. J. Hamilton

of smaller solid islands not yet enclosed in endothelium are best studied (Pl. 1, fig. 2). In this region the yolk-sac wall consists of two well-defined layers, the entoderm and splanchnic mesothelium, between which in the loose mesoblastic tissue are empty spaces, some of which are interstitial and others are lined by

endothelium. There are also small solid masses of proliferating cell elements, '

haemocytoblasts and primitive erythroblasts, which represent the earliest stage of blood formation (Pl. 4, fig. 10). Sometimes the blood cells project from one side into the lumen of an endothelial-lined space into which at a later develop- mental stage the freed blood cells will be discharged (Pl. 4-, fig. 11). The cell outlines of the future haemocytoblasts are, in the earliest stages, absent or only very indistinctly visible (Pl. 1, fig. 2). In the more advanced stages the cell elements of these plasmodial masses differentiate and become separated from each other and are finally liberated in the form of free haemocytoblasts showing irregular protoplasmic processes. These by further division form the first generation of embryonic red blood cells or primitive erythroblasts. Some of the cells show signs of degeneration, e.g. the formation of vacuoles, the contents of which when discharged may contribute to the formation of the embryonic plasma.

In the later stages of development seen at the abembryonic pole of the yolk sac, when the endothelial walls of the blood spaces are completed (Pl. 1, fig. 1), the vessels are much increased in size and communicate with each other, forming the capillary plexus from which the vitelline vessels will afterwards be formed in the early stages of somite development, as has been demonstrated in the Ingalls (1921) three-somite human embryo. In these vessels we can recognize the following types of cell (Pl. 1, fig. 1): haemocytoblasts, tran- sitional stages between these and primitive erythroblasts, primitive erythro- blasts, intermediate erythroblasts, late prirnitive erythroblasts, degenerating erythroblasts, non-nucleated primitive erythrocytes, and binucleated giant cells. '


It is proposed to compare the present embryo only with those that are nearly related to it in development. It is at approximately the same stage of development as W9. 17 (Grosser, 1931) and would, therefore, be at a. later stage than the Manchester embryo (Florian & Hill, 1935) and younger than the Pehl-Hochstetter (Rossenbeck, 1923).

If we accept the patient’s history of coitus, the embryo has a coital age of 26 days, much older than the corresponding embryos of this stage of develop- ment. The embryo, indeed, appears to be much younger than its supposed coital age, being at about the 17th or 18th day of development, and as it was

procured as the result of an operation there is no reason to suppose that it ,

had been dead for some time. A presomite human embryo 31

Embryonic disc

After comparing the shape of the dorsal views of graphic reconstructions, Hill & Florian (1931) have suggested that the embryonic discs of embryos may be divided into two varieties, as Rabl (1915) has done for the rabbit, a narrow type, e.g. the embryos Bi 24 (Florian, 1934), Manchester (Florian & Hill, 1935), Wa 17 (Grosser, 1931), and Dobbin (Hill & Florian, 1931), and to these may be added the embryos, Kl 13 (Grosser, 1913), Ingalls (1918), Carnegie 5960 (Heuser, 1932), and HR1 (Johnston, 1940), and a broader type, e.g. the Hugo (Stieve, 1926), Scho (Waldeyer, 1929) and Pehl-Hochstetter (Rossenbeck, 1923) embryos; to this list may be added the Thompson-Brash (1923) and H0 (Fahrenholz, 1925) embryos. -

The embryo Shaw would belong to the latter group. In this connexion it 1

may be pointed out that the head process of this embryo is at an early stage of development, and this may partly account for the fact that the disc is broad; on the other hand, the Pehl-I-Iochstetter embryo has a well-developed head process.

The distinct dip on the embryonic disc in front of Hensen’s knot is a

feature of this embryo, and it- is partly due to the proliferation of the cells at’

the node. In 'the HO embryo (Fahrenholz, 1925) there is also a distinct elevation at the knot. In the Scho embryo (Waldeyer, 1929) there is an

- elevation in front of Hensen’s knot and the disc slopes gently away towards

the anterior edge. The Carnegie 5960 (Heuser, 1932) embryo shows a distinct slope some distance in front of the node. There is no marked convexity of the disc in thesagittal plane as described in HR, by Johnston (1940); this author is of the opinion that this convexity is the result of abnormal growth. The concavity in the Shaw embryo, in our opinion, is normal and is the result of the rapid proliferation at Hensen’s knot. It may, however, be an artefact due to fixation. The arrangement of the cells in the ectoderm is typical for an embryo of this stage of development.

Primitive streak, cloacal membrane, allanto-enteric diverticulum and yolk we

The arrangement of the layers in the primitive streak is essentially similar

to that described in embryos of a corresponding stage of development and will‘

not be discussed further.

