Paper - The equivalence of different homatopoietic anlages by method of stimulation of the different stem cells 2

From Embryology
Embryology - 16 Jul 2019    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

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

Danchakoff V. The equivalence of different homatopoietic anlages by method of stimulation of the different stem cells II. (1918) Am. Jour. Anat. 24(2): 127- 189.

Online Editor 
Mark Hill.jpg
This historic 1918 paper by Danchakoff is a historic early study of the origin of blood cells in the embryo from stem cells.

See also: Danchakoff V. The equivalence of different homatopoietic anlages by method of stimulation of the different stem cells I. (1916) Am. Jour. Anat. 20:
Danchakoff V. The origin of blood cells. (1916) Anat. Rec. 10(5): 397-414.
Thiel GA. and Downey H. The development of the mammalian spleen, with special reference to its hematopoietic activity. (1921) Amer. J Anat. 28(2): 279 - 339.

Our current understanding blood development is a significantly more accurate description.

Modern Notes: Blood Development

Cardiovascular Links: cardiovascular | Heart Tutorial | Lecture - Early Vascular | Lecture - Heart | Movies | 2016 Cardiac Review | heart | coronary circulation | heart valve | heart rate | Circulation | blood | blood vessel | blood vessel histology | heart histology | Lymphatic | ductus venosus | spleen | Stage 22 | cardiovascular abnormalities | OMIM | 2012 ECHO Meeting | Category:Cardiovascular
Historic Embryology - Cardiovascular 
1902 Vena cava inferior | 1905 Brain Blood Vessels | 1909 Cervical Veins | 1909 Dorsal aorta and umbilical veins | 1912 Heart | 1912 Human Heart | 1914 Earliest Blood-Vessels | 1915 Congenital Cardiac Disease | 1915 Dura Venous Sinuses | 1916 Blood cell origin | 1916 Pars Membranacea Septi | 1919 Lower Limb Arteries | 1921 Human Brain Vascular | 1921 Spleen | 1922 Aortic-Arch System | 1922 Pig Forelimb Arteries | 1922 Chicken Pulmonary | 1923 Head Subcutaneous Plexus | 1923 Ductus Venosus | 1925 Venous Development | 1927 Stage 11 Heart | 1928 Heart Blood Flow | 1935 Aorta | 1935 Venous valves | 1938 Pars Membranacea Septi | 1938 Foramen Ovale | 1939 Atrio-Ventricular Valves | 1940 Vena cava inferior | 1940 Early Hematopoiesis | 1941 Blood Formation | 1942 Truncus and Conus Partitioning | Ziegler Heart Models | 1951 Heart Movie | 1954 Week 9 Heart | 1957 Cranial venous system | 1959 Brain Arterial Anastomoses | Historic Embryology Papers | 2012 ECHO Meeting | 2016 Cardiac Review | Historic Disclaimer


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

=The equivalence of different homatopoietic anlages by method of stimulation of the different stem cells II. (By Method of Stimulation on their Stem Cells)=

Vera Danchakoff

Anatomical Laboratory of Columbia University

Eight Plates


1. Introduction and statement of the problem

2. Arrangement of results obtained

A. Grafts of adult splenic tissue on the allantois

3. Changes in the graft consequent to the treatment of the grafted material

4. Early changes observed in the region of the graft

5, Further changes in the adult splenic tfisue growing in the allantois

6. B. Response of the allantoic tissues to the graft

1. Introduction and Statement of the Problem

One of the most striking changes which appear after a graft of splenic tissue of a hen on the allantois of a chick embryo is a considerable enlargement of the spleen in the host. The microscopical changes which underlie the enlargement of the embryonic spleen have been shown (Danchakofi"12) to consist chiefly in an intense stimulation and proliferation of the cells in the splenic anlage and in a splitting off of innumerable stemcells from its Inesenchyme accompanied by their granuloblastic differentiation.

Parts of the host spleen or even thewhole spleen have been transformed in many cases into a homogeneous mass of granuloblastic tissue; free granuloblasts and granulocytcs (lcucocytes) formed huge accumulations, and cysts appeared ‘consequently, which consisted of large agglomerations of leucocytes. They gradually disintegrated and formed numerous foci of necrotic tissue.

In cases, in which the intensified isolation of stemcells in the mesenchymal splenic anlage corresponded in time with the development of venous sinuses, intense erythropoiesis could be observed in the host spleen. A great number of free amoeboid stemcells appeared within the vascular lurnina and underwent here erythroblastic differentiation. (Erythroblastic differentiation of the stemcells in the normal development of the spleen is observed occasionally only.) Deficiencies in the development of the splenic vascularization with consequent alterations in the further differentiation of the splenic tissue have been also described.

Briefly it has been shown that the development of mcsenchyme and lymphoid hemocytoblasts can be intensely stimulated and that the respective boundaries in which the differentiation of certain cell groups usually takes place can be shifted (displaced), in other words, that the prospective potency of the blood stemcells is greater than their prospective (actual) value. The fact that under certain conditions nearly the Whole amount of the mesonclhymal cells of the spleen anlage may undergo a granuloblastic differentiation, that under other conditions they show a fibroblastie, or an erythroblastic differentiation, and so forth, is the natural sequence of the polyvalency of the mesenchymal cells in the spleen anlage. The early process of segregation must have led in the spleen anlage to a production of one group of numerous homogeneous stemcells. which under different conditions split off variously differentiated cells.

The stimulation of the mesenchyme and of the stem cells observed in the spleen of the host with subsequent granuloblastic differentiation is not a process confined to the splenic mesonchymc only. As shown by Danchal<off,1°’ 11 the stimulation of the mesenchyme and of the hemoblasts, which is the true cause of the enlargement of the embryonic host spleen after grafts of adult splenic tissue, extends throughout the whole mesenchymc of the host. The cells in the loose connective tissue of the host body, in the adventitia of the vessels, in the perineural sheaths of the nerves, in the central axis of the feathers, in the interstitial tissue of the various glands (liver, kidney, sex glands, adrenal, pancreas), and between the muscles display an intensive proliferation followed by a granuloblastic differentiation.

One of the problems of hematology is to determine how long in a developing organism young polyvalent stemcells can be found capable of qualitatively different development. This problem forms merely a special case of a more general problem of biology, which endeavors to determine the potentialities of Various tissues.

The results of the above—mentioned study clearly demon~ strates the polyvalency of the mesenchymal cells in the early splenic anlage. These cells, under definite experimental conditions, which can be repeated at Will, manifest potentialities qualitatively different from those, which are displayed by them in normal embryonic development. The hemoblasts, .'i.e., the isolated mesenchymal cells, underthe conditions existing within the developing venous sinuses, form hemoglobin in their cytoplasm. At the time when the venous sinuses develop in a normal splenic anlage, hemoblasts are present in its tissue in small numbers, therefore they appear Within the lumen of the sinuses in small numbers also, and here transform into erythro~ blasts. The development of e hemoblasts in the splenic anlage is greatly increased after grafts, and the tissue of the splenic anlage may appear densely infiltrated by free ameboid hemo~ blasts at the time. of the development of the sinuses. In such cases large groups of them appear Within the sinuses and here undergo erythroblastic differentiation.

Now, these groups of cells developed from the splenic mesenl chyme, under normal conditions, WO11ld‘I10t differentiate into erythroblasts, but would » have followed another line of devel~ opment. The polyvalency of the mesenchymal cells of the splenic anlage is thereby proved. The ubiquitous granuloblastic differentiation of the mesenchymal cells in the loose connective tissue outside of the vessels, which is also observed after grafts, merely further demonstrates that the Wide range of potentialities is inherent in all embryonic mesenchymal cells. The hypothesis holding that invisible differences in mesenchymal cells spredeterrnine their further differentiation (Stockardggz 3“) into various blood cells is thereby deprived of all material basis.

Closely related to the results of the experiments above cited are those obtained by Marie Daiberf She extirpated the anlages of the spleen in a number of larvae of Siredon pisciformis. Subsequently to this intervention numerous spleens developed in the larva. Around the region where the original anlage of the spleen was situated, new centers of splenic tissue arose which finally fused together. In one of the cases, the most significant in my judgment, instead of a local regeneration, the mesodermal wall of the stomach, the muscularis and the submucosa were ‘splenisiert.’

These remarkable results prove that the embryonic mesenchyrnal cells possess larger developmental potencies than they actually display during normal development. I After the devel~ opment of the splenic anlage, the mesenchyme in this regionand also in the regions more distant potentially remains capable of further identical activity, and it actually displays it as soon as the required» conditions appear (in the particular case they are effected by the extirpation of the already developed splenic anlage). The differentiation of the spleen, therefore, is not conditioned by specific potentialities inherent in the limited group of mesenchymal cells, from which it actuallyarises, but is due to the local exercise of certain potentialities appertaining to the Inesenchyme in general. Such results as found in Daiber’s experiments can be obtained only if other embryonic mesenchy— mal cells than those which actually form the splenic anlage are endowed with the same potencies. However, the question whether polyvalency Ts retained by any part of the mesenchyme throughout its existence is thereby not answered.

The recent investigations tend to demonstrate that at least certain kinds of cells, which formerly were considered to be completely differentiated, actually are capable of undergoing further progressive changes. V. D. Stricht“ assigns to cartilaginous cells a potency of developing into osteoblasts after they are freed from the intercellular substance. Studying the development of bone marrow in the chick3 I also could observe in cartilaginous cells progressive changes leading them to mitotic division. Lately Streeter“ found in his studies on histogenesis of the otic capsule “a wide progressive and retrogressive adaptability of the cartilaginous tissue;” he considers that the cartilage is capable of dedifferentiation into precartilage and reticulum. The cartilaginous cells therefore may be considered as a group of cells which can exist as such during considerable periods of time and ultimately resume a greater activity.

There exist certain indications that the reticular cells in blood-forming organs belong to a similar group of cells. Laguesse“ refers to the splenic tissue as to a “réliquat du mesenchyme embryonnaire.” Ziegler“ calls it, “ein Residuum des embryonalen Mesenchyms.” Downey and Weidenreich” consider the reticular cells of the adult spleen as the source of the lympho-granulopoietic differentiation of this organ. Kijono” derives the imacrophages partly from the reticular and partly from the endothelial cells.

However, numerous authors consider the cellular reticulum in the spleen as a specifically differentiated tissue. Daiber4 states that “ Sie (die mesenchymalen Anlagezellen der Milz) specializieren sich zu Retikulum und Endothelzellen und verzichten damit definitiv auf andere Entwicklungsmoglichkeiten.." Neither does Babkina from Maximov’s laboratory” find a development of lymphocytes at the expense of any sessile cells in aseptic inflammation or normally in an adult spleen. However, the latter statement does not exclude a possibility of further differentiation of the sessile elements under other more favorable conditions.

Investigations on increased cellular activity during aseptic inflammation Were frequently made in order to determine the dormant potentialities of various kinds of cells. Though they did not definitively solve, at least they helped to formulate a number of problems concerning the reciprocal relations of the mobile and the sessile cells of the vascular and connective tissues and of the blood.

Interesting results are found in the work of Evans” and Ki~ jono.” An intensive isolation of the reticular cells has been observed by Kijono in different hematopoietic organs after the introduction in the organism of carmine by repeated intravenous injections. The presence of carmine served as a stimulus for a widespread development of histiocytes. These results, however, do not indicate that the isolation of the histiocytes is the only potency of the cellular reticulum in the spleen.

The method of cultures in vitro, to which I also took recourse, expecting to see splenic cells manifest new potencies under these entirely new conditions, did not lead to decisive results. Neither do we find in Maximov’s latest culture work" results indicating a possibility of a new line of differentiation in cultures for any of the splenic cells.

