TEVIBES Reason

New York State Veterinary College

ITHACA, NEW YORK

Sabin, Florence R.

The origin and development of the
lymphatic systems 1913.

Cornell University Library
QL 971.S110

“Hci

24 61

CORNELL UNIVERSITY

THE
Flower Veterinary Cibrary

FOUNDED BY

ROSWELL P. FLOWER

for the use of the

N. Y. STATE VETERINARY COLLEGE
1897

DATE DUE

GAYLORD PRINTEDINU.S.A.

CORNELL UNIVERSITY
THE

Flower Peterinary Library |

FOUNDED BY
ROSWELL P. FLOWER

for the use of the

N. Y. STATE VETERINARY COLLEGE
1897

AL
471 meee OU
Slls

CONTENTS.

PAGE

T. UNtrodUction: «. caieiy.25 dctin sivoy hatte tes iS 9 a Gere sed eeaslanl a oe eae deed «
1. Morphology of the vascular system, the angioblast ......... L
2. Morphology of the lymphatic system...................... 4
ATS LISCOV ICED 505 es Seccs Sarg wach ge Splints yer r Saaghale s Bae eae toe ated 4
A VASE: SOROS asics Pe mig alam Paws ek wed eoe ne caches canned. 4
2. Lymphatics with open mouths............... 0000-00000 eee 4
8. Discovery of endothelium....................cccceeeeeeee 6
ELL. “Bm Dry QlOey® ee sictese's sce adidas tue eaud cod 4 x hue ars west tee. g aiden geneva PRL 7
Le Tt OGUCt ORD sie: seed cedars dojaern eg be ecg WR eae REA LA ow: 7
2. Lymphatics as dilated tissue spaces...............-.00000. 8
3. Analysis of Budge’s work............. 0.000 e eee ees 8

4. Statement of the three arguments for the origin of lymphat-
ics from the veins............. eee cee ccc eeeeeaee 10
A. Growth of lymphatic capillaries by sprouting......... 10
B. Lymphatic and non-lymphatic zones................. 18

C. Direct budding of the lymphatic vessels from the
WEIMS > vce Sac these besuara: oda wad WaWine Laldalad aoeaders 21
IV. Primary lymphatic system in different forms ......... oxide a
1. Primary lymphatic sacs in the pig..................0000008 24
A. General summary of the sacs.............0...000000- 24
B. Origin of the jugular sacs................. 00 cece eee 25
C. Origin of the renal sacs......... 0.0.00 cc cece eee eee 35
a. The retroperitoneal sac............... 00000000 35
Dy "Phe Wiae: ‘SaCSia ss cats sats doce wee oo a Se ok Re 36
¢; ‘The. cisterna ehyli. wis s2.caveud ia eyes veae ee ced 36
D;. ThE: Thoracic Guct.. coc sca cas aR sea eke cam eae daa ee 40
2. Primary lymphatic sacs in human embryos................ 45
3. Primary lymphatic sacs in rabbit embryos................. 47
4. Primary lymphatic sacs in cat embryos.................... 48

5. Primary lymphatic system in birds, posterior lymph heart
and jugular lymph sac............. 02. ee ee ee eee 50

6. Primary lymphatic system in amphibia, anterior and poster-
jor LyMph. Weartsis.. see ie nase cccwans ecg edad 55
V. Peripheral lymphatic system ............ 00-0. cece e eee eee 59
1. In the pie’. iceese seed ceded sea eeeeek ned ies ea veaw ome sete 59
A. From the jugular sacs............-. 00 e eee ee eee eee 59
B. From the renal sacS.........- 2c eee ee cece eee eee 63
2 In other LOLS sis siseeg ee degi's gee ears hae arm Gaeans Gy BIOe Hag A LA SMe 65

VI. Origin and development of lymphatic glands and their relation
to primary lymph sacs ..............-. eee eee 65

Digitized by Microsoft®

ii CONTENTS.

.

PAGE

VII. Comparative morphology of primary lymph sacs, lymph hearts,
lymph glands and amphibian lymph sacs......... 70

VIII. Various other theories in regard to the origin and development
of the lymphatic systemM............. 0c cece eee 72
1. Mayer-Lewis Anlagen ............. ccc eee eee eee eee e nee 72
2. Extra-intimal and perineural spaces and fenestration...... 75
3. Growth of lymphatics by the addition of tissue spaces...... 77
TEX. ‘OOnelusiOns): cc. ears ea baa Pee ow ae ane et eet te 82
Ke Literature vce seeedee sss ss enae eee wen emeoumuleas anatase penta s 83

Digitized by Microsoft®

THE ORIGIN AND DEVELOPMENT OF THE LYMPHATIC
SYSTEM.

By FLORENCE R. SABIN.

(From the Anatomical Laboratory of the Johns Hopkins University,
Baltimore.)

I. INTRODUCTION.
1. MorpHolocy or THE VascuLar SYSTEM, THE ANGIOBLAST.

Until the past few years our knowledge of the morphology of the
lymphatic system was in a very unsatisfactory state. Our con-
ceptions of such main questions as the origin of the first lymphatics,
the time of their appearance, the relations of the lymph hearts
and sacs of the amphibia to the lymph glands and ducts of higher
forms, the relation of the lymphatics to serous cavities and to the
various forms of tissue spaces, and the development of the lym-
phatic vessels within an organ have been so vaguely understood
that the opinions of no two investigators approached agreement.

A clear conception of any system must be based on an under-
standing of its fundamental morphology, as, for example, the
morphology of the arterial system was placed upon a satisfactory
basis by the discovery and comparison of the aortic arches in dif-
ferent animals. An example more closely in touch with our knowl-
edge of the lymphatic system is the study of the fundamental
morphology of the blood vascular system. Our knowledge of this
is in the process of being built up, but a foundation has been laid
for the clear understanding of the vascular system in a series of
discoveries." The first and most fundamental of these is that the
blood vessels arise as blood islands in the extraembryonal mem-
branes. The earlier embryologists on the other hand believed that
the first vessels were spaces without walls, which the pressure of
the circulating blood hollowed out. The best description of the

1 Aided by a grant from the Baltimore Association for the Promotion

of University Education of Women and by the Carnegie Institution of
Washington.

Digitized by Microsoft®

2 Florence R. Sabin.

origin of blood islands and the best analysis of the meaning of this
discovery is to be found in two works of His, “ Untersuchungen
ueber die erste Anlage des Wirbeltierleibes” in 1868 (pp. 95-103)
and in “ Lecithoblast und Angioblast” in 1900 (pp. 268-295). The
discovery of blood islands, however, dates back to the work of
Wolff (154) and Pander (101), who introduced the name, and
perhaps no one subject in embryology has a more extensive litera-
ture. Certainly a most interesting account of the development of
the vascular problem can be followed through the pages of von
Baer (7%), Prévost and Lebert (108), Remak (124-125), Reichert
(121-123), Koelliker (67-71) and Thoma (148). The second
great advance was the discovery that blood vessels throughout are
lined by endothelium (Hoyer 48), which followed soon after the
corresponding discovery in lymphatics by von Recklinghausen. ‘The
third discovery involves the proof that blood vessels grow by the
sprouting of their endothelium, Prévost and Lebert (108), His
(47), Rouget (126) and Arnold (5); the fourth that the main
vessels of the body wall, including the posterior part of the aorta
(Evans 33-35), even the anterior part of the aorta and lateral heart
anlagen, arise as a capillary plexus or as solid angioblast cords
(Bremer 15), which invade the body from the extraembryonal mem-
- brates. Finally a complete conception of the development of the
vascular system is based on the theory that the blood vessels of the
extraembryonal membranes invade the body wall (His 47), and
that within the body wall these capillaries of endothelium gradu-
ally invade or spread over the body so that there are definite vascu-
lar and non-vascular layers and zones. His (47) discovered the fact
that blood vessels grow into the central nervous system; but the theory
of vascular and non-vascular zones which is essential to an understand-
ing of the development of the vascular system we owe to Mall (80-87).
It has been worked out by his pupils, notably Evans (33-35). Its
meaning can be grasped by studying figs. 4, 5 and 6 in connection
with fig. 437 (Evans 35), which all show that the skin dorsal to the
central nervous system is a non-vascular zone, which for a long
time is not reached by the blood vessels in the centrifugal growth.

This series of discoveries, notwithstanding the gaps and uncer-
tainties in our knowledge of the early stages which were well
brought out by Minot (98, pp. 483-485) and by Evans (35, pp. 551-
567) in 1911, uncertainties in part since removed by the work of

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 3

Bremer (15), offers a consistent as well as a constructive theory
of the vascular system which goes far to modify the force of the
following words of Koelliker quoted by Riickert and Mollier (128,
p- 1019): “Ueber die erste Bildung der Gefasse und des Blutes
herrschen wohl ebenso viele Ansichten, als Forscher sich ueber
diesen Gegenstand ausgesprochen haben, was auf jeden Fall be-
weist, dass derselbe zu den schwierigsten gehért.” The theory
thus outlined is that the fundamental morphology of the vascular
system is based on the specificity of the endothelium, or in the
words of His (47, p. 325), on the fact “Zu den am friihesten sich
sondernden Gewebsanlagen gehort der Gefasskeim oder Angioblast.
Seine Sonderung erfolgt sehr scharf, und sein Wachsthum geht
nach durchaus eigenthtimlichen Gesetzen vor sich.” It includes
the discovery of Mall (83-87) that endothelium may produce reticu-
lum, a process which he has shown takes place both in the liver
and in the heart; but maintains that there is an early differentia-
tion of two tissues, namely, endothelium and mesenchyme, so that
the angioblasts once formed give rise to all the vascular endothelium
of the body.

The opposing theories in the varying forms of the origin of blogd
vessels from tissue spaces, their growth by the addition of tissue
spaces or by the addition of connective tissue cells, and the differ-
entiation of endothelium from mesenchyme over extensive areas
within the body wall find their most recent evidence in the works
of Riickert (127), Riickert and Mollier (128), Hahn (40), Bon-
net (14) and Maximow, 1909 (90, p. 511). These views, however,
must be traced through the earlier works of Gotte (37) and Reich-
ert (121-123). The evidence for the continued origin of blood
vessels from the mesenchyme is for the most part from the inter-
pretation of appearances in sections. Hahn’s work, however, in
connection with the difficult point of the origin of the heart and
aorta is experimental. He removed the vascular membranes on
one side of early chick embryos and obtained a heart anlage and
aorta on both sides. He thought that he could entirely exclude a
growth from one side to the other as well as remnants of the mem-
branes on the same side and concluded that both heart and aorta
arise in situ from the mesenchyme and not from the endoderm of
the yolk sac not from an ingrowth from the extraembryonal mem-
branes.

Digitized by Microsoft®

4 Florence R. Sabin.

2, MorpHoLogy oF THE LyMPHATIC SYSTEM.

It will be proved in the following pages that the lymphatic problem
is closely connected with, or rather is a fundamental part of, the
vascular problem, so that the study of the one throws light on the
other, or. to put it more strongly, that the same kind of evidence is
needed to solve both problems. The fundamental morphology of the
lymphatic system has, however, been put on a more satisfactory basis
than that of the vascular system, for it has been seen in the living
embryo that the first lymphatics bud off from the veins. Moreover,
lymphatics begin at a comparatively late stage, long after the forma-
tion of the blood islands has ceased.

II. HISTORICAL.
1. Vasa SEROSA.

Until the primary origin of any system is known our cenceptions of
it are necessarily hazy and vague and this is nowhere better illus-
trated than in connection with the lymphatic system. The views of
the eighteenth century may well be summed up in the hypothetical
“ vasa serosa” of Boerhave (13), Haller (41) and others, which were
tiny channels too small to allow corpuscles to pass, supposed to connect
arteries, veins and lymphatics at their tips. The conception which
underlies vasa serosa may be traced back to the experiments of Nuck
(100), who injected air into the arteries and fownd it returning in
the lymphatics, as may be represented in the diagram of fig. 1. Vasa
serosa meant the idea that the arteries finally branched into vessels too
tiny to carry the corpuscles, but the term likewise represents a whole
series of vague conceptions, such as Bichat’s (12) absorbents and ex-
halants, which sought to make definite some idea of the nature of
lymphatics.

2. LyMPHATICS WITH OPEN MovuTHS.

In the latter half of the eighteenth century the conception of definite
“vasa serosa” was modified through the work of William Hunter (50)
and Munro, who believed that lymphatics began with open mouths.
The views of these English observers, as shown by Cruikshank (23),
involved the idea that the mouths of the lymphatics opened directly
onto the surface of the body, into the cavity of the intestine and the
air sacs of the lungs, as well as into the connective tissues. The theory
was also involved with discussions of the lymphatics as the exclusive

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 5

organs of absorption for the body, and the physiological theories of
filtration.

Fic. 1—Diagram to show the “vast serosa” of Boerhave, Haller and
others. The veins are striped, the lymphatics are dotted, and the “ vasa
serosa’ which connect them are cross-hatched. The connective tissue is
indicated by lines.

The doctrine, or as His (44) terms it, “the dogma of lymphatics
with open mouths,” was first combated by a definite counter theory
by Schwann (139), following his discovery of the cell in the animal
body. The theory of Schwann and notably of Virchow (150) was that
in place of vasa serosa and hypothetical connections which did not
exist hollow connective tissue cells spanned the gaps between blood
vessels and lymphatics, fig. 2.

Fic. 2.—Diagram to show the theory of Schwann and of Virchow that
the blood vessels are connected with lymphatics by hollow connective tissue
cells. The veins are striped, the lymphatics dotted and the hypothetical
lymph vessels are cross-hatched. The nuclei denote the fact that cells have
been discovered.

Digitized by Microsoft®

6 Florence BR. Sabin.

3. Discovery oF ENDOTHELIUM.

In contrast to these hypothetical relationships of lymphatics, the
work of von Recklinghausen (119) gave the first definite theory which
was actually based on the interpretation of direct observations of the
lymphatic capillary. Through the use of silver nitrate solutions re-
duced in direct sunlight, he discovered that the lymphatics are lined by
a definite cell layer of endothelium. This is clearly one of the most
important discoveries in histology. Von Recklinghausen interpreted
the silver pictures to mean that the finest lymphatic capillaries were
lined by an epithelium, but that these capillaries communicated by
wide openings with Saftcandle or lymph radicles, which formed
the real rootlets of the lymphatics. Subsequently he described more
definite openings or stomata in the serosa of the diaphragm (120).

Fig. 3.—Diagram to show von Recklinghausen’s theory of the relation of
lymphatics to tissue spaces. All of the tissue spaces are cross-hatched to
indicate hypothetical lymphatics.

The von Recklinghausen figures seemed to offer a perfectly satisfac-
tory demonstration of the relations of open lymphatics, especially in
view of the fact that the great majority of anatomists already believed
in open lymphatics. But it is of great interest to follow the writings
of His at this period. His (44), who was just publishing in favor of
open lymphatics, and saw von Recklinghausen’s paper as his own was
going to press, at first recognized in von Recklinghausen’s descriptions
a confirmation of his own views. On careful studying of the silver
pictures, however, His (45) saw that the discovery of endothelium was
the very step which, instead of proving that lymphatics are open and

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. i

without definite limits, tended rather to throw the balance of proof in
the opposite direction ; that the discovery of endothelium gave the con-
ception of lymphatic capillaries not as indefinite spaces without walls,
but as limited by a definite membrane. To quote his words (45, pp.
457-458), “ Wenn ich nach den eben gemachten Auseinandersetzungen
der einen Aufstellung v. Recklinghausen’s, namlich derjenigen vom
Vorkommen eines Epithels in den feineren Lymphgefissen des
Diaphragma vollkommen beistimme, so bin ich durchaus anderer
Ansicht als jener Autor hinsichtlich der Saftcanalchen und ihrer Bezie-
hung zu den Lymphgefassen ; fiirs erste naémlich leugne ich das Vor-
kommen von Saftcanalchen in dem Sinn von v. Recklinghausen, und
zweitens halte ich die Verbindung der von ihm als Saftcanalchen
gedeuteten Theile mit den Lymphgefassen nur fiir eine scheinbare,”
and later in his interpretation of the silver picture he says: “Je
diinner und blasser die supervasculare Schicht ist, um so leichter wird
es den Anschein haben, als ob die hellen Figuren der intervascularen
Bindegewebskérper unmittelbar in die Lymphgefisse selbst einmiinde-
ten, um so eher wird man uebersehen, dass in Wirklichkeit der Zusam-
menhang jener Figuren nicht mit den Gefassen, sondern mit den
Gefissen neberlagerten verzweigten Kérpern stattfindet.” Further he
adds that if the openings pictured by von Recklinghausen exist, there
would be no reason why any injection of a finely divided mass should
not fill practically all of the Saftcandlchen or tissue spaces as ex-
pressed in fig. 3, and this is obviously not the case. Thus after a care-
ful study and interpretation of the silver reaction on the lymphatic
vessels, on the albuminous fluid of the interspaces, and on the cell
bodies of the connective tissue corpuscles, His was led to believe that
the significance of the von Recklinghausen discovery was that lym-
phatics are lined throughout by an endothelium rather than that the
silver method demonstrates open lymphatics.

Ill. EMBRYOLOGY.
1. InTRODUCTION.

It is, I think, clear that the problem in this state could only be
attacked successfully by embryology with the purpose of establishing
the fundamental morphology of the system. The importance of this
is best expressed by His (44, p. 223), “Von den mancherlei offenen
Fragen, die seit mehr denn 200 Jahren in den anatomischen Schriften

Digitized by Microsoft®

8 Florence R. Sabin.

ueber das Lymphsystem discutirt zu werden pflegen, bietet wohl keine
ein so unmittelbares physiologisches Interesse dar, als gerade die Frage
nach dem Ursprung der Lymphgefisse in den Kérperorganen. Mag
man sich ueber die Bildung der Lymphe und ueber die bei ihrer
Fortbewegung wirksamen Krafte eine Theorie machen, welche man
will, so fordert diese als ganz unerlissliche Grundlage eine praezisere
Vorstellung von dem anatomischen Verhalten der ersten Wurzeln des
Systems, edi nun eine solche Vorstellung wirklich aus der Beobachtung
entsprossen, sei sie nur hypothetisch angenommen:” It is without
question that a comprehension of the relations of the lymphatic capil-
laries depends on an understanding of the fundamental morphology
of the system.

2. LympHitics as Di,atep TIssuE SPACES.

The relation of the lymphatics to tissue spaces formulated by von
Recklinghausen was entirely in agreement with the early crude investi-
gations of fcetal conditions recorded by Breschet (16). It was the
practically universally accepted idea up to 1900 that lymphatics arose
by the dilatation of tissue spaces caused by the fluid that exuded from
the blood vessels; that this dilatation of the spaces began in the
periphery, and that the vessels gradually approached the veins and
joined them. Thus the growth of the lymphatic vessels was thought
to be from the periphery to the center. This view was well stated in
1894 by Gulland (39), who found the lymphatics first in the subcu-
taneous tissue and then along the extremities in foetuses, human,
bovine and rabbit, between 3 and 4 cm. long, and thought that these
vessels subsequently joined the veins. He thought that the force of
the fluid in the tissues formed the lymphatics, so that they arose by an
entirely different method from his conception of the origin of blood
vessels (p. 467).

3. Bupcr’s Work.

An account of the newer work on the origin of the lymphatic
system must begin with Budge (17-22). Budge was led to a study of
the lymphatic system in birds through noticing the great variations
of the posterior lymph hearts in the adult (21). The posterior lymph
hearts had been seen in birds by Panizza (103), A. F. J. Mayer
(quoted by Stannius) and Stannius (142). Budge had already begun

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 9

to study lymphatics in the adult liver (17) and in bone (18) by means
of injection, and in 1880 (19) he described studies of a system of
canals which he could inject in mesoderm of three-day chick embryos.
These injections, which can be easily repeated, giving the exact pat-
terns of Budge, were really made into the extraembryonic coelom, as
has been shown by Mall (81) and myself (129), and have nothing to
do with the lymphatics. They are not, as Budge himself noted, lined
by endothelium ; indeed Budge himself was most cautious in his inter-
pretation, for he says (19, p. 325), “Mit Absicht habe ich den Aus-
druck Lymphgefassystem fiir das eben beschriebene Canalsystem
vermieden, weil mir hierfiir noch nicht Anhaltspunkte genug gewon-
nen zu sein scheinen. Und doch liegt der Gedanke hieran nahe.” In
1881 Budge (20) announced the important discovery that there was
a plexus of lymphatic vessels accompanying the allantoic arteries of
chicks easily injected from the sixteenth to the eighteenth day of incu-
bation, and that these injections ran to the thoracic duct. The next
year (21) in a fuller paper he described that these allantoic lymph-
atics not only entered the thoracic duct but drained into the posterior
lymph heart, which played a great réle in relation to the allantois.
He injected the lymph heart through the allantoic vessels by the tenth
day, and noted that the posterior lymph heart was preceded, as seen
in the living chick (8 to 12 days), by veins, and that subsequently the
lymph heart appeared, which could be distinguished by the fact that it
pulsated with a different rhythm from the blood vessels. After
Budge’s death his work was brought together by His (22) and formu-
lated somewhat in this manner: There are two lymphatic systems, a
transitory and a permanent system, the first consisting of spaces in the
extraembryonal membranes analogous to the coelom and found only in
early stages. The second, consisting. of the true endothelial-lined
vessels which accompany blood vessels—which Budge so successfully
injected. The thoracic duct he thought arose from the first system and
was the only permanent part of the first system. Thus Budge’s work,
which was a reaching out into a dark field and had genuine important
discoveries, was in line with the prevailing theory that lymphatics are
related to tissue spaces, and introduced a misleading conception that
lymphatics form after the same manner as the coelom and indeed in
part as an extension of the ccelom as far as the primativ system and
thoracic duct are concerned.