The cloacal membrane is easily recognized in this embryo. It involves the posterior part of the embryonic disc and the proximal part of the allanto- enteric diverticulum. The junction between the_two parts makes a distinct angle as in the Pehl-Hochstetter (Rossenbeck, 1923) embryo. There is no indication of the separation of the cloacal membrane into two parts by the appearance of mesoderm between ectoderm and entoderm as described by Florian (1930) in the Bi II and Sternberg (1927) embryos, and by Wybum (1987) in the McIntyre embryo. The present embryo does not throw any light 32 R. J. Gladstone amid W. J. Hamilton

on whether the cloacal membrane first appears between theamniotic ectoderm and entoderm, as in the Beneke embryo (Florian & Beneke, 1930-1) and HR1 embryo (Johnston, 1940), or within the embryonic disc, as in the Bi I and Fetzer embryos (Florian, 1933). ' '

The proliferation of the ectoderm which gives rise to the primitive streak in the Edwards-Jones-Brewer embryo (Brewer, 1938), which is well preserved, seems to us to settle the controversy as to whether the primitive streak appears before or after the cloacal membrane. In the Brewer embryo the primitive streak is at the earliest stage described so far for the human subject, and there is no indication yet of the cloacal membrane. Further, in this embryo the primitive streak is a crescentic mass, as in the pig (Streeter, 1927) and the ferret (Hamilton, 1937), and it lies at the posterior part of the disc and not at some distance in front of the edge of the disc as shown by Florian (1933) for the OP and W0 embryos of von Mollendorff, and the Fetzer embryo.

The mesodermal cells which arise from the posterior part of the primitive streak in the Shaw embryo pass around the cloacal membrane to the con- necting stalk. The cells are closely packed together as described by Florian & Beneke (1930-1), and Hill & Florian (1931) in the Dobbin embryo.

It is not our intention to enter into a detailed discussion on the develop- mental history of the cloacal membrane since, in the present embryo, it is already well developed. We are, however, unable to accept the interpretation of Florian (1933) and Wyburn (1937) for this structure. Florian states: ‘ If we realize that the cloacal membrane and the primitive streak represent parts of the blastopore in lower vertebrates, we must accept the View that the homo- logue of the blastopore in Man originates in two separated areas which grow together only during later development. This fact has perhaps caused the difficulties in interpretation of the structure which I regard as cloacal mem- brane.’ Wyburn writes: ‘The appearance of the primordium of the primitive streak representing the fused lips of the blastopore acclaims the formation of secondary mesoderm separating the ectoderm and entoderm in this region’, and ‘ Caudal to the fused lips of the blastopore and reaching to the attachment of the stalk remains a relatively extensive area of contact, the anlage of the cloacal membrane. . . .The cloacal membrane would thus be the homologue of the ventral lip of the blastopore and adjacent area.’ There is apparently no phylogenetic reason why the cloacal membraneshould be regarded as part of the blastopore. In Amphiowus and Amphibia the blastopore is recognized as an invagination and the cells around it are proliferating rapidly to give rise to axial structures and mesoderm. In reptiles, birds and mammals the ventral lip of the blastopore is represented by the primitive streak and, as such, gives rise to mesoderm. The cloacal membrane in man and mammals, on the other hand, is a region where ectoderm and entoderm are in contact, or actually fused, and gives rise neither to axial structures nor to mesoderm. It is diflicult, therefore, to see why it should be regarded as part of the primitive streak or of the blastopore. If Rabl’s (1915) opinion is accepted, that gastrulation (formation A presomite human embryo 33

of head process and mesoblast) is a process of growth which allows certain presumptive organs to reach their deflnitive positions, then it seems to us. that the cloacal membrane must be regarded as lying outside this field of growth. We are _rather of the opinion of Keibel (1896) that the cloacal membrane is a secondary structure. The matter is only to be decided by experimental methods as it has been in lower vertebrates. Whether the area in the connecting stalk where mesoderm is in contact with the amniotic ectoderm can be regarded as partof the blastopore, as it gives rise to mesoderm, is another question. The origin of the primitive mesoblast from trophoblast, as described by Hertig (1935), introduces further and fundamental dilficulties in accepting the view that the area described by Florian is part of the blastopore.

N otochordal process and chorda, canal

The presence of a chorda canal has been described in embryos at approxi- mately the same stage of development as the present embryo.

The youngest embryo described with a head process (length 0-06 mm.) is the Meyer (1924) embryo; according to Florian (1928) the structure identified as the head process was erroneously interpreted. Johnston (1940) describes a. head process (length 0-04 mm.) consisting of a mass of cells derived from what he interprets as Hensen’s knot in the HR1 embryo. It should be pointed out that Florian does not recognize a head process in the HR1 embryo (see Johnston’s paper). As far as can be’ ascertained from the photographs pub- lished in J ohnston’s paper we are of the opinion that this author’s interpre- tation is the correct one. One of us (Hamilton, 1937) has found appearances similar to that described and figured by Johnston in the early stages of the development of the head process in the ferret.

An undoubted head process is present in the embryos H. Schm. 10 (Grosser, 1931), Hugo (Stieve, 1926) and Bi 24 (Hill & Florian, 1931; Florian, 1934). In the Hugo embryo the process (length 0-07 mm. according to Hill & Florian (1931), and 0-09‘ mm. according to Stieve) is composed of irregular cells which are never columnar; they are, however, in continuity with the intra-embryonic mesoderm. The entoderm is missing underneath the head process. In the H. Schm. 10. the head process (length 0-1 mm.) is composed, in one section, of columnar cells which surround a space which Grosser recognized as the chorda (Lieberkuhn’s) canal. Thus at an early stage /the cells are typical of later stages of development when a well-defined canal is established, as in the embryo described by Heuser (1932). In the Bi 24 embryo the head process (length 0-1 mm.) is more differentiated than in the Hugo. It consists of a thickened median cord ‘and two. thinner lateral wings of mesoderm. The underlying entoderm is indistinct andis represented in some places only by. scattered cells, many of which have pycnotic nuclei. The pycnosis of the nuclei, in our opinion, may be regarded as the first sign of the disappearance of this entoderm. The median cord of cells ends in continuity with thickened entoderm which is recognized as the prochordal plate. The next stage in the development of the