The method of grafts used by Rous and Murphy in their studies on cancer seemed to me well adapted to the comparative study of changes in the hematopoietic organs of the host as well as to a study of potencies of variouscells by culturing them in vivo. The allantois proved to be indeed a medium far more favorable than any of the artificial media used for cultivation in vitro. Of course I could not observe the multiplication and differentiation of cells directly under the microscope; however, it was easy to ascertain that the transplanted cells proliferated and underwent further progressive differentiation. The study of the adult splenic tissue in such cultures has revealed in "ehe reticular splenic cells potentialities which usually are not assigned to them. The study of the reactions manifested by the various cells of the adult spleen in the new environment of the allantois and of their changes during twelve days after transplantation, forms the first part of the present paper. These results have been partly reported at the anatomical Meeting, 1916.

The remarkable stimulation of the mesenchyme in the host subsequent to the grafting is incited by some processes developing in the graft. A further determination of thestimulus is not easy, because the effects of the applied procedure are complex. The mechanical injury, the partial necrosis and disintegration of the material grafted, the proliferation of one definite or of various kinds of cells in the graft are merely a few agents with which thechanges in the host might be related. Only a number of control experiments and a thorough investigation of the processes occurring in the graft tissue itselfmight throw some light on the nature of the stimulus of the remarkable changes in the host.

The grafting itself is without doubt accompanied by a mechanical injury of the allantois. Sometimes considerable, it is, however, not causally related with the peculiar tissue proliferation found in the region of the graft, neither with the widespread stimulation of the mesenchyme in the host. Control experiments consisting in an injury of the allantois by an incised wound or cauterization have shown that a temporary proliferation of the mesenchymal cells in the allantois regularly accompanies the healing of the wound, but it is not in the slightest degree comparable to the growth of the graft into a considerable tumor. Nor does a simple injury of the allantois call forth the extensive changes found in the different parts of the host.

The grafting of a living tissue only is followed by the development of a tumor on the allantois. Aseptic dead tissue of a hen’s spleen, deposited on the allantois of the chick embryo, is usually not incorporated in the allantoic tissue, and after a few days is found much disintegrated under the egg-shell membrane. The changes found in such cases in the al‘antoic tissues of the host are not different from those which accompany a simple injury of the allantois; neither does it stimulate the mesenchyme of the host. And yet not all of the living tissues of an adult fowl, if grafted on the allantois, will call forth a proliferation of the mesenchyme in the host. The stimulating action upon the mesenchyme of the host seems to be a property of definite adult organs only belonging to the same species (spleen, liver, bone-marrow). Embryonic chick tissues will grow on the allantois without producing any noticeable change in the tissues of the host.

The grafting of a piece of an adult chick spleen on the allantois of a chick embryo introduces into the economy of the embryo a group of various cells of an adult animal. The conditions in the allantois are favorable for the proliferation and differentiation of a considerable part ofthe grafted material; a focus of rapidly growing tissue arises in the region of the graft. In most cases this tissue is easily recognized as the grafted tissue by its well-defined limits, which frequently are also indicated by the ectodermal tissue of the serosa of the allantois. The proliferation of the various cells of the adult animal, deposited on the allantois, leads to the formation of a mass of new tissue which soon exceeds many times the volume of the grafted material. This sudden inauguration of new synthetic processes in tissues commonly regarded as specifically differentiated, cannot take place without absorption of embryonic substances by the growing tissue and without yielding various products of its metabolism to; the embryo. The (substances received from the graft by the host and vice versa, at least in the early periods of the growth of the graft, seem to be different from those which were hitherto present and circulated in the embryo and in the adult spleen, their immediate effects being of a nature normally not encountered in the embryo, nor in the tissue of the adult spleen.

The analysis of the progressive and regressive changes of the various kinds of cells observed during the growth of the grafted spleen might also give some suggestions for the further study of the nature of the stimulus applied and it forms the first part of the present paper. The changes which occur in the allantois in the immediate neighborhood of the graft, as well as those which appear in it as a part of the reaction common to the entire loose mesenchyme of the host, have been considered in the second part.

2. Arrangement of Results Obtained

The splenic tissue of an adult animal previous to its deposition on the allantois was mashed in the same manner as described in my previous paper.” The mashing of the splenic tissue breaks up the normal relations of the constituent parts of the spleen and is followed by an extensive phagocytic activity shown by the reticulo—endothelium of the graft. This activity depends upon the new reciprocal relations of the cells and will be described under a separate paragraph.

Intensive cellular activity resulting in proliferation and heteroplastic differentiation is observed not only in the transplanted tissue, but in many cases throughout the entire allantoic membrane. During the first forty-eight hours following the grafting, progressive changes are confined to the region where both tissues, the grafted and those layers of the allantois, upon which the grafted tissue is deposited, come into contact. The adult cells of the graft and the embryonic cells of the allantois react to the environmental conditions, new for both of them. The various surviving cells of the transplanted spleen and the ectodermal layer of the allantois with its capillary respiratory net are the first involved in a long process of gradual manifold changes which finally extends to all parts of the host. The early reaction by these tissues are not of complex nature and, so to speak, preparatory to the later developmental processes and will be described jointly under the heading, “Early changes observed in the region of the graft.” ,

The further existence of the adultsplenic tissue within the embryonic allantois results in a manifestation of definite new potentialities in its surviving cells. They will bedescribed in a separate paragraph under the heading, “Further changes in the adult splenic tissue in the allantois.”

The further presence and development of the adult splenic tissue Within the allantois evokes in it definite changes which appear in the mesodermal as well as ectodermal and entodermal layers of the organ in the vicinity of the graft. Finally, as above intimated, the mesenchyme of the allantois may be involved in the general process of stimulation with the f OllOWing ubiquitous granuloblastic differentiation A observed in other parts of the host and insnumerous cases it responds in the same way, as does the mesenchyme in the body of the host. The numerous interesting details of these processes will be considered in the paragraph, “Response of the allantois to the graft of adult‘ splenic tissue. .


3. Changes in the graft consequent to the treatment of the grafted material

The splenic tissue of an adult animal previously to its grafting is passed through a syringe with a sievebottom. This was done in order to free the constituent parts of the adult spleen from their reciprocal relations, which in an adult organ are definitively established. The injury done by the pressing of the spleen through the sieve chiefly affects the tissue and not the cells. The walls of the sinuses of thepulp are broken, their contents are intimately intermixed with the free cells, normally situated in the reticulum of the spleen, but the individual cells, as a thorough investigation has shown, remain intact. Even numerous arterioles andsmall follicles pass undamaged through the holes of the sieve and can be easily identified as such in preparations of the mashed spleen. . Numerous small lymphocytes in preparations of mashed splenic tissue are found in large groups around the follicles, but a number of them still remain in the reticulum of the follicle and later their presence here served as a good crite-g rion for the identification of the follicles in the growing graft. The dissolution of the normal relations between the constituent parts of the spleen allowed. them, moreover, an easier access into the allantois.

The dissolution of the definite relations, which in a normal organ constitute its characteristic structure, brings a number of various cells within the grafted material under new conditions. Erythrocytes and granulocytes normally streaming in organized channels now stop and inter-mix with other cells. A great number of mesenchymal cells, which in the normal spleen form a tissue of a syncytial nature, now appear as isolated wandering cells. A striking antagonism between the different kinds of cells is soon displayed in certain parts of the grafted tissue and deserves to be reported with a few details.

The cells of the mashed organ,aif the latteris kept in the incubator, gradually die and undergo processes of karyorrhexis and cytolysis. If a certain quantity of the mashed splenic tissue is deposited upon the allantois. of a chick embryo and the egg returned to the incubator and allowed to develop further, the conditions are not equally favorable for allthe parts of the grafted tissue. The grafted tissue forms on the allantois a layer of a definite thickness and the parts which lie in immediate contact’ with the allantois are certainiy in more favorable conditions than those, which are nearer the eggshell membrane. The thin ectodermal layer of the serosa, which at later stages is traversed by a rich capillary net, does not impede an efficient exchange of substances needed for the maintenance and growth of the grafted tissue. The cells of the layer immediately adjacent to the serosa therefore soon manifest an intense activity. The more distant parts of the grafted tissue, however, are in no better conditions than if they were left in a humid chamber, and they soon disintegrate into a fine granular mass.

The parts of the grafted tissue between these two extreme layers often show the above—mentioned antagonism between the various cells which are there situated. Though the spleen is regarded as an organ of complex structure, its cellular elements are simple and well determined. All of mesodermal origin, they partly appear in the form of a delicate cellular web——the cellular reticulum, which is reinforced by a fibrillar net, product of cellular activity. A large part of‘ the splenic tissue, the most intricate in its composition, is represented by the vascular tissue. The relations between the arterioles and the veins and of both to the sinuses have been repeatedly investigated and discussed Without bringing about a definite agreement. The mashing of the splenic tissue dissolves the existing arrangement of these parts, breaks up the sinuses, arteries and veins, transforms the distribution of the originally free cellular elements into a more uniform one.

Numerous cells of the reticular syncytium become mobile after the mashing of the splenic tissue; their bodies are no longer thin vacuolized processes of cytoplasm, fused with those of the adjacent cells, but have now often a Well-defined though variable shape. If the constituent cells of a normal spleen can be divided into sessile and mobile elements (though the latter may be merely astage of the first), the sessile cells of the splenic tissue after mashing all appear, if not actually mobile, at least easily mobilized. They possess a various degree of motility, some of them being highly amoeboid, others more slow in their movements. None of them, however, have in the mashed splenic tissue the facilities of leaving the organ through the normal circulation.

As above indicated, the conditions in the graft are not equally favorable for all the parts of the grafted tissue and not all the cells of a definite zone show themselves equally resistant to conditions not altogether favorable. The cells further removed from the allantois soon show a marked difference in their viability and resistance. The red blood corpuscles are the less viable elements among the various cells found here, their nuclei are soon fragmented, their bodies lose their characteristic shape, the haemoglobin is gradually dissolved and lost, and their remnants are easily ingested by other cells. The greatest activity is shown by the reticular cells of the spleen as early as twelve to twenty-four hours after the graft is made. Whether isolated or still forming small syncytial agglomerations, they manifest now an intense phagocytic activity, which, if displayed after a hemorrhage or another injury in some part of the organism, would give the impression of being a purposeful process for the benefit of the whole organism. Just as after intravenous injection of dyes, the mobilized reticular and reticulo-endothelial cells in this case also become phagocytic. Their activity is directed now against cellular elements. Figure 4 gives merely an approximate idea of the ingestive capacity of these cells. The reticular cells, when just isolated from the common reticulum, are not larger than a blood corpuscle, their nucleus is usually eccentric and somewhat flattened, their cytoplasm is slightly basophilic and vacuolized. These cells, however, are capable of ingesting very numerous erythrocytes, of apparently digesting them, and of attaining large dimensions. It is not a seldom occurrence to find a cell, in which in an optical section twenty to thirty cells can be seen incorporated in its cytoplasm. In such cases the nucleus of the cell appears as a fusiform body flattened against the periphery of the cell. There is no doubt, that these active and resistant cells are the histiocytes of Kiyono or the macrophages of Evans. In a normal organ they exert the same activity against any foreign body; in a mashed organ their phagocytic activity is intensified to such a degree, as to give a strong impression of the existence of an actual antagonism between various cells.

