View source for Talk:Paper - Further evidence on the origin of the lymphatic endothelium from the endothelium of the blood vascular system (1908)

===Further Evidence on the Origin of the Lymphatic Endothelium from the Endothelium of the Blood Vascular System===

{{ref-Sabin1908}}

By

Florence R. Sabin.

From the Anatomical Laboratory, Johns Hopkins University.

The question of the origin of the lymphatic system has now resolved itself into the question of the origin of its endothelial lining. Is the lymphatic endothelium angioblastic or does it come from the less differentiated mesenchyme cells? By the one theory the endothelinm of the blood vascular system has become a specific tissue before tlie lymphatic system begins and is capable of giving rise to all the endothelium both of the blood vascular and of the lymphatic systems. By the other theory, the lymphatic endothelium is formed out of mesenchyme cells, just as the lining of the primitive blood islands had been formed at a much earlier stage. Thus the question is one of the specificity of tissues, and means how early does endothelium become specific.

The theory that the lymphatics arise from tissue spaces and that the lining comes from flattened-out mesenchyme cells is well expressed by Gulland. He studied sections of human and other mammalian embryos and found spaces in the connective tissue of the liml) buds, and thought that these spaces gradually dilated, flowed together and formed lymphatic ducts. ^

Budge studied the lymphatic system in chick embryos. He injected the coolom in chicks 2^/2 days old and found that the fluid passed by blunt processes out to the limits of the area vasculosa, where it formed a peripheral sinus. These spaces of the area vasculosa, he thought, made a primitive lymphatic system, later supplanted by the definite Ivmphatic system of which he made beautiful injections along the allantoic vessels in chicks 18 days old. The most recent exponent of this theory is Sala, who has also studied the lymphatics in the chick, and finds a series of sacs along the aorta

'Gulland. .Tmu-nal of Pathology and Bacteriology. Vol. II. 1894, p. 4G6. ^Budge. Ai-eli. f. Anat. u. Pliys., Anat. Abth., 1880 and 1887.



The Anatomical Eecord. 47

.>lnoh he eonAlers to be lymph hearts. From these mesenehyme spaces

the thoracic dnct is formed.^ ^ »P<ii-tJs

The first link in the chain of evidence for the vascular origin of the ymphahc endothelium was the discovery by Langer in 18.8 th^ h lympha ics an the skin of the frog grow by means of blind sprouts o endothehal cells. That is, the lymphatic capillaries grow by Te sLe

covered by Eanvier m studymg developing lymphatics in the skin of

that the lymphatics grow from the veins '

If this hypothesis be true, it should be possible to trace the growth o the lymphatic capillaries into the different tissues and organs This has been done m the case of the skin, where it was shown that in embryos under 20 mm. there are no lymphatic capillaries "the skin At about that time the lymphatics rea.h the skin in the neck an" gradually spread over the body. Thus it is possible to mark ut successive zones containing lymphatics. Along the margin of these zones

ts'lt T 7ri T "^^ '-'^'"'^^ ""^ -^^ non-lvmphatic area . Tl e lymphatics first cover the skm as a single primary plexus

m the subcutaneous tissue. Then a secondary and more superficia plexus develops from the first and from this second plexus the final capillaries enter the papilkT.

In studying vascular injections of pig embryos which are to b. demonstrated at this meeting, we have noted that there are non-blood vascular areas m the skin as well as non-lymphatic areas. In the iniected specimens of embryos 15 to 20 mm. long it will be seen that the deep capillaries oyer the central nervous system have met in the mid-dorsal line while the superficial capillaries in the skin do not pass bevond the myotomes. Thus there is a non-vascular area of the skin dorsal to the centra nervous system, which is narrow over the spinal cord and wider over the brain. In the case of the head the non-blood vascular area persists late, so that it is still present in embrvos 5 to 5.5 cm. Ion- The skm of the head is also late in receiving its lymphatics, and at this

DJ^'e'oo/fT' "■ 'f- '"• ""^""^ ■ '■ ^'"'^•- '^^' 0^- ^-n- i^>.

p 453 ^^^'^^^^d "^ Archives Italienne de Biologie, Tome 34. 1900,

J.anger. Ueber das Lymphgefass-System des Frosches, LVIII Bd der Sitzb. d. K. Akad. d. Wissensch.. I Abth.. 1868. ^Ranvier. Comptes Rendus, 1895 and 1896; Archiv d'Anatomie. Tome 1



48 The Anatomical Eecord.

stage the lymphatic vessels which are growing up between the eye and the ear are at least 3 mm. behind the invading blood capillaries. This makes one of the best places in the body to study growing blood vessels and growing lymphatics in their relations to new territories.