Anatomy 76 V 3 34 R. J. Gladstone and W. J. H (tmilton

head process (length 0-12 mm.) is found in the Manchester embryo (Florian & Hill, 1935); it has a definite Hensen’s knot extending forward from which is a short head process which terminates in the prochordal plate; no mention is made of the presence of a canal in this head process." In the embryo Wa 17 (Grosser, 1931) there is the beginning of a chorda canal in the head process (length 0-18 mm.). Whether a ventral opening connects the chorda canal with the yolk sac appears to be doubtful. In the Shaw embryo, with a head process 0-18 mm. in length and, therefore, at approximately the same stage of develop- ment as the Wa 17, the dorsal opening of the chorda canal and the caudal part of the canal are distinct. We have not been able to find a ventral opening into the yolk sac. If entoderm is underlying Hensen’s knot and the head process it is indistinguishable from the ‘tissue of the knot and head process. In the embryo Thompson-Brash (1923) the head process (length 0-3 mm.) has not yet acquired a lumen, but the dorsal cells are arranged fanwise so that the appearance of a lumen is imminent. In the Pehl-Hochstetter embryo (Rossen- beck, 1923) the head process (length 0-6 mm.) has a distinct canal. There is some doubt as to whether there are one or more ventral openings into the yolk sac (see Hill & Florian, 1931). In the.Kl 13 embryo (Grosser, 1913) the chorda canal (length 0-2 mm.) communicates by means of several openings with the yolk sac. This embryo is smaller, but apparently more differentiated, than the Pehl-Hochstetter, as is also the Dobbin embryo (Hill & Florian, 1931). There

‘is, therefore, considerable variation in the time of appearance of the canal. It is

distinct in the Shaw embryo, with a head process length of 0-18 mm., and has not yet appeared in the Thompson-Brash (head process length 0-3 mm.-). There is also variation as to the stage of development when it communicates with the yolk sac. .

The line of demarcation between.the posterior edge of Hensen’s knot and the anterior part of the primitive streak can be recognized in the present specimen. The cells in the roof of the chorda canal are tall and columnar, with clear cytoplasm towards the lumen, and are, therefore, typical for embryos of this stage of development (compare Rossenbeck, 1923; Heuser, 1932; Johnston, 1940). The obliquity of the section in the present embryo does not permit us to give an opinion as to the amount of mesoderm, if any, arising from the side of the head process. That the head process becomes intimately connected with the underlying entoderm in man is, however, borne out by the present specimen. _

The prochordal plate

There is now an extensive literature dealing with the prochordal plate in man and in mammalian forms (Hubrecht, 1890; Bonnet, 1901; Adelmann, 1922; Bryce, 1924; Hill & Tribe, 1924; Hill & Florian, 1931; Heuser, 1932; Hamilton, 1937; and others).

In the present embryo the thickened entodermal cells form,a horseshoe- shaped plate resembling somewhat that described by Florian & Beneke ( 1930-3 1 ) in the Beneke embryo. The anterior tip of the head process lies between the A presomite human embryo 35

two horns which project backward. The chromophilic granules first described

by Bonnet (1901) in the prochordal (completion) plate in the dog have been '

recognized in the cells of the entoderm in that part of the disc that we interpret as prochordal plate. We have not found pouches or_ ‘cell groups’ in the pro- chordal plate, as described by Bryce (1924) in the McIntyre embryo (Sternberg, 1927; Heuser, 1932).

Connecting stalk

As stated earlier, we have followed Florian (1930) in recognizing two parts in the connecting stalk, an amnio-embryonal and an umbilical part. We find in the Shaw embryo, as in nearly all human embryos at this stage of develop- ment, that the amniotic cavity runs into a distinct amniotic tip at its dorso- caudal part where it comes into contact with the mesoderm in the basal part (i.e. near the chorion) of the umbilical stalk. In some of the cells of the meso- derm, entoderm, and ectoderm in the anterior part of the umbilical stalk there are conspicuous chromatic granules. Florian regards these granules as a sign of degeneration in the cells. After discussing at some length the changes occurring in the stalk tissue he comes to the» conclusion that the amniotic cavity extends backwards into the umbilical stalk at the expense of the tissue of the stalk, as first described by von Mollendorff (1921), and that the cell dissolution is a preliminary to this extension. ,

We find that there is some irregularity of the ectodermal amniotic cells in the front of the umbilical stalk which would support the hypothesis of back-

ward extension; we did not find, however, the sequestration of a large mass of mesoderm as described by Florian.

The vascular primordia found in the stalk are discussed under angiogenesis.

Allanto-enteric diverticulum

The allanto-enteric diverticulum in the present embryo is found to be associated at its tip with a solid mass of cells which we interpret as entodermal.

In the Mateer embryo.Streeter (1920) describes the allantois as separated into two parts. In the proximal part of the distal separated portion there is a lumen. In the Sternberg embryo (Sternberg, 1927) a separated epithelial vesicle is associated with the allantois. Hill & Florian (1931) found, in the Dobbin embryo, that the terminal part of the allantois passed into a terminal vesicle. We are in agreement with the opinion expressed by them that the allantois may undergo separation into two parts during‘ development, the distal part being subject to early degeneration. In the present specimen the cavity, if ever present, has disappeared.