The greatest number of cellular elements ingested by the isolated reticular cells are erythrocytes, degenerated or apparently intact; however, occasionally granular leucocytes and even small lymphocytes may be found within the phagocyte. Figure 4 illustrates without description the further fate of the ingested cells. The cause of the evident predilection shown by the reticular cells to the erythrocytes can only be surmised; whether this cause lies in the greater inertness of the erythrocytes, which are not amoeboid as are the small lymphocytes or the granular leucocytes, or whether the predilection is determined by some chemical afiinity between these two kind of cells, cannot be definitely decided at present. However, the fact that the erythrocytes can also be ingested by a hemoblast (fig. 3, c, C1) or even occasionally by a granular leucocyte (fig. 3, e) seems to indicate that the inertness of the erythrocytes at least exposes them to the attack of any kind of free amoeboid cell. The very mobile small lymphocytes, as will be seen later, are attracted by the living tissue of the allantois and rapidly migrate into the host tissue. The granular leucocytes, though also mobile, continue to crawl around in the grafted tissue and occasionally are incorporated into the amoeboid cells (fig. 3, c). y or

The phagocytic activity is chiefly manifested in the regions of the grafted spleen, in which the reticulum has been separated into single cells or into small syncytial groups of cells. Numerous follicles, as above mentioned, remain almost intact, as shown by preparations of mashed splenic tissue as well as in developing grafts, and in early stages consist chiefly of a cellular reticulum. Though sometimes erythrocytes are found lying free in this reticulum, they are here extremely rarely ingested, especially not at an early stage.

The reticular tissue, which in a normal organ is often considered a mere supporting tissue, when artificially separated into single cells, displays under definite conditions a phagocytic activity of great intensity. This observation confirms and strengthens the views of Kiyono” and Evans” on the reticuloendothelium as being capable of mobilizing at least a part of its cellular elements as phagocytes. The phagocytic activity of these cells can no longer be regarded as a purposeful process instituted for the benefit of the whole organism. It can be compared to the ingesting power of an amoeba, which in the presence of definite cells or particles of a definite kind will ingest them, disregarding, whether this will benefit or will remainindifferent for the whole culture, of which it forms a single individual.

4. Early changes observed in the region of the graft

(1,. Early response shown by the seroso. The first to react actively to the transplantation of the splenic tissue is ‘the ectoderm of the serosa. At the time of the grafting experiments (seven to nine days of incubation) the serosa has a double layer of ectodermalcubic cells. It is soon traversed by the numerous meshes of the capillary net, which form a rich vascular network under the egg-shell membrane above the eotoderm (Danchakoff14) . More or less abundant hemorrhages resulting from the injury to this vascular net may or may not immediately follow the transplantation of a tissue upon the allantois. In a number of cases, in which the grafting was made during the seventh and the eighth day of incubation, not the least sign of a hemorrhage could be seen. The eotoderm, which at this stage of egg development has not yet its capillary net fully developed, six to twelve hours after the intervention may remain as an intact layer under the deposited tissue. Hemorrhages from injury of the embryonic vessels are invariably seen in the region of the grafts in later stages. From the twelfth day of incubation on it is hardly possible to remove the egg-shell membrane without injuring the capillary net beneath. Tissue grafted at this stage adheres however more effectively and completely to the allantois, because of the fragmentation of the ectodermal membrane beneath it.

The ectodermalcells of the serosa, whether traversed by the capillary net or not, when brought into contact with the grafted tissue invariably manifest a strong I reaction. They begin to proliferate with great rapidity; mitoses, which normally are not unfrequently observed among them, increase in numbers, and it is not a rare occurrence to see in the small area occupied by the tissue under a 2—mm. immersion, three to six mitoses, all in one field (fig. 17, 25). Often mitoses of various value are encountered, some of them very regular and similar to those found normally; others, in which the chromatin material is undoubtedly increased and forms very beautiful large and occasionally pluripolar figures. (fig. 17). It seems that not every mitotic figure found in the cell is necessarily accompanied by the formation of two nuclei or two cells. This suggestion is much enforced by the fact that cells with very large nuclei, containing giant nucleoli (fig. 16, g. ep.c.) are encountered or cells containing two and three nuclei.

The ectodermal layer of the serosa begins to grow atypically not only in that the mitotic figures of its constituent cells are irregular but also that the tissue resulting from its proliferation is differently arranged. In regions where the ectodermal layer of the serosa is in immediate contact with the transplanted tissue, the ectodermal cells begin to grow actively towards the graft. Numerous irregular excrescences of the epithelial ectodermal tissue in the form of large and short or of narrow and long papillae are formed. T Their connection with the original layer of the serosa is not interrupted. Such epithelial outgrowths are seen within the grafted tissue, if the grafts were made at early stages before the development of the capil— lary respiratory net. i The hemorrhages which usually accompany the grafts in later stages destroy at least partially the serosa in the region of the graft and the response of the ectodermal cells just described is then much ilessaevident. It is uncertain, whether such epithelial papillae are entirely formed by the apposition of the newly divided epithelial cells, or whether their formation is partially due to an amoeboid activity of the cells analogous to that observed in cultures, by which cells crawl along resistant surfaces. It is remarkable that the form of the cells in such epithelial out growths is considerably changed, and many of them transform from a cuboid or a cylindric shape into a long and fusiform body (fig. 18).

The first change in the serosa after the deposition of mashed splenic tissue upon it is a proliferation of the ectodermal cells. This reaction is most evident in cases, in which grafting was made at an early stage of egg development. It is equally pronounced, though in a modified way, in grafts at later stages. At the eleventh or twelfth day of incubation the ectoderm is traversed and partly covered by a rich capillary net (Danchakofi").”= This delicate structure, if not destroyed by the lifting of the shell membrane, is more easily injured by contact with a living tissue than the epithelial cells of the ectodermal membrane. The thin endothelial membranes, which on account‘ of the blood pressure are under definite tension, now in contact with a relatively large mass of transplanted tissue, easily give way to additional pressure and lose their continuity. More or less considerable hemorrhages follow and exert further deleterious influence upon both the adjacent meshes of the capillary net and the ectodermal membrane. The latter is thereby broken (fig. 11) and no longer forms an uninterrupted layer. Epithelial cells may be found now arranged in larger or smaller groups surrounded by the grafted tissue or by the mesenchyme of the allantois (fig. 1 Ect. P.). In addition to the mechanic injury by the apposition of the graft, the amoeboid activity of the innumerable free cells in the graft contributes not in a small degree to the loosening and to a disintegration of the ectodermal membrane as such. The ectodermal epithelial cells, appearing now as separate islands, also manifest an intensive proliferation. Mitotic divisions are frequently observed and lead to a rapid enlargement of these cell agglomerations. They are frequently traversed by amoeboid cells, which loosen but do not dissolve completely the characteristic arrangement of the epithelial tissue.

The immediate results produced by the apposition of the splenic tissue on the serosa are the following: A partial destruction of the capillary net followed by hemorrhages and an occasional fragmentation of the ectodermal membrane. A stimulation of the ectodermalepithelial cells, which intensively proliferate, no matter whether they remain in the form of an intact membrane or appear as isolated islands.

5. Early response shown by the grafted f23ssuc.' The first reaction observed in the adult cells of the splenic graft after its apposition to the allantois is different from that manifested by the embryonic cells of the serosa. The phagocytic activity of the mesenchymal, respectively of the reticulo-endothelial cells, described in a previous paragraph, depends, in my judgment, upon the radical changes in the reciprocal relations of the various cells brought about by the mashing of the tissue. Phagocytic activity is manifested by the surviving; cells under conditions not altogether favorablefor them and decreases in intensity as soon as the favorable conditions of the environment are more evenly distributed to the grafted tissue. On account of the extensive destruction of the epithelial membrane by hemorrhages in graftings at later stages, the grafted tissue is often seen to occupy the deeper layers of the allantois. ; In such cases the

p cells of the reticulum and those of the endothelium, if situated in

the midst of the allantois mesenchyme, exercise only occasionally phagocytic activity. The phagocytic activity, though an inherent potency of the reticulo—endothelial cells of the spleen, doesnot invariably characterize a grafting, and therefore has been described separately. .

The first response to the new conditions, to which the grafting experiment submits different parts of the splenic tissue, is shown by the small lymphocytes. They manifest a very intense amoeboid activity. _ These cells, the most mobile among a group of free and easily mobilized cells of the splenic tissue, move all in one definite direction, toward the embryonic tissues. The various substances presentin the lembryonictissue produce a striking chemotactic effect upon the small lymphocytes. They leave their places of origin and crawl through the narrow spaces between the other splenic cells toward the layers of the graft in contact with the embryonal tissue; they traverse the ectoderm of the serosa, both if the latter still forms an uninterrupted membrane or if it is transformed into isolated islands of epithelial tissue. They appear in great numbers within the mesenchyme of the allantois twelve to tWenty—four hours after grafting (fig. 12). The first small lymphocytes can be found within the mesenchyme of the allantois as early as six. hours after grafting. The path traversed by these cells before they appear Within the mesenchyme of the allantois is easily traced. Large accumulations of small lymphocytes are early observed just above the serosa; these cells came from all parts of the grafted tissue, which consequently becomes loosened.

The epithelial membrane of the serosa, even though not damaged, is not an obstacle for the immigration of the small lymphocytes into the mesenchyme. Numerous small lymphocytes can be demonstrated in fixed preparations within the epithelial membrane. The -very mobile lymphocytes protrude

small amoeboid processes between adjacent cells and crawl

through the membrane. If a capillary within the epithelium is in their way, they pass sometimes around it, in other instances they traverse the endothelial membrane and can be found within the lumen of the capillary. The meshes of the reticulum formerly occupied by the small lymphocytes now appear vacant, the reticular cells ' more evident and their structure more conspicuous.

The appearance of small lymphocytes within the mesenchymal tissue of the allantois is very conspicuous in the early stages of the graft. Aftertwo or three days, however, they are found only as single isolated cells. Numerous small lymphocytes degenerate in the allantois, their nuclei appear then as spherical bodies of homogeneous chromatic substance, surrounded by a faintly stained rim of cytoplasm. They finally break into two or more particles. Thecytoplasm, which in small lymphocytes is distinctly basophilic, now changes its reaction, and in many cases is stained by eosin. Groups of such degenerated small lymphocytes are often observed in the mesenchyme of the allantois. Degenerated small lymphocytes are readily ingested by wandering cells. A great number of small lymphocytes undoubtedly undergo rapid regressive changes after their immigration into the allantois. s A thorough study, however, makes it possible to trace also further progressive changes in the small lymphocytes. They appear later and will be described together with the differentiation of other structures of the transplanted spleen.

As a direct result of the emigration of the small lymphocytes, the transplanted splenic tissue subsequently appears less dense. It presents quite a different aspect after a great number of mobile cells, which in a normal spleen occupy the interstices, have left it.’ The nuclei of the cellular reticulum, which by the compression of great numbers of mobile cells in a normal spleen were transformed into elongated and fusiform or even polygonal bodies, now appear as oval and sometimes even round structures. They contain distinct nucleoli and small basophilic granules of chromatin. The cytoplasm around these nuclei becomes more evident, but on account of numerous large spaces now left free between the cell processes, this tissue deserves still the name of reticulum. Numerous thin cytoplasmic processes effect the continuity between various cells of the reticulum and now appear distinctly basophilic. The syncytial nature of the cellular reticulum in the spleen is easily recognisable in such parts of the graft which have not been damaged bypassing through the sieve. As above stated, the follicles of the spleen are but slightly affected by this procedure, hence their frequent appearance in the graftsas small areas of mesenchymal syncytium. Figure 6 demonstrates such a follicle at a little later stage. It still shows clearly the structure of the splenic cellular reticulum after the emigration of the small lymphocytes. Only a few of them are present in the meshes of the tissue; new cells appeared from outside and also developed locally. A very distinct feature of the follicle reproduced in this figure is, however, the syncytial arrangement of a large number of its cells.