Digitized by Microsoft®

10 Florence R. Sabin.

4, STATEMENT OF THE ARGUMENTS FOR THE VENOUS ORIGIN OF THE
LYMPHATIC SYSTEM.

I shall now take up the proof of the theory that the lymphatic
system is derived from the venous system, and will subsequently
analyze the opposing theory, which, in its most recent from, is that
lymphatics grow on the addition of tissue spaces (Kampmeier 66 and
66a).

The essential elements in the establishing of the venous origin of
the lymphatics have been (1) the proof that lymphatic capillaries
grow by the sprouting of their endothelial wall and not by the addition
or hollowing out of connective tissue spaces. This argument is known
as the method of growth by sprouting. (2) The proof that lym-
phatics gradually invade the body from center to periphery, establishing
the point of lymphatic and non-lymphatic zones. This means that
the main ducts grow as capillaries by the same method as the per-
ipheral capillaries. This argument is known as the one of lymphatic
and non-lymphatic zones. (3) That the original lymphatic ducts
bud off directly from the veins. This is known as the origin of the
lymph sacs by budding.

Thus the study involves three elements: The nature of the lym-
phatic capillary, the nature of the lymphatic ducts, and the nature of
the first lymphatics or the lymph sacs. Each one of these three lines of
proof involves the study of endothelium; the first lymphatics are buds
from the endothelial lining of the veins; these endothelial buds invade
the body down to the ultimate capillaries by the cell division of their
endothelial wall.

A. GROWTH OF LYMPHATIC CAPILLARIES BY SPROUTING.

A study of the method of growth of lymphatic capillaries is neces-
sarily associated with the study of the method of growth of blood
capillaries, for as far back as the time of Schwann (139), who first
saw the capillaries in the living tadpole’s tail and who vaguely dis-
tinguished those which carried blood from those which had no cor-
puscles, there is a clear recognition that the two forms of capillaries
grow by the same method. This method Schwann thought to be by
the addition of hollow connective tissue cells.

The term growth by sprouting was introduced quite recently by His,
but the first description of the process dates back to the year 1844, to

Digitized by Microsoft®

Lhe Origin and Development of the Lymphatic System. 11

the work of Platner (106) and of Prévost and Lebert (108). Platner
described the growth of capillaries as seen in the living tadpole’s tail.
He analyzes Schwann’s idea of the addition of cells as necessarily
merely an inference from the appearance of the vessels in one stage, for
when actually seen growing new cells are never added on. He says
that rather each new vessel is a process of a preceding one and
describes these processes as long, thin Ausléiufer, which form loops
and soon show a double contour. He says that the tiny processes
never contain nuclei and are not cells, so that the interpretation of
Schwann of the addition of cells could not possibly hold. Platner
says that the same process of growth in the living tadpole had been
observed by Prévost and Lebert. Indeed, in the same year (108) they
published a series of four papers, in which they describe the growth
of capillaries, not only in Batrachians, but more in detail in connec-
tion with the blood islands in the chick. They definitely use the term
centrifugal growth and describe the process in the following graphic
terms (p. 239): “Les vaisseaux poussent des saillies latérales par
décollement partiel des lamelles du feuillet vasculaire, saillies plus
ou moins arrondies ou pointues, allongées, formant des espéces
déperons qui souvent finissent par se rencontrer, provenant de deux
cétés différents, et établissent ainsi des vaisseaux de communication.”
They thought that the processes or “spurs” were for the most part
hollow from the start and state that the larger vessels grew by the
same method as the smaller.

In 1846 Koelliker published two papers (67 and 68), in one of
which he announced the discovery of lymphatic capillaries in the
tadpole’s tail (67). He described the characteristic differences as seen
in the living form between the two types of capillaries, the irregulari-
ties and more numerous processes of the lymphatics as well as their
connection with the superior and inferior caudal lymph trunks. In
the other paper he gives a most valuable analysis of the views of the
early embryologists, Wolff, von Baer, Reichert and others on the prob-
lem of the growth of vessels. He takes up the various views of growth
(p. 118) which were then being discussed, namely, growth (1) by the
addition of unbranched cells, or (2) by branched or stellate cells,
or (3) by spaces hollowed out by the force of the heart beat, or (4)
from spaces produced by fluids in the tissues as well as (5) by the
then new work of Platner, Prévost and Lebert, showing centrifugal

Digitized by Microsoft®

12 Florence R. Sabin.

growth, and concludes that the vascular and lymphatic capillaries
grow by the addition of branched or star-shaped cells.

Four years later, 1850, Remak (125) rediscovered the lymphatics
in the tadpole’s tail by watching the living embryo, and, without
knowing of the preceding work, he hurriedly published an even more
graphic description of the form and growth of lymphatics than Koel-
liker’s. Subsequently, after seeing Koelliker’s work, he added a
note (p. 182-183) to emphasize the fact that if the growth be watched
in the living form star-shaped cells are never added—neither to the
blood capillaries, nor to the lymphatic capillaries, nor, he adds, to the
growing nerves. Many years later, 1886, Koelliker (71, p. 34), in
an article mainly on nerves, corrected his view of the growth of the.
lymphatics by the addition of connective cells in favor of the theory
of growth by sprouting.

The years from 1863 to 1867 represent a great advance in our con-
ceptions of the lymphatic system. The historical article of His (44)
on the lymphatic system must form the basis of any work which deals
with the development of the problem. In 1865 Stricker (143) pub-
lished a paper in which he expresses a doubt in regard to Koelliker’s
discovery of lymphatics in the tadpole’s tail, on account of the fact that
undoubted blood capillaries had also the blind sprouts. He, however,
submitted the blood capillaries to an experimental study of great
interest. He stimulated them with the vapor of ammonia and gives
the following description of his results:

Ich sah ein Gefassstiick, welches seinem Baue nach unzweifelhaft zu den
Kapillaren gezahlt werden musste, sich im Laufe von fiinfzehn Minuten
zweimal bis nahe zum Verschwinden des Lumens verengern und bis zum
Breitendurchmesser eines Blutkérperchens wieder erweitern. Ich sah
varicése Gefasse sich so verinderen, dass der Bauch an Stelle des Halses zu
liegen kam, und umgekehrt. Mit den Formverdnderungen ging gewoéhnlich
noch eine andere Veranderung Hand in Hand. So wurde beispielweise eine
eingeschnirte Kapillargefassstrecke so blass, wurden deren Conturen so
undeutlich, dass ich sie kaum mehr von der Umgebung unterscheiden

konnte, wahrend nach bis zwanzig Minuten das Gefiiss wieder in seiner
ursprtinglichen Deutlichkeit zu sehen war.

Thus he discovered the contractility of endothelium. He found that
electricity was a still better stimulant for observing this contractility.
He discusses his results from the standpoint of determining the nature
of the capillary wall, especially in connection with the silver prepara-
tions of Aeby, Auerbach and Eberth (quoted from Stricker 143),

Digitized by Microsoft®

\

The Origin and Development of the Lymphatic System. 13

which has already given the suggestion of a lining of flattened cells
for blood capillaries. Stricker saw clearly that the wall of the capil-
lary was of living protoplasm, but was confused in regard to whether
the blood stream was inter- or intra-cellular. He was perplexed also
in regard to the existence of cells or not, since he was dealing with
the growing tip, which we now know to be a syncytium. This idea
occurred to Stricker, as is plain in these words, “ Dass bei dem Aufbau
von Kapillargefassen ueberhaupt Zellen mit einander verschmelzen,
und dass solche Verschmelzungsspuren spater einmal durch Silber-
niederschlige kenntlich werden kénnen, lisst sich ferner auch nicht‘
leugnen.” Stricker also noted the sprouts of blood vessels as indi-
cating their method of growth. ;

I have already taken up the fact that von Recklinghausen’s work
on the endothelium of lymphatic vessels, studied by means of the
silver nitrate method, settled the question of the cellular nature of
the lining of lymphatics, especially in connection with the analysis of
the work by His, who had regarded lymphatics as vessels without
walls (44, p. 229), and saw by a thorough study of preparations made
‘by the silver method that lymphatics were lined throughout by endo-
thelium.

In 1867 Langer (75), as a conclusion to a series of studies on the
lymphatic system in amphibia, published a study of the lymphatics
in the tadpole’s tail. He was able to make complete injections of the
blood vessels and the lymphatics and gives a most interesting analysis |
of specimens as observed in alcohol from the standpoint of weighing
the evidence on the method of growth. He describes the uninjected
capillary as often appearing to have a single contour, while the injected
one had the complete wall. He could not find in the alcoholic speci-
men the tiny processes previously described in thé living form. He
noted, however, the larger sprouts, the loops and blind processes, and
says that the nuclei are definitely within the contour of the -wall of
the capillary, so that he makes the distinction between the vessel and
the surrounding tissue. He concludes that the various forms of the
processes must indicate the process of centrifugal growth for the lym-
phatic capillary as well as for the blood capillary.

Again in 1873 the growth of the lymphatics and blood vessels by
the method of sprouting was observed in the tadpole’s tail and very
clearly described by Rouget. By the use of curare he was able to keep
the larva still. He also emphasizes the great number of tiny filaments

Digitized by Microsoft®

14 Florence R. Sabin.

on the lymphatics and the fact that they never join with the con-
nective tissue cells.

Thus the observation of the growth of capillaries by sprouting
was made over and over again where there was the chance of seeing
the process in the living form or of obtaining injected specimens,
and yet during all these years the observations were not generally
accepted, since as the fundamental morphology of the lymphatic sys-
tem was unknown, the significance of the observations could not be
understood.

It is, however, not true that every observer who studied the living
tadpole’s tail described the process of the sprouting of the endothelial
wall of blood vessels and of lymphatics as showing the method of
growth. Indeed at least three observers have doubted whether there
are two different kinds of capillaries in the tadpole’s tail. I have
already noted that Stricker doubted the presence of lymphatic capil-
laries in the tadpole’s tail.

Wysotzky (155), whose work, since it is published in Russian,
I quote from Mayer (89), thought that the capillaries without blood
were merely young blood capillaries.

Sigmund Mayer (89), in an interesting and valuable paper, “ Ueber
die Blutleerengefisse im Schwanze der Batrachierlarven,” gives first
an excellent analysis of the literature on the subject of the capillaries
in the tadpole’s tail. He brings out the fact that it was the presence
of an occasional red corpuscle in the “lymphatic” which was the
stumbling block in accepting the presence of lymphatic capillaries.
He then records an observation which in reality cleared up the entire
difficulty in regard to the “empty capillaries.” As HE. R. Clark (26)
has shown, Mayer, however, failed to see the bearing of his own
observations. Mayer used curare and the electric current to ans-
thetize the larva and then covered the tail with a cover slip. When he
sucked the water out from under the cover slip he noted that the
blood would gradually stop flowing or even go in the reverse direction
in the blood capillaries and that the vessel would collapse, so that it
looked like a solid cord. Thus he also observed the contraction of
endothelium. Furthermore, he noted that the “other empty vessels ”
(namely, lymphatics), contracted more frequently than the blood
capillaries. When, however, he studied these other empty vessels he
thought that the occasional cells in them flowed now toward the center,
now toward the periphery, and he ended with the somewhat vague

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 15

and erroneous conclusion that the empty vessels, though they might
have some of the functions of the lymphatics of the adult, could not
properly be grouped with them. We now know that the two sorts of
empty vessels are (1) the collapsed blood vessels which Mayer himself
observed and (2) the true lymphatics (E. R. Clark, 26).

The work of Mayer, however, has another point of great interest
to us beside the important observation of the contractility of endo-
thelium, namely, what we call Mayer-Lewis anlagen. Mayer noted
the long strands of cells (endothelium) which often span the gap
between two capillaries; indeed, he figures them, some that look
entirely solid and some that have, here and there, a patent lumen. In
reality both blood capillaries and lypmhatic capillaries have these
strands and sprouts, and it is one of the most remarkable qualities
of the two forms of endothelium that no sprout from a lymphatic
capillary ever joins a blood capillary and vice versa. However, Mayer’s
mistaken idea that these strands of cells connected with two different
kinds of capillaries, namely, the vessels which are always empty, in
other words, the lymphatics and the blood capillaries, aided in his
confusion of the lymphatics with the blood vessels. Besides these
strands of endothelium he saw small hollow vessels, having the same
lining as the capillaries, which he thought to be entirely isolated.
These he interpreted as follows (p. 236) :

Unseren Erfahrungen nach sind nun die beschriebenen Bilder nur gewisse
Stadien der Ausschaltung von Bestandtheilen des Blutgefassystems aus der
normalen Circulation..... Bei genauerem Studium der letzteren For-
mationen ergibt sich, dass die Wandungen derartiger mit freien Spitzen in
das Gewebe hineinragenden, und ausserhalb jeder Kommunication mit
Gefadssréhren stehenden Fragmente bezitiglich des Aufbaues ihrer Wand
entweder mehr den: Typus der blutleeren oder demjenigen der bluthaltigen
R6hrchen sich n&hern.

Most of the apparently isolated anlagen seen in sections are due
to the collapsing of the vessels which connect them with the rest of
the capillaries through the contractility of the endothelium which
Stricker and Mayer observed. In sections they may occur in great
numbers through the contraction of the endothelium due to the fixing
agent. The possibility of entirely isolated vessels and the meaning
of the apparently isolated vessels are fully discussed later in connec-
tion with the theory of the growth of lymphatic capillaries by the
addition of Mayer-Lewis anlagen in section VIII.

Digitized by Microsoft®

16 Florence R. Sabin.

If it be established that the peripheral capillaries grow entirely
from preceding capillaries, then it is a perfectly logical hypothesis
that the lymphatics grow from the veins, from center to periphery,
rather than from the periphery to the veins. This hypothesis was
advanced by Ranvier (113), based on the evidence of the method of
growth of the lymphatic capillaries within the peripheral plexus.
Ranvier (110-118), in a series of studies between the years 1895 and
1897, made injections of lymphatics in a variety of embryos. For
example, he injected the exceedingly abundant plexus in the mesentery
of a fetal pig 10 cm. long, as well as in the rat and rabbit. He also
injected the lymphatics in the amphibia. These injections gave him an
extensive knowledge of the form of the developing lymphatic plexuses,
the valves looking like collarettes in the walls of lymphatic ducts, the
irregular capillaries with blunt terminal knobs, of which especially
large ones are often found in the mesentery as well as the slender
sprouts, either ending free or as bridging a gap in the network to
another vessel. From all of these forms of growing lymphatics Ran-
vier saw that the growth of lymphatics was from center to periphery,
and even though he thought that the lymphatics did not begin in a
mammalian embryo until it was about 10 cm. long, a stage at which the
embryology of the system may be called complete, nevertheless he went
a step farther than all of the preceding observers, for he deduced the
hypothesis that the lymphatics grow from the veins. He likened their
growth to the growth of a gland (113). In my own work I was led
to the same view by tracing the lymphatics in younger and younger
embryos back to the veins, and the next section will include the evi-
dence which I regard as essential to establish the hypothesis of Ranvier
as the correct theory of the origin of the lymphatic system. The ac-
count of the evidence for growth by sprouting is, however, not yet
complete. As far as the evidence can be obtained by the observation
of injected specimens, it was especially well described and figured
by MacCallum (79). He added the observation of watching the in-
jection under the microscope and showed that extravasations are due
to explosive rupture of the lymphatic wall.

Injected specimens of developing lymphatic capillaries are now
well known; excellent figures are given by Bartels (10, figs. 9 and 10,
p. 43). They show a large number of sprouts, because they are taken
from the skin of a pig embryo 6.5 cm. long. This is the exact stage
when the secondary, superficial, fine-meshed plexus is just growing

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 17

from the deeper coarse-meshed or subcutaneous plexus, and, since the
secondary plexus is growing actively, it shows many sprouts.

The: proof of growth by sprouting from injections is always an
interpretation of appearances, and though it is a logical inference,
one must turn to the observations of the living form for conclusive
proof. This has been given by E. R. Clark (25-26), who has again
restudied the classical object, the tadpole’s tail. The introduction
of chloreton anesthesia is one of the factors that have made this study
possible, since the same specimen can be kept alive for weeks, notwith-
standing repeated doses. Dr. Clark’s devise of an upright chamber,
so arranged that the tadpole can be kept upright without the pressure
of the cover slip, and the finding of larva of the form Hyla picker-
ingii in the spring of 1910, a form which has few pigment cells, have
all combined to make his studies so valuable and convincing. He
has described not only the complete history of a lymphatic capillary,
watched through periods of weeks, but has followed every connective
tissue cell in the neighborhood (30) through several generations. Dr.
Clark has shown that the wall of the lymphatic capillary is in ceaseless
activity. In the living form the wall is in part hyaline, in part granu-
lar. The nuclei are surrounded and obscured by granular protoplasm,
so that they are clearly seen only when dividing. The wall is of
irregular thickness, often extremely delicate, and from the sides and
tips are sent out numerous tiny processes, some hyaline and some
granular, which may be well described as amceboid. A few of these
persist and grow into permanent lymphatics. Most of them are with-
drawn. These tiny sprouts, indicating the functional activity of the
vessel, do not disappear in alcoholic specimens (26, p. 403). They
are at first without nuclei, but nuclei wander into them from the
parent stem; indeed, two nuclei may pass each other as they advance
or recede. This proves that the growing amceboid wall of the lym-
phatic capillary is a syncytium and explains the failure to obtain the
silver markings at the growing tip. The silver markings indicate
the more stable lymphatic capillary.

E. R. Clark’s most recent observations (30) show that increased
activity of the lymphatic wall in the sending out of many of the tiny
processes is a sign of growth, so that an area can thus be selected to
watch the processes of the formation ‘of new vessels.

As far as the relation to the surrounding connective cells is con-
cerned, the growing, and, we might also say, the functioning, ameeboid
tips avoid the cell bodies and processes of the mesenchyme cells.

Digitized by Microsoft®

18 Florence R. Sabin.

Thus there can be no question in regard to the method of growth
of the peripheral capillaries. Both blood and lymphatic capillaries
grow by the activity of the protoplasm of their walls. The endo-
thelium at the growing tip is a syneytium, the processes advance and
recede; they are tiny, hair-like vessels, and they receive their nuclei
from the parent stem. There are differences in the appearance of the
two types of capillaries probably associated with the presence and
absence of the circulating blood and the difference of function. The
lymphatics have very many more tiny sprouts, only a few of which
are permanent; that is, their wall is much more ameeboid and the
vessels much more irregular. Thus on positive evidence, namely, the
observation of the process in the living form, which is the best evi-
dence known to anatomical science, we are justified in the conclusion
that capillaries grow by sprouting.

B. LYMPHATIC AND NON-LYMPHATIC ZONES.

The second argument in favor of the venous origin and the centrif-
ugal growth of lymphatics by their own endothelium is that it can
be shown that lymphatic vessels gradually invade the body from certain
limited centers. Thus if injections of vessels are made from these
centers in larger and larger embryos an ever-increasing zone of
lymphatic capillaries can be demonstrated. Moreover, for each stage
the zone of vessels is constant.

I began my work with the study of the development of lymphatic
glands in the embryo pig and to obtain injections of them introduced
the needle into the foot pads. By taking younger embryos a stage
was soon reached when no injections of the foot pads ever entered
lymphatics, but at these stages one could still inject lymphatics in the
skin over the body. This was the beginning of the proof that lym-
phatics bud off from the veins and grow peripheralward instead of the
reverse (Sabin 129).