Yollc sac The details of the histological appearances of the cells in the yolk sac are

discussed under Angiogenesis. The formation of the foregut by anterior, and partially by lateral, foldings is clearly shown. A folding is shown in the Heuser

3-2 36 R. J. Gladstoneiaml W. J. Hamilton

(1932) embryo but the formation is regarded as a fold in the yolk sac; Heuser states that a recognizable foregut has not yet developed. ‘

Angiogenesis and haemapoiesis

The study of this specimen affords evidence in support of certain con- clusions which have previously been reached from observations made on other human embryos of approximately the same age and stage of development. It also throws light upon the following problems which are still sub judice:

(1) The origin of the angioblastic tissue of the primary or chorionic meso- derm. V .

(2) The origin of the first haemocytoblasts and primitive blood cells.

(3) The relation of the blood islands of the yolk sac and umbilical segment of the body stalk to the endothelial walls of the vessels which surround them.

(1) The origin of the angioblastic tissue of the primary or chorionic mesoderm. It is almost universally accepted that both blood cells and blood vessels arise from (the mesenchyme. This View is based chiefly on the observations of Maximow (1909, 1927), Bloom (1938) and Bloom & Bartelmez (1940).

It seems best, in our opinion, to approach the subject of the early stages of blood-cell formation and development of blood vessels in the human subject and other vertebrate animals by an examination of the first stage of the differentiation of these elements rather than to work backward from later stages of blood formation, when differentiation of other cells from morpho- logically similar cells is taking place at the same time in different situations. Apart from the difiiculty of deciding whether cells are developing intra— or extra-vascularly, the examination of later stages is complicated by the fact that there is a mixing in the general blood stream of cells which may be derived from different sources.

This difficulty has to a large extent been overcome by Stockard’s (1915) experimental work on the origin of blood and endothelium in Fundulus larvae. Heconcludes that the origin of the blood cells and endothelial lining of the heart, aortic arches and dorsal aortae is independent. Stockard succeeded in rearing Frmdulus larvae in weak solutions of alcohol; although a heart and intra-embryonic blood vessels lined by endothelium were developed in these larvae, the connexion of the vitelline veins with the heart was never established and consequently no circulation of either blood cells or plasma could take place. Moreover, the blood developed in the intermediate cell mass at the posterior end of the embryo, and, found in the blood islands in the vascular area of the yolk sac, remained in its original situation and was not drawn into the heart and intra-embryonic vessels. Furthermore, neither the heart with its endo- thelial lining nor any portion of the aortae were, at any stage of development, seen to contain an erythroblast or erythrocyte. Stockard believes that the blood islands in Fundulus embryos are formed by wandering mesenchymal cells which migrate from the intermediate cell mass which lies at the posterior end of the embryo, between the notochord and the caudal end of the gut; he A presomite human embryo 37

further states‘ that the only mesodermal portion of the yolk sac in Fundulus

. is made up of disconnected wandering mesenchymal cells, some of which group

themselves to form the blood islands, others give rise to the yolk vessel endo- thelium, and still other wandering cells develop into the chromatophores. Stockard thus ranges himself alongside those authors, such as Maximow, Bloom and Bartelmez, who believe that both erythroblasts and endothelial cells arise from the mesenchyme. He does not, however, believe that endothelial cells once differentiated are capable of producing erythroblasts and erythrocytes, since, in his series of embryos which developed without a circulation, he never found that the endothelial-lined heart or the vessels leading from it, such as the ventral aorta, aortic arches and dorsal aortae, contained developing blood cells; nor does he believe that the mesenchyme in general, that is to say in other places than the known haemopoietic foci, gives origin to developing red corpuscles.

In the chorionic mesenchyme, including that of the villi and that at the chorionic end of the connecting stalk, in the human subject, specialized angio- blastic tissue appears at an early stage of development. This tissue gives rise, by vacuolation of primarily solid protoplasmic strands, to endothelial-lined spaces and vessels which before the circulation is established contain no primitive blood cells, although these are abundant in the blood islands of the yolk sac and are present in the umbilical stalk in the vicinity of the allanto— enteric diverticulum. The formation of vessels in the chorionic ‘mesenchyme has been specially studied in primates and in the human subject by Hertig (1935). He concluded that there is ‘a simultaneous origin of angioblasts and primary mesoderm by a. process of delamination and differentiation from the chorionic trophoblast of macaque and human ova’, and that the isolated vascular primordia thus formed soon possess the power of independent growth and lumen formation, resulting not only in the vascularization of the 'chorion and primary villi but of secondary villi as well. Further, after discussing the question of haemopoiesis in the chorionic blood vessels of early human embryos, such as the Mateer (Streeter, 1920) and the Heuser (1932) presomite specimen, Hertig points out that ‘blood elements begin to appear in the chorionic and villous circulation with regularity only about the time the embryonic circu- lation is becomingofunctionally established’. The generally empty condition of these chorionic vessels which we have observed in the Shaw embryo, and which was also‘ noted by Stieve (1926) in his 13;-day human embryo, confirms Hertig’s conclusions and, in the light of Stockard’s experimental work,‘is significant. It may be noted herethat the few formed elements, which appear to be arising from the endothelium of the vessels shown in P1. 4, fig. 9, differ considerably in size and character from the primitive blood cells and erythro- blasts formed in the blood islands of the yolk sac, and although they may be interpreted as being a generation of developing red blood cells arising from the endothelium of vessels which have been developed in the chorionic mesenchyme at the base of the connecting stalk, it is by no means certain that 38 . R. J. Gladstone and W. J. Hamilton

they are such. Apart from this question it appears that angioblastic tissue is differentiated at a very early stage from the mesenchyme in which it lies, and which arises by delamination, either directly or indirectly, from an ectodermal layer, namely, the cytotrophoblast.