The small arteries are another structure of the grafted spleen, which deserve special mention in the study of the early response shown by the grafted tissue (fig. 5). These structures are very re— sistant and can be easily identified two or three days after grafting. In the sections they often contain a couple of erythrocytes, their endothelium becomes slightly swollen, but is still formed by individual cells; the structureless membrane surrounds the endothelium as a closed ring, it is only gradually affected and then separates into numerous fibers of a homogeneous substance similar to that, g of which the original membrane was built. The changes shown by the cellular elements of these vessels in later stages will be described in detail together with other progressive changes undergone by the various elements of the grafted tissue. ‘ T '

The study of the early response manifested by the tissue of the graft and by that of the allantois which is in immediate contact with the graft, has shown that most of the cells of these tissues respond rapidly to the new conditions, in which the grafting has placed them. The ectodermf of the serosa, on which the graft is placed can be partly destroyed, but the remaining parts, whether in the form of a continuous membrane or of isolated islands, proliferate vigorously and form masses of new epithelial ectodermal tissue. The various cells of the grafted tissue respond also rapidly. The small lymphocytes leave the graft in great numbers and immigrate into the allantois. The isolated Inesencymalcells of the ‘splenic cellular reticulum, which appear as wandering cells, exercise a great phagocytic activity. The granular leucocytes are not numerous in the adult spleen, neither are they conspicuous in the tissues of the graft in the early stages. Erythrocytes remain inert and are ingested in large numbers by other cells. Parts of the splenic tissue, even after mashing, appear undamaged and easily recognizable in the grafts; these are the small arterioles and islands of reticular tissue of the follicles. The arterioles retain their structure for a considerable length of time, the follicles appear after the emigration of the small lymphocytes as areas of a loose syncytial mesenchyme.


a. Changes in the cellular reticulum. Grafting of splenic tissue has been made for two different purposes; first, in order to study the changes produced by them in the various hematopoietic anlages of the embryo (Danchakoff1°- 11* 12) and, second, in order to study the changes undergone by the grafted tissue itself, in this particular case the spleen. The changes in the tissues of the embryo are more conspicuous in the cases when the grafting is made at early stages, at the seventh or eighth day of incubation, therefore a number of grafting series were made at early stages. However, the grafted tissue develops better, if transplanted in later stages of incubation, about the ninth to the twelfth day. , This difference is easily understood, if the changes in the structure of the allantois at various periods of incubation are taken into account. In early stages the serosa is covered by a compact layer of epithelial tissue, in later stages the ectoderm is traversed by a rich net of respiratory capillaries. The thin ectodermal membrane seems to be a good protective tissue for the layers beneath it, and a greater number of grafts made at earlier stages do not take, if compared with the results of grafting at later stages. n h .

The ectodermal layer of the serosa is at least partially destroye in later stages by the apposition of the grafted tissue, and the graft thereby. is brought into immediate contact with the mesenchymal tissue of the allantois. The mesenchyme of the allantois easily gives way to the grafted tissue, which sinks and is often found to occupy the deeper layers of the allantois. In such conditions the graft is surrounded by mesenchyme for the greatest part of its circumference, and only on the surface toward the egg-shell membrane it is not in touch with living embryonal tissue. The grafted tissue in these conditions develops particularly well and the results described further apply chiefly, to grafts made in later stages of incubation (nine to twelve days).

Among the cellular elements of the adult spleen, situated now in the midst of the allantoic mesenchyme, the splenic follicles and the endothelium of the small arteries are those, which offer a special interest in our effort of determining potentialities inherent in cells, though under laormal adult conditions’ not revealed. The cellular reticulum of the splenic tissue becomes very visible after the emigration of the small lymphocytes and appears as a syncytial tissue. This tissue is now less dense. Its meshes no longer contain accumulations of small lymphocytes. The meshes of this tissue are widely opened toward the embryonic mesenchyme, and the spaces between the cells of the reticulum are now filled with substances, which have free access from the embryonic tissue to the reticulum. The conditions, under which a syncytium of mesodermic cells (which in Laguesse’s conception is un réliquat du mesenchyme embryonnaire—~“) existed in the spleen of an adult organism, have been fundamentally changed. It has been relieved of the presence of large accumulations of small lymphocytes. These. cellular units with active metabolism have been replaced by embryonal amorphous substances very different from those, by which the splenic cellular reticulum was surrounded in the normal adult organ.

i The results of the interaction between the cells of the splenic reticulum and the new environment, in which they are placed, are soon seen in the form of a well-pronounced cellular hypertrophy- Figures 6 and 7 demonstrate it. The small lymphocytes have entirely evacuated the syncytium. The cells of the reticulum are characterized by their relatively large nuclei. Previously elongated and in a normal spleen transformed by the compression of a large number of free cells into irregular fusiform bodies, the nuclei now appear in the section oval and

i even round. This is only partly due to the cessation of the coin pression. As the study of the further changes in such reticular areas shows, it is due in great part to an active growth and a further proliferation of theindividual nuclei. Not only nuclei grow, but also the cytoplasm around them. The thin processes of the reticular cells, which in normal splenic preparations are difiicult to demonstrate, now form a net, the meshes of which appear as cytoplasmic bands of considerable thickness. Around the nuclei the cytoplasm condenses and forms large rims. The mass of cytoplasm increases here steadily‘. Thesubstances contained in the meshes of the syncytiumpare gradually assimilated, the vacuoles becoming smaller and being replaced by the masses of newly formed cytoplasm. Finally the cytoplasm is seen to contain merely few and small vacuoles and form rather a continuous mass, in which numerous nuclei are situated. . We witness a formation of plasmodial masses out of a mesenchymal syncytium.

The development of plasmodial. cell accumulations from a mesenchymal syncytiumi has been frequently reported and forms a common process in the development of blood and vessels. The blood islands which appear in the early embryonic stages are also plasmodial cell agglomerations Within the mesodermic mesenchyme. A development of similar blood islands has been observed also within the embryonic body. Under the name of blood islands one understood originally such condensations of meso dermic tissue, at theexpense of which red blood cells developed as one of the products of their differentiation. The blood islands appear as more or less well delimited islands and soon transform into cell cords, which anastomose reciprocally. The

i term of ‘blood islands’ for these structures came into general use, and an attempt to change this term in hematology could only be compared with an attempt to change in the general biology the term of ‘cell.’ Such an attempt Was recently made by Dr. Sabin.” She proposes to call the structures generally known under the name of blood islands, angioblasts, and to keep thername of blood islands for the dense collections of cells, which before becoming isolated cells, are situated Within the newly-formed vessels. She remarks however, that the masses which she calls blood islands “are attached to the Wall and hence are not, strictly speaking, islands,” and that, “as isolated masses or islands develop the more primitive masses,” which she calls not blood islands, but angioblasts. Angioblasts are, according to her, the blood anlages together with the vascular anlages which after their definite formation retain the potency to further split off blood cells; bloodislands are angioblasts minus vascular wall. The confusion, which may result from substituting His’s. term “angioblast” for the term “blood island,” is aggravated by the proposal to retain the ancient term “blood island” and to apply it to a difierent structure. a g g

Now, is the confusion warranted, which undoubtedly will result from applying old terms designating well-defined concepts to other different phases of development? Dr. Sabin’s observations confirm very definitely by means of the method of cultivation in vitro the well-known fact, that the structures originally known as the blood islands differentiate from the mesoderm and grow by proliferation as Well as by the addition of cellular elements from the mesoderm; that they develop into blood cells and an endothelial membrane, the cells of Which, as recently confirmed by Emmellg and J ordan,21» 92 retain hemoblastie potencies for a certain length of time. A very interesting detail in her observations is, that in cultures of blastoderms made in Locke’s solution, the cells of the blood islands, after separating a vascular wall, develop in their totality into erythroblasts, while in normal development a stock of undifferentiated stemeells is always preserved. Her observations confirm our contentions, that the cells of the blood islands are all endowed with homeblastic potencies, whether they appear as cell masses within the vessel or arrange themselves in the form of a specialized endothelial membrane. The fact that blood island cells situated Within the vessels or cells split off into the lumen by the endothelial membrane differentiate even in Locke’s solution only into erythrocytes, strengthens my view-point that the erythroblastic differentiation is determined by intravascular conditions. ‘There only Willa hemoblast develop into erythroblasts and into erythroblasts only.

To deny to the blood islands other hematopoietic potencies than the erythroblastie one, on the ground, that they do not develop into other cells within the vessel, is unwarranted. I have shown that cells in the splenic anlage, which have the morphological structure of lymphoid hemoblasts or stemeells and which normally develop into white blood corpuseles, will differentiate into erythroblasts if situated within a developing sin us.”

Dr. Sabin states that the cell agglomcrations originally known as blood islands are morphologically islands. Histogenctically derived jrom the mesoderm, they are anlagcs for blood and endothelium. The endothelium at least in part retains, during the embryonic period, a hematopoietic function as a blood anlage; the structures originally known as blood islands are therefore chiefly hematopoietic. The cells of the endothelial membrane themselves must be considered as a temporarily specialized form of mesodermal cells which continue to exercise a hematopoietic function.

After this little more need be said to show thata structure, the immediate derivatives of which are mainly blood forming, deserves more to retain its ancient name of blood island than to be called angioblast, a term which in anatomical literature has a definite and vvell—defined meaning (His). This is especially true when we consider that syncytial and plasmodial agglomerations or blood islands can develop within the mesenchyme and not be followed by a differentiation of an endothelium and still function as hematopoietic anlages. Such cell agglomerations which also are properly called blood islands separate into. single cells, but, situated here outside the vessels, they differentiate into white blood corpuscles and not red ones. On the ground of these considerations I shall continue to use the term of blood islands in the sense employed in my previous papers. Under this term I understand agglomerations of mesenchymal cells in the form of dense syncytia or plasmodia, which will develop into blood cells of any kind. If the cells of the external layer of these agglomerations specialize into an endothelial membrane and the cell agglomerations are further incorporated into the vascular system, red blood corpuscles are developed as seen in the area opaca. If they separate into single cells and remain outside the vessels, White blood corpuscles are formed, as seen in the area ppaca as Well as in numerous regions Within the body of the embryo.

The name of blood islands in the grafted tissue is fully deserved by the parts of the cellular reticulum, which after the emigration of the small lymphocytes hypertrophy into plasmodial accumulations. A beginning of the development of such blood islands can be seen in figure 6. The proliferation together with the growth of the individual cells transforms areas of syncytial tissue in the graft into blood islands. The structure of cytoplasm and nuclei in such cellular agglomerations is very similar to that in the original blood islands encountered in the earlier stages of embryonic development. Like those cell groups, which in the area vasculosa remain outside the newly formed vascular net, they appear for the most part in small groups of three to ten or twelve cells. They soon develop, however, into large plasmodial cell agglomerations. Their nuclei are large and contain well-developed true nucleoli and small particles of chromatin.

The plasmodial agglomerations in later stages of their development become more basophilic, and the cells in their outer layers show numerous amoeboid processes, by which they finally separate from the agglomerations and become isolated.