In an embryo pig 5 cm. long (Sabin 130, fig. 5) a simple plexus of
lymphatic capillaries has almost entirely covered the body. I say capil-
laries advisedly, adopting Ranvier’s (118, p. 74) criterion of valves
for lymphatic ducts in contrast to their absence in the capillaries.
The body is first invaded by lymphatic capillaries, and the primary
"plexus shown in fig. 5 (Sabin 130), which is the anlage of the deep,
subcutaneous plexus, was injected by a single puncture of the needle.

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 19

Subsequently this plexus develops valves, but at this stage the entire
lymphatic system can theoretically be injected from any one vessel,
because there are no valves, or, in other words, because it is entirely
a capillary system. Nearly the same stage is shown for a human
embryo in figs. 505 and 506 (Sabin 134). This was a remarkable spec-
imen, in which air filled the lymphatics, and it has the additional in-
terest of showing the beginning of valves in the vessels of the skin,
namely, those of an occipital duct and in the axillary and inguinal
ducts.

If earlier stages of embryo pigs are taken the cutaneous vessels, as
shown by injecting them, gradually recede, as seen in figs. 1 to 4
(Sabin 130), to two points, one in the neck and one in the groin.
The earliest lymphatics reach the skin in the neck of an embryo pig
about 18 mm. long. All injections of the lymphatics of the skin of
the anterior part of the body between the stages of 2 and 5 cm. long
can be seen to run into two symmetrical sacs in the neck close to the
jugular vein. These anterior lymph sacs, which have now been identi-
fied in a very great number of forms, can be found in pig embryos in
earlier stages than the first cutaneous lymphatics. In my first paper
I traced them back to the stage of 14 mm., when they are small empty
sacs, close against the anterior cardinal vein. In reality they begin
in the pig of 10 to 11 mm. and their method of origin will be discussed
in the next section.

The argument of lymphatic and non-lymphatic zones thus consists
in the fact that the first lymphatics are sacs close to the veins, and
from this stage onward lymphatics can be injected in wider and wider
zones until they cover the entire surface of the body.

It is necessary to prove that each zone represents the limit of the
growth of the lymphatics for a given stage. To prove an injection is
complete for any stage such as one shown in fig. 6 (Sabin 135), it is
essential to have abundant material and to show by injection that the
lymphatic tips always rupture along a given line when the pressure
is increayed. The shape of the lymphatic vessels at the line of growth
will vary according as the vessels are progressing rapidly or not; for
example, the rounded ends of fig. 2 (Sabin 129) are from an area
where lymphatics grow slowly, while the margins of fig. 6 (Sabin 135),
with numerous pointed sprouts, are from a rapidly growing zone. Fur-
thermore, sections of the skin in the lymphatic and non-lymphatic zones
show a sharp contrast in the presence or absence of the large lymphatic

Digitized by Microsoft®

20 Florence R. Sabin.

capillaries, fig. 6 (Sabin 130), which are easy to identify. The method
of sections is, however, inadequate to define the limits of growth,
except for the large vessels, for the tiny lymphatic sprouts can only
be found in sections when the plane of cutting is favorable.

This progression of the lymphatics in the skin has been confirmed
by Polinski (107) in bovine embryos. The patterns which he shows
make a most interesting comparison with those of pig, cat and human
embryos. In the bird the same progression of zones is shown to some
extent in the injections of Mierzejewski (96).

Beside the definite zones the lymphatics grow into definite layers.
Ranvier (118) noted that the superficial lymphatics are deeper than the
blood vessels. They follow the vessels into the skin and each plexus is
deeper than the corresponding blood vascular plexus (Sabin 130). The
development of the lymphatics into the intestine, as worked out by
Heuer (43), illustrates the point of the definite layers into which the
lymphatics grow particularly well. The lymphatics for the intestine
grow from the retroperitoneal sac, which arises from the renal veins
(Baetjer, 8). The vessels form a rich plexus in the mesentery and
from this plexus a series of vessels grow into the intestinal wall and
penetrate to the submucosa. Here they form at first a series of units
shown by Heuer in fig. 10 (43), which has been copied as fig. 508
(Sabin 134). These primary vessels soon form the submucosal plexus
of ducts, from which the mucosal capillary plexus and lacteals develop
on the one hand and the serosal vessels on the other.

This fact, that the body is invaded by lymphatics, has an especial
interest, because it was the clue which enabled us to trace back the
systems of lymphatic ducts to their beginnings and show that they
arise from the veins and grow toward the periphery. It is, further-
more, the key by which the development of the lymphatic system
within each organ may be traced to the ultimate capillaries. It is
considered again in relation to the development of lymph sacs, for
there it can be taken up in connection with its exact relations to
vascular and non-vascular zones. Both the blood vascular and the
lymphatic systems invade the body by the growth of their capillaries.
Each system has its definite zones and layers which can be followed in
orderly progression. The invasion of the body by the blood capillaries
is more complete than by the lymphatic capillaries; for example, the
lymphatic capillaries do not grow into the central nervous system, nor
into the interspaces between muscle fibers.

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 21

C. DIRECT BUDDING OF THE LYMPHATICS FROM THE VEINS.

As has been stated, it was the study of the decreasing zones of lym-
phatic capillaries which led to the discovery that the first lymphatics
are sacs close to the veins, In mammalian forms the first lymphatics
are the anterior lymph sacs which bud off from the anterior cardinal
veins. These jugular lymph sacs were noted by Saxer (138) in 1896
in his extensive study on lymph glands, but he did not recognize
their significance. They can readily be made out in sections of pig
embryos 14-20 mm. long, or in sections of any mammalian embryo
of the corresponding stages. The fact that they are the first lymphatics
in mammals and that they lie close to the vein, separated only by the
double endothelial wall, led me to the conclusion that the lymphatics
budded directly off from the veins and it is gratifying to find that
this reasoning has been justified by further investigation (Sabin 129,
E.R. & E. L. Clark 29 and Hoyer 49a).

To follow the development of our knowledge in historical sequence
the next step was taken by F. T. Lewis (76) in a series of excellent
reconstructions of the anterior lymph sacs in rabbit embryos, the best
.. figures we have had up to the present time of the form of the early
sac. Lewis carried the study still further back and showed that the
definite sac was preceded by a blood-filled capillary plexus. This blood-
filled plexus has proved to be a very important observation. From the
presence of blood Lewis made the logical inference, with the evidence
at. his command, that these vessels were blood capillaries and that
hence lymphatics were transformed blood capillaries. The observation
of the blood-filled vessels has been confirmed by Huntington and
McClure (54) on the cat and by myself in human and pig embryos
(132-133) and we have all accepted the theory of transformed capil-
laries until new evidence has now shown us the correct explanation
of the presence of the blood.

This new evidence has been furnished by the work of E. R. and
E. L. Clark (27-29). E. L. Clark began with the study of the allan-
toic lymphatic vessels in chick embryos. From the allantoic vessels
she was led back to the posterior lymph hearts and the lymphatics of
the body wall. The injections of Mierzejewski (96) show that the
primary line of growth for the lymphatic vessels in the skin of the
chick is along the thoraco-epigastric vein (96, figs. 1 to 4). EH. L.
Clark noted that in early stages the lymphatic vessels along this line

Digitized by Microsoft®

22 Florence R. Sabin.

were filled with blood and that the blood in them was stagnant, in
striking contrast to the rapidly circulating venous blood. The fact
that the lymphatics invade the skin in the chick while they are in the
blood-filled stage gives at once an advantage. By means of fine glass
cannulas E. L. Clark was able to inject individual vessels of this blood-
filled plexus in the living chick and watch the ink flow to the pul-
sating lymph heart and thence to the veins. These blood-filled vessels
have the usual characteristics of lymphatic capillaries, in that they
are larger and more irregular than blood capillaries, but the essential
point in E. L. Clark’s work is that by using a form in which the lym-
phatics could be seen in the living embryos she discovered a new
criterion for recognizing lymphatics, namely, the stagnant blood in
contrast to the circulating blood.

With this important criterion she has followed back the origin of
the posterior lymph hearts. Since Sala (137) it has been known that
in the chick the posterior lymph hearts develop opposite the lateral
branches of the first five coccygeal veins (137, p. 269). Sala described
the process as beginning during the sixth day by a hollowing out of
spaces in the mesenchyme, but by watching the living chick embryo
during the fifth day E. R. and E. L. Clark have seen that the skin
over the first five coccygeal veins is a comparatively non-vascular zone,
so that brilliant direct illumination enables them to see in the depth
a series of tiny blood-filled buds close to the main coceygeal vein and
its branches. The blood in these buds is of slightly different color
from the circulating blood. They proved that these buds are always in
connection with the parent vein, for a direct injection of them always
runs over into the vein, but no injection of the peripheral blood capil-
laries ever fills them. Hence they bud off from the veins and are filled
with blood from a back-flow from the parent vein. It is of course evi-
dent that this process cannot be seen in the body wall of the chick with
the clearness with which every cell division can be followed in the
tadpole’s tail ; nevertheless, the stagnant blood has been proved by injec-
tion to practically fill the developing vessels, so that it is an adequate
criterion of the extent of their development.

By selecting a chick which shows these primary lymphatic buds and
keeping it under observation under high power in a warm chamber
B. R. and E, L. Clark (29) have been able to watch these blood-filled
lymphatic buds join with each other to form a deep circumscribed
lymphatic plexus, the anlage of the posterior lymph heart. It is clear

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 23

from Sala’s description that it was this plexus which he described
as the beginning of the lymphatics. While the deep lymphatic capil-
lary plexus is being transformed into a lymph sac and pulsating
lymph heart, lymphatic vessels are already growing toward the skin,
where they gradually extend. They grow first along the lateral line
to anastomose with the jugular lymphatics. There is a continuous
peripheral growth of the lymphatics from the time of the first buds
on, and the entire blood-filled stage takes but a few hours. Neverthe-
less, it is long enough to enable one to watch the buds form a plexus;
the plexus becomes a sac and lymph heart, and the lymphatics from it
spread out in the skin. Injections prove that in every stage of their
development, from the primary buds onward, the lymphatics connect
with the parent veins; thus there is no separation from the vein and
subsequent rejoining, as all of the recent workers, Lewis, Huntington
and McClure and myself, have thought from the difficulty of finding
the connections with the veins in serial sections.

~The posterior lymph heart in the chick is especially advantageous
for study, since it is in the first place sufficiently superficial to be
seen in the living embryo, and in the second place it is covered by a
non-vascular zone. In mammalian embryos no caudal lymph hearts
develop. It is very interesting to note in injected embryo pigs that
the corresponding area in the skin at the root of the tail, instead of
being a non-vascular zone, is an area of especially wide blood capil-
laries. This area, seen in fig. 467 (Evans 35), is constant, and, as
it almost always retains a little blood, can be injected when a partial
injection of the veins of the lower part of the body is desired.

The work of the Clarks adds the final proof of the theory that the
lymphatics bud off from the veins. Moreover their method is so simple
that anyone may convince himself of the essential results by a few
hours observations. Their injections have corrected two errors,
namely, the idea that the lymphatics are transformed blood capillaries
and that they temporarily separate from the veins.

Thus I regard as proved the three points which have been stated
as essential in establishing the origin of the lymphatics from the
veins. Stated in reverse order they are: First, the lymphatic system
begins as a series of endothelial buds which sprout out directly from
the veins as a new type of vessels, namely, lymphatics. These buds
unite to form plexuses which develop into sacs. These sacs may become
lymph hearts. Second, from these lymph sacs or hearts lymphatic

2

Digitized by Microsoft®

24 Florence R. Sabin.

capillaries gradually invade the body in orderly sequence in definite
and characteristic zones and layers. The growth ‘is always in the
capillary bed, that is, all lymphatics develop as capillaries, and the
earliest ones to develop become the first lymph trunks or ducts. Third,
the method of growth throughout is by the sprouting of the proto-
plasm and the nuclear division of the endothelial of the capillary wall.
The ultimate capillaries are distinguished by the absence of valves.
In a word, there ig a continuous growth and invasion of the body by
lymphatic capillaries from the primary sprouts which bud off from
the veins to the ultimate peripheral lymphatic capillaries.

IV. PRIMARY LYMPHATIC SYSTEM IN DIFFERENT
FORMS. :

1. Priaary Lywpratic Sacs iN THE Pic.
A. GENERAL SUMMARY OF TIE SACS.

I shall begin the section on the special anatomy of the lymph sacs
by a detailed description of the origin of the lymphatics in the pig,
since that is the best known form. I can now describe the origin of
all of the lymphatic sacs in the pig and give the relation of the main
ducts to all of these primitive sacs. On the basis of our knowledge of
the lymphatic system in this form I shall give what is known of the
lymphatic system in other mammals, in birds, reptiles and amphibia,
and shall then be in a position to compare the primary lymph sacs in
mammals with the amphibian lymph hearts and sacs.

In the pig the lymphatics bud off from the veins in two places, from
the anterior cardinal veins and from the veins of the Wolffian body.
There are two sets of paired sacs, the jugular and the iliac; and two
unpaired sacs, the retroperitoneal and the cisterna chvli. In the most
genera] terms the jugular sacs drain the anterior half of the body;
the iliac sacs drain the posterior half of the body, while the retroperi-
toneal or pre-aortic sac drains the viscera. The cisterna chyli with
the thoracic duct connects the jugular and renal lymphatics. In more
specific terms the jugular lymphatics bud off from the anterior cardi-
nal veins, form large sacs in the neck, from which lymphatics grow
to the head and neck, the foreleg and thorax, and to the heart and
lungs. From the left sac arises the jugular stem of the thoracic duct.
All the rest of the lymphatic sacs bud off from the vena cava, where it
is a part of the median mesonephritic vein, from the median mesone-

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. — 25

phritic vein and the veins in the dorso-medial edge of the Wolffian
bodies. The lymphatics which grow from the ventral surface of the
mesonephritic veins make the large retroperitoneal or pre-aortic sac;
those which grow along the dorso-medial edge of the Wolffian bodies,
lateral to the aorta, form the iliac sacs, and those which grow dorsal
to the aorta make the cisterna chyli-and thoracic duct. The thoracic
duct is formed in small part from a duct from the left jugular sac;
in larger part from a plexus which surrounds the aorta from the
cisterna chyli. The retroperitoneal sac drains the diaphragm and the
abdominal viscera, except the permanent kidney; the iliac sacs drain
the permanent kidney, the abdominal walls, the hindlegs, tail and em-
bryonic membranes.

B. THE JUGULAR SACS.

As has already been stated in part III, the spreading of the primary
superficial lymphatic capillaries in the skin of the embryo pig takes
place while the embryo is growing from 20 to 50 mm. long. The first
lymphatic buds are found when the embryo is about 11 mm. long
(fig. 6); the sac is well formed at 18 mm. (fig. 7), and reaches its
maximum size in an embryo 30 mm. long (fig. 8). It remains as a
large sac until the embryo measures 50 mm., when it begins to be
transformed in part into lymph glands (fig. 16).

An understanding of the development of the jugular sac depends
on a knowledge of the veins of the region. Indeed, I was not able to
inject the early stages until I had so mastered the pattern of the
blood vessels that a little blood in any of them enabled me to repro-
duce the whole picture and so avoid the blood vessels.

The study of the form of the early anterior cardinal vein is shown
for the chick by Evans (34) in his figs. 1 to 3. These figures show
that the primitive deep dorsal segmental branches drain the ventro-
lateral surface of the spinal cord. They bring out the gradual curve
of the anterior cardinal vein and the more marked bend of the pos-
terior cardinal vein, where the two join the duct of Cuvier. This
stage is shown for the chick in cross section by Evans (35) in his
fig. 437 B, as well as in (Evans 34) his fig. 4.

In fig. 4 is given an injection of the blood vessels in an embryo pig
7 mm. long, a stage just before the lymphatics begin. The curve of
the dorsal border of the cardinal veins toward the duct of Cuvier is

Digitized by Microsoft®

26 Florence R. Sabin.

slight, not as marked as in the chick, nor as it is on the left side of this
same embryo.

Arising from the cardinal veins there is a series of seven deep
segmental branches extending to the nervous system between the
area of the primitive posterior cerebral vein (Mall 143), which arches

va % tf te

Fic. 4.—Injected specimen to show the anterior cardinal vein and its
branches in an embryo pig which measured 7 mm. in its greatest diameter.
Magnified about 28 times. This is a stage before the lymphatic vessels
have begun. A =aorta arches; D. C.= duct of Cuvier; V. c. p. = vena
cerebralis posterior; V. 5 = fifth segmental vein.

downward over the medulla and cord and the cerebral end of the arm
bud. The first two of these branches drain the medulla, and hence
there are five spinal or segmental branches proper. Each of these
branches has a corresponding artery which lies slightly caudalward
and nearer the midline. Only the lower four segmental arteries are
injected in this specimen. The relative position of artery and vein

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 27

shows in fig. 437 (Evans 35). Each of the five spinal veins drains
a plexus of wide capillaries, which marks the presence of the develop-
ing spinal ganglia.

It is, however, the superficial veins which are of more interest from
the standpoint of the lymphatic system. Along the dorso-lateral bor-
der of the cardinal veins there is a series of intersegmental sprouts,
varying from one to four or five in each interspace. These sprouts
drain a superficial capillary plexus which is developing in a groove
in the neck which is a continuation of the groove dorsal to the Wolffian
ridge. The arteries for this superficial capillary plexus come from the
deep segmental arteries (Evans 35, fig. 437). The connections of
the plexus with the cardinal vein are most abundant near the poste-
rior cerebral vein. ‘This superficial capillary plexus of the body wall
has a most interesting development. For the area between the limb
buds it has been worked out by Dr. Helen W. Smith (141), whose
figures are copied by Evans (35). This superficial plexus in the tail
region shows a constant zone of widened capillaries (35, fig. 467)
corresponding to the position of the posterior lymph hearts in birds.
For the neck region it is necessary to know this plexus in order to
locate the jugular lymphatics, and therefore its development will be
followed in further stages.

The ventral branches of the anterior cardinal vein are represented
at this stage by one large pericardinal vein, which receives branches
from the gill arch region. The common stem for the branchial and
pericardial vessels connects with the anterior cardinal vein, but it
drains directly into the duct of Cuvier (see Grosser and F. T. Lewis
as quoted by Evans 35, p. 660) on the right side of this embryo,
while on the left side of the same specimen it opens into the anterior
cardinal vein.

The earliest lymphatic buds have been found in embryo pigs
measuring from 10 to 11 mm. long. All of my measurements mean
the greatest length and are made on the fresh specimen. The stage
of the first lymphatics is shown in two figures, fig. 5, a total mount
of an embryo injected with India ink through the umbilical artery
in a specimen 11 mm. long, and fig. 6, a cross section 250 4 thick,
through the lymphatic area in an embryo measuring 11 mm. The
plood vessels were injected with India ink, and the lymphatics are filled
with blood.

Digitized by Microsoft®

20 Florence R. Sabin.

\

‘N ach

Fic. 5.—Injected specimen of the anterior cardinal vein and its branches
in an embryo pig measuring 11. mm. in its greatest diameter. Magnified
about 33 times. This is the stage at which the lymphatic vessels begin to
develop. The vessels were injected with India ink through the umbilical
artery. D.C. = duct of Cuvier; V. c. p. = vena cerebralis posterior; F. u. v.
= primitive vena ulnaris; V. 5 = fifth segmental vein.

Digitized by Microsoft®

The Origin and Development of the Lymphatic Syslem. 29

The blood vessels show a great development. The five spinal seo-
mental veins are clear, draining the large plexuses around each spinal
ganglion; the ganglia are now farther yentralward than in the pre-
ceding stage. The superficial capillary plexus shows the most marked
changes. There is a continuous sheet of superficial capillaries in the

Fic. 6.—Section through the third segmental vein in an injected embryo
pig which measured 11 mm. in its greatest diameter. The section, which is
250 w thick, shows the relation of the jugular lymphatic sprouts or buds
to an anterior cardinal vein. Magnified about 40 times. The blood vessels
were injected with India ink through the umbilical artery. The lymphatic
sprouts, L. s., have a natural injection of blood; these lymphatic sprouts lie
in a non-blood-vascular zone dorso-lateral to the anterior cardinal vein.
A. c. Vv. = vena cardinalis anterior; g — groove.

neck, anastomising with the branchial vessels, with the pericardial
vessels and with the abundant capillary plexus of the arm bud.
Opposite the medulla the connections of this superficial plexus with
the anterior cardinal vein are very numerous, while opposite the heart
a few veins connect it either directly with the anterior cardinal vein
or more frequently with the main segmental branches. It is this

Digitized by Microsoft®

30 Florence R. Sabin.

zone, opposite the third, fourth and fifth segmental veins, that is the
lymphatic area.