(2) The origin of the first haemocytoblasts and primitive blood cells. The results of an investigation of the early stages of development of the blood vessels and of the heart were published by Wang (1918). This research was on ferret embryos and commenced with the stage in which the embryo measured 1-15 mm. and in which no intra—embryonic blood vessels could be recognized, nor any indication of a heart rudiment. In this paper not only are the earliest stages of development of the vessels, heart and pericardium described, but there is a description of the author’s observations on the earliest stages of development of the blood cells, preceded by a discussion of the literature on the origin of blood vessels and blood cells. This discussion is so complete that it will be quite unnecessary to do more" than refer readers to Wang’s paper and we need only’ give a summary here of some of his conclusions. At that time there .was a diversity of opinion with regard to the germ layer from which the vascular endothelium and blood cells arise. In 1887 Ziegler expressed the view that ‘the system of blood vessels and that of- the lymphatic vessels are produced in their first fundaments from remnants of the primary body cavity (blastocoel) which at the general distribution of the primitive tissue (mesoderm) remain behind as vessels’. This view, therefore, harmonizes with B1'itschli’s hypothesis that in all Metazoa the blood-vascular system originates from the blastocoel. This theory, which may be referred to as the origin of the blood- vascular system from the primary mesoderm, has ‘received a considerable amount of support from recent writers such as Hertig (1935) although, as mentioned under the first heading of our discussion, Hertig limits his de- ductions to the origin of the blood vessels (angioblast) only, and he traces that back to the simultaneous origin of the vascular rudiments and the chorionic mesenchyme‘ from the cytotrophoblast by a process of delamination. It is worth while here to draw attention to the statement made by Wang ‘that in no case has an author stated that the blood cells and vascular endothelium are derived from different germ layers’, and he furnishes evidence; based on his observations on early ferret embryos, in support of his opinion that ‘whilst blood cells and vascular endothelium are closely related to each other, and are found invariably between the mesoderm and endoderm, there is evidence to show that, in the ferret, the origins of these two vascular elements are separate and distinct—the blood cells arising from the endoderm, and the vascular endo- thelium from the mesoderm ’. ‘ Blood cells develop at first extra-embryonically in the area vasculosa in the form ‘of clusters of spheroidal cells which are

provided with large and round nuclei and with a comparatively small amount ~

of protoplasm. These are for the most part found adherent to the endoderm in the neighbourhood of their origin, before they are engulfed by the endothelium, and are in most instances identical in structure with the endodermal cells Apresomite human embryo 39

where the contact is intimate.’ ‘On the other hand, the cells which form the endothelial rudiment are mesodermal in origin.’ I

Wang’s_microphotographs of sections of early ferret embryos show features which are very similar to those of the early stages in the development of the blood islands in the presomite phase of human embryos, and we believe that it may be safely concluded that the process of blood formation is essentially the same in the ferret as it is in man, and that any differences in the con- clusions which may be drawn are due to differences in interpretation rather than to differences in species.

Another supporter of the entodermal origin of the first embryonic blood cells is Piney (1927), who bases his argument in its favour on the following facts: (1) ‘at the time of earliest blood formation thereare no formed organs in the embryo and, therefore, transport of the yolk substances is very essential’. It may be noted that in the Shaw embryo vacuoles are frequently visible in the cytoplasm of the early blood cells as well as of the cells forming the walls of the blood islands and the entodermal cells: (2) ‘The endodermal masses become surrounded by mesodermal endothelium which, therefore, takes no part in this form of haematopoiesis.’ (3) It is very striking that this extra—embryonic form of haematopoiesis is later seen in the liver of .the embryo.’ With reference to Maximow’s claim that these first-formed blood cells are the ancestors of all the types that develop later, Schridde and Piney contend that ‘they are all haemoglobiniferous elements’. Maximow holds that some of these primitive cells become haemoglobiniferous and act as temporary carriers of oxygen (primitive erythroblasts), while others retain all their haemopoietic potencies, whereas Piney considers that ‘these first formed cells have only a short life in the embryo but that they are all nucleated red cells (megaloblasts) which normally persist longest in the liver of the embryo from which they disappear before birth’. Piney agrees with Sabin $1920) that certain amoeboid cells develop from, or near, the primitive endothelium, and quotes Ferrata as stating that ‘all the white cells of primitive blood in the embryo are of such endothelial origin (haemohistioblasts), i.e. are not derived from ‘the first formed intravascular islet cells’. The marked difference in the shape and character of the nuclei belonging to the early or primitive blood cells in the yolk sac of a human embryo from those of the endothelial cells of the vessel which encloses them is evident in his figure showing mitoses of the blood cells. The nuclei of the blood cells are rounded in form, whereas those of the endothelial cells are elongated and spindle or lens—shaped. The difference is seen to be even more pronounced in the figure showing the cross-section of a vessel of the area vasculosa of a rabbit embryo (five somites) by Maximow & Bloom (1934), who, however, explain the difference as being due to rounding off of the endothelial cells and their transformation into primitive blood cells. This ‘rounding off’ of cells which are afterwards set free in the lumen of the vessel, as seen-in Pl. 4, fig. 9, certainly appears to take place in the mesenchyme at the chorionic end of the connecting stalk, and the appearances seem to support the 40 R. J. Gladstone and W. J. Hamilton