The second place where the test of the invasion of lymphatics into new territories has been made is in the intestine. Dr. Heuer, whose work is to be presented at this meeting, has followed the gradual growth of the lymphatics through the mesentery to the wall of the intestine. He has followed the growth of the lymphatics by means of injections of successive stages and has shown that after entering the intestinal wall they first grow into the submucosa, making there a primary plexus. Then a secondary plexus forms beneath the mucosa and lastly the terminal lacteals grow into the villi.

In Dr. Flint's recent paper on the development of the lung he has shown that the lymphatics follow the same general law, — appearing first in the hilus of the lung and gTadually invading the lung itself.® Thus, as far as developing lymphatics have been studied in relation to their growth in organs, they first approach each organ from without and then gradually invade its structures. Therefore, the hypothesis of the growth of the lymphatics from center to periphery, from the veins rather than to the veins, is sustained by the study of the growth of the lymphatics into the organs.

The second series of experiments in regard to the origin of the lymphatics has been the tracing of the finished lymphatics back to the earlier stages and concerns the study of the lymphatic sacs. Everyone who has worked on the lymphatic system in early stages is aware of the extreme difficulty in identifying lymphatics. There are so many structures with which they may be confused in sections that it seems as if almost the only clue with which to meet the difficulty is to start with the finished lymphatic system and trace it 1)ack to the younger stages. The finished lymphatics in the skin of the embryo are identified with great ease by means of injections ; their characteristics are so well known that they cannot be mistaken. In younger stages the character of the lymphatic capillaries more closely resembles the blood vascular capillaries, so that it becomes even more necessary to have injections that will show whether the vessels in question are connected with the blood vascular system or with the lymphatics. By a great number of lymphatic injections it has been shown that the lymphatics in the skin of

•^Flint. Amer. Jour, of Anat., Vol. VI, 1906.



The Anatomical Kecord. 49

the pig run eitlier to the neck or to the groin. In the neck these vessels run either to a sac or into the group of lymph nodes into which this sac differentiates.

In studying the cervical heart, the first anlage of the lymphatic system, I was able to trace it back, first by lymphatic injections, and, where they failed, by vascular injections, to a stage where it consisted of a small sac against the jugular vein. The opening of this sac was difficult to find in sections even in later stages, where one could watch the injecting fluid pass from the lymphatic duct into the vein. In these injected specimens, however, it was possible to find the valve in sections, and they showed that the lymphatics entered the vein at a very oblique angle. Having found how the valves looked in later stages, it was possible to find them in earlier stages. I followed the origin of the lymph heart back to a stage in which it is a simple sac connected with the vein, and interpreted this as a budding out from the vein. This sac is to be found in embryo pigs about 15 mm. long. Dr. Lewis, of Harvard, approaching the subject after a careful study of the veins, was able to trace the origin of the cervical sac still further.'^ He showed that there is first a plexus of veins in the region of the sac; these veins are in free connection with the jugular vein. Then the plexus of veins is cut ofl^ from the jugular and appears full of blood, but without venous connections. Later the plexus forms a sac and rejoins the main vein. After it has joined the vein it becomes empty of its blood. This discovery of Dr. Lewis I can entirely confirm. The methods of injecting embryos have been much perfected in our laboratory, so that we have many series of sections of complete vascular injections, including embryos from 10-30 mm. in length. In one specimen of an embryo 15 mm. long, I have the symmetrical sacs in the neck filled with blood in both sides. On one side a very small amount of the ink injected into the vein has run over into the sac, showing that it has a small opening on the one side and iione on the other. Thus the origin and relations of the cervical lymph heart have been traced.