To locate the lymphatic area definitely it is necessary to note the
details of the superficial veins more carefully. The superficial plexus
along the lateral groove is finer meshed and of smaller capillaries than
the rest. The groove is an important land mark and it shows best in
the cross section, fig. 6. From the plexus in the groove a sheet of wide
capillaries is growing dorsalward external to the myotomes, and loops
from this sheet connect with the deep plexus around the spinal cord.
The skin over the entire dorsal surface of the cord is an entirely non-

vascular area at this stage. From the ventral border of the super-
ficial plexus in the groove a sheet of wider and more open-meshed capil-
Jaries covers the surface of the embryo over the anterior cardinal
vein. The surface contour of this, the lymphatic area, is a swelling
seen in fig. 6, which I shall call the jugular lymphatic ridge. The
blood capillaries of this area are very superficial and drain into the
cardinal vein in four ways: (1) Through the plexus in the groove and
its dorso-lateral superficial veins: (2) in small part through direct
branches to the lateral surface of the cardinal vein; (3) through
branches of the ventro-lateral surface of the anterior cardinal vein
which drain the pericardium (fig. 5); and (4) through anastomoses
with the capillaries (fig. 5) of the primitive ulnar vein. The primi-
tive ulnar vein has developed from the diffuse capillary arm bud
plexus of the preceding stage. It enters the lateral surface of the
posterior cardinal vein directly opposite the fifth segmental branch.
Thus a part of the blood of the branchial region and of the pericar-
dium drains through the superficial plexus over the lymphatic ridge
into the primitive ulnar vein. In a little older embryos, namely, those
12 mm. long, a second vein has developed from the arm bud plexus
just cerebralward to the primitive ulnar, which now receives the blood
from the plexus of the lymphatic ridge. The complete description of
this superficial plexus involves a study of the origin of the external
jugular vein which I do not wish to go into until I can illustrate it
adequately. I bring up the point here because A. H. Clark (24) has
shown that occasionally a lymph trunk in the neck in older pigs opens
directly into the external jugular vein near its point of union with the
internal jugular vein. It is therefore of importance to note that the
capillary plexus in the arm bud, from which this part of the internal
jugular vein comes, is present in the embryo when the lymphatics are

Digitized by Microsoft®

Lhe Origin and Development of the Lymphatic System. 31

budding out, though the internal jugular vein as a whole is formed
considerably later.

For the present the interest in the superficial blood capillary plexus
is that it covers the lymphatic ridge. It will be seen in fig. 6 that the
lymphatic ridge includes a non-vascular area. This non-vascular area
is bounded by the anterior cardinal vein and its dorsal branches, the
plexus of the groove, the wide-meshed superficial plexus and the
ventro-lateral branches of the anterior cardinal vein. An occasional
direct lateral branch of the cardinal vein cuts through the otherwise
non-vascular area. Within this area are seen the blood-filled lymphatic
buds. They lie in the angle between the anterior cardinal vein and its
dorso-lateral branches and connect with both. They are opposite the
third, fourth and fifth dorsal segmental veins, and extend a distance
of 1.5 mm. from the primitive ulnar along the anterior cardinal vein.
The tiny lymphatic buds are already sending sprouts away from the
parent vein.

The presence of stagnant blood, which has proved so valuable a
criterion in the living embryo, must be used guardedly in studying
sections, unless there is a complete vascular injection. In the lym-
phatic area [ have noted in the fresh embryo that the blood capillary
plexus of the groove is often empty, while the superficial plexus over
the lymphatic area and over the pericardium tend to retain some
blood. The blood vessels of an area where lymphatics are budding
must be thoroughly known before one can be sure of the lymphatic
buds. The early lymphatic buds are packed with blood to an extent
not common for the veins, which perhaps shows best in figs. 490 and
491 (Sabin 134), for the jugular lymphatic buds in a human embryo.

I have not yet succeeded in injecting the first jugular lymphatic
buds in the pig as E. L. Clark has done for the early stages of the
posterior lymph heart of the chick. In the embryo 12 mm. long they
can be seen in the fresh embryo, and sections show that they have
formed a plexus along the margin of the anterior cardinal vein. To
see the lymphatic buds at 12 mm. the embryo should be placed in warm
Locke’s solution while the heart is still beating, and it should be viewed
under the high powers of the binocular microscope with the direct
sunlight focused upon it. The lymphatic area is wider in the dorso-
ventral direction than in fig. 6, but its dorsal boundary is clearly
marked by the surface groove. If the superficial blood capillary
plexus is empty the lymphatics can then he seen Icoking like a cluster

Digitized by Microsoft®

32 Florence R. Sabin.

of dull red grapes against the cardinal vein; in the older specimens
some of the buds are markedly larger than others, while at 14 mm.
the largest of them makes a definite sac with large sprouts projecting
dorsalward. At this stage, namely, 14 mm., the sac can be injected

ce Me BE Heat

Fic. 7—Injection of the left jugular lymph sac in an embryo pig which
measured 18 mm. in its greatest diameter. The specimen was fixed in 10
per cent formol and cleared by the Spalteholz method. Magnified about 30
times. The glass canula which shows piercing the upper border of the arm
bud shows the point of injection. A. c. — anterior curvature of the lymph
sac; A. s.=apex of the lymph sac; I. = glass canula used for injection;
S. s.= stalk of the lymph sac; V. j. i.=vena jugular interna.

by direct puncture. The preceding cr plexus stage is shown by E. L.
Clark for the corresponding sac in the chick in fig. 15. The beginning
sac is shown for the human embryo in fig. 491 (Sabin 134).

It is not easy to get perfect injections by direct puncture, but by
the time the embryo is 16 mm. the dorsal sprouts from the sac are

Digitized by Microsoft®

Lhe Origin and Development of the Lymphatic System. — 38

WE Ure wes

Fic. 8.—Injection of the right jugular lymph sac in fetal pig measuring
3.5 cm. Magnified about 10 times. After A. H. Clark (24); this specimen
was shown as fig. 3 in 130; it has since been cleared by the Spalteholz
method so that it shows the relation of the superficial lymph vessels to the
jugular sac. It shows a complete injection of the suprascapular and oc-
cipital plexuses, but an incomplete injection of the superficial cervical
plexus. S. s. = the stalk of the sac showing faintly through the shoulder;
V. 1. f. = vena linguofacialis.

Digitized by Microsoft®

3t Florence R. Sabin.

sufficiently large so that the sac can be injected indirectly through
them.

The process by which a lymphatic plexus becomes a sac has been
called confluence or cavernization by Ranvier (117-118). He says
that when two lymph vessels lie in contact the endothelial wall between
them disappears (possibly it is retracted) and thus large sacs develop.
It is exceedingly interesting that this process does not take place when
lymphatic endothelium rests on venous endothelium, as is the case
of the jugular lymph sac.

The most interesting stage in the formation of the jugular sac is
shown in fig. 7 from an embryo 18 mm, long. To inject the sac at
this stage it is again important to note the plan of the blood capillaries
In fig. 6 it will be noted that there is a tiny blood capillary which cuts
across the lymphatic area from the superficial plexus to the deeper
veins and divides the lymphatic area into two parts, a ventral jugular
part and a dorsal part. By the time the embryo is from 16-20 mm.
long the path of this slender vessel is occupied by a considerable plexus
of blood vessels and by nerves as well. The sprouts from the main
jugular sac (fig. 7) have grown up into the dorsal lymphatic area and
there developed into an abundant plexus. It is this plexus which can
be injected, as is shown by the glass tube which pierces the arm bud.
The original sac is shown just lateral to the shadow of the internal
jugular vein. This is difficult to inject by direct puncture, because
the blood capillary plexus over it has become double and is excessively
abundant. Moreover the sac is now just mesial to the developing
external jugular vein and lies very close to it. The blood capillary
plexus of the groove shown in fig. 6 is now deeper in, and the entire
area of the groove so dense a vascular area that no canula can enter the
zone without filling the blood capillaries. But just ventral to the
groove in the dorsal lymphatic area almost every puncture will fill the
lymphatics, provided it avoids the superficial blood capillaries. This
dorsal lymphatic area is destined to be the posterior triangle of the
neck ; even in this early stage it looks translucent and continues to do
so, for it is always uncovered by muscle, lying between the trapezius
and the sternocleidomastoid muscles (fig. 17).

A description of the form of the sac in fig. 7 will make the basis
for the description of the peripheral lymphatics of the head, neck and
thorax worked out by A. H. Clark (24) and given in section V. The
primitive sac is the portion above the arm bud and lateral to the

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 35

internal jugular vein. This becomes the internal jugular trunk; the
cerebral end of this portion is an abundant lymphatic plexus, the
anlage of a single gland in the pig (A. H. Clark) which drains the
pharynx. In the dorsal lymphatic zone the rest of the sac is a com-
plete arch from which develop the lymphatics of the shoulder, head,
neck, face, arm and thorax. The dorsal arch of the sac becomes by
far its largest portion, as will be readily seen in fig. 8, from an embryo
3.5 em. long. This is the same specimen shown as fig. 38 (Sabin 130).
It has since been cleared by the Spalteholz method, so that it shows.
the sae as well as the superficial lymphatics.

The transition stages between fig. 7 and fig. 8 are very easy to
inject and can be readily imagined. The further development of the
sac, especially with reference to the lymphatics along the external
jugular vein, is brought out in the sections on the peripheral lymphatics
and glands.

C. THE RENAL SACS.

The study of the rest of the lymph sacs in the pig may well begin
with the interesting and valuable work of Silvester (140). Silvester
has shown that in the monkey all of the lumbar and mesenteric lyn-
phatic ducts drain not by the thoracic duct into the jugular veins, but
directly into the renal veins or the surface of the inferior vena cava
near the renal veins. This he shows in a number of figures from
beautiful injections.

A. RETROPERITONEAL SAC.

It had previously been discovered by Lewis (76) in a study of
rabbit embryos that a lymphatic sac develops just ventral to the median
mesonephritic vein. Baetjer (8) proved that this, the retroperitoneal
or pre-aortic sac, arises from the ventral surface of the large vein
which connects the two Wolffian bodies in embryo pigs measuring from
17 to 23 mm. long. Baetjer’s fig. 3, from an embryo 19 mm. long,
shows the blood-filled lymphatic buds which have been slightly injected
from the veins. These buds rapidly form a large sac which lies in the
root of the mesentery and is the place of origin for all of the mesen-
teric lymphatics (Heuer, 43); those that grow to the stomach and
intestine, the liver, the capsule of the Wolffian bodies and the repro-
ductive glands. The blood-filled lymphatic buds can be seen in the
fresh embryo pig 19 to 20 mm. long, if the intestine is pushed to the

3

Digitized by Microsoft®

36 Florence R. Sabin.

side and the veins are emptied of blood by injecting them with salt
solution.

The retroperitoneal sac becomes the largest of all the lymphatic sacs
in the pig. As shown in Heuer’s fig. 3 (43) it spreads out just behind
the rectum and covers the entire area in the root of the mesentery
between the two Wolffian bodies. It spreads over the ventral surface
of the Wolffian bodies as far as the edge of the reproductive glands.
It was from injecting this sac that Heuer was able to study thé devel-
opment of the lymphatics of the intestine.

This sac would give an excellent opportunity to study the process of
cavernization. Injections of it with silver nitrate give beautiful
specimens, showing the endothelial-covered trabecule that cross its
lumen. Total preparations of the silvered sac remind one of the trabec-
ule in the wall of the cavity of the heart. These trabecule, which
are the beginning of the process of transferring the sacs into lymph
glands, show especially well in sagittal sections of pig embryos 20-25
mm. long.

2. Intac Sac—CisTERNA CHYLI.

I can now bring some evidence to show that the iliac lymphatics
which drain the legs, tail and abdominal wall, and the cisterna chyli
which forms the lower part of the thoracic duct, arise together as buds
from veins of the Wolffian bodies. A complete account of this process
needs a more extensive illustration of the blood vessels of the region
than I can give at this time. In the pig the lymphatics which bud
off from the veins of the Wolffian body and grow forward dorsal to the
aorta to form the cisterna chyli and caudalward along the edge of the
Wolffian body to form the iliac lymphatics, do not begin until the
embryo is 22 mm. long. I found that they arise from the mesone-
phritic veins because a direct puncture of the blood-filled buds entered
the main veins and not the blood capillaries. In a litter of pigs which
measured 23 mm. I washed out the blood vessels with Locke’s solution,
and then opened the specimens and pushed one of the Wolffian bodies
over toward the midline, A plexus of blood-filled lymphatics was
then readily seen dorsal to the aorta. I succeeded in puncturing some
of the larger vessels of the plexus forming the cisterna chyli in three
specimens and saw the ink run from the lymphatics into the mesone-
phritic veins. In all of these injections the cisterna chyli is obscured
by extravasations at the point of injection, but they all show the

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 3

iliac lymphatics. For the present I shall describe the renal lymphatics
from sections which can be done in the blood-packed stage.

The description of the iliac lymphatics, however, needs an outline
of the blood vessels of the region. If the veins of the prevertebral
space between the level of the median vein connecting the Wolffian
bodies and the root of the tail be injected it will prove that they

Fic. 9.—Section through the hilus of the Wolffian bodies to show the place
of: origin of the iliac lymphatic vessels and the abdominal part of the
thoracic duct in an embryo pig 23 mm. long. (Specimen 23b.) Magnified
50 times. A.= aorta; G. s. s. c.=gl. suprarenalis s. corticalis; L. v.=
lymph vessels filled with blood; M. v. — vena mesonephritica; R. s. — retro-
peritoneal lymph sac, a part of which is filled with blood, a part empty;
W. b. = edge of the Wolffian body.

are enormously abundant. In the embryo measuring 20 mm. the
segmental veins draining the cord and the body wall make an
extensive network, which drains in part into the median vein of the
Wolffian body and thence directly to the vena cava, while farther
caudalward this same plexus -drains directly into the large surface

Digitized by Microsoft®

38 Florence R. Sabin.

veins of the Wolffian body. Opposite the lower pole of the Wolffian
body this plexus is directly continued into the iliac ves. Cross
sections of injected embryos at this stage, namely, 20 mm., show that
the entire area lateral to the aorta and between the two Wolffian bodies
is practically filled with a plexus of these segmental veins. By the
time the embryo is 22 or 23 mm. long, however, there is a small non-
vascular area along the dorsal-medial edge of the Wolffian bodies, and
into this small area the iliac lymphatics bud. The earliest specimen
in which I have found these blood-filled buds measures 22 mm.; the
specimen is of interest because it shows clearly that the buds lie against
the veins of the Wolffian body itself as well as the median mesone-
phritic vein. The place of origin of the cisterna chyli and iliac lym-
phatics can be best seen in fig. 9, which is a cross section of an embryo
measuring 23 mm. through the median mesonephritic vein. Just
lateral to the aorta are the masses of the sympathetic nervous system
through which runs the great plexus of the segmental veins, and the
cortex of the adrenal bodies. Ventral to the aorta is the mesonephritic
vein, and the retroperitoneal sac, some of the buds of which are full
of blood and some now partly empty. In the dorso-medial edge of the
two Wolffian bodies is the plexus of blood-packed lymphatic buds which
arches across the midline, and, as shown in fig. 10, which is farther
cerebralward, forms a definite blood-filled cisterna chyli in the midline.

My various series of cross sections of pigs of about the same meas-
urement are interesting, because they show the progression of the
buds. In a specimen (23a) the iliac lymphatics extend only as far as
the hilum of the permanent kidney which lies dorsal to the lower
pole of the Wolffian bodies. The iliac buds lie throughout their
course in the same relative position to the Wolffian bodies and the aorta
as is shown in fig. 9. In this specimen (23a) the iliac lymphatic
plexus is spreading over the capsule of the permanent kidney, showing
that it receives lymphatics while they are in the blood-filled stage.
In later stages I have injected lymphatics from the iliac sac around
the pelvis of the kidney. It is thus clear why the lymphatics of the
kidney drain into the iliac lymph glands, while those of the repro-
ductive glands which develop ventral to the Wolffian body drain
into the pre-aortic glands. In a series 23b from the same litter as 23a
the iliac plexus is complete, extending from the mesonephritic vein to
a dilated sac opposite the bifurcation of the aorta. It was this swollen
end of the iliac lymphatics that I identified in embryos 25 mm. long

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 39

and called the posterior sac in 1901 (Sabin 129). The space for the.
iliac chain in the pig is small, much smaller than in human embryos,
and the iliac vessels in the pig are therefore very inconspicuous in cross
sections, except in the blood-filled stage. They can be found, however,
when one knows that they lie between the segmental veins and the
dorso-medial edge of the Wolffian bodies or kidneys. Sagittal sections
are, however, brilliant for the iliac lymphatics and cisterna chyli in

Fic. 10.—Section through a plexus of vessels filled with blood which
marks the beginning of the cisterna chyli in an embryo pig 23 mm. long.
(Specimen 23b.) Magnified 40 times. The lymphatics are filled with blood,
the blood vessels are empty. A. — aorta; C. c. — cisterna chyli; G.s.s.c.
= GI. suprarenalis s. corticalis; G. s.s.m. = GI. suprarenalis s. medullaris;
M. v. = vena mesonephritica; S. v. = segmental vein; W. b. = edge of the
Wolffian body.

the blood-filled stage. In a sagittal series of a specimen 23 mm. one
can see the entire iliac chain in a single section, from the mesone-
phritic vein to the blood-filled sac just caudal to the Wolffian body.
From this caudal swelling, from which a large group of iliac glands
develops in the pig, three sets of vessels grow out: (1) a plexus
which completely surrounds the umbilical arteries, (2) the femoral
vessels, and (3) the ileo-inguinal vessels.

Digitized by Microsoft®

40 Florence R. Sabin.

D. THE THORACIC DUCT IN THE PIG.

The thoracic duct is easy to inject after it is once fully formed,
but in the early stages exceedingly difficult. In later stages it can be
injected indirectly either from the retroperitoneal sac or from the iliac
sac. Moreover it lies in the edge of the wall of the aorta, which guides
the needle for a direct injection. In the lower thoracic region it
forms such an abundant plexus around the aorta that if the needle

Fic. 11.—Section through the 7th cervical vertebra of an embryo pig
measuring 19 mm. Magnified 45 times. The blood vessels have been injected
with India ink through the umbilical arteries. This is the stage just pre-
ceding the development of the thoracic duct. B. v. p. = plexus of blood
vessels ventral to the vertebra; E. = esophagus; J. 1. s.=jugular lymph
sac; N. s.=nervus sympatheticus; T. = trachea; V. j. i. = vena jugularis
interna.

avoids the azygos vein it is almost certain to enter lymphatics. Pensa’s
(105) series of injections in different forms, as well as fig. 7 (Heuer
48),-show that it is a constant characteristic of the lower thoracic part
of the duct to be a complete plexus around the aorta, while the upper
thoracic portion consists of one or two ducts.

The thoracic duct forms from two places: (1) a duct which grows
downward from the left jugular sac, and (2) a plexus which buds
out from the mesonephritic veins and completely surrounds the aorta.

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 41

Before the thoracic and renal segments have met, however, it is most
difficult to inject them.

The early jugular trunk I have never injected directly and only
once indirectly from the jugular sac. It will be readily seen that ink
injected into the jugular sac usually takes the line of least resistance,
the physiological path into the veins. Once in an embryo 23 mm.

Fie. 12.—Section through the point of origin of the jugular part of the
thoracic duct in an embryo pig 23 mm. long. (Specimen 23a.) Magnified
40 times. The left jugular lymph sac and the beginning thoracic duct
emptying into it were injected in the specimens; in the drawing the veins
are arbitrarily shown as injected and the lymphatics as empty. The thick-
ness of the endothelial lining of the lymphatics has been exaggerated. A. =
aorta; D. t. = ductus thoracicus; E. = esophagus; N. s. = nervous sym-
patheticus; N. v.—nervus vagus; J. l. s.=jugular lymph sac; V. j.i-=
vena jugularis interna.

long I filled the sac, and on pressing the head forward the ink shot
into the thoracic duct. The pressure obviously could not be regulated.
In 1911 I reported this specimen (23a) at the meeting of the Ameri-
can Association of Anatomists in connection with an unfinished study
of the thoracic duct. Dr. McClure requested the specimen for Mr.
Kampmeier, who was also studying the thoracic duct in the pig at

Digitized by Microsoft®

42 Florence R. Sabin.

the time, and he has published, with my permission, a valuable recon-
struction of my specimen in the Anatomical Record for 1912 (Kamp-
meier 66).