View expressed by Piney that ‘ the embryo is supplied for a short time with two varieties of blood; one derived from the entoderm, and composed purely of haemoglobiniferous cells, and the other of mesenchymal origin and consisting of both red and white cells’. T

Quite recently a description of haemopoiesis from the clinical standpoint has been published by Gilmour (1941). He has made a special study of early presomite and somite embryos prior to the establishment of the circulation, as well as of older embryos, foetuses, and infants up to 21 days after birth. From his study of the earlier stages of development he concludes that ‘vessels arise from mesodermal cells independently in three areas, the yolk sac, the chorion—perhaps at first limited to the body stalk—and the embryo. The vessels in each of these areas unite to form nets or systems. The three systems later unite with each other and the complete circulation is established.’ ‘ Blood islands form constantly in "the yolk sac and consist of separate vascular units containing blood cells. The cells are first haemocytoblasts which arise from the vessel wall. Later, the cells are almost entirely primitive erythro- blasts but a few haemocytoblasts persist and a few differentiate into histio- cytes.’

It may be noted here that this account differs from the usual text—book description in which it is stated that the primarily uniform mass of rounded cells which forms a ‘blood island’ or ‘angioblastic cord’ in the earliest stage of development becomes differentiated into a superficial layer of flattened cells—the endothelial wall of the blood vessel—which becomes separated by a cleft containing embryonic plasma from the central cells, which retain their rounded form and become the primitive blood cells.

Space will not permit us to review adequately this paper as a whole; we may state, however, that we are in agreement with Gilmour’s conclusions regarding the ipdependent development from mesoderm of vessels in the three areas mentioned and their subsequent union prior to the establishment of the circulation. _We cannot, however, wholly concur with his statement that the first haemocytoblasts arise from the vessel wall, as there are numerous instances of small islets which are not enclosed in an endothelial wall, the cells of which cannot have arisen from differentiated endothelium by transformation, although, as Piney suggests, this does not exclude the formation of other cells of a different type from this source.

(3) The relation of the blood islands of the yolk sac and the umbilical stalk to the endothelial walls of the vessels which are formed around them. The frequent appearance of syncytial masses, or small groups .of rounded cells, which lie in the interval between the entoderm and the mesoderm of the yolk-sac wall (Pl. 4, fig. 10), and which are either isolated or in intimate relation with the entoderm, and also of spaces which are bounded, on the inner side, by a single layer of entodermal cells (Text-fig. 11), and, on the outer side, by undiffer- entiated mesenchyme, suggests that the cells which comprise these groups, and which are often indistinguishable from the entodermal cells, may be primarily A presmnite human embryo 41

derived from the entoderm. The vessel wall, on the other hand, appears to be developed by differentiation of the neighbouring mesenchyme into endothelium either before or subsequently to the formation of the island, or the primitive blood cells may grow outward as a rounded mass into a mesodermalspace the waH of which is simultaneously becoming transformed from undifferentiated mesoblast into a. continuous endothelial wall.

This embryoals'o presents some interesting features in connexion with the early ‘development of the heart and its relationship to the anterior part of the wall of the yolk sac. The description and discussion of these in conjunction with the development of the pericardium, floor .of the foregut, and intra- embryonic vessels, we propose to defer for consideration in a later communi- cation. A


1. A description is given of a well-fixed presomite human embryo at a stage of development with a blastopore, notochordal process and chorda canal.

2. The age of the embryo is discussed in relation to embryos at approxi- mately the same stage of development. V \

3. A well-defined cloacal membrane is formed which involves the posterior part of the embryonic disc and the proximal part of the allanto-enteric diverticulum.

4. A description is given of the formation of blood vessels "from specialized angioblastic tissue in the ehorionic mesenchyme. The majority of these vessels at this stage of development contain no blood cells.

5. The vascular spaces are partly developed by the fusion of small vacuoles, which are formed in solid angioblastic cords (intracellular spaces), and partly by direct transformation of mesodermal cells into flattened endothelium, which may either enclose the blood islands of the yolk sac, or form the walls of vascular spaces, which atfirst empty and incomplete, become secondarily filled with blood cells and enclosed by a continuous membrane.

6. The earliest generation of blood cells (haemocytoblasts and primitive erythroblasts) are formed in the wall of the yolk sac and in the umbilical segment of the connecting stalk in close connexion with the entoderm of the allanto-enteric diverticulum, and in the situation of the future umbilical vessels. A few rounded cells of endothelial origin were, however, found in the

. mesenchyme at the base, or amnio-embryonic segment, of the connecting

stalk. These differ in type from the former cells which arise in close association with the entoderm of the yolk sac and its diverticulum.

It is a pleasure to record our indebtedness to Mr A. K. Maxwell for the beautiful natur-treu figures reproduced in P1. 1, and for Text-figs. 1+3. We wish to thank our technicians, Messrs Westwood and Park, for their invaluable assistance. ‘ 42

R. J. Gladstone and W. J. Hamilton


ADELMANN, H. B. (1922). Amer. J. Amt. 31, 55. ALLEN,’E., PRATT, J. P., NEWELL, Q. U. & BLAND, L. J. (1930). Contr. Embryol. Carney. Inatn, 22, 45. ‘ '

BLOOM, W. (1938). Handbook of Haematology. Ed. by Downey. N.Y. : Hoeber. Vol. 2. BLOOM, W. & BARTELMEZ, G. W. (1940). Amer. J. Amt. 67, 21.