Dr. Lewis discovered a similar sac in the root of the mesentery of a pig embryo 20 mm. long and identified it as a part of the lymphatic system. We were working on this mesenteric sac at the time, but had not been alile to interpret it. Mr. Baetjer has traced the origin of Dr. Lewis' mesenteric sac and its relation to the thoracic duct. He

'Lewis, P. T. Amer. Jour, of Anat., Vol. V, 1905.



50 The Anatomical Eecord.

will report his work at this meeting and Avill show that the mesenteric sac arises from the veins of the renal anastomosis, independently of the thoracic duct; the receptaculum chyli arises as still another sac, adjacent to the azygos veins, and in an embryo 3 cm. long the mesenteric sac and the receptaculum chyli have joined and can be injected. The details of the origin of the receptaculum chyli Mr. Baetjer is not yet ready to report.

In regard to the posterior lymph sacs, I can demonstrate them in sections, but have not studied them further. Thus far it has been shown that there are six primitive sacs which form the anlage of the hmphatic system. Four of these sacs are paired, two in the neck and two in the groin, while two are unpaired, one at the root of the mesentery and the other behind the aorta forming the receptaculum chyli. All of these sacs arise independently, all have a complete lining of endothelium and the three that have been studied with care, namely, the two cervical sacs and the mesenteric sac have been shown to arise from the veins.

"We must now take up the question brought up by Dr. Huntington and Dr. McClure as to whether these sacs are the first lymphatics. At our meeting last year, Dr. Huntington and Dr. McClure showed a series of models representing the development of the lymphatic system. Some of the later models showed the true cervical lymphatic sac, easily identified as the sac which I had published as the first anlage of the Ivmphatic svstem, a point which had been confirmed by Dr. Lewis. In disagreement with both Dr. Lewis and myself, these models carried the origin of the lymphatics back to much earlier stages, showing them in cat embryos 8 to 8.5 mm. long. Dr. Huntingion and Dr. McClure describe oval or spindle-shaped spaces arising along the veins as the result of shrinkage and condensation of primitive redundant precardinal vein channels. These spaces they describe as situated without the intima of the veins in the peri-intimal adventitious tissue and are not to be confounded with the mesenchymal spaces of Sala, since they develop in territory formerly occupied by the veins themselves. From this it is not clear what forms the lining of these spaces ; since the veins at tjiis stage have only a lining of endothelium Avhich rests on a connective tissue svncytium, these spaces must be lined either by endothelium or by the connective tissue, unless one imagines them as formed by a sagging of the endothelium away from the surrounding tissue so that the spaces are bounded by endothelium on the one side and the connective tissue on the other.



The Anatomical Eecord. 51

I have cut many new sets of serial sections of various stages of pig embryos, from 10 to 16 mm. long, and am convinced that these early lymphatics cannot be demonstrated. Spaces exist, but they are not part of the lymphatic system. In the first place, there are a very interesting series of spaces, which encircle many of the nerves and may be called peri-neural spaces. These peri-neural spaces, which were excluded from the lymphatic system in my earlier work, since they are easily injecte-i but never connect with the lymphatic system, are large and prominent in early embryos. Inasmuch as the nerves are often close to the veins, the peri-neural spaces often touch the veins. I have many specimens showing that these peri-neural spaces are like the arachnoid from which they can l)e injected, and I will later show their development. They are especially large around the growing tips of the nerves and they will be demonstrated at this meeting. These peri-neural spaces are to be distinguished from lymphatics by the absence of endothelium, by the fact that they arise from the destruction of connective tissue and bv their following the nerves. They have undoubtedly great physiological significance, and may supply the place of a lymphatic system to the nervous system, since it never receives true lymphatics.