The thoracic duct needs to be considered in relation to the veins
of the region. Fig. 11 is a section through the seventh cervical verte-
bra of an embryo pig 19 mm. long, which is before the thoracic duct
begins. There is a complete vascular injection. It will be seen that
there is a vascular zone ventral to the vertebra, and from this zone
veins surround the sympathetic nerves and enter the dorso-medial
border of the internal jugular vein. The esophagus has a plexus of.
blood vessels in the submucosa, but dorsal to the esophagus is a non-
vascular zone of loose connective tissue. The jugular stem of the
thoracic duct grows into this non-vascular area dorsal to the esopha-
gus, as is shown in fig. 12 from 23a. In this tracing of the section
the veins are arbitrarily made black and the lymphatics are shown
empty, though in the section the left lymphatic sac and its ducts were
injected. The thoracic duct lies in the margin of the vascular zone.
As’ Kampmeier’s reconstruction of this specimen shows, there is a
considerable plexus of lymphatics dorsal to the esophagus near their
place of origin from the jugular sac. This plexus connects with the
left jugular sac in three places. From the injected plexus a short
duct follows the left cardinal vein and I think grows to the heart
and lungs. A longer vessel crosses to the right side and is the jugular
segment of the thoracic duct. This crossing of the duct behind the
aorta I have frequently, though not always, noted in older stages.

The right lymphatic duct curves ventralward and grows to the heart
and lungs. Its course is shown in fig. 13 for an embryo 25 mm. long.
The asymmetry of the thoracic duct is confined to the jugular portion
and corresponds with the asymmetry of the aorta. It is interesting
to note that Sala (137, Taf. 14, Fig. 16) and Pensa (104, Taf. 15,
Fig. 2) picture a symmetrical thoracic duct in the bird.

The position of the cisterna chyli has already been shown in figures
9 aud 10. The cisterna chyli and lower part of the thoracic duct
arise in common with the iliac sacs from the mesonephritic veins on
either side, as shown in fig. 9. These lymphatic buds from the two
sides meet dorsal to the aorta and grow both cerebralward and caudal-
ward along the dorsal wall of the aorta. This makes an abundant
plexus of blood-filled lymphatics along the course of the abdominal
aorta. Any sections of pig embryos of this stage will show that the

Digitized by Microsoft®

"he Origin and Development of the Lymphatic System. 43

‘round the aorta are very abundant, so that the non-vascular
s small. Just opposite the adrenal anlage, however, there is a
ion zone which is less vascular. It marks the end of the azygos
Below this level the segmental veins drain through the Wolffian
and here the lymphatics form a wider plexus, which becomes
terna chyli, fig. 10. As in the adult there is a gradual tran-

13.—Section through the upper thoracic region of an embryo pig
ing 25 mm. in which the blood vessels are injected, to show the
of the right lymphatic duct approaching the heart. Magnified 40
The injection of the blood vessels was made through the umbilical
A. = aorta; D. C. — duct of Cuvier; D. t. — ductus thoracicus; E.
hagus; L. = lympb vessels to the heart from the right lymphatic
N. v.=nervus vagus; P.—pericardium; T.—trachea; V. a=
ygos.

between the cisterna chyli and the rest of the aortic plexus.
ie part which extends caudalward and the part which extends
lward. Figure 10 shows the cisterna chyli while it is still a
and not a sac. The series shows that in many places this dorsal
of lymphatics is sending sprouts around to the ventral pre-
sac. All of the abdominal sacs, the retroperitoneal, ventral to
ta, the two lateral iliac sacs and the dorsal plexus connect with

Digitized by Microsoft®

44. Florence R. Sabin.

each other. In following the series cerebralward from the level of fig. 9
it is clear that the blood-filled lymphatics not only arch across the
midline dorsal to the aorta, but they follow the border of the Wolffian
bodies lateral to the aorta. Above the adrenal bodies the Wolffian
bodies approach very close to the aorta, and some of the lymphatic
buds curve around the ventral surface of the aorta. Thus in a speci-
men 23 mm. long there is a plexus of blood-filled lymphatics surround-
ing the aorta and extending into the lower thorax as far as the cerebral
pole of the Wolffian bodies. At the cerebral end of the Wolffian bodies
there are large veins which curve ventral to the aorta and connect
the veins of the capsule of the Wolffian bodies. Some of the lymphatics
are near these median veins.

The finding of the renal lymphatics in the blood-packed stage makes
it quite certain that the thoracic duct arises in two places: First,
the jugular stem, which can be injected from the left jugular sac and
develops long after the jugular sacs are entirely empty of blood; and,
secondly, the renal plexus, which surrounds the aorta and forms a defi-
nite cisterna chyli.

These two observations, namely, (1) the presence of blood in one
part and its absence in the other, and (2) the fact that injections
prove a connection of the jugular stem with the jugular sac, and of
the abdominal plexus with the mesonephritic veins, make it quite cer-
tain that the two portions of the thoracic duct are distinct from each
other in embryo pigs measuring 23 mm., and hence that the thoracic
duct begins in two places. Practically the entire question at issue
between those of us who think that the lymphatics grow by their own
endothelial wall and those who do not centers around the question
of how the thoracic duct develops, or, to put it more specifically, how
these two lymphatic anlagen become connected. The discussion centers
around my specimen 23a, which Kampmeier studied and which he
uses as his most conclusive evidence of the theory of the growth of
lymphatics by the addition of tissue spaces. On the other hand, it
is my theory that the thoracic duct grows from these two anlagen from
the veins by the same method by which other lymphatic capillaries
can be seen to grow in a living specimen, namely, by the sprouting of
their endothelial wall. It is true that I have not yet sufficiently
mastered the difficulties of injecting the renal lymphatics to demon-
strate the progression of the thoracic duct and indeed it may not
prove possible to get as conclusive proof of the origin of the deep lym-

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 45

-phatic sacs as of the more superficial ones; yet the inference that
they arise in the same manner is in harmony with our modern knowl-
edge of morphology. I wish to postpone a discussion of Kampmeier’s
evidence until I have described the thoracic duct in a human embryo,
and discussed the development of the theories opposed to my own—of
which Kampmeier’s work is a part (Sec. VIII).

2. Primary Lympuatic System In Human Empryos.

On this description of the lymphatic system in one mammalian form
I shall base an account of what is known of the primary system in other
forms. All of the primary lymphatic sacs have been found in the
human embryo. The jugular sac is the earliest to appear. It has
been found as a few blood-packed buds in two embryos measuring 8
and 9mm. (Mall collection, No. 397 and No. 163, Sabin 134.) They
are on the lateral surface of the anterior cardinal vein near the duct of
Cuvier and were described as veins (Sabin 133, 134), since at that
time the blood-filled buds were thought to be blood capillaries. The
blood-packed buds extend along the lateral surface of the anterior car-
dinal vein and very early develop sprouts that project dorsalward.
These are plain in an embryo measuring 11mm. (Mall collection, No.
353, figs. 7 and 8, Sabin 133, and figs. 490 and 491, Sabin 134.) The
figures of this specimen, together with the reconstruction of Lewis
(Harvard collection No. 1000, measuring 16 mm.; No. 189, measuring
11.5 mm.; and No. 1322, measuring 16 mm., figs. 1-3, Lewis 78),
show that the jugular lymphatics in the human embryo not only
bud from the anterior cardinal veins, but from the posterior cardinal
vein, and the common stem or plexus which forms the origin of the
primitive ulnar and thoraco-epigastric veins. A large extension of
the jugular sac along the primitive ulnar vein is characteristic of
human embryos (fig. 12, Sabin 133, or fig. 493, Sabin 134, and figs.
2 and 8 after Lewis 78).

In a human embryo 10.5 mm. long I have found a small sac only
partly filled with blood (figs. 3-6, Sabin 133, and figs. 488, 489, Sabin
134). This specimen has also a beginning thoracic duct. These two
facts show that there is considerable variation in the rate of growth
of the lymphatic system of different specimens.

The study of human embryos has also brought out the fact that
the openings of lymphatics into the veins can only be made out when

Digitized by Microsoft®

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 47

In tracing the lymphatics cerebralward, from the level of the hilus
of the kidney, there is a small lymphatic capillary plexus in the place
of the cisterna chyli and the two lateral sacs become a plexus of small
ducts which can be traced the entire length of the permanent kidneys
and then along the dorso-lateral surface of the aorta, just ventral to the
azygos veins. This small plexus of ducts is nearly empty on the left
side, but full of blood on the right side. The two plexuses can be
followed almost to the level of the bifurcation of the trachea. Some
blood in the plexus on the left side just before the vessels end makes .
it possible that the renal part of the thoracic duct does not yet connect
with the jugular part of the duct.

This particular specimen is interesting in regard to certain zones
of dilated tissue spaces which are conspicuous along the aorta. There
are some of these large spaces near the bifurcation of the aorta, in
the root of the mesentery near the retroperitoneal sac, and along the
ventral surface of the aorta, especially opposite the mesentery of the
stomach and opposite the bifurcation of the trachea. I have studied
them with care to see if they have any definite relation to the develop-
ing lymphatics and am sure that they have not. Sometimes they are
near the lymphatics and sometimes not. Those in the thorax are not
adjacent to the lymphatic plexus. They can be distinguished from
the endothelial-lined lymphatics, but if one were convinced that lym-
phatics came from tissue spaces one might imagine transition ‘pictures
between the tissue spaces and the ducts. This is particularly true when
the endothelium of a lymphatic vessel sags away from the surrounding
tissue, a picture familiar to histologists in sections showing lymphatics
in adult tissues. The further discussion of the thoracic duct is in
part VIII.

All of the primary lymphatic system is present in a human embryo
measuring 30 mm. (Mall collection No. 86, fig. 12, Sabin 133, copied as
fig. 493, Stricker 144). This stage represents the maximum size of the
jugular sac. The dorsal arch is large and its line of separation from
the jugular part is indicated by the perforation for the cervical nerves
and blood vessels. The very small mesenteric sac and larger iliac sacs
are distinct as well as the complete thoracic duct.

3. Primars Lympnatic System IN Rassir Emsryos.

In 1906 F. T. Lewis (76) published a valuable paper on the lym-
phatic system in rabbit embryos. Besides excellent reconstruction of

Digitized by Microsoft®

48 Florence R. Sabin.

the primary lymphatic system in rabbits, the cat and the pig, this paper
has three important points: First, the discovery that the early lym-
phatics are filled with blood, which has proved to be of such value,
now that its meaning is understood. Second, the discovery of the
retroperitoneal sac which he described as lying adjacent to the mesen-
leric veins. He also showed the iliac sacs and cisterna chyli; all of
which we now know come from the renal veins (Sabin 136). Third,
he noted that lymphatic ducts when they are reconstructed from sec-
_tions appear as a row of beads. Since these rows of beads occurred
along the veins in his reconstructions and since he could not find the
connections of the early sacs with the veins he was led to suggest that
lymphatics might arise from multiple anlagen split off from the veins
along their course. The apparently isolated endothelial-lined vessels
found in serial sections we call Meyer-Lewis anlagen. They are dis-
cussed in section VIII.

The jugular sac in rabbits has an early extension along the primitive
ulnar vein like that of the human embryo. The dorsal portion of the
sac, for the posterior triangle of the neck, arises from a large stalk
from the primary jugular portion between the level of the 4th and 5th
cervical nerves.

4, Primary LyMpPHATIC SYSTEM IN THE Cart.

The primary lymphatic system in the cat will be described from the
extensive work of Huntington and McClure (51-58 and 91-94), one
reconstruction of Lewis (76) and a few injections of my own. The
work of Huntington and McClure has been mainly on the development
of the jugular lymph sacs and the thoracic duct in the cat and their
observations are for the most part brought together in two extensive
papers (54 and 58). Their work is difficult to analyze and cannot
be done entirely accurately without seeing their sections. They have
relied on the appearances of sections and the appearances of wax models
from them to differentiate lymphatics, and since these are not ade-
quate criteria it is impossible to be sure that all the structures shown
as lymphatics are lymphatics; in fact, one can be certain that some of
them are not. Their theories in regard to the origin and development
of the lymphatic system are discussed in sec. VIII. In the article
(54) on the jugular lymph sac of the cat figs. 8 to 21 may be disre-
garded as evidence, since they are diagrams. Figure 22, which they

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 49

: group under “ Karly Venous Stages,” is, I think, a reconstruction of
the cardinal veins before the lymphatics begin (and is therefore to
be compared with Evans’ (34) fig. 3, and with my fig. 4 in this paper).
In their fig. 22 the fourth segmental vessel is shown and described
(page 226) as “ formed by the confluence of two extended para-neural
channels.” This is intended to indicate the beginning of the lym-
phatics, but since their fig. 48 from an embryo 10 mm. long shows
undoubted lymphatics in the blood-packed stage, I think it is prac-
tically certain that the specimen measuring 5 mm: does not show lym-
phatics. It is not necessary to prove that veins do not open into the
tissue spaces around nerves.

The description of the stages between 5 and 10 mm. is bound up
with the discussion of “fenestration.” In reconstructing the veins
Huntington and McClure have noted the pattern of the developing
veins, which is especially well shown in fig. 4, in connection with the
pericardial veins. In this specimen the pericardial veins drain into a
loop which connects the anterior cardinal vein with the duct of Cuvier.
The formation of this loop is a part of the process by which the peri-
cardial veins shift from the duct of Cuvier to the anterior cardinal
vein. Numerous small loops are seen along the anterior cardinal
veins in figs. 4 and 5. They are characteristic of the veins. They
suggested to Huntington and McClure a vague conception of a rela-
tion to the lymphatics which they expressed as follows (p. 204 in 54):
“Tn a subsequent stage confluence of these ‘fenestre’ results in the
more or less complete separation of the primary vein into two parallel
secondary channels, which appears to be a uniform principle in the
development of parallel venous trunks. Or, as in the case of the jugu-
lar lymph sac, further extension of the same process may result in
separating from the main venous channels elements which unite to
form a closed sac entirely distinct from the vein from which it arose.
The term ‘ fenestration ’ means, therefore, in the sense in which it is em-
ployed in this paper, one of the last stages in the definite crystalli-
zation of the venous system out of an indefinite antecedent plexiform
condition, and the determination of an important element in lymphatic
organization, closely associated with the embryonic venous system.”
The lymphatics which they thought formed by this process they termed
“veno-lymphatics.” Both terms “ fenestration ” and “ veno-lymphat-
ies” should drop out; there are no veno-lymphatics, for the lymphatics
bud directly from the veins as lymphatics, and fenestration is a term

Digitized by Microsoft®

50 Florence R. Sabin.

that describes models better than the actual process of growth by sprout-
ing by which the veins are formed.

Some of the figures of the models from embryos measuring 7 mm.,
for example fig. 33, represent lymphatics in the early plexus stage;
I do not, however, find any mention of the presence of blood. Figure
46 shows the beginning of the enlarging of the plexus into sacs. From
their later stages figs. 46 to 65, but much better from Lewis, fig. 6
(76), can be made out the especial characteristics of the jugular
sacs in the cat. They are (1) that the jugular sac in the cat arises
from the posterior cardinal vein as well as from the anterior cardinal
vein. This point is well shown in Huntington and McClure figures.
(2) The cerebral end of the ventral or jugular portion of the sac is
very large. This is the part that drains the larynx. (3) The internal
jugular trunk is small. (4) The dorsal apex of the sac which lies
in the posterior triangle of the neck is large. (See Lewis’ fig. 6.) It
is connected with the cerebral end by a slender vessel. In injections
of the sac in embryos between 30 and 60 mm. long it looks very much
like the sac in a bovine embryo as shown by Polinski (107) in his fig.
7. It lies nearer the arm than the corresponding portion of the sac
in the pig. (5) There is a large extension though possibly a dis-
tinct sac (Lewis, fig. 6). along the primitive ulnar and thoraco-epi-
gastric veins. This sac is very conspicuous in older stages, extending
into the axilla and receiving large trunks from the thoracic wall.
This makes the jugular lymphatic sac in the cat more like the human
than the pig (sec. V).

The segmental divisions of the sac as indicated by colors in the
Huntington and McClure figures are arbitrary and not as valuable
as the divisions I have just given, which correspond to the function
of the different glands which develop from the sac.

Lewis’ figure also shows a very interesting view of the retroperito-
neal and iliac sacs as a continuous plexus, which again emphasizes
the fact that the renal sacs anastomose with each other.

5. Primary Lympyatic System 1n Birps, Posrertor Lympu
Heart anp Jucutar Lympu Sac.

The history of our knowledge of the lymphatic system in birds is
most interesting. The early discoveries of the posterior lymph hearts
by Panizza, A. F. J. Mayer, and Stannius, as well as the work of

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 51

Budge on the relation of the heart to the allantoic vessels, have already
been given. From that time on there was no advance until the work
of Sala (137), who studied the development of the posterior lymph
hearts and thoracic duct. He described the posterior lymph hearts
as arising in relation to the lateral branches of the first five coccygeal
veins, during the seventh day of incubation. We now know that
the posterior lymph hearts bud from the veins during the 5th day
(E. R. and E. L, Clark, 29), so that Sala was describing the trans-
formation of the plexus into the sac. He noted the connections with
the veins but described the sac as arising from excavations in the
mesenchyme. His work, published a short time before my paper on
the origin of the lymphatic system from the veins (Sabin 129), was an
advance over the preceding work, for it placed the first lymphatics
near the veins instead of in the periphery. Sala figured the thoracic
duct in the bird as a symmetrical structure (Taf. 14, Fig. 16), and
this is also shown in Pensa (104, Taf. 15, Fig. 2). Sala described
the early stages of the thoracic duct as being solid cords of cells. This
observation, which has been confirmed by Mierzewski (96), makes it
necessary to restudy the thoracic duct in the chick, which is now pos-
sible through the great improvement in our methods of injection.

The most recent work on the lymphatic system in birds has been
done by Mierzewski (96), Jolly (59), Miller (97) and E. R. and
E. TL. Clark (27-29).

I have already brought out the fundamental importance of the
Clarks’ work, showing that the lymphatics bud off from the veins in
a non-vascular zone, and show a continuous growth to the periphery.
Mierzewski had shown that the early lymphatics grow along the lateral
line of the embryo, as shown in his fig. 3. It was these lymphatics
which grow to connect the posterior lymph heart with the jugular
lymph sac that E. L. Clark observed in their blood-filled stage and
which led her to watch the process of budding in the living embryo.

The process may best be described in a quotation from p. 254 (29):

The first evidence of lymphatics in the tail region of living chicks is the
appearance of a number of separate knohs, filled with stagnant blood, a
little darker in color than the circulating blood, just lateral to several of the
most anterior of the dorsal intersegmental coccygeal veins. The connections
with the veins cannot be seen, since the knobs lie between them and the ob-
server, but ink injected into the knobs can be seen to pass directly into the

main intersegmental veins. Between the separate knobs no anastomoses
can be seen, nor can any be discovered by injection.

4

Digitized by Microsoft®

52 Florence R. Sabin.

Soon after these knobs appear (in about fifty-five minutes), similar ones
develop about them which have fine connections with them, thus forming
a small cluster. The new ones are located partly on either side of the first
and partly superficial to them. Their injection now shows discreet tiny
clusters, somewhat like bunches of grapes, connected, as were the earliest
knobs, with the same intersegmental veins. These clusters are still separate
from one another.

Fic. 14.—After Miller. Diagram from an reconstruction of the veins and
nerves of the cervical and upper thoracic regions of a chick embryo 13.5
min. long, after 5 days and 10 hours of incubation; right side. 1 Pre-
cardinal vein. 2. Post-cardinal vein. 3. Duct of Cuvier. 4. Intersegmental
peoreet) arse bees piel of oe islands and veno-lymphatic

; i Tvical) nerves. a. chi i
Amer. Jour of Anat., 1912, Vol. 12, fig. 12) PexUS. After Miller,

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 53

FLCCceeCG

=| Emit
= 4
S— TD

Fic. 15.—After E. L. Clark. Lateral view of the jugular lymphatic plexus
of the right side of a chick embryo 14 mm. long after an incubation of 5
days and 20 hours. The measurement was the greatest length and was
taken after fixation. With a very fine canula (the lumen of the point being
about 15 w in diameter) diluted India ink was injected under a binocular
microscope into one of the superficial lymphatic capillaries in the skin
between the fore and hindlegs. From here the ink filled the deep jugular
lymphatic plexus, and from there a few granules entered the vein through
five connections (C). The drawing was made. with the help of a camera
lucida with a Zeiss binocular microscope (oculars No. 4, objective A,).
Magnified 67 times. A. = vessel which connected the superficial lymphatic
canillaries with tha inonlar lwmnh nlovue and which was iniected: A.C. =

Digitized by Microsoft®

54 . Florence R. Sabin.

There is a rapid extension of these blood-filled structures, and soon, in
about an hour and a half after their first appearance, connections between
neighboring clusters may be seen. Injection at this stage with India ink
shows an anastomizing plexus, connected, as before, with the intersegmental
veins. Injection with silver nitrate shows distinct endothelial markings
in the walls of the plexus.

During this plexus formation there is a steady extension toward the sur-
face, and by the time anastomoses have formed between neighboring clusters
sprouts have grown to the surface and started to extend in the region
superficial to the plexus and also ventralwards. It now becomes possible
to study with more minuteness the changes which are going on, since these
sprouts are quite superficial and are developing in a plane parallel with
the surface. Observation and successive records of these sprouts in the
living chick reveal a rapid extension ventrally and also anteriorly, ac-
companied by a plexus formation. Two or three sprouts are seen to lead and
soon numerous connections develop between them. Various portions of the
irregular plexus thus formed enlarge and become more densely packed with
blood which continues to back up from the vein. Then new sprouts grow
out in advance and the same process of extension accompanied by plexus
formation is repeated. If a single sprout is selected and frequent careful
drawings are made the changes are seen to be rapid and striking. The
sprout becomes wider and longer. Branches appear, and they in turn in-
crease in width and length. From a branch a connection forms with the
original sprout, thus forming a loop. New branches and connections are
formed, making a plexus. Branches from neighboring loops or plexuses
meet one another and anastomose. The several parts of the plexus are
quite irregular in size. Most of the lymphatic vessels are several times as
wide as.a blood capillary, while some of the connections are as small as,
or even smaller, than a blood capillary. Throughout the blood in these new-
forming lymphatics is markedly darker in color than the circulating blood.

This work thus corrects the idea that lymphatics are transformed
blood capillaries. They bud out from the veins as lymphatics and
gradually invade the body as a new type of vessels. ;

Miller’s work (97) on the anterior lymph sac is chiefly valuable
in giving an excellent chance to contrast the method of reconstruction
with the method of injection in an area where an injection can be
obtained. Figure 14 is copied from Miller’s fig. 6, and fig. 15 is from
an injection by E. L. Clark’s (27, fig. 2) of the plexus in the same
stage. The points of the apparently isolated islands of lymphatics
and the separation of the sac from the veins show by Miller are errors
which are cleared up by E. L. Clark’s work. The jugular lymphatic
plexus is connected with the vein in a number of places and is a con-
tinuous endothelial-lined plexus of lymphatics,

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 55

The jugular lymph plexus develops into a small sac (‘Miller’s fig. 6)
which becomes a single jugular gland. This small, fusiform gland
has been studied by a number of observers and has been worked out
in all its relations by Jolly (59) in his interesting paper on the lymph
glands in buds.

Miller (97, page 489) quotes Mierzewski as having seen rhythmical
contractions of the jugular lymph sacs. A reference to Mierzewski’s
paper, however, will show that he consistently speaks of posterior
lymph heart and jugular lymph plexus and his reference to pulsation
(page 479) is a confirmation of Sala’s observations in the posterior
lymph hearts. Miller notes the absence of muscle and the point should
be emphasized, because the jugular lymph sac in birds becomes not
a lymph heart, but a lymph gland.

The iliac lymphatics in birds are very well brought out by Jolly
(59) in his article on lymph glands. He does not consider their
origin, but his fig. 15 shows their distribution extremely well, and his
fig. 11, pl. XI, gives the exact histological picture of a lymph sac with
connective tissue bridges for a bird on the 18th day of incubation.
This is the method by which the primary sacs become lymph glands,
as is shown in section VI.

6. Primary LympHatic SystEM IN AMPHIBIA, ANTERIOR AND
Posterior LympH Hearts.

It is not possible at this time to give a complete account of the
development of the lymphatic system of the amphibia and reptilia,
but enough evidence is at hand to show that the lymphatics are derived
from the veins. The remarkable subcutaneous lymph sacs in the frog
have always attracted interest and stimulated a curiosity to unravel
their development as was brought out by Langer (75) in 1868. Some
recent experiments of Abel (1) which combine anatomical and pharma-
ecological methods add to this interest.

Meltzer (95) had published experiments to show that a smaller dose
of a drug produced convulsions when injected into a cardiectomized
frog than into a normal one. He concluded that the drug traveled
through the tissue spaces.

Abel found that a dye, acid fuchsin, when injected into the lymph
sacs of a normal frog was distributed widely over the body by the
blood vessels. In a cardiectomized frog a small dose injected into the

Digitized by Microsoft®

56 Florence R. Sabin.

muscle (where there are no lymphatics, only tissue spaces) traveled
exceedingly slowly; on the other hand, a small amount of the fluid
injected into a lymph sac traveled rapidly, but in a zone limited ex-
clusively to the lymph sacs, according to their anatomical conections.
When the dye reached the lymph heart in the cardiectomized frog
it was pumped, not through the heart to the systemic vessels, but
through the vertebral veins to the nervous system. Thus is explained
the marked effect of the small dose. If the lymph hearts were killed
the convulsions did not take place.

It has been shown that the anterior lymph hearts of the amphibia
bud off from the vetebral veins (Hoyer 49), or in a more primitive
position from branches of the earlier segmental veins which are going
to form the vertebrals (Knower 74). They develop earlier than the
posterior hearts and have been found in R. palustris, R. sylvatica and
R. virescens measuring about 6.5 mm., and in R. temporaria and
Bufo vulgaris measuring about 4.5 mm. Hoyer noted the presence
of blood in the early stages of both anterior and posterior hearts.
Knower brings out the fact in connection with the anterior hearts
that they lie on the myotomes from which they derive the striated
muscle of their wall. This is, I think, a very important point. The
particular myotome Knower has shown may vary.

The pulsations of the anterior lymph hearts can be readily seen in
the dorsal surface of the tadpole, just behind the pronephros, and thus
they can be injected.

From the anterior lymph heart develop two symmetrical vessels.
one of which runs forward toward the head, the other backward toward
the tail. The posterior duct as described by Hoyer divides into two
branches, one running to the tip of the tail dorsal to the myotomes,
the other passing back along the ventral margins of the myotomes
to form the ventral caudal trunk. It is the branches of these two
caudal trunks which have been studied so much in the living specimen.
There is also a lateral branch which runs caudalward from the an-
terior lymph hearts on the lateral surface of the myotomes half way
between the dorsal and ventral branches. This lateral trunk subse-
quently connects with the posterior lymph heart.

Wieliky (153), Jossifov (62-65), and Favaro (36) thought that the
posterior lymph heart arose from the dilation of the caudal lymph
trunks which grow from the anterior lymph hearts, -and Jourdain
(61) describes them as being formed by a rapid destruction of con-

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 57

_ nective tissue. Knower and Hoyer, however, have found that they
bud off from the posterior vertebral veins considerably later than the
anterior lymph hearts, and they have been well worked out by
Baranski (9). Barafiski shows that they arise by several buds from
the endothelium of the posterior vertebral vein and its branches, His
fig. 1 shows particularly well the heaping up of the endothelium so
that it looks like a solid mass at the point of origin. Practically
nothing is known of the origin of the deep lymphatics in amphibia.

It has thus been shown that the lymph hearts of amphibia arise as
buds or sacs from the endothelial wall of segmental veins. They lie
on the myotomes, from which they derive the striated muscle of their
wall.

The subcutaneous lymph sacs of the anura are secondary structures,
This was found out by Ranvier (116) in 1896, and also by Knower
and Hoyer. They are derived from ducts which grow from the
lymph hearts. They have been comparatively little studied. Hoyer
(49) found that the large sac on the ventral surface of the head (sace.
submaxillaris) developed from a branch of a duct from the anterior
lymph heart, and that the lateral sacs are derived from the lateral
lymph trunks.

This point has been just shown in connection with the sacs of the
extremities by Frl. Goldfinger (38), who has injected the primary
ducts and then the increasingly denser lymphatic plexus up to the final
lymph sacs. This process of cavernization could be studied here by
silver nitrate injections.

Since the above was written Hoyer (49a) has published an article
in which he gives a general review of the origin of the lymphatic
system in vertebrates, and Hoyer and Udziela have given the first
comprehensive description of the lymphatic system in a urodele (49b).
In a salamander larva the lymphatic system opens into the veins
first through symmetrical axillary lymph sacs or sinuses, and secondly
through a series of subcutaneous lymph hearts which pump the lymph
into the vena lateralis (Meyer 88a and Marcus 87a).

There are six longitudinal lymph trunks, four superficial and two
deep (Hoyer 49b, Taf. XII, Figs. 1,2 and 5). (1) The median, dor-
sal, longitudinal lymph trunk, which extends from the point of the tail
to the middle of the head; (2) a medial, ventral caudal trunk, which
branches around the cloaca and opens into the inguinal sinus as is
shown in fig. 2 (49b) ; (3 and 4) symmetrical lateral trunks, which

Digitized by Microsoft®

58 Florence R. Sabin.

lie external to the myotomes and open in part into the segmental
lymph hearts and in part into the axillary sinus, and (5 and 6) the
deep subvertebral lymph trunks. The subvertebral lymph trunks are
paired vessels with many anastomoses which extend from the tip of
the tail to a point opposite the stomach, where they unite to form a
cisterna chyli. From the cisterna chyli four vessels run forward to
the axillary plexus, the paired thoracic duct and the two paravertebral
lymph trunks.

The axillary lymph sacs open by three or four branches into the
cardinal veins and receive the following vessels: (1) The lateral
superficial lymph trunks; (2) the paravertebral vessels; (3) the thor-
acic duct; (4) vessels of the cranial part of the stomach and esopha-
gus; (5) a vessel which runs in the groove between the pericardial
and peritoneal sacs; (6) a vessel which comes from the base of the
skull; (7) superficial vessels from the side of the head, and (8) the
lymph vessels of the forelegs.

Besides the axillary sac there are two other sacs or sinuses, one of
which lies dorsal to the aorta at the base of the heart, namely, the sinus
lymphaticus cordis (Hoyer 49b, fig. 5), and is in the course of the jug-
ular lymph trunks and the other of which makes the paired sinus of
the inguinal region.

The segmental lymph hearts receive three groups of vessels: (1)
branches from the subvertebral lymph vessels (which agrees with
the findings of Marcus 87a, in Gymnophionem; (2) dorsal and ven-
tral segmental branches, and (3) the longitudinal lymph trunk.
They open into the vena lateralis. The segmental lymph hearts and
the sinus lymphaticus cordis have striated muscle in their walls. It
is clear that a knowledge of the origin of this system would be of
great value, as Hoyer says: “Erst wenn die beriihrten noch zweifel-
haften Punkte durch weitere Untersuchungen vervollstandigt sein
werden, wird eine gewisse Grundlage zu vergleichenden Betrachtungen
des Lymphgefassystems der niederen Wirbeltiere geschaffen sein”
(49b, p. 555).

Stromsen (145) and Huntington (57) have worked on the de-
velopment of the lymphatic system in reptiles. Stromsen shows that
the posterior lymph hearts develop in relation to the coccygeal veins.
He describes the process as a combination of veins and dilated spaces
similar to Sala’s decription for the corresponding hearts in birds.
This method of formation has now been disproved by the observation

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 59
of the process in the living chick. He notes that the striated muscle
of the heart comes from the myotomes. Huntington (57) states that
in the reptiles there are jugular lymph sacs like those of mammals.
The lymph vessels of reptiles he describes as arising from tissue spaces.

VY. PERIPHERAL LYMPHATIC SYSTEM.
1. IN tre Pia.
A. FROM THE JUGULAR LYMPH SAcs.

J shall again base the description of the peripheral lymphatics on
the vessels in the pig, since more of the main trunks have been worked
out in that form than in any other. The description is taken from
the work of A. H. Clark (2+), of Heuer (43) and myself.

The development of the jugular sac in the pig has already been given
in section IV. A study of the distribution of the lymphatic vessels
from the jugular sac must be based on the form of the jugular sac
as seen in figs. 7 and 8. In fig. 7 there is a primary sac between the
internal and external jugular veins and a complete arch of lymphatic
capillaries extending into the dorsal lymphatic area. This entire
plexus becomes transformed into the sac, so that, as shown in fig. 8,
the sac consists of a stem along the internal jugular vein and a com-
plete dorsal arch. Through following the fate of this sac A. H. Clark
divides it into three parts, the sac stalk; the anterior curvature along
the internal jugular vein; and the apex, in the posterior triangle of the
neck.

The sac stalk becomes the deep jugular lymphatic trunk. It is
usually one large vessel, but may give off lymphatics which run directly
to the pharynx. The anterior curvature becomes a lymph gland and
it sends vessels to the pharynx and Eustachian tube and to the nose.

The apex of the sac, as is seen in fig. 8, becomes the largest part
of the jugular sac. This part lies in the posterior triangle of the
neck and develops in the pig into one gland, the largest in the
neck (Clark 24). This corresponds to the group of glands in the pos-
terior triangle of the neck particularly well shown in Bartels (10)
fig. 30 from new born child.

The fact that the jugular lymph sac in the pig develops into two
lymph glands and the deep jugular lymphatic trunk is well brought

Digitized by Microsoft®

60 Florence R. Sabin.

out in fig. 16 from a pig 5.5 cm. long. The sac is shown in the depth
and the swollen anterior curvature (deep jugular gland) and apex
(gland of the posterior triangle) stand out through the relatively small
size of the connecting vessel.

Fic. 16.—Injection of the jugular lymph sac, the superficial cervical
plexus, and the superficial lymphatic vessels in the neck of a fetal pig
5.5 cm. long. After A. H. Clark (24). The specimen was fixed in forma-
lin 10 per cent formol and made transparent by the Spalteholz method.
Magnified 7 times. The figure is to be compared with fig. 5 in the Amer.
Journ. of Anat., 1904, Vol. 3, p. 118, which shows a complete injection of the
superficial lymphatics of the same stage. A. s. = apex of the lymph sac
which forms the lymph gland of the posterior triangle of the neck; A. c. =
anterior curvature of the lymph sac, which forms the deep jugulo-pharyngeal
lymph gland; C. p. = superficial cervical lymphatic plexus; S. g. = lympho-
glandular submaxillaris; S. s.= stalk of the jugular lymph sac.

From the apex of the sac superficial lymphatic vessels grow both
from the dorsal and from the ventral border. Those from the dorsal
and caudal border make a large plexus of superficial lymphatics which,

Digitized by Microsoft®

The Oriyin and Development of the Lymphatic System. G61

as shown in fig. 8, grow (1) over the scapular region and (2) over
the occipital region. The lymphatics from the ventral border of the
apex are most interesting and complex. They grow ventralward as
shown in figs. 8, 16 and 17 and form a very extensive plexus along the
course of the external jugular vein, lateral to the sternocleidomastoid
muscle. This is the superficial cervical plexus and it becomes a large
group of superficial lymphatic glands in the adult pig (Clark 24).

The cervical plexus not only connects with the apex of the sac, but,
as is shown faintly in fig. 16, its caudal end has a group of vessels
that connects with vessels that follow the external jugular vein. Ina
mesial sagittal section of injected pigs 5 to 6 cm. long it is plain that
a very abundant plexus of Imyphatics from the root of the sae stalk
completely surrounds the external jugular vein, and this external
jugular plexus has numerous connections with the cervical plexus at
its caudal end. The plexus around the external jugular vein empties
mainly into the sac stalk, but a single vessel may join the external
jugular vein. It must be said in general that the cervical plexus
develops from the jugular sac in two places: (1) from the apex of
the sac over the sternocleidomastoid muscle, and (2) from the sac
stalk along the external jugular vein.

From the cervical plexus there are seven sets of lymphatic capillaries
(fig. 16). From the cerebral end of the cervical plexus vessels grow
(1) behind the ear, and into a new facial plexus from which vessels
grow (2) between the eye and the ear, and (3) over the face. From
the ventral border of the cervical plexus there is (4) a set of sub-
maxillary lymphatics which supplies the lower jaw and tongue and
(5) superficial lymphatics for the skin of the neck. From the caudal
end of the cervical plexus grow (6) the superficial lymphatics of the
arm, and (7) the superficial lymphatics of the thorax.

Thus to sum up the superficial lymphatics for the jugular sacs: from
the sac stalk and anterior curvature arise the pharyngeal, Eustachian
and nasal vessels, from the apex of the sac arise the suprascapular
and occipital lymphatics, while from the cervical plexus arise the pos-
terior auricular, temporal, facial, submaxillary, anterior cervical and
the superficial lymphatics of the arm and thorax. These groups of
lymphatics begin as distinct plexuses of vessels which arise from the
sac or from the secondary cervical plexus which in turn comes from
the sac, In fig. 8 is shown particularly well how distinct the groups
may be at the start. Subsequently all of the groups anastomose and

Phia da aw Smmnnland os ntact

Digitized by Microsoft®

62 Florence R. Sabin.

In fig. 8 it will be seen, for example, that the large vessels of the
suprascapular region drain into the apex of the sac, but some of them
have small connections with the cervical plexus. In fig. 16 these
connections are so much more abundant that they exceed in number

Fig. 17.—Injection of the jugular lymph sac and the cervical lymphatic
plexus of a fetal pig 7.5 cm. long to show the relation of the developing
lymph glands to the jugular sac. After A. H. Clark (24). Magnified |
times. Fixed in 10 per cent formol and cleared by the Spalteholz method
A. s. = apex of the sac, which is now a lymph gland in the posterior triangl«
of the neck; the anterior curvature of the sac, which is now a deep jugulai
lymph gland, lies behind the sternocleidomastoid muscle. C. p. = super
ficial cervical lymphatic plexus, which will become a group of lymph glands
At the cerebral end of the cervical plexus a developing facial lymph gland
S. g. = lympho-glandula submaxillaris.

those that pass to the sac itself. This point of the anastomoses o

the dfficult groups of vessels is well shown in fig. 5 (Sabin 130), wher.
practically all of the superficial lymphatics in the embryo pig 5.5 em

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 63

mg have been injected from one puncture. This is a complete injec-
ion, which it is clear that fig..16 is not. In fig. 5 (Sabin 130) for
xample the lymphatics between the eye and ear are terminal ducts,

Good examples of the anastomoses of different groups of lymphatics
re shown in fig. 16 in the vessel which connects the submaxillary
nd the facial plexus; moreover the thoracic, cervical and brachial
lexuses are continuous.

In the embryo pig from 5 to 5.5 cm. long there are no valves in the
ucts. It is also the stage of the simple primary plexus of lymphatics.
n fig. 16 a secondary plexus of finer lymphatics is growing more
uperficially and by the time the embryo is from 6 to 7 cm. long
here is a deep plexus with valves and a finer-meshed superficial plexus.

The relation of the jugular lymph sacs to lymph glands is shown in
ig. 17. Here it is clear that the sac’stalk and anterior curvature lie
eneath the sternocleidomastoid muscle and that the apex of the
ac lies in the posterior triangle of the neck. That the cervical plexus
s now a lymph gland is clear, and there is a facial and submaxillary
Jand also. The cervical plexus becomes a group of glands.

The account of the peripheral lymphatics from the jugular sacs is.
\ot complete without mention of the lymphatics of the heart and lungs.
(hese lymphatics have not yet been worked out, but it can be said that
here are two sets of lymphatics for the lungs, the deep and the super-
icial or pleural. The deep lymphatics develop from the jugular sacs;
hey follow the trachea and are present in a pig measuring 23 mm.
nd a human embryo measuring 20 mm. I have not injected them and
vhen reconstructed they show the Mayer-Lewis anlagen; that is, in
econstruction they split into a chain of beads. The pleural lymphatics
- have injected from the retroperitoneal sac, through the diaphragm
o the caudal surface of the lung. Injections of the thoracic duct as,
or example, fig. 7 (Nuck 100), also shows vessels from the thoracic duct
o the lung. Therefore the lung seems to have a double supply of lym-
hatics, part from the jugular sac and part from the renal sacs by
vay of the diaphragm and by way of the thoracic duct.

B. THE PERIPHERAL LYMPHATICS FROM THE RENAL SACS.

The retroperitoneal sacs give rise to the lymphatics of the abdominal
iscera except the kidneys. The sac as it lies in the root of the mesen-
ery has been figured by Heuer (43). It spreads over the ventral

Digitized by Microsoft®

64 Florence R. Sabin.

surface of the Wolffian bodies and supplies their capsule as well a
the reproductive glands. From the sac vessels grow to the stomac]
and the entire intestine, including the rectum. The lymphatics o
the diaphragm can be injected from the retroperitoneal sac. The de
velopment of the lymphatics of the diaphragm, liver, capsule of th
spleen, adrenal and pancreas has not yet been worked out.

Heuer (43) has followed the lymphatics of the intestine from th
retroperitoneal sac to the ultimate lacteals. While the embryo 1
growing from 3 to 4 cm. long a very abundant plexus of capillarie:
spreads out in the mesentery. The lymphatics reach the wall of the
intestine by the time the embryo is 4 cm. long. They first enter thi
submucosa and form then a primary plexus. Heuer’s figures 9 anc
10 show that the early vessels in the submucosa have a segmenta
arrangement, which is, however, nearly lost as the complete plexu:
forms. The plexus of the mucosa, and of the serosa are both second.
ary. The lacteals develop from the mucosal plexus and are present
in a pig measuring 9 cm. The mesenteric vessels and the submucosa.
plexus develop valves.

The iliac lymphatics and the thoracic duct: The lymphatics whict
arise from the mesonephritic veins on the two sides form a very
simple pattern. Starting from the veins at the hilus of the twc
Wolffian bodies they grow caudalward along the edge of the Wolffiar
bodies to make the iliac sacs, and cerebralward along the dorsal sur-
face of the aorta to form the cisterna chyli. The lymphatics dorsa:
to the aorta grow in two directions: (1) caudalward to form a chain
of preevertebral lymph nodes and (2) cerebralward to form the thoracic
duct.

The iliac lymphatic sacs are two long symmetrical sacs extending
from the hilus of the Wolffian bodies to the level of the bifurcation
of the aorta. In an embryo 23 mm. long they drain into the mesone-
phritic veins. By the time theembryo is 25 to 27 mm. long this
connection is lost and-the two iliac sacs converge into the median
cisterna chyli dorsal to the aorta.

The caudal end of the iliac lymphatics in the pig is a sac of con-
siderable size, from which three sets of lymphatics can be injected:
the ilio-lumbar, the femoral and a plexus which surround the umbil-
ical arteries. The ilio-lumbar superficial lymphatics are very con-
spicuous in the pig, as shown in figs. 4 and 5 (Sabin 130). This is ix
marked contrast to the human embryo, where they form part of the
inguinal group.

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 65

2. PERIPHERAL LymMpuatics In Oruer Forms.

The pattern of the superficial lymphatics in the human embryo is
ywn in the two figs., 505 and 506 (Sabin 134). This was a re-
ikable specimen of an embryo 5.5 cm. long which had been kept in
malin some time. On transferring it to freshly made up alcohol,
‘filled the entire superficial lymphatic system. It shows the primary
serficial lymphatic plexus. Valves in a few ducts, namely, the
vipital, the thoracic and the inguinal, indicate the oldest vessels.
ie differences between the superficial pattern in the human and in the
xy are mainly the fact that in the human embryo the thoracic lym-
atics are much more prominent, draining into the axilla, while in the
; they drain into the superficial cervical plexus, and all the posterior
nphatics drain into the inguinal region in the human embryo. In
3 human embryo the deep lymphatics for the arm grow from an
tension of the jugular sac along the primitive ulnar veins, and the
ep lymphatics for the legs grow from the iliac sac along the femoral
ins. This shows in fig. 493 (Sabin 134).
Rabbit and cat embryos are like the human in having a marked
velopment of the jugular sac along the primitive ulnar vein (Lewis
). In this respect they differ from the pig where the axillary lym-
aties are smaller. In fig. 7 is shown a deep vessel from the jugular
2, which is, I think, the axillary duct in the pig. In the rabbit and
e cat the axillary sac may possibly arise independently of the jugular
2, and in the rabbit a conspicuous chain of peripheral vessels grows
mg the thoraco-epigastric vein (fig. 8, Lewis 76). Injections of cat
ibryos show that the axillary lymphatic trunk is very large.
The superficial pattern of the lymphatics in bovine embryos, as
own by Polinski (107), is much like that of the pig.
In the chick the primary lymphatic vessels in the skin connect
e posterior lymph heart with the jugular lymph sac along the lateral
te in the course of the thoraco-epigastric vein Mierzewski (96)
d BE. L. Clark (27 and 29).
(. ORIGIN AND DEVELOPMENT OF LYMPHATIC GLANDS
AND THEIR RELATION TO PRIMARY LYMPH SACS.

It is easy to prove that the lymphatic sacs in mammals and the
gular lymph sac in birds become in part lymph glands. I shall define -
imary lymph glands as those which are derived from the primary
mph sacs.

Digitized by Microsoft®

66 Florence R. Sabin.

From the jugular lymph sac in the pig it is clear in fig. 16 t
two lymph glands develop: (1) the deep jugular lymph node wh
drains the pharynx and (2) the node in the posterior triangle of
neck which drains the skin of the anterior part of the body. A.
Clark (24) has shown that these two glands are single glands in
neck of the adult pig and that they become the largest of all the cervi
glands. In the human embryo these two glands are represented
groups of glands, the glands of the posterior triangle and the d
jugular glands. In the human embryo also the group of deep axill.
glands comes from an extension of the jugular sac along the primit
ulnar vein and hence they are primary lymph glands. Thus the 1
mary nodes for the anterior part of the body are (1) deep jugular noc
(2) the nodes of the posterior triangle, (3) the axillary nodes. 1
relation of the primary lymph gland, both to the various structu
of the neck and to the stalk of the jugular sac, that is, to the jugu
lymph trunks, is shown especially well in fig. 17 from pig 7.5 ¢
long. The anterior curvature and the sac stalk lie behind, that
medial to the sternocleidomastoid muscle. Of the primary nodes
the neck the deep jugular nodes drain a restricted area, namely, ©
pharynx and nose; the axillary lymph nodes drain the arm, and ~
large node of the posterior triangle drains all the rest of the head, fa
neck and thorax, either directly or through the superficial cervi
plexus. The nodes of the superficial cervical plexus are secondary
the primary lymph sac nodes. It has been brought out that th
groups of lymphatics do not remain distinct as they arise, but all |
come connected with each other through anastomoses of the lym
vessels.

From the retroperitoneal sac develop the retroperitoneal lym
nodes. From the iliac sacs in the pig there is a chain of small no:
lateral to the aorta and a large group of glands on either side oppos
the bifurcation of the aorta. These are primary iliac nodes. Dor
to the aorta is a chain of nodes from the lower end of the ciste1
chyli to the bifurcation of the aorta. Thus the renal lymphatics g
rise to three groups of primary lymphatic glands: (1) pre-aortic
retroperitoneal, (2) symmetrical iliac nodes lateral to the aorta, «
(3) pree-vertebral nodes dorsal to the aorta.

The secondary lymphatic nodes develop along the lymphatic vess
The most extensive group of secondary nodes in the embryo pig is
one which comes from the superficial cervical plexus along the exter

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 6%

igular vein. In fig. 17 this plexus is shown as one large lymph
land. ‘From the cerebral end of the plexus vessels are forming a
land at the point where the lymphatics divide to form the facial and
ie temporal lymphatics. The cervical plexus in the adult pig becomes
group of at least a dozen small lymph glands. In both figs. 17 and
8 there is a developing submaxillary lymph gland along the course
f the facial branch of the linguo-facial vein (fig. 8). The small facial
ode and the submaxillary node are tertiary nodes compared with the
igular lymph sac. They show how lymph glands develop at points
here lymph ducts radiate out on their development. The primary,
xcondary and tertiary nodes are constant in the pig.

The mesenteric glands are secondary for the retroperitoneal sac.
‘he secondary glands from the iliac sac are very simple, for there are
aly two of them: (1) The ileo-inguinal gland, which is very charac-
ssristic of the pig and which remains as a single gland. Its position
i readily made out in fig. 4 (Sabin 130), for the superficial lymphatics
f the posterior body wall radiate to it. Like the superficial cervical
lands it is superficial. (2) The inguinal glands in the pig become
large group of superficial glands, the position of which can be made
ut in fig. § (Sabin 180). The lymph ducts from the inguinal lymph
lands develop to the leg, the ventral abdominal wall and the external
enital organs. The efferent vessels of the inguinal group of glands
nd ileo-inguinal gland converge to the large group of primary iliac
lands opposite the bifurcation of the aorta.

The histological development of lymph glands is now well under-
ood. The work of Saxer (138) on developing lymph glands considers
rimarily the development of lymphocytes which I shall not attempt
) discuss in this paper, confining the work to the development of -the
land as a whole. From the time of Breschet (16) it has been known
1at lymph glands begin as a plexus of lymphatic ducts. Their de-
alopment can be followed through the work of Ranvier (117), Bartels
10), Gulland (39), Kling (72), Jolly (59), Lewis (77), and Sabin
131).

The primary lymph glands, which, with the exception of the post-
ortic, develop out of sacs, begin by a bridging of the sac by bands of
mmective tissue covered by endothelium. In the case of the retroperi-
meal sac these bands are never absent (figs. 498 and 499, Sabin 134).
or the jugular lymph sac they show well in human embryo 30 mm.
mg in fig. 495 (Sabin 134). That these bridges are entirely covered

Digitized by Microsoft®

68 Florence R. Sabin.

by endothelium is best brought out by total mounts of silver nitrate in
jections. ‘This has been done for the retroperitoneal sac in the pig
When the primary sacs are thus completely bridged by these band
they are practically a dense plexus of lymphatic capillaries and ar
therefore in the first stage of the development of lymph glands. A
this stage the connective tissue septa are undifferentiated and contain
only mesenchyme and blood capillaries.

We are now in a position to consider the question of which are th:
first lymph nodes in the embryo. This point has been discussed b:
Lewis (7%) and Sabin (131 and 134). It depends on the criterion
used. If the criterion is recognizing the formation of a plexus o
lymphatic capillaries with undifferentiated connective tissue bridge
as the anlagen of a gland, then it is clear that the primary lymp]
glands, that is, those that come from the lymph sacs, are the firs
lymph glands to begin for each region of the body. Thus the jugular
subclavian sac in the human embryo is extensively bridged at 30 mm
and is therefore a plexus of lymphatics, the anlage of a lymph glanc
At this stage we are agreed there are no lymphocytes (Kling (72)
Lewis (77) and Sabin (131)). If, on the other hand, the criterio:
is the development of the first lymphocytes in the body, we mus
say that this point has not been reinvestigated since all the primar
lymph nodes of the embryo have been determined. Saxer (13¢
p- 381) notes that lymph nodes are beginning in the neck, the pos
terior mediastinum, the retroperitoneal tissue, the outside of bot
hips and along the Wolffian bodies in bovine embryos 4.5 cm. long
and in sheep embryos of the same length describes undoubted wandei
ing cells in the same areas. Kling (72) has no specimens of huma
embryo between 31 mm., where, he says, there are no lymphocytes i
the axillary glands, and 70 m., when lymphocytes are present in man
glands. Lewis (77) finds lymphocytes in human embryos 42 mn
long. He finds that they occur practically simultaneously in th
glands around the internal jugular vein (primary lymph nodes
and certain “ isolated subcutaneous lymph glands,” of which he figure
one or two along the linguo-facial vein and its branches. The relatio
of these glands to lymphatic vessels will be readily made out by
comparison of fig. 8 and fig. 16. Figure 8 shows the ducts from tt
sac along the linguo-facial vein at a stage even younger than Lewis
isolated glands, namely, a pig measuring 3.5 cm. Moreover the b
ginning cervical plexus from which these submaxillary vessels con

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 69

1ows plainly in fig. 7 from an embryo much smaller, namely, one
easuring 18 mm., and they are readily injected in embryos 30 mm.
ng, fig. 2 (Sabin 130). Thus it is almost certain that the ducts for
ewis’s early facial glands in human embryos were present, but could
ot be found, in sections. It is probable that the lymphocytes begin
| human embryos which measure about 40 mm. and they probably
svelop almost simultaneously in the primary and in the secondary
‘ands,

The plexus stage of developing lymph glands'has been shown by
‘construction by Kling (72) and Lewis (77), and by injection in
x, 10 (Sabin 131).

The whole question of the structure of a lymph gland may be
immed up in a word by a quotation from Ranvier (p. 1038 in 117):
Un ganglion lymphatique est un angiome caverneux lymphatique qui
été d’abord angiome simple.” The development of a gland involves,
ywever, two processes: (1) the formation of lymph sinuses out of
mph plexuses, and (2) the formation of lymph cords and lymph folli-
es in the trabecule. A comparison of Jolly’s (59) plates from lymph
lands in birds, Kling’s (72) figures of human lymph glands, and
y own (131) from the developing glands in the pig, will show that
1e relative proportion of these two parts varies exceedingly in differ-
it glands.

The formation of lymph sinuses can be best understood by beginning
ith Jolly’s figures. He shows that in the birds, instead of the mam-
alian sinus, there is a diffuse plexus of lymphatic vessels with lym-
aocytes in the septa between the vessels. He also brings out the
ct that very large vessels may pass directly through the gland. Thus
is clear that the sinus of birds is a simpler structure than that
mammals. On this account the complete covering of endothelium
very plain in Jolly’s figures. Out of a plexus of capillaries such as
‘ly shows, the mamalian lymph sinus is formed, as can be seen
fig. 15 (Sabin 131), by an increase in the density of the lymphatic
exus until the septa are reduced to a framework of reticulum covered
‘endothelium. In this figure all the stages of the formation of a
aus can be seen in a single section. In the lower part of the specimen
ery lymphatic vessel is plain with its complete lining of endothelium,
nile in the upper part the septa are in places already reduced to a
1e. This stage is from an embryo pig 24.5 cm. long, which is the

Digitized by Microsoft®

70 Florence R. Sabin.

best stage for studying the formation of the sinuses. The question
of the development of reticulum has been taken up by Mall (84).

In regard to the formation of the lymph cords and follicles, the funda-
mental point is the question of the origin of the lymphocytes themselves.
This, however, I shall not undertake to discuss. The lymphocytes may
occur diffusely in all of the trabecule of the developing node and this
is the simplest form of a gland. This form is shown by Jolly (59)
in his plate IX, fig. 3, for the lumbar glands in birds, by Kling (72),
plate XXVII, fig. 11, for the axillary glands in human embryos and
in fig. 9 (Sabin 131) from the primary jugular lymph glands in
the pig. It will be noted that all of these figures are from primary
lymph glands. The definite follicles may develop in the center or in
the periphery of a node, and there is every possible variation in
the proportion of the diffuse cords and the definite follicles. There
are two processes in the development of the follicle: (1) an increase
in the number of lymphocytes forming a definite clump, and (2)
the formation of a tuft of blood capillaries. Both the cords and the
follicles form along the blood vessels, the follicles coming at the cap-
illary bed. ‘

As I have said, the early glands which come from the primary sacs
pass through the form of diffuse distribution of the lymphocytes.
Follicles which develop later, however, may begin at once as clumps
of lymphocytes, making very definite follicles surrounded by a lym-
phatic plexus. A number of such tiny follicles are to be seen in fig.
ié6 (Sabin 131), which is the developing group of inguinal lymph
glands in a pig 24.5 em. long.

Lymph glands may be either simple, consisting of one follicle with
a peripheral sinus, or compound, with many follicles and cords, and
both peripheral and central sinuses.

The subject of hemolymph glands has been taken up by Helly in
the Ergebnisse fiir Anatomie and Entwickelungsgeschichte for 1902.

VII. COMPARATIVE MORPHOLOGY OF THE PRIMARY
LYMPH SACS, LYMPH HEARTS, AMPHIBIAN LYMPH
SACS AND LYMPH GLANDS.

We are now in a position to consider the comparative morphology
of lymph sacs and lymph hearts. I shall not attempt to analyze the
work of Favaro (36a) and of Allen (2-+) on lymphatics in fishes,

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 71

both of whom state that vessels may function now as veins and now as
phatics, but beginning with the amphibia, it is certain that lym-
phatics bud off from the veins in certain definite areas and form
plexuses which are transformed into sacs. The essential structure of
these sacs is the endothelial lining. A primary lymph sac is there-
fore one which buds off from the endothelium of a vein.

In the amphibia, probably in the reptilia, and in the case of the
posterior lymph sacs in birds, the primary sacs bud off from segmental
veins and rest on the myotomes. They derive striated muscle from
the myotomes, which has been shown at least in the case of the
anterior lymph hearts in amphibia (Knower 74), and the posterior
lymph hearts in reptiles (Stromsen 145) and pulsate rhythmically; in
a word, they become lymph hearts.

The lymphatic sacs for the anterior part of the body in birds and all
of the lymphatics in mammals do not come from the vertebral and
coccygeal veins, but from the anterior cardinal veins in the neck, and
the renal veins in the abdomen. These sacs do not lie on the myotomes,
they do not receive striated muscle, but rather are transformed into
lymph glands. Thus the bird represents a transition stage having a
posterior lymph heart and a jugular lymph gland. I shall keep,
therefore, the term anterior and posterior lymph hearts, and use the
terms jugular and renal lymph sacs. The embryological classification
is:

(1. Lymph hearts.
a. Anterior, amphibia.
b. Posterior, amphibia, reptilia and birds.

Primar
aa 2. Lymph sacs (glands).
Sacs , a. Jugular—birds and mammals.

Retroperitoneal.
pb. Renal 2 Iliac.
Cisterna chyli.

The subcutaneous and deep lymph sacs of the amphibia are not
primary, but secondary. They are transformed lymphatic ducts.
Thus the amphibia have primary lymph sacs which become lymph
hearts, lymph ducts which become secondary sacs, and lymphatic capil-
‘aries. Mammals have primary lymph sacs which become primary
lymph glands, lymphatic ducts which develop valves and along which
secondary glands are formed, and thirdly, lymphatic capillaries without
valves.

5

Digitized by Microsoft®

92 Florence R. Sabin.

VIII. VARIOUS OTHER THEORIES IN REGARD TO THE
ORIGIN AND DEVELOPMENT OF THE LYMPHATIC
SYSTEM.

1. Mayzr-Lewis ANLAGEN.

In the preceding pages has been given a consistent account of how
the lymphatic system buds off from the veins and gradually invades
the body. The places where the lymphatics bud off vary in different
forms; in the amphibia and reptiles, and in the case of the posterior
lymph hearts in birds the primary lymph sacs lie in the myotomes and
receive striated muscle and become lymph hearts. In the higher forms
the anterior lymphatics come from the anterior cardinal vein and its
branches, the caudal lymphatics do not develop (mammals) and the

Fic. 18.—Diagram to show the theory of Ranvier and myself that the
lymphatics arise from the veins. The veins are striped, the lymphatics
dotted.

posterior lymphatics come from the vena cava and renal veins. From
these points of origin can be traced all the lymphatics of the body
down to the ultimate capillaries.

This theory has not been developed without opposition. In section
I it was made clear that the first theory of the origin of the lymphatics
was that they arose in the periphery as dilated tissue spaces and grew
towards the center, Gulland (39). This was questioned by Sala
(137), showing that the first lymphatics in birds were against the
veins, and by my proving in 1902 that the jugular lymphatics are
the first lymphatics in mammals and that they bud off from the
anterior cardinal veins. Thus I substituted for the theories shown in

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 13

' figs. 1, 2 and 3 the simple hypothesis of Ranvier, shown in fig. 18.
This indicates that lymphatics bud off as lymphatics from the veins
and have the same relation to tissue spaces as have blood capillaries.

The most difficult obstacle in regard to this theory is what we call
Mayer-Lewis anlagen. Mayer (89) noted in the tadpole’s tail certain
isolated vessels which he thought were evidences of the degeneration
of blood vessels.

Ranvier (114, p. 578) in studying the plexus of lymphatics of the
great omentum in a new-born cat noted that a great many of the
vessels terminated in cul de sacs. Some were attached by extremely
slender threads, so that they looked as if they were becoming iso-
lated by the atrophy of the intermediate part. Since in the adult cat
there are no lymphatics in the omentum, Ranvier interpreted this as
a method of degeneration of lymphatic vessels.

The study of the method of degeneration of blood and lymphatic
capillaries is an important one. It is obvious that there must be a
destruction of capillaries, since a single vessel or a few vessels come
from a plexus of capillaries. In Clark’s observations (25, p. 191) the
usual method of degeneration is by retraction of the processés, but
he has also observed (26, p. 410) together with Mayer and Ranvier,
that an occasional segment of a blood capillary or lymph capillary
may become detached and be left to atrophy. This is in accord with
some of the findings of pathology. We must, therefore, conclude that.
detached, degenerating vessels do occur, though they do not represent
the usual method of degeneration.

F. T. Lewis (76) found that in reconstructions lymphatics looked
like chains of beads, and to explain this phenomena he suggested that
lymphatics might grow by small detached portions of blood vessels.
These hypothetical lymphatic anlagen we call Lewis anlagen.

It is clear in watching lymphatics grow, that they never pick up
isolated vessels, but grow by the increase in the protoplasm of their
own wall; yet it is clear also that reconstructions of uninjected lym-
phatics appear like rows of beads.

To study this point it was necessary to submit the method of recon-
struction to comparative tests with the method of injection and when
possible with the method of growth in the living form. This has been
done three times (Clark, E. R. 26; Clark, E. L. 27, and Sabin 135).

The best opportunity for studying this point is the tadpole’s tail,
where, in the living specimen, or in the tail fixed in alcohol, the entire

Digitized by Microsoft®

74 Florence R. Sabin.

lymphatic system can be seen and drawn. This was done by Clark.
The specimen was then cut in serial sections and both blood capillaries
and lymphatics were reconstructed. Two reconstructions were made,
one with the 4 mm. Zeiss objective, and the other with a 2 mm. Zeiss
oil immersion lens. Both reconstructions show that neither blood
capillaries nor lymphatic capillaries can be reconstructed completely.
More is obtained with an oil immersion lens, but both powers show
capillaries in the form of rows of beads (figs. 5, 6 and 7, Clark 26,
copied as figs. 513 and 514, Sabin 134).

This test of Clark’s is the best possible test, because it is a recon-
struction of exactly the same specimen from which the original draw-
ing was made. This point cannot be made in testing the method of in-
jection and that of reconstruction. For this test I (135) used, however,
symmetrical plexuses in the same embryo. An embryo pig 2% mm.
was chosen in which there was an almost complete injection of the occip-
ital superficial lymphatics. Many of the sprouts on the margin had been
ruptured. The injected plexus could of course be reconstructed, while a
reconstruction of the empty lymphatics on the opposite side showed the
entire plexus split wp into isolated vessels (figs. 6, 7 and 8, Sabin 135).
On the injected side there was just one lymphatic vessel which did not
receive the injection mass, and there was an extravasation just at its
base. This is readily explained by the fact that vessels are often con-
nected by very slender strands to the main plexus, as, for example, the
vessel near the point of injection in fig. 18, and a rupture might readily
occur in such an area before the end of the vessel was reached.

The test of the two methods has now been made a third time by
Mrs. E. R. Clark (27) in her injection of the jugular lymphatic
. plexus in a stage corresponding to the one which Miller (97) has
reconstructed. The two results are shown in figs. 14 and 15.

A comparison of the injected jugular sac in a pig 18 mm., shown in
fig. 7, with reconstructions of the corresponding stages in the cat
(figs. 60 to 62, Huntington 55), also brings out the same point that,
wherever the injection method can be applied, it demonstrates more
continuous lymphatics in an area than can be reconstructed, even as
isolated vessels.

Undoubtedly an occasional blood capillary or lymphatic capillary
may separate from the main plexus and atrophy, but the apparently
isolated vessels found in serial sections along the course of grow-
ing lymphatics connect in life. Lymphatics do not grow by de-

Digitized by Microsoft®

Lhe Origin and Development of the Lymphatic System. 5

tached blood vessels, and hence the hypothetical Mayer-Lewis anlagen
do not exist, that is to say, they are not anlagen of lymphatics,
but are parts of a continuous vessel. It is clear that reconstructions of
growing lymphatics are valuable, even though they show the lym-
phatics as rows of beads, provided only the true lymphatics are in-
cluded; that is to say, the endothelial-lined vessels. For example,
Lewis (76, figs. 7 and 8) shows that the early lymphatics in the rabbit
grow along the thoraco-epigastric vein.

2, EXTRA-INTIMAL AND PERINEURAL SPACES AND FENESTRATION.

Huntington and McClure (51) began their work on the lymphatic
system in 1907 with a study of the jugular lymphatics in the cat.
They thought that lymphatics began as clefts between the intima,
the vein and the surrounding mesenchyme, so that for a time the
wall of a lymphatic was half venous endothelium and half mesen-
chyme. These spaces they called “extra-intimal” lymphatics. As
far as the jugular lymphatic sacs were concerned they subsequently
abandoned this idea, accepting the theory that the jugular sacs are
derived from the veins (Huntington and McClure 54), but they have
since revived the extra-intimal theory in connection with the thoracic
duct (Huntington 58).

Their work on the jugular sacs has been given in section JV. In
general they accept the theory of the venous origin of the jugular
sacs, but they confuse the picture (1) by the inclusion of some pert-
neural spaces (fig. 22, Huntington and McClure 54) and (2) by the
idea of fenestration..

The perineural spaces are an interesting phenomenon. I con-
sidered them in my first paper ‘in 1902. Along the entire central
nervous system and following the peripheral nerves are dilated tissue
spaces. Within the spinal canal these spaces are the anlage of the
spaces of the pia-arachnoid. Along the nerves we call them perineural
spaces. All sections of embryonic tissue show them. They are of
undoubted significance for the physiology of the growing nerves, but
they never bear any relation to the lymphatics. Injections of them
outline the nerves and never enter lymphatic vessels—occasionally
they can be injected from the pia-arachnoid, but usually the dense
tissue between the vertebre prevents this.

The question of fenestration was considered in section IV. Paral-
lel veins are formed by the same process of sprouting of endothelium

Digitized by Microsoft®

16 Florence R. Sabin.

by which all other vessels are formed, and the term fenestration,
giving ihe vague suggestion of the splitting of the wall of a vein
once formed, does not describe the actual process and is misleading.
We now know that the jugular sacs bud directly from the anterior
cardinal vein as lymphatics and grow by the sprouting of their endo-
thelium.

As has been said, Huntington and McClure believe that the jugular
lymphatics come from the veins, but that they remain only as a means
of communication between the veins and the'lymphatic ducts. The
lymphatic ducts they think develop in a variety of ways.

The entire question of the method of growth of the lymphatic
system is now concentrated on a study of the methods of growth of
the thoracic duct. It is probable that the thoracic duct arises in two
places; that it is formed in mammals by an asymmetrical down-
growth from the left jugular sac and by a plexus which arises sym-
metrically from the renal veins and grows along the course of the aorta.
T'his second portion arises later than the first and is in the blood-filled
stage (pig 23 mm.) when the jugular stalk isempty. These two anlagen
grow as do all other lymphatics and join by the same process by which
any lymphatic plexus is formed.

The opposing views are brought out in two papers, one by Hunting-
ton (58) and one by Kampmeier (66). Huntington describes a compli-
cated method of origin of the thoracic duct, including (1) a jugular
stalk; (2) extra-intimal spaces, and (3) general tissue spaces. Kamp-
meier, on the other hand, gives a clear and excellent presentation of
the theory of the origin of lymphatics by the addition of tissue spaces.

I will begin with the work of Huntington (58). His memoir on
the peripheral lymphatics has excellent photographs of sections, so that
it is easy to see what he is considering as lymphatics. For example,
figs. 1 to 9 are tissue spaces, figs. 10 to 12 are extra-intimal spaces,
and figs. 13 to 19 are probably lymphatics. Figure 25 shows the iliac
sac labeled 76. Numerous examples of extra-intimal spaces are shown
in figs. 105 to 147. Perhaps the best figure of an extra-intimal space
is No. 111A. Figures 148 to 158 are lymphatics.

The theory of the origin of lymphatics from extra-intimal spaces
has been especially described in an article by Huntington in 1910
(55). In a diagram on page 409 he gives his idea that a space which
forms around a degenerating vein eventually includes the entire vein,
which then disappears so that the lymphatic is left with a wall of
mesenchyme.

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. %%

There are numerous points which rule out the theory of the origin
of the lymphatics by extra-intimal spaces. (1) These spaces can be
varied at will by changing the fixation. In my experience they are
much more common with lymphatics than with veins. We have sec-
tions of adult human tonsils, in which all surrounding lymphatics
have extra-intimal spaces. In the human embryo, 460, the large
jugular segment of the thoracic duct has one area where the endo-
thelium has sagged from the surrounding tissue. There is, I think,
no question but that the extra-intimal spaces are artefacts. This is
made the more certain by the fact that they do not occur in the living
tadpole’s tail, but are found along the veins after fixation.

(2) In connection with the presentation of the extra-intimal
spaces, Huntingion gives no proof whatever that the structures he
figures are degenerating veins. Most of the extra-intimal spaces he
shows are along the line of the pulmonary and cardiac lymphatics
arising from the jugular sacs. I think it probable from their size
and position that the structures he is dealing with are shrunken
lymphatics and not veins. The pulmonary and cardiac lymphatics
(fig. 13) will, when reconstructed, give the same appearance of iso-
lated lymphatic vessels as all other lymphatic trunks.

(3) The third point against the theory that lymphatics grow by
extra-intimal spaces is this: the growing lymphatic tip always keeps
as far as possible from the blood capillaries, just as in the adult the
ultimate lymphatic capillaries are as far as possible from the blood
capillaries; as, for example, in the relation of the central lacteals
and peripheral blood capillaries in the villus. The point of the avoid-
ance of the blood capillaries and of the mesenchyme cells as well by
the growing lymphatic tip has been well described by Clark (26).
Moreover, in certain tadpoles the lymphatics grow out ahead of the
blood capillaries where no blood capillaries have ever been. Therefore,
lymphatics do not grow by extra-intimal spaces, for extra-intimal
spaces are artefacts; there is no evidence that lymphatics follow
degenerating veins and on the other hand the growing lymphatic
tip, far from following the blood capillaries, avoids them as much
as possible.

3. GrowrH or LYMPHATICS BY THE ADDITION or TISSUE SPACES.

With these various structures as anlagen of lymphatic capillaries,

Huntington and McClure have included certain of the tissue spaces
Tetna alane tha aanrea af the Iwmnhaties. This theory Huntington

Digitized by Microsoft®

78 Florence R. Sabin.

\

brought out especially in a study of the lymphatic system in reptiles
(57). After a confused account of the origin of the anterior lymph
hearts he describes the development of the peripheral lymph vessels
out of tissue spaces. He says that whereas in mammals the greater
number of the lymphatic vessels come from extra-intimal spaces (57,
p. 272) in reptiles, on the other hand, the lymphatics come from
tissue spaces without relation to the veins.

The theory that in the pathway of developing lymphatic vessels
certain tissue spaces enlarge and are added onto the growing tips
has been worked out by two pupils of McClure, Kampmeier (66a) and
Stromsen (146a), both of whom published their work in 1912.

Kampmeier’s first article (66) is a preliminary report of the second
(66a). He has worked on the development of the thoracic duct in the
pig and bases most of his conclusions on a reconstruction of a speci-
men of mine. Kampmeier thinks that the thoracic duct develops in
an anterior-posterior direction by the addition of certain connective
tissue spaces which enlarge in the pathway of a developing vessel.
This is a return to the view of the earlier embryologists, except that
they believed that the growth was from the periphery toward the center.

To a certain extent Kampmeier still holds to the theory of lym-
phatics from extra-intimal spaces; that is, he believes that a part of
the thoracic duct follows veins that are degenerating. A vein which
lies in the pathway of a developing lymphatic vessel he calls a “ veno-
lymphatic.” This is a different use of the term from that of Hunting-
ton and McClure, who used the term to mean a vein which was trans-
formed into a lymphatic vessel. This latter use of the term can well
be given up, since veins do not become lymphatics. The replacing
of degenerating veins Kampmeier does not regard as a fundamental
process in the development of lymphatics, since they may or may not
follow such veins. He thinks that in the development of the thoracic
duct about half the duct comes from extra-intimal spaces (66a, p.
434). It is interesting to note in connection with the degeneration
of veins in Kampmeier’s work that he speaks of the plexus of veins
medial to the nervous sympatheticus, fig. 11 (or fig. 11 in 66a), as
veno-lymphatics; that is, as veins which disappear as the thoracic
duct forms, and yet this same plexus of veins can be readily injected
in embryo pigs 27 to 30 mm. long; stages in which the thoracic duct
is well formed. It is true that the blood vessels may not show in
uninjected specimens, just as the plexus of blood capillaries around

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. %9

the air cells of an adult lung cannot be reconstructed from sections
of uninjected specimens. The degeneration of blood vessels can only be
followed by means of complete injections of different stages.
Kampmeier finds another example of lymph vessels which grow
by the extra-intimal replacement of veins in the region where the
thoracic duct empties into the jugular sac. In his fig. 8 from an
embryo 20 mm. long he shows vessels in which there is a considerable
sagging of the endothelium from the surrounding tissue, making the
so-called extra-intimal spaces. Kampmeier’s embryo measuring 20
mm. is about the same stage as mine, measuring 23 mm., since
mine was measured before fixation, and hence his figure can be com-
pared with fig. 12, in which a lymphatic injection makes it easy to
separate veins from lymphatics. As Hoyer says (49b, p. 536) : “ Den
‘Verlauf der Lymphgefasse nur an nicht injizierten Serienschnitten zu
verfolgen, ist eine miihevolle und zeitraubende Arbeit, deren Ergeb-
nisse, wie wir uns selbst tiberzeugt haben, hdchstens nur fiir grdssere
Lymphstémme, deren Lage man kennt, als sicher zu bezeichnen sind.
Ist der Verlauf von Gefassen und deren Veriastelung noch unbekannt,
so lassen sich auf Grund von Serienschnitten nur Vermutungen
iiber Verlauf und Veriastelung derselben austellen. Erst wenn man
tiber Gefasse und ihre Aeste mit Injektionsmasse gefiillt vor sich sieht,
erhalt man einen guten Hinblick in die Verteilung derselben, deren
Verfolgung selbst an nicht injizierten Serienschnitten dann keine
wesentlichen Schwierigkeiten mehr bietet”’ Although it is not pos-
sible to be entirely sure in regard to Kampmeier’s sections, yet it is
at least probable that he is dealing in large part with lymphatics and
not with veins at all. The sagging of the endothelium from the
surrounding tissue I have found more often with lymphatics than
with veins, and yet in tissues fixed in Zeuker’s fluid, which Kamp-
meier uses exclusively, it may occur in any cf the veins and even in
the aorta. Kampmeier says, with reference to this group of vessels
(fig. 8 and fig. 29 in 66a, p. 460), that they are mesenchymal, peri-
vascular spaces into which the jugular sac opens freely; that is, he
thinks that the jugular sac in an embryo 20 mm. long opens freely
into the tissue spaces. This point is easy to disprove, for in contrast
to the early stages of the thoracic duct the jugular sac in this stage
is easy to inject. If this were true then every injection of the sac in
this stage would show extravasations. This is, however, not so. In
fig. 7 is an injection of the sac in an embryo 18 mm. long and I can

Digitized by Microsoft®

80 Florence R. Sabin.

state with great certainty on the basis of many injections that the
jugular sac has a complete endothelial wall and is a closed vessel.
It is thus clear that Kampmeier has not demonstrated that this upper
part of the thoracic duct replaces degenerating veins.

The chief point, however, in Kampmeier’s work is not that some
of the lymphatics replace degenerating veins, but that they develop
~ out of tissue spaces. As the chief proof of this theory he uses a
reconstruction of an injected specimen of mine. Notwithstanding the
fact that this injection is the only one which has yet been made in a
mammal in this early stage, Kampmeier does not hesitate to call it
complete. It was injected indirectly through the jugular lymph sac,
from which the injection mass ran into the thoracic duct. At a cer-
tain point in the injection there is an, extravasation (Kampmeier,
fig. 18, line 15), and in exactly the same position in the neat section
is a large endothelial-lined empty space. It is therefore merely an
arbitrary decision whether the empty vessel was actually connected
with the injected part or not, that is to say, there is as much evidence
for the one view as for the other. No one who has had experience with
the injection method would be sure that the first injection in a new
region was a complete one. The spaces which Kampmeier has shown
as lymphatics in my specimen are lined by endothelium; that is,
they are the spaces with which Lewis (76) has made us familiar;
they are not the mesenchyme spaces which Kampmeier and Strom-
sen regard as the anlage of lymphatics.

Stromsen (146a) has injected the prevertebral lymphatics in turtles
and finds that in advance of the injected zone there are enlarged
tissue spaces which he thinks are going to become lymphatics. Kamp-
meier says that it is easy to select tissue spaces which are going to
become lymphatics for, “histologically, all incipient lymphatic an-
lagen, whether they are spaces independent in position or spaces fol-
Jowing, transforming and expanding the discarded pathways of
redundant venous channels, are decidedly different from either an
active vein or a mature lymphatic ” (Kampmeier 66a, p. 430). Strom-
sen (146, p. 354) adds to this ability to select tisste spaces which are
going to become lymphatics this further point, that such enlarged tis-
sue space occur only in the pathway of developing lymphatics. These.
two points can be easily disproved by anyone who has access to serial
sections of embryos. For example, why were not the enlarged tissue
spaces in Kampmeier fig. 3 selected as lymphatics? Good examples

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 81

of such spaces are the constantly occurring spaces in the parietal
pleura on either side of the vertebra. They are always present in the
stages in which the thoracic duct is developing; they are adjacent to
the vascular zone just internal to the ribs and they never have any
relation to lymphatics. A definite, constant group of such spaces is
also to be found in the subcutaneous tissue of the mid-dorsal line of
certain stages and they never become lymphatics. The pia-arachnoid
and perineural spaces likewise never become lymphatics. In fact,
large as well as small groups of such spaces are a constant occurrence
in sections of embryos; sometimes near lymphatics and sometimes far

Fic. 19.—Diagram to show scme of the various structures which Hunt-
ington and McClure have described as lymphatics. a.=true lymphatic
capillaries; b. = Mayer-Lewis anlagen, also true lymphatic capillaries; c.
= extra-intimal hypothetical lymphatic capillaries; d. — tissue spaces.
The perineural spaces are not shown. The veins are striped, lymphatic
capillaries dotted, and hypothetical lymphatic capillaries are cross-hatched.

from them. The selection of certain tissue spaces as destined to
become lymphatics is an arbitrary selection and brings us back to the
confused standpoint of the earlier embryologists; a standpoint which
would indeed justify the view of Bartels that the question of the
relation of the lymphatics to the tissue spaces is “ eine philosophische,
keine anatomische Frage.”

The difficulty of finding out the method of growth by observing
sections is best illustrated by following the different theories of their
growth in the pages of Huntington and McClure. Some of these
methods are shown in the diagram, fig. 19. This development of the

Digitized by Microsoft®

82 Florence R. Sabin.

subject has been necessarily confusing to those who have not followed
the subject carefully. However, now that the origin and method of
growth of the lymphatic system has been cleared up, so that the funda-
mental morphology is understood, the controversy has had this great
value, that it has brought up for analysis and discussion every con-
ceivable method of growth. Lymphatics do not arise as dilated per-
ipheral tissue spaces after the manner of the ccelom as the earlier
embryologists thought; they do not grow by the addition of hollow con-
nective tissue cells, as Schwann and Virchow thought; they do not rise
as perineural spaces, nor by fenestration of a vein, nor by extra-intimal
clefts, nor by the progressive addition of connective tissue spaces, nor
by the addition of detached blood vessels, but they bud from the veins
and grow by the sprouting of their endothelial wall.

IX. CONCLUSIONS.

The most important result of this study on the morphology of the
lymphatic system is the emphasis it throws on the importance of
endothelium as a tissue. The angioblast is one of the early tissues
to be differentiated ; it is not an inert lining for vessels, but an actively
growing functioning tissue. In its place of growth it is a syncytium
of actively amceboid protoplasm. Mollier (99) has shown that in the
spleen it may be reticular; Mall (86 and 87) has shown that it may
give rise to reticulum. Undoubtedly the further development of our
knowledge of endothelium depends on the development of the new
experimental anatomy.

The lymphatic endothelium buds off from the veins. It is always
a little different in appearance from the endothelium of the veins, and
the lymphatic capillary is different in size and form from the blood
capillaries. The growing lymphatic tip has the remarkable character-
istic that it avoids the blood capillaries, while it is attracted by other
lymphatic capillaries.

Endothelium is the essential tissue of the lymphatic system. In the
lower vertebrates lymph hearts are formed by the addition of striated
muscle to primary lymph sacs. In the higher forms lymph glands
are formed by the development of lymphocytes around the ducts.
This takes place not only in the wall of the primary lymph sacs, but
along plexuses of ducts, so that there are primary and secondary
lymph glands.

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 83

The fundamental morphology of the lymphatic system has been es-
tablished, but there remain many gaps in our knowledge of the system
as a whole. The deep lymphatics in amphibia and reptilia are almost
unknown; in fact a complete account of the lymphatic systems in
both of these groups would be of great value. In birds the origin of
the iliac lymphatics and the growth of the thoracic duct would be most
interesting. In mammals the gaps in our knowledge are especially
in regard to the development of the lymphatic system within many
of the organs; for example, the heart and lungs, the liver, spleen,
kidneys and reproductive organs. A study of the embryology of the
lymphatic system makes it more certain that the central nervous
system has no lymphatics. The lymphatics invade the body, but not
completely; the nervous system is a permanent non-lymphatic area.

It is now possible to define the lymphatics. Lymphatic capillaries
are tubes of endothelium; they are derived from the endothelium of
the veins, and they have the same relation to tissue spaces as have
blood capillaries,

LITERATURE.

1. Aset, J. J.: On the action of drugs and the function of the
anterior lymph hearts in cardiectomized frogs. Journ. of
Phar. and Exp. Ther