This is again a process analogous to that found between the Vessels in the area Vasculosa of the birds and reptiles (Dancha— koff 5, 6, 9), as Well as in the other blood-forming anlages (7, 8, 9) . In the area vasculosa extravascular lymphoid hemoblasts developed from small blood islands left outside the Vascular net. The number of these cells increased here by a process of a continued separation from the mesenchyme and also by a proliferation of endothelial cells, some of which passed into the spaces between the Vascular net. In the grafts both separation of the plasmodial masses into single cells and intense splitting off of free cells from the splenic syncytium can be seen. Numerous cells of the syncytium hypertrophy, break up their continuity with other cells, and creep freely Within the graft. They develop a large rim, of basophilic cytoplasm, are intensely amoeboid, frequently divide mitotically, and form large accumulations of free amoeboid cells, which finally infiltrate the syncytial tissue to such a degree as to transform it again into a reticulum (figs. 7, 8, ,9). Onlynow, if compared with the reticulum of an adult organ, its meshes are distended by large basophilic cells, the lymphoid hemoblasts and not by the small lymphocytes as previously. Large parts of the cellular reticulum of the spleen are transformed into such tissue about three or four days after grafting. In the regions, in which the grafted tissue has been broken into single cells by passing it through the sieve, the cells hypertrophy even more rapidly. The mesenchymal cells, which after their forced isolation, appear as Wandering histiotopic cells, transform rapidly into lymphoid‘ hemoblasts. Finally, large areas of grafted tissue assume a more embryonal character and in many respects can be compared to the tissue of the splenic anlage before the development of the small lymphocytes. Recent studies on the reaction shown by various tissues in cultures seem to yield results analogous to the above described. Champyfi Avroroff and Timofeevsky,‘ and recently Lewisflfi find that tissues cultured in vitro assume embryonal characters, and they speak therefore of cellular dedifferentiation. A cursory comparison of the results obtained by growing splenic tissue in the allantois with those gained by cultures of various tissues in artificial media may lead to the conclusion, that the results of both series of experiments are similar. However, the process, by which the adult splenic tissue assumes embryonal character is entirely different from that described by the above—mentioned authors. There is no dedifferentiation of a differentiated cell in the culture of the spleen in the allantois, as Lewis=5 describes for striated muscle fibers, or Champy3 for kidney tissue, and Avroroffl even for granular leucocytes. Most of the differentiated cells of the splenic tissue, if grafted on the allantois, show regressive changes; the erythroblasts are ingested, also some of the granular leucocytes; the small lymphocytes leave the tissue of their origin and immigrate into the allantois. The greatest part of them succumb and do not show dedifferentiation. Only the reticular tissue of the spleen hypertrophies and finally resolves itself for the greatest part into free amoeboid cells of the type of lymphoid hemoblasts oi stemcells for blood development. This process is not a dedifferentiation of the reticular tissue, neither of any of the splenic cells. The cellular reticulum of the spleen which is regarded by Laguesse‘-’4 as a ‘réliquat du mésenchyme embryonnaire’ transplanted into the allantois proves to be truly an embryonal tissue in the sense of being capable of further difierentiation. The hypertrophy of its cells, the development of plasmodial cell agglomerations or blood islands, the separation of the constituent cells of these blood islands into lymphoid hemoblasts are not processes of dedifferentiation of differentiated cells, but are processes of further differentiation of young undifferentiated cell groups retained in the spleen from the embryonic period in the form of its cellular reticulum ; of course, the results of the study of the spleen grafts on the allantois can not be applied to cultures of different tissues in artificial media. However, the recent studies by Maxi1n'ov2"' of splenic cultures in vitro do not support the views of Champy on the process of dedifferentiation of various growing tissues in vitro.

The development of large numbers of lymphoid hemoblasts within the grafted tissue is soon followed by their further differentiation into granuloblasts. Figures 8 and 9 demonstrate this process, which usually is observed within three to five days after the grafting.

The figure 1 which is a slightly schematic figure, with the number of cells somewhat reduced shows a graft at this stage. It represents a small part of the graft developing in the allantois. The inner layer of the allantois is formed by a thin membrane of fiat entodermal cells s(Ent.). The mesenchymal part of T the allantois appears oedematous and contains numerous vessels. The serosa in this particular graft has been depressed (E’ct., E'ct.P.) and formed a deep invagination, which contained the grafted tissue. Figure 1 shows a part of this invagination with a follicle of the grafted. spleen (Gr. S .25.). The grafted tissue consists for the most part of well-developed cells. In the proximity of the allantois they mobilize in large numbers and on account of the basophilic cytoplasm give to the tissue in preparations stained by eosin-azure a deep blue color. At a certain distance from the allantois, the grafted splenic tissue has taken the arrangement of the fibrous connective tissue. Only 3.-SII1.}.ll amount of the. grafted tissue did not survive and appear in figure 1, (N.t.), as an acidophilic necrotic mass with numerous small basophilic granules. The ectodermal layer of the serosa, which at this time is fused with the allantois, has lost its continuity and appears in the form of numerous islands of epithelial tissue Ect. P. These islands f01‘1I1. a well-marked boundary between the mesenchyme of the allantois and the grafted tissue. Also the striking contrast between the loose tissue of the allantois and the dense structure of the graft assures the identification of the grafted and of the allantoic tissues. A communication between these two tissues is established in the intervals between the ectodermal islands. Through these spaces small vascular ‘branches grew from the allantois. into the grafted tissue.

The further differentiation of the lymphoid hemoblasts into granulo blasts, is associated with the establishment of circulation in the grafted tissue. An intimate relation of granuloblastic differentiation with localisation in a richly vascularized region is also seen in the area vasculosa. Single lymphoid hemoblasts can transform into granuloblasts also at greater distance from the vessels, as observed throughout the mesenchyme of the embryo body7, but large accumulations of granuloblastic tissue invariably develop around i vessels. The accumulations of granuloblastic tissue between the sinuses in the yolk sac appendages“, inthe bone-marrowg, around the large venous sinuses of the embryonic spleen”, also around any vessel in the embryo body after splenic grafts can be cited as examples. The development of granuloblastic tissue around the vessels in the yolksac appendages is so intense that in numerous regions vessels are provided by uninterrupted sheaths of granuloblastic tissue. At the time of the discovery of the hematopoietic function of the vessels in the yolk sac appendages, I was led to believe that large accumulations of granuloblastic tissue signified their possible participation in the absorption of yolk and its transmission into the vessels. Later experience has taught me to regard the structure of a certain tissue, even if strikingly purposeful, as an unavoidable result of interaction. between the physicochemical constitution of cells and environment. , The differentiation of hemoblasts into granuloblasts in all the above—cited regions takes place regardless of whether the results ensuing from this differentiation might prove useful for the organism (development of granuloblasts in the yolk sac) or indifferent and even harmful (development of large cysts of granulocytes with consecutive necrosis of the cells in the embryo after splenic grafts).

The granuloblastic A differentiation of the lymphoid hemoblasts begins soon after the ingrowth of the embryonic vessels into the graft. Large numbers of lymphoid hemoblasts develop within the grafted tissue before the appearance of embryonic vessels. This alone would suffice to determine their local origin. However, an analysis of the contents of the vessels should be made in order to see whether some hemoblasts were carried in by the blood current. The vessels as seen in figures 8 and 9 contain at this stage only differentiated cells, red blood corpuscles and granular leucocytes. Only occasionally a small or a large lymphocyte is found Within the developing capillary net in the graft, and this certainly cannot account for the large accumulations of lymphoid hemoblasts in the grafted tissue. The embryonic vessels around the graft, especially the smallest branches, contain numerous granular leucocytes, which soon infiltrate the regions around the necrotic parts of the graft.

At the time of the granuloblastic differentiation of the hypertrophied and isolated reticular cells of the spleen, the grafted tissue has a characteristic and not altogether uniform appearance. Larger areas of surviving cellular reticulum show marked structural difference in their central parts and in those adjacent to the allantoic mesenchyme. A dense mesenchymal syncytium can be situated in the center of such an area, while numerous isolated cells occupy the more distant parts. A granuloblastic differentiation can already start in the lymphoid hemoblasts at the periphery of a follicle, while its central parts may still show characteristic features of the reticulum of the adult spleen.

The gradual changes undergone by the lymphoid hemoblasts during its granuloblastic differentiation have been already described in detail and need no further comment. Many beautiful pictures of the gradual differentiation of granuloblasts are seen now in the grafted tissue. Acidophilic granules appear usually in a hemoblast with a large mass of cytoplasm. The periphery of a slightly acidophilic centrosphere is often seen to be the first place of origin of these acidophilic granules in hemoblasts. They increase in size and number and soon occupy the greatest part of the previously homogeneous basophilic cytoplasm. Many young granuloblasts and even some of the hemoblasts contain vacuoles in Zenker-Formol preparations, the contents of which are preserved in alcohol and appear in the form of a basophilic substance. In further differentiated granuloblasts these vacuoles disappear, ‘and such cells contain eosinophilic granules only.

The differentiation of the granuloblas-ts is completed when it transforms into a typical granular leucocyte with rod—shaped eosinophilic granules and a polymorphic nucleus. During the whole process of differentiation the granuloblasts proliferate mitotically and their number increase .both by differentiation from the hemoblasts and by their own proliferation.

In many cases nearly the entire surviving graft was transformed into granuloblastic tissue. The results approached the changes known in pathology under the name myeloid metaplasia of the splenic tissue. The excessive development of granuloblastic tissue in a lymphoid organ is observed during various diseases. Erythropoiesis and leukopoiesis develop simultaneously in mammalian lymphatic organs during myeloid metaplasia. These changes in lymphatic organs were explained from three diflerent points of view. It was believed by a number of pathologists that specific granuloblasts were brought in these diseases by the blood current from the bone-marrow into the spleen. They settled here, proliferated, and brought the lymphatic tissue to an atrophy. Others held the views that the proliferation of granuloblastic tissue in lymphoid organs was autochthonous and due to a stimulation of dormant specific cells retained in the spleen from embryonic periods. a Only a few scientists looked upon the myeloid metaplasia in the spleen as upon a manifestation of one of the various potencies inherent in the splenic cells, but not revealed in normal. adult conditions.

The study of the gradual transformation of theadult splenic graft into granuloblastic tissue offers a good opportunity for defining the source of the granuloblasts (myelocytes) in the adult splenic tissue. On the basis of the data described above, the hypothesis of metastasis is easily disproved. Nor could the existence of definite cells endowed with ‘specific potencies to granuloblastic differentiation only find any support. On the contrary, the whole weight of the evidence obtained by the present study supports the view that the granuloblastic metaplasia in the adult splenic organ is due to the existence of a tissue endowed with various potencies. In normal conditions it develops into small lymphocytes; if transplanted within the allantois it develops into granuloblasts. This tissue is the cellular reticulum of the spleen. In the present experiments its granuloblastic potency is manifested in an unquestionable way. The small lymphocytes normally present in great numbers in its meshes, evacuated their places of origin, the further development of small ' lymphocytes ceased, and the cellular reticulum became most apparent. It gradually transformed into a granuloblastic tissue.

The extension of the myeloid metaplasia in the graft varied somewhat in differentcases. It was invariably very intensive if the graft was amply provided with blood-vessels. The metaplasia was more complete and rapid if the grafted parts were smaller.

One case of grafting deserves special mention. The graft after six days of growth attained considerable dimensions. -The grafted tissue was situated within the allantoic wall, the ectoderm regenerated above it in a complete epithelial membrane of partly cornified cells. Necrotic tissue was entirely lacking. Only a few small vascular branches traversed the graft. Its cells proliferated intensely. The process of hypertrophy and splitting off of amoeboid cells transformed the grafted tissue into large accumulations of wandering cells. These cells, however, did not differentiate further into granuloblasts. They remained as typical histiotopic wandering cells similar to those found in the fig. 2 g, h, i, k and exercised occasionally active phagocytosis (fig. 3 f It is difficult to say whether this remarkable arrest of further differentiation was due to the absence of a Well-developed circulation in this particular case.