Besides these peri-neural spaces, there are true rounded spacs in the mesenchyme which undoubtedly contain lymph but which are not a part of the lymphatic system. There are also veins, or blood capillaries, which are exceedingly difiicult to trace in relation to other blood capillaries in iminjected specimens. Just as in adult tissues one cannot trace a capillary plexus completely without injecting it, so one meets the same difficulty in the embryonic stages. From a careful study of new sets of sections of early stages we feel that the true lymphatics cannot be demonstrated before the appearance of the lymph sacs, and therefore think that the importance of the lymph sacs as the first lymphatics is fully emphasized. It seems to me that it is now possible to define the lymphatic system as a series of vessels lined by endothelium which arise from transformed veins and eventually open into the veins. All tissue spaces bordered only by connective tissue and all the serous sacs are to be excluded from the lymphatic system morphologically. Previous to the formation of the lymph sacs, the anlage of the lymphatic system, there are no spaces which have been proved to become a part of the true lymphatic system.^

^Inasmuch as the discussion which followed the I'eading of the papers of the Lymphatic Sj-niposium of lOOS will not he pnhlished, I should like to



52 The Anatomical Eecord.

We pass now to the third important point in connection witli the origin of the lymphatic system, namely, the series of ))lind spaces, definitely lined with endothelium, which Dr. Lewis has pictured in a series of most careful reconstructions of the development of the lymphatics in the rabbit.^ In a rabbit embryo of 11 mm., or 14 days, he pictures the lymphatic system as consisting simply of the cervical sac. Later, in embryos nearly 15 and 17 days old, he pictures isolated spaces along the external mammary and umbilical veins. Dr. Lewis does not picture any of these spaces in the pig of 20 mm., where the entire system consists of the cervical sac, the mesenteric sacs and the receptaculum chyli, but if the later stages of the cat and rabbit embryos show them, one should be able to find them in pig embryos older than the stage of 20 mm. Thus Dr. Lewis agrees in regarding the sacs as primitive anlagen of the lymphatic system, but disagrees in finding multiple endothelial anlagen associated with the growing system of ducts. Since these spaces are lined with a definite endothelium, they form a much more serious obstacle to the theory of growth of the lymphatics from the endothelium of the veins than the more indefinite spaces to be found in earlier embryos, and I cannot but think that if these multiple endothelial-lined isolated spaces do exist along the veins in the later stages, they would form serious evidence against the theory of the origin of the lymphatics

add one or two footnotes to my paper to brine; out the points in whicli we now all agree and those in which there are still differences of opinion. In this paper, if I seem to lay too much emphasis on the method of the origin of the lymph sacs, it is because that was the chief point of controversy at the meeting of 1907. Since we have all come to agree that the lymph sacs arise from the veins, that they are preceded by a plexus of veins which separate from the main veins, then unite to form lymphatic sacs and subsequently join the vein again, it is clear that the most fundamental and important evidence for the venous origin of the lymphatic system still stands. We may then regard as established by the recent American work on the lymphatic system that the primary lymph sacs are derived from the veins, that their endothelial lining is derived from the venous endothelium, and that the permanent openings of the lymphatics into the veins are secondary rather than primary. In regard to the primary lymphatic sacs or hearts, in mammals, only the jugular lymph sacs and the mesenteric lymph sac have been thoroughly investigated. The posterior lymph sacs, the receptaculum chyli and probably symmetrical lymph sacs along the subclavian veins for the lymphatics following the deep vessels of the arm and the external mammary veins need further study. ^Lewis, Ihiil.



The Anatomical Eecord. 53

from the veins. Or at least if the lymphatics, in their growth, do pick up isolated endothelial-]ined spaces, we shall again be left without a clue as to the origin of the lymphatic system. These multiple anlagen would then need to be traced back to their origin before any generalization in regard to the lymphatic system could be made.