Bommm, R. (1901). Amt. Hefte, 16, 231.

Bnnwnn, J. I. (1938). Ccmtr. Embryol. Carney. Imtn, 27, 87.

BRYCE, T. H. (1924). Trans. Roy. Soc. Edirtb. 53, 533.

B1'i'rscnL1, O. (1882). Morphol. Jb. 8, 474.

CLARK_, E. R. (1909). Amt. Rec. 3, 183.

CLARK, E. R. & CLARK, E. L. (1932). Amer. J. Anat. 4-9, 441.

—-—— — (1937). Amer. J. Amt. 60, 253.

FA1mENHoLz, C. (1925). Z. milcr. Amt. 8, 249. ,

FLEW, J. D. (1941). Brit. med. J. 11 Jan., p. 66.

FLOBIAN, J. (1928). Z. milcr. Amt. 13, 500.

—— (1930). J. Anat., Lond., 64-, 454.

—— (1933). J. Amt.,\Lond., 67, 263. ,

— (1934). Bratial. lelcdr. listy, 14-, 299.

FLORIAN, J. & BENEKE, R. (1930-31). Amt. Anz. Erg. 71, 229. momm, J. & Hm, J. P. (1935). J. Anat., L<md., 69, 399.

GILMOUR, J. R. (1941). J. Path. Bact. 52, 25.

GBOSSEB, O. (1913). Amt. Hefte, 4-7, 649.

—— (l930—1). Anat. Anz. Erg. 71, 135. I —— (1931). Z. ges. Amt. 1. Z. Amt. EntwGeach. 94-, 275.

HAmL'1'oN, W. J . (1937). Trans. Roy. Soc. Edinb. 59, 165.

HEBTIG, A. T. (1935). Contr. lfhnbryol. Carneg. Irwtn, 25, 39.

HEUSER, C. H. (1932). Contr. Embryol. Carney. Instn, 23, 251.

HILL, J . P. & FLORIAN, J . (1931). Philos. Tram. B, 219, 443.

HILL, J . ,P. & T3131}, M. (1924). Quart. J . micr. Sci. 68, 513.

HUBBECHT, A. A. W. (1890). Quart. J . micr. Sci. 31, 499.

INGALLS, N. W. (1918). Contr. Ernbryol. Carney. Inatn, 7, 111.

— (1921). Contr. Embryol. Carney. Instn, 11, 63.

JOHNSTON, T. B. (1940). J. Amt, Lond., 75, l.

KEIBEL, F. (1896). Arch. Anat. Phyaz'ol., Lpz., p. 250. _ KNAUS, H. (1934). Periodic Fertility and Sterility in Women. Vienna: W. Maudrich. MAx1Mow, A. (1909). Arch. milcr. Amt. 73, 444. ‘

—: (1927). V. M6]1endo1'fl"s Handb. Miler. Anat. Merw, 2. Berlin. MAXIMOW, A. & BLOOM, W. (1934). Textbook of Histology. London: Saunders. MEYER, P. (1924). Arch. 122, 38. ‘ M5LL1a:NDonn-Ir, W. voN (1921). Z. gas. Amt. 1. Z. Amt. EntwGeach. 62, 406. PINEY, A. (1927). Recent Advances in Haematology. London: Churchill. RABL, C. (1915); Arch. mikr. Anat. 88, 1.

ROSSENBECK, H. (1923). Z. gee. Amt. 1. Z. Anat. EntwGeach. 68, 325. SABIN, F. R. (1920). Contr. Embryol. Carneg. Irwtn, 9, 213.

SHAW, W. (1925). J . Physiol. 60, 193.

—— (1932). Brit. med. J. 5 Mar., p. 411.

SIEGEL, W. (1916). Munch. med. Wschr. 63, 748.

STEBNZBERG, H. (1927). Z. gee. Amt. 1. Z. Amt. EntwGeach. 82, 142, STIEVE, H. (1926). Z. milcr. Amt. 7, 295.

STOCKARD, C. R. (1915). Amer. J. Amt. 18, 227.

Smnnmnn, G. L. (1920). Contr. Embryol. Cameg. Imtn, 9, 389.

— (1927). Contr. Embryol. Carney. Irwtn, 19, 73.

THOMPSON, P. & BRASH, J . C. (1923). J . Anat., Lond., 58, 1. Fig.






. A presomite human embryo 43

VENNING, C. H. & Bnowrua, J. S. L. (1937). Endocrinology, 21, 711. WALDEYEB, A. (1929). Z. gas. Amt. 1. Z. Anal. E'n.twGeach. 90, 412. WANG, C. C. (1918). J. Am1.i., L¢md., 52, 107.

WYBUBN, G. M. (1937). J. Anat., L¢md., 71, 201.

ZIEGLEB, H. E. (1887). Arch. milcr. Anal. 30, 596.