Another case of graft also presented considerable interest in that the small lymphocytes failed to evacuate the grafted tissue. They partly remained in large numbers within the meshes of the reticulum, and formed small irregular accumulations around the grafted tissue. This graft was studied after eight days of growth in the allantois. Small vascular branches grew into the graft and the whole tissue after eight days was healthy and well developed.

Two principal zones could be distinguished in this graft, The innerzone was formed by a uniform splenic reticulum containing numerous small lymphocytes. Among them many showed an increase in size, especially of the cytoplasm, an intensification of the basophilia of the cytoplasm and a transformation of the nucleus characteristic for the small lymphocytes into a ‘Radkern.’ The small lymphocytes through these changes approached the structure of the plasmacells. The reticulum showed a marked hypertrophy in those regions only, where the small lymphocytes were less numerous, and clusters of large hemoblasts could be here discerned. It was rather curious to miss in this tissue any indication of the previous existence of small arterial brariches. Small capillaries traversed the graft, but these vessels were branches of the embryonic vessels. s The adult small arteries, which are provided with a characteristic elastic coat, disappeared completely with no sign of necrotization. The fate of these vessels and the changes of their constituent elements will be studied later.

The second zone, the outer one, consisted chiefly of granuleblastic tissue and in stained preparations was easily recognized even with a low power. It occupied about a third of the diameter of the grafted tissue. Its passage into the inner zone was gradual. The granuloblastic zone consisted of a dense tissue, which contained hemoblasts in various stages of their differentiation into granular leucocytes. Younger granuloblasts were more numerous toward the center of the graft, While the more mature cells occupied preferably its peripheral parts.

Accumulations of small lymphocytes were aggregated around the graft, and among them numerous plasmacells were found. The presence of transitional stages made it easy to ascertain that the plasmacells developed here from the small lymphocytes. No evidence was found in favor of a possible transformation of the small lymphocytes into hemoblasts. The fact that the small lymphocytes do not transform into phemoblasts in the allantois does not completely exclude this possibility. It shows, however, distinctly» that the conditions, in which the hemoblasts rapidly develop from the mesenchyme, proliferate and differentiate further, are not sufficient to make a small lymphocyte develop into a hemoblast. It is not my intention to describe the many individual peculiarities which various splenic grafts showed, my principal d subject being the study of the developmental potencies inherent in the elements of the adult . splenic tissue and of the new differentiation lines manifested by polyvalent cells in accordance to new environment. .6. Changes "fry, the crnvdothclrlum. proper. Besides the cellzlar reticulum of the spleen, the true endothcliums of the smallest arterial branches manifest in the graft very interesting changes. These branches appear usually grouped together and are surrounded by a Inesenchymal tissue. The smashing of the splenic tissue before grafting has partly disconnected them from their stembranches, but their individual structure is for the most part well preserved for a considerable length of time. Figure 5 shows a group of such vessels three days after grafting. One of the vessels still contains twoerythrocytes. The endothelium of the vessels consists of cellsewith largeoval nuclei, identical with those of the mesenchymal cells in the immediatesvicinity of the vessels. The elastic-membrane is still present, but it begins to resolve itself into bundles of T fibrillae. The mesenchyme between the group of a small arterioles consists of a syncytium of cellsswith afslightly basophilic vacuolized cytoplasm. Some of these cells separate, hypertrophy, and transform into more basophilic wandering cells. The same process is observed in the endothelial cells of the vascular walls. The endothelium of the vessel at is undergoing‘ such transformation. It is formed in this section bytwo cells only, which are connected together not only at the circumference of the vessel, but a band of vacuolized cytoplasm extends from one eellto the other through the lumen. One of "them, if compared with the endothelial cells of the adjacent vessels, resembles them. The other, however, exhibits. the structure of a wandering cell, has already partly separated. from the wall and sends a long cytoplasmic process outward, through the loosened elastic membrane. The limited space does not allow me to fully illustrate this process of transformation in the graft of endothelial cells into wandering cells. It becomes veryconspicuous at the time when the elastic mem— brane partly disintegrates. It is not a rare occurrence that, after the disintegration of the elastic membrane, groups of endothelial cells join the mesenchymal syncytium and so become an in-y tegral part of it. Morphologically at least, it is no longer possible to differentiate it from the mesenchymal cells. A similar loosening of the vascular endothelium in the embryonal tissues of the allantois and the persistence of its cells under the form of mesenchymal cells has been “already reported in my last paper.” The present results give new experimental evidence of the morphological adaptability of the mesoepithelium, not only in early embryonic, but also in adult stages. This fully confirms Huntington?“ views, that the endothelial cell is merely an adapted form of an indifferent mesenchymal cell. Whether or not an adult endothelial cell is capable of further heteroplastic differentiation cannot be decided on the basis of the present study. An endothelial cell can no longer be recognized‘ as such in the splenic graft after it becomes an integral part of the mesenchyme.

From that time on its potencies cannot be determine diseparately from those of the mesenchyme. The embryonic endothelium, however, after grafting, exhibits in the allantois, as will be seen in a later paragraph of this paper, a heteroplastic differentiation in the highest degree. This membrane shows progressive changes in all its parts, chiefly, however, in the mesodermic derivatives, and the study of their changes forms the second part of the paper.

An important conclusion can be deduced from the reactions exhibited by the constituent elements of the adult splenic tissue grown in the allantois. The changes undergone by the living matter during development are not always specific. They may lead to a specialization of tissues without difi”e.rentiating them specifically. The difference between these two processes con-— sists in that specialization does not imply a limitation of potencies in the cell, while specific differentiation is a process, by which the constitution of the cell is changed irrevocably andtitsrpotencies to development are narrowed. The distinction between these two processes would make it unnecessary to introduce a new concept of dedifferentiation in order to understand certain phenomena.

During the gradual development of tissues in the organism, some of them acquire a definite structure and show themselves univalent, no matter under what conditions they are placed; such are granuloblasts (or myelocytes) and erythroblasts, to cite only examples from the field more familiar to me. Similar anlages, tissues and cells are specifically differentiated or univalent and their metabolism not reversible. The latter is modified to such a degree that under no conditions do such tissues change their line of development. Their further differentiation can stop, if the conditions be unfavorable, but it cannot be interchanged for another line of development. The changes which a specific tissue is undergoing are predetermined. To surmise the existence of a process of dedifierentiation is to destroy What is obtained by establishing the concept of differentiation. Differentiation or specific transformation of a polyvalent structure into a univalenti one does not take place, if a possibility of, dedifferentiation still exists. A univalent structure is not univalent if it can again become polyvalent. An erythroblast and a granuloblast can neither interchange nor assume an altogether new line of development, no matter What conditions they are subjected to; they will either die or develop into their specific products. Such structures are specific or univalent, they are specifically differentiated.

There exist in the organism other kinds, of tissues, which in the general economy take up distinct functions and appear mor« phologically as well determined structures, at least as far as the reciprocal arrangement of their cells goes. Under normal conditions in the embryo their progressive changes seem to be restricted to a moderate proliferation. In the adult organism, in which a stable equilibrium between different tissues is attained, they seem to be rather inert. The results of their constructive metabolism are given off and not accumulated Within the cytoplasm of the cell, therefore they retain rather a simple structure.

The cellular splenic reticulum and the vascular endothelium are examples of such tissues. Both tissues have been considered as specifically differentiated or univalent structures, especially in the adult organism- It has been shown this paper, however, that neither of them is specific. Under normal conditions in the embryonic stages the cellular reticulum of the spleen is a source of manifold specific products. In normal adult conditions, however, neither of them seems to manifest any appreciable differentiating power; their metabolism becomes steady and their structure therefore remains unchanged. Transplanted into the allantois they both rapidly change their metabolism and new structures arise as an unavoidable result. The cells of the reticulum proliferate and differentiate further into granuloblasts. The granuloblastic potency is retained by the cellular reticulum of the adult spleen though not exhibited under normal conditions. It has also been shown, that endothelial cells may loosen their reciprocal relations and enter into the composition of the allantoic mesenchyme.

On the basis of these facts neither of these adult structures can be considered as being specifically differentiated. They are only specialized structures and both (endothelium in embryonic stages and cellular reticulum in ’ embryonic and adult stages) retain the potency to various differentiation, therefore must be considered plurivalent. The granuloblastic potency of the splenic reticulum is not altogether lost in ‘adult conditions, but only inhibited, it is manifested anew as soon as the tissue is transported into favorable conditions. If endothelium and cellular reticulum differentiate further into specific structures,

I itis not by a process of dedifferentiation, but by the process of further differentiation. The terms ‘specialization’ and ‘specific differentiation’ might in my opinion give the distinction between the two processes above outlined. They are qualitatively different, the one being a limitation of potencies, the other an inhibition of them for a certain length of time. Since under ‘differentiation’ tout court any progressive change is usually meant, the term ‘specific differentiation’ could be used for designating a process, by which an anlage, or a tissue, or a cell is rendered univalent. There would be no need to invoke a mysterious process of dedifferentiation in order to-explain a further differentiation of a presumably differentiated but in reality only specialized structure.


The figures 10 and 11 on plate 4 represent two photographs of

(well-developed splenic grafts on the allantois. In one case (fig.

10), the allantoic membrane showed a well—marked local thickening around the graft; in the other (fig. 11), the whole membrane underwent considcraiible changes and besides the thickening around the graftcontained in its walls innumerable small tumors (pp. 135 and 165). In both cases all the constituent parts of the allantois responded to the grafting. The response in the first was, however, a local one, while in the second case it extended to the entire membrane. ( , , I

a. Response of the ectoderm. The early reaction shown by the ectoderm of the serosa around the graft has already been described. It consists in a proliferation of the ectodermal cells and is evidentlydue to the irritation produced by the apposition of the graft. The proliferation in the ectoderm persists in later stages. The doublelayer of cubic ectodermal cells, if not damaged by the grafting, are soon transformed into a stratified epithelium. This transformation may extend to a considerable distance around the graft. If the ectoderm is fragmented by the injury during grafting, the cells of the ectodermal islands still proliferate and form solid sprouts and typical oornified pearls (figs. 1s and 18).. The proliferation of the ectoderm proceeds atypically in many places. Very large cells develop (fig. 16) and may contain giant nuclei. Also two or three equatorial plates can be seen within an undivided mass of cytoplasm. Strands of epithelial tissue, as seen in figure 25, can from the deep layers of theectoderrn creep along the vessels, which traverse at this stage the ectoderm of the serosa. The cells of the ectoderm become highly polymorph, their form being the result of external (mechanical factors. From apolyhedrio form they readily change into elongated and even fusiform bodies (fig. 18)». The .ectodermal epithelium transforms even into a reticulum—like tissue, if; it is densely infiltrated by amoeboid cells. Such regions in the graft have a striking resemblance of the thymic tissue,; the epithelial pearls (figs. 1 and 18) simulating the Hassals corpuscles.

31. Response of the entoderm. The entoderm is not directly affected by the process of grafting. It is separated from the grafted tissue by the mesodermal parts of the allantois and does not react until two or three days after grafting. Its reaction also consists in a proliferation which may transform the thin layer of flat entodermal cells into a r stratified epithelium of cylindrical cells. The proliferation of the entoderm is strictly limited to the region, in which themesodermal parts of the allantois are affected. The entoderm in these parts transforms into a more or less thick layer of cylindrical cells not only in the part of the allantois upon which the graftingtis made, but also at any point where proliferation of the allantoic mesenchyme occurs.

c. Response of the mesodermals parts of the allantois. The reaction of the allantoic mesenchyme to the grafting can be regarded as a part of the responseof the whole embryonic mesenchyme to the resultingstimulation. This stimulation has led the splenic anlage of the host to hypertrophy after grafting andto develop into an organ which many times exceeded the normal dimensions. The stimulation of the allantoic mesen-T chyme does not differ in principle from that of the embryonic body and soon leads to i very striking changes in the structure of this membrane. Figure 11 shows that the whole surface of the allantois may becovered With small whitish dots, which areas many foci of active proliferation in the mescnchyme of the ‘membrane with subsequent granulopoietic differentiation. These peculiar changes do not always accompany a graft, even a Well-growing one. In the case illustrated by figure 10 two welldeveloped splenic grafts grew on the allantois, but the mesonchyme responded specifically’ only in the immediate vicinity of the grafting, causing thereby a tliickeninggof the membrane, which

in thelphotograph appears as a whitish area around the grafts.

In one of my previous papers” I was inclined to attribute the peculiar reaction of the allantois, represented in figure 11, to the local effect of the grafted tissue, if the latter was in some way dispersed upon the allantois. Though the dispersing of the grafted material is usually accompanied by the development in the allantois of miliary proliferating centers, it seems, however, that these foci may be also the result of some agents brought by the circulation. I must reserve myvfinal judgment concerning this point until later. p g

The endothelium of the allantoic vessels and its loose mesonchyme are the structures which after splenic adult grafts react in an intensive and Well’-determined way. Plate 5 illustrates the reaction of the mesodermal elements of the allantois in the immediate vicinity of the graft. The normal mesenchyme of the allantois consists of rather small many-branched cells. 1 In the vicinity of the graft the mesenchyme of the allantois is infiltrated by small lymphocytes emigrating from the graft. The mesenchymal cells here hypertrophy (fig. 12) and split off

numerous amoeboid cells with the structure of the lymphoid.

hemoblasts. The latter proliferate and form considerable accumulations (fig. 15). 1 1 s Similar groups of lymphoid hemoblasts are formed also at the expense of the endothelium of the small vessels in the vicinity of the graft (fig. 13). Numerous small capillaries are injured here during grafting; the circulation in many of them must have stopped; the lumen in such vessels might be distended by degenerating erythrocytes, or contain only a few of them. Their endothelium manifests interesting changes. In some cases the endothelial cells separate and join the mesenchymal syncytium of the allantois (fig. 14) in other cases they transform into lymphoid hemoblasts and as such migrate and infiltrate the mesenchyme around the vessels. 1 It is extremely difficult to decide whether or not any of the small lymphocytes, emigrated from the spleen graft and now present in the allantoic tissue, may transform into lymphoid Ehemoblasts. Occasionally one finds in the vicinity of the graft groups of lymphoid hemoblasts much smaller than they usually appear. Such a group is represented in figure 15. Among a number of small lymphoid hemoblasts a few typical small lymphocytes are seen. Numerous small cells in mitotic division are usually encountered in such groups. As seen in figure 15, the size of some of the lymphoid hemoblasts is not larger than that of the small lymphocytes. Their structure is, however, distinctly different. In my opinion, the small size of the lymphoid hemoblasts in such foci is due rather to the high rate of proliferation as exhibited in the great number of mitotic figures than to their genetic connection with the small lymphocytes. The preservation of the very distinct structure in both of these cell groups in spite of their nearly equal size can be considered as a further argument for their distinct and independent nature.

Numerous small lymphocytes perish in the allantois, distintegrate, or are ingested by mesenchymal cells. There is, however, a possibility that small lymphocytes maymanifest further progressive changes. The mesenchyme around the graft contains amoeboid cells not only in the form of large lymphocytes, but also of numerous typical histiotopic wandering cells. A series of intermediate stages between a small lymphocyte and a large wandering cell can be easily drawn, as shown in figure 2, and the possibility of a further transformation of a histiotopic wandering cell into a hemoblast is most evident. I doubt. how’ever, whether this is sufficient. evidence for establishing a genetic link between the small lymphocytes and the various differentiation products of the lymphoid hemoblasts, which develop in the allantois after grafting. T

The peculiar changes shown by the allantois in figure 11 depends upon the activity both of the endothelium and of themesenchyme.* Both manifest themselves as active hematopoietic respectively granulopoietic anlages. Plate 7 illustrates the origin of a great number of granulopoietic centers in the allantois. Figure 19 shows a typical small vessel in a normal

'-The erythropoietic foci, which, no doubt, contribute to the peculiar transformation of the allantois, as represented in figure 11, were made the subject of a special study (16) and will no longer occupy us in this paper. allantois. Figure 20, without any further comment, illustrates the proliferation of the endothelium, the separation from it of numerous amoeboid cells in the form of lymphoid hemoblasts and the infiltration by these cells of the mesenchyme a surrounding the vessel.

Figure 21 illustrates the further differentiation of the hemoblasts split off from the vascular endothelium. The endothelium of the vessel has undoubtedly been the ultimate source of the greatest number of the large accumulation of hemoblasts represented in this figure. The endotheliumto the right and low corner of the figure is still separating groups of hemoblasts. Among them cells are seen joined in common plasmodial masses, very similar to true blood islands.

The hemoblasts easily change their metabolism, what in our

preparations is reflected in their rapid structural changes. The

hemoblasts proliferate intensely when their metabolism. does not change qualitatively in a marked Way. They differentiate into granuloblasts outside of the vessels as soon as their metabolism begins to change qualitatively. This occurs more rapidly in groups of hemoblasts situated around vessels, especially around such with thin Walls. Only in the regions distant from the vessels do groups of hemoblasts retain unchanged their structure for a certain length of time. In such regions they even develop gradually structural features closely resemblingthose of the histiotopic wandering cells, and can be here regarded as representing elements in a phase of less active metabolism. They may moderately multiply, but they do not intensely differentiate here. Their differentiation into granuloblasts is closely associated with theirlocalization around the vessels.

An interesting feature in the granuloblastic differentiation is, that it starts in isolated cells only. Not until a cell is freed

from the mesenchymal syncytium and its reciprocal relations to

the surrounding cells are broken, does it develop a metabolism which transforms it into aagranuloblast. The association of cells in syncytia has an absolutely unfavorable influence upon the synthesis. of chemical products characterizing various blood cells. While the cell is a part of the syncytium, it is evidently regulated in its metabolism by the common reactions taking place in the Whole colony. As soon as the cell loses its physical continuity with the neighboring cells, it becomes more independent in its metabolism and at the same time, if the conditions are favorable, begins its heteroplastic differentiation. The entire surface of a free cell which has been separated from the mesenchyme is accessible to interaction with the environment. The previous continuity of the cytoplasm with that of other cells of the syncytium no longer inhibits an independent metabolism, and it is only natural to see suchcells hypertrophy and follow a further differentiation. Their metabolisms is not a self-perpetuating process, and changes according to the environment. This is reflected in various lines of qualitatively different development of the hemoblasts (erythroblasts and granuloblasts) and in a long range of gradual quantitative changes, occurring in each line of differentiation (various stages of differentiation of hemoblasts into erythrocytes and granulocytes). The association of the granuloblastic differentiation with the isolation of the cell as a hemoblast is particularly well soon in the allantois, as represented in figure 21.

Any part of the allantoic mesenchyme may become a granuleblastic center. The gradual development of such foci .immediately under the ectoderm is shown in plate 8. In figure 22 the beginningof such development is shown. Figure 23 illustrates the further growth of ya granuloblastic center situated immediately under the ectodermal membrane. Mesenchymal cells proliferate, -hypertrophy, Withdraw their processes, and appear as wandering cells. To What extent such a wandering cell can occasionally hypertrophy is shown in figure 24, representing a giant hemoblast creeping among other wandering cells in the mesenchyme of the allantois.

The iectoderm above the proliferating mcsenchyme responds by a proliferation also. Mitoses are frequently seen in the ectodermal cells and the vvholetsectodermal layer soon becomes stratified epithelium. s Obliteration of the respiratory capillary net follows subsequent to the cornification of the ectodermal cells. Ectodermal cells are also seen to proliferate as solid sprouts from the deep surface of the ectoderm and follow the small vascular branches. which unite the respiratory capillary net with the more deeply situated vessels (fig. 25).

The proliferation of the mesenchyme is the primary and the most important feature in the development of the large number of the small tumors in the allantois. This proliferation, as seen in figures 21 and 23, extends often over the entire mesenchyme between ectoderm and entoderm. It results finally in the formation of very numerous small tumors. At first they contain only hemoblasts, then the greater part of these differentiate into granuloblasts and granular leucocytes. Some of the small vessels traversing these tumors are seen to contain a great number of granular leucocytes. It is an obscure point for me as to whether the granular leucocytes seen in these vessels are immigrating into the vessels from the granulopoietic centers or vice versa. It seems only natural to suppose a flow of granular leucocytes into the vessels, and still I have data which do not allow me to pronounce a final judgment upon this point. The infiltration of the tissue by granular leucocytes might be so intense as to cause a liquefaction of parts of granulopoietic centers in the allantois. -The granular leucocytes are in larger granulopoietic centers compressed to such a degree, that finally small necrotic centers arise. Whether the cells are here killed by their own products of metabolism or die on account of other unfavorable conditions is difficult to decide.

The principal effect of the splenic graft is exerted on the mesodermal elements of the allantois. The response of the entodermal layer and that of the ectoderm of the serosa is local and

secondary. It seems to be due to a local and mechanical agent. The pressure of the apposed graft upon the ectoderm of he serosa incites proliferation of its cells. Pressure of the numerous small tumor-like proliferations within the allantois stimulates both the ectoderm of the serosa and the entoderm of the allantois. The changes in the ento-» and ectoderm have clearly shown that the form of the cell is here chiefly due to external factors. The flat cells of the entodermal layer of the allantois readily change their form into a cylindrical one, the cubic ectodermal cells of the serosa develop into squamous stratified epithelium. The ectoderm of the serosa follows in development the line elsewhere characteristic for its development. ~ Like the ectoderm of the body surface, it undergoes cornification. It proliferates and transforms into typical pearls, if pushed into the midst of the allantois. p

The response of the mesodermal parts of the allantois is a part of the common reaction shown by the entire mesenchyme of the host after spleengrafts. This response reveals in the allantoic mesenchyme and in the endothelium potencies which were never suspected. The results obtained in the allantois by its stimulation have shown the allantoic mesenchyme and endothelium to be potential hematopoietic anlages.

The differentiation of large T accumulations of granuloblastic tissue developing equally from endothelium and mesenchyme in the allantois brings these two tissues closely together. The differentiation by both tissues of identical products, under equal conditions, makes these tissues almost identical at least in their potencies. Moreover, the direct observation of the persistence of the endothelial cells under the form of mesenchymal cells in spleengrafts as well as in the allantois has lifted the visible barrier between these tissues and has joined them into a group of young undifferentiated polyvalent tissues.

On the other hand, the observations concerning the intense separation of hemoblasts from the endothelium throughout the allantois, followed by their granuloblastic differentiation, has markedly emphasized the truth of the monophyletic interpretation of bloodorigin. The erythropoietic power of endothelium has been worked out, besides more ancient authors, by Maximov,“ Danchakoffj J ordan,31~‘” and Emmel,” and lately definitely confirmed by Sabin” in cultures. The endothelial cell becomes, therefore, a true anlage for development of various blood cells. This polyvalency characterizes both rnesenchyme and endothelium and again draws these two tissues together.

Finally, the determination of the development of polyvalent stemcells by external factors of the environment is clearly brought out. The endothelium of a vessel also appears to be a polyvalent structure. It splits off into its lumen cells, under-going an erythroblastic differentiation and at the same time, smay proliferate and give off cells outside, which will differentiate into granuloblasts. The lines of differentiation are hereby determined by the intra- respectively extravascular conditions.


Avnonov, P. AND T1MoFEnvsKY, A. 1915 Cultivation of white blood corpuscles outsidethe body. Russk. Vratch. Petr., V. 14. I

BABKINA, H; 1910-11" Uber Veranderungen der ‘blutbildenden Organe bei asept. Entzund. In., Diss. Petersburg. _

CHAMZPY, 1913 La diiierenciation des tissus culstivé-s en dehors de 1’organi°sIne. Bibl. Anat. T. 23.

Di1IBER, NI. 1907 Znr Frage nach der‘ Entstehung und Regenerations fahigkeit der Milz. Jen. Zeitschr. fur N aturwiss.

DANCHAKOFF, V. 1908 Tiber die Blutbildung im Dottersack des Huhnchens. Verhandl. Anat. Gcs. Berlin. 1908 Untersuchungen fiber die Entwicklung des Blutes und Bindegevvebes bei .den Vogeln. Anat. I-Iefte. Bd. 37, H. 3. 1908 Untersuchungen fiber die Entwicklung von Blut und Bindegewebe bei Vogeln. Das lockere Bindegevvebe des Hiihnchens im fetalen Leben. Archiv. fur Mikr. Anat., Bd. 73.. 1909 Tiber die Entwicklung des Knochenmarks bei den Viigeln und iiber dessen Veranderungen bei Blutentziehungen und Ernahrungsstorungen. Archiv. fiir Mikr. Anat., Bd. 74. _ 1916 Uber die Entwicklung des Blutes in den Blutbildungsorganen (Area vasculosa, Dottersackanhiinge, Knochenmark, Thymus, Milz und lockeres Bindegewebe) bei Tropidonotus natrix. Archiv. fiir microskopische. Anatomic. Bd. 87.’ 1916 The wandering cells in the loose- connective tissue of the bird and their origin. Anat. Rec., vol. 10, no. 7, May." 1916 . Concerning the conception of potentialities in the embryonic cells. Anat. Rec., vol, 10, no. 5, March. . 1916 Equivalence ofndifferent hematopoietic analages (by method of stimulation of their stem cells). I. Am. Jour. Anat., vol. 20, no. 3, November. ' 1917- Potentialities of the lymphoid hernoblasts of the adult spleen. AI1at..Rec., vol. 11, no. 6, January. 1917 The position of the respiratoryyascular net in the allantois of the chick. Am. Jour. Anat., vol. 21, no." 3, May. 1917 Differentiation by segregation and environment in the developing organism. The American Naturalist, vol. 51, July. 1918 Cell potentialities and differential factors considered in relation to erythropoiesis. Amer. Jour. Anat., vol. 24, May. 17













30 31



.S'rocKAnD, C. R.



DOWNEY, H., UND WEIDENREICH 1912 Uber die Bildung der Lymphocyten in Lymphdri'1s'en und Milz. Arch. fur Mikr. Anat_., Bd. 80.

EMMEL, 1916 The cell clusters in the dorsal aorta of mammalian embryos. Am. Jour. Anat., vol. 19, no. 3, _May.

EVANS, H. 1915 The macrophages of mammals. Amer. Jour. of Physic-l., V. 37.

HUNTINGTON, GEORGE S.‘ 1914 The development of the mammalian jugular lymphsac, of the tributary primitive ulnar lymphatic and of the thoracic ducts from the Viewpoint of recent "investigations of ‘vertebrate lymphatic ontogeny, together with a consideration of the genetic relations of 1ympl_1at_ic'and haemal vascular channels in the embryos of amniotes. Am. Jour. Anat., vol. 16, no. 3,'Ju1y. '

J ORDAN, H. E. Anat. Rec., vol. 10, no. 3, March.

1917. Aortic cell ‘clusters in vertebrate -embryos. National Academy of Science, vol. 3.

KIYONO, K. 1914 Die vitale Karminspeicherung. Ein Beitrag zur Lehre V011 der vitalen Fiirbung mit besondere Beriicksichtigung der Zelldifierenzierungen im entziindeten Gewébe. Gustav Fischer. '

LAGUESSE, E. 1890 Recherches sur le développement de la rate chez les poissons. Jour. de l’,Anat et de la Physiol. norm. et path. de l’homme et des anim, 26 Année. _ _

LEWIS, W. 1917- Behavior of cross striated muscle in tissue cultures". Am. Jour. Anat., vol. 22, no. 2.1 _

Maxmov, A.’ 1909 Unt-ersuchungen iiber Blut und Bindegewebe. I. Die friihesten Entwicklungsstadiren der-Blut und Bindegewebszellen, etc., Arch. f. Mikr. Anat., Bd., 73.‘ I 1916 The cultivation of connective tissue of adult mammals in vitro. Arch. Russes d’Anat. d’Hist. et d’Embr.-, T. I, face. I.

SABIN, FLORENCE R. 1917 Preliminary notes on the difierentiation of angioblasts and the methods by which they produce blood—vessels, blood-plasma and red blood cells as seen in the living chick. Anat. 'Rec.,- vol. 13, no. 4.. _ _

STOCKARD, C. R. 1915 The origin of blood and vascular endothelium in embryos without a circulation of the blood and in the normal embryo. Am. Jour. Anat., vol 18, no. 2, September.

1915 Science Octob. _

STREETER, GEORGE L. 1917 The development of the scala tympani, scale vestibuli and perioticular cistern in the human embryo. Am. Jour. Anat._, Vol. 21, no. 2, March.

STRICHT, O. VANDER, 1890 Recherches sur le cartilage articulaire den oiseaux. Arch. de. Biologie, T. X.

ZIEGLER, H. E. 1892 Uber die embryonale Anlage des Blutes bei den Wirbeltieren. Ver-h. Z001. Gee.

Procecdings of the 1916 Evidence of hemogenic capacity of endothelium.


Explanation Of Figures

Figures 10 and 11 are photographs of grafts on allantois, kindly taken by Mr. Schmidt, of the Rockefeller Institute. All the other figures were drawn with the camera lucida at stage level with Zeiss Apochromat 2—mm. oil-immersion obj ective. The compensatory ocular 8 was used for figures 2, 3, 16 and 17, ocular 6 for all the others.


1 Section of a small part of the graft (:1) ‘of figure 10. The grafted splenic tissue Gr. S.t developed 011 the allantois. The ectoderrn of the serosa, Est, is interrupted and appears within and below the grafted tissue as islands or strands of epithelial tissue Ect.P.

.2 Series of amoeboid cells found in the allantois around the graft. a, b, a, small lymphocytes, 9*, h, 2', Is, histiotopic wandering cells; the nature of the cells d, e, f, is not certain. See text‘, page 167.

3 a, b, lymphoid hemoblasts ; c, d, the same as phagocytes; e, granular leucocyte in phagocytic "activity; J’, histiotopic wandering cell as phagocyte.

4 Group of phagocytes (Evans’ macrophages; Kiyono’s histiocytes), in the mashed splenic tissue. See text, pages 138-139. ' .

5 Group of small splenio arterioles in the graft. The endothelial cells (a:) emigrate into the adjacent mesenchyme.


6 A part of the grafted reticulum three days after grafting. Ms.St-., mese11ehymel syncytium hypertrophies; some of its eells H bZ., become isolated; others fuse into plssmodial agglomerations. Only a few small lymphocytes, Sm.Lmc, are left in the meshes of the reticulum. Grcn, granulocytes.

7 Numerous hemoblasts develop at the expense of the cellular reticulum in the graft, others are in the process of separation; nu.mer'ous proliferating hemoblasts, _ -Hbl. Small lymphocytes are scarce. Numerous granuloeytes are brought in. '


8 A part of the splenic graft. four days after grafting. Numerous vessels (1'). 6.) grew into it. They contain for the most part difi”eren.tiated cells-+erythr0cytes and granuleeytes. The greatest part of the reticular cells have trallsformed into hemoblasts, and now many of them differentiate into granuloblasts (myeloeytes).

9 Further differentiation of the granuloblasts into gra,m1.1ocytee (granular polymerphonuelear leueoeytes). Gradually sheaths of granuloblastic tissue develop around the small vessels.


10 A double graft of adult splenic tissue was made in this case. Both grafts were found five days after developed" into two tumors, G’r.a and Gahb. The a,11an— toic melnbrane shows 3. thickening in the immediate vicinity of the grafts only.

The allantoic nlcmbrane remained otherwise unchanged. 11 Graft on the allantoic membrane after five days of growth. The whole allantois dotted by small hemo- respectively granule-poietic tumors.


12 A part of the allanteic meseneliynie surrounding the graft. Small lym'phocytes S~m.Lmc. densely infiltrate it- Heinclylasts, HbZ., develop at the expense of the mesencliyinal cells, Hblxc, and proliferate, ‘H53’.

13 Small vessels in the vicinity of the graft, D.V., begin to disintegrate, the erythrocytes show in the preparation the reaction characteristic of stagnant blood cells. The eridotlielium of such vessels Ends. wander out into the allantoic inesenchyme in the form of hemeblasts, H51.

14 Vessels around the graft and in the oedematous parts of the allantois disintegrate; erythrocytes are found free i11 the mesenehym_c,‘t-he cells of which Helm. often develop into hernoblasts. End0thelia.l sells, End.;i:, join the surrounding mesenchyme. V

15 A hetnoblastic focus intensely proliferating in the Inesenchyme of the allantois. Hemoblasts reduced to the size of small lymphocytes. Each , of these cell kinds retain their eharzicteristie structure.


16 Structure of the ectodermal epithelium grown into the graft t-issue from the surface of-the serosa; note the size difference in the cells, Ect.ep., among which an epithelial giant cell, G.ep.c., is present.

17 Ep.c’, a cell in which three equatorial plates could be seen in the prepara— tion ; only two could be drawn in the figure.

18 Proliferation of the eetodermal epithelium, if separated into islands. A eornified pearl, O’.P., is seen in the figure. Epc.:z:, epithelial cells in the form of fueiform bodies. Hemololaets, H £71., and mesenchymal cells, M3,, traversing the epithelial tissue and transforming it into a reticulum.


19 A typical small vessel in a. normal allantois with its endothelium E-mi. and adjacent 1'11eseI1c11yma,l cells, Ms.

20 A vessel of the same calibre as drawri in figure 19 in an allantois containing miliary tumors (fig. 11). An int-ense pf91ife1'a.t.ion of the endothelium takes place. The endothelial cells 9-r2,d.m. proliifem-te outward into the adjam-rt}-t. mesenchyme, separate here in the form of hemoblasts, Hbl, and pro1ifer.21.te, H bl’. ' . 21 The same as in figure 2-0 in 3 Izmter stage. Gr2muIol31a_st.ic differentiation of the hemoblasts given off by the vascular endotheliuxn.


22 A beginning development of a hemogranulopoietic center under the ectoderm of the ' serosa, act. The mesemehymal cells, M3,, here proliferate, h3—-'pert.ropl1y, and separate in the form of he1m)hla.ets, Hbl.

23 A hemogranulopoietic center under the serosa. in later stages. Numerous hemoblast-S developed from the Iiieseneliyliie show an inte1‘1sifi-::at.ion of the basophilie 1‘eaot.ioI1,Hbl. A beginning of the further granulopoietio diiTereI1tiation is seen in some {if them Grbl.

2-4 A creeping giant hemoblast, G’. H62.

25 A strand of epithelial tissue from the deep surface of the ectoderm, act, proliferates intensely along a capillary.