I have not had the same material as Dr. Lewis with which to test his numerous anlagen, but if they be general, one should be able to find them, whenever lymphatics are invading a new territory following the veins. I have, therefore, tested the point in the zone of the skin previously described where both invading blood vessels and invading lymphatics can be studied. The test was made in this way : The lymphatic plexus between the eye and the ear in pigs 50 to 55 mm. long was injected. Some of the injections were made complete, which can be done in this way : after withdrawing the needle the finger is laid on the plexus and gentle pressure applied until the terminal sprouts just rupture. Other injections were left purposely incomplete, and some specimens were not injected at all. Serial sections of all the specimens were made, cutting parallel to the surface of the skin, and then the specimens were studied with reference to whether there are isolated vessels in front of the growing lymphatics. The first important point is to be gained from watching the injections. It will be often seen in making an injection that large, long sprouts run out in front of the plexus ; often from such a long sprout a slender side channel will suddenly shoot out and open into a large space which might easily appear isolated in sections. When this second large vessel is full a second slender channel will open up from it and carry the injection mass back to meet the main plexus. Thus it might have been injected at first from the main plexus, but the mass runs easiest into the wide channels and the very slender, collapsed vessels open up only under pressure. From these experiments one may readily expect difficulties in tracing the connections of such spaces in uninjected specimens.

The serial sections bring out three points : First, that in complete injections there are no vessels which have not received the injecting mass. Secondly, in partial injections and iminjected specimens there are endothelial-lined vessels which can be traced only with difficulty or not at all to the primary plexus. Thirdly, since some of the injecting fluid often flows out of the vessels when the needle is withdrawn, one can find vessels containing ink and which must have been injected from the plexus, and yet one cannot trace their slender connections. From this



54 The Anatomical Eecord.

evidence we doubt if such isohited vessels really exist in front of growinj^ lymphatics. Of the accuracy of Dr. Lewis' observations on these anlagen we have no question; they undoubtedly exist in serial sections, so we question not the observation, but the limitations of the method. If in adult tissues sections without injections are wholly inadequate to show all of the blood capillaries, it is obvious that they must fail more conspicuously with lymphatic capillaries where the walls are thinner and the channels are often excessively small and collapsed. Thus, where I have been able to put the test of injection, there are no isolated anlagen of endothelial-lined spaces preceding the growing lymphatics.^*'

Thus the reworking of the subject of the origin of the lymphatic system seems to me to have emphasized the fundamental distinction between tissue spaces in the mesenchyme and the endothelial-lined lymphatics. The first lymphatic endothelium is derived from the venous

"As was stated iu the first footnote, we are all agreed in regard to the method of origin of the primary lymph sacs as far as they have been studied. The differences of opinion have wholly to do with the growth of the lymphatics after the lymph sacs have been formed. There are before us three possibilities ; first, one suggested by the isolated endothelial lined spaces shown in Dr. Lewis' reconstructions along the course of veins near growing lymphatic zones. These might be interpreted as the formation of multiple lymph hearts from the veins. There is no theoretical objection to this interpretation, but, if I .iudge correctly, Dr. Lewis wisely refrains from making it definite for lack of positive evidence. My work offers some suggestion that a more adequate method than that of serial sections alone may eventually show that these isolated rings of endothelium are in reality connected with the lymphatic endothelium. The second possibility is that to the lymph sacs are added tissue spaces lined by mesenchyme and that these tissue spaces form tjie lymph trunks. This position seems to me wholly untenable, it means that the lymphatics first arise from venous endothelium which is growing by the method of sprouting; then the system grows by additions of a new kind of lining cell, and subsequently returns to its primitive method of growth, namely, the sprouting of endothelium. This seems to me out of harmony with the laws of growth as far as we know them. The third possibility, which seems to me the most likely to be correct, is that the lymphatics grow from the various primary lymph sacs by the sprouting of endothelium and gradually spread over the body. Wherever it has been possible to test this method it has been found to hold. I believe that it is only the difticulty of making lymphatic injections in the early stages immediately after the formation of the lymph sacs that makes this point still an open question. It can undoubtedly be settled by further studies of growing lymphatics, and these studies must be based on the crucial point, which is to find the source of all of the endothelium of the lymphatic system.



The Anatomical Eecord. 55

endothelium, not hy a simple process of budding, but by a transforming of veins into the primitive lymphatic sacs. To recapitulate, (1) all the lymphatic vessels arise by a process of sprouting from the lymph sacs, which in turn are formed from the veins; (2) no lymphatics have been demonstrated in embryos before the lymph sacs are formed; and (3) the additional anlagen of Dr. Lewis can be injected from the lymphatic system and therefore must also arise from it. In uninjected specimens the connections cannot alwavs be seen.