EXPLANATION OF PLATES 1-4 The photomicrographs in Pls. 2-4 have been reproduced without retouching. -


1. A drawing of part of the wall of the yolk sac in the abembryonic region showing the diiferent types of developing blood cells lying within, and in the neighbourhood of, a vascular space which is lined with endothelium. Section no. 31, the sections having been numbered from the anterior end of the embryonic disc. x c. 750. Cell no.: 1, haemocytoblast; 2, extra- vascular haemocytoblast; 3, small haemocytoblast; 4, transitional haemocytoblast; 5, early primitive erythroblast; 6, intermediate primitive erythroblast (with indented nucleus); 7, binucleated primitive erythroblast; 8, intermediate primitive erythroblast; 9, late primitive erythroblast; 10, disintegrating erythroblast; ll, non-nucleated primitive erythrocyte; 12, binucleate giant cell; 13, remains of early blood cell.

2. A drawing of part of the wall of the yolk sac near to its attachment to the embryonic disc. A well-formed blood island is seen in the right of the drawing. Cell no. 1 is a haemo- cytoblast and no. 9 a late primitive erythroblast. No. 14 shows the formation of an angiocyst in relation with the mesothelium. No. 15 shows a cell undergoing vacuolation (? transfbrma- tion into an endothelial cell), the contents being discharged to contribute to embryonic plasma. A large vacuole is formed in one of the entodermal cells; other entodermal cells have ‘ragged’ edges suggesting that they have discharged their vacuoles into the yolk sac. Section no. 31. x c. 750.

- PLATE 2 3. A section through the umbilical stalk in the region of the allanto-enteric diverticulum. Chromatic particles are conspicuous in many of the mesodermal cells, and are also present in some of the ectodermal and entodermal cells. Ventro-lateral to the allanto-enteric diverti-

culum is a group of primitive erythroblasts some of which are undergoing Section no. 108. x 560.

. 4. Section through the blastopore and beginning of the chorda canal. The anterior lip of the

blastopore lies to the left of the photograph. The cells are columnar both in the superficial ectoderm and in the roof of the chorda canal. The nuclei are deeply stained and are, for the most part, situated deeply, near the basement membrane. The cytoplasm is clear and the cell boundaries are mostly well defined. Section no. 46. x 560.


5. A section of the definitive cloacal membrane. Itshows the continuity of ectoderm and entoderm. The mesodermal cells pass around the side of the membrane. The amniotic cavity is situated above in the figure. Section no. 96. x 560.

6. A section through the prochordal plate. The entodermal cells are seen to be cuboidal. Chromophilic granules are seen in some of the cells and between the plate and the ectoderm. Section no. 28. x 700. .

7. The section shows an empty capillary vessel, cut transversely, which is lying in the mesen- chymal tissue of a chorionic villus; it is bounded on one side by angioblast which sends pro- cesses outward into the surrounding mesenchyme. An isolated angioblastic strand lies below the vessel, but does not appear to take part in the formation of its wall. Section no. 93. x 350.

. 8. In the upper part of the figure, maternal red blood corpuscles are seen in the intervillous

space. Below this is the trophoblast which shows a distinct division into plasmodial and cytotrophoblastic layers. In the chorionic mesoderm angioblastic strands and an early stage in the development of capillary vessels may be seen. Section no. 30. x 350. 44




R. J. Gladstone and W. J. Hamilton


9. The section passes through the basal or chorionic end of the connecting stalk,‘ where the supporting mesenchymal tissue resembles, and is continuous with, the parietal layer of the chorionic mesoderm. A well-developed vascular channel crosses the lower part of the photo- graph. The endothelial wall of this‘ channel shows various stages in the ‘rounding up’ and liberation of endothelial cell elements, some of which are seen lying free in the lumen of the vessel. Note the contrast between these cells and those in P1. 4, Hg. 11. Note also extension of solid angioblastic strands into the mesenchyme on the left. The group of epithelial cells in the upper central part of the figure probably represent a degenerate remnant of the distal portion of the allanto-enteric diverticulum. Section no. 144. x 350.

10. Section through a portion of the wall of the foregut which is shown at a lower magnifi- cation in Text-fig. 8. A small blood island is seen between the entoderm lining the floor of the foregut and the mesothelial layer covering it superficially. In the roof of the foregut opposite the blood island is a triangular space which can be traced through a series of sections. The base of the triangle is formed by entoderm; the inner and outer walls, which converge to the apex of the triangle, closely resemble the angioblastic tissue found in the chorionic mesenchyme and the basal segment of the connecting stalk. Section no. 10. x 560.

11. Segment of wall of yolk sac showing entoderm on the right side and mesothelium on the left. In the upper part of the figure the entodermal cells show signs of degeneration; in the lower part the cells appear to have been shed, leaving the endothelial wall of a Vascular space exposed. In the mesoblast is a blood island which is covered, in the greater part of its extent, by endothelium; it projects below into the vascular space and appears to be breaking through previous to discharging free erythroblasts into the lumen of the space. V x 560. Journal of Anatomy, Vol. 76, Part 1 Plate 1

’_ ENTODE_/?/‘1


I 2'<‘3MM. .

GLADSTONE AND HAMILTON——A rnmsomrm HUMAN EMBBYO Journal of Anatomy, Vol. 76, Part 1 Plate 2

GLADSTONE AND HAMILTON———A PRESOMITE HUMAN EMBRYO Journal of Anatomy, Vol. 76, Part 1 Plate 3



Cite this page: Hill, M.A. (2020, March 30) Embryology Paper - A presomite human embryo (Shaw) with primitive streak and chorda canal with special reference to the development of the vascular system. Retrieved from

What Links Here?
© Dr Mark Hill 2020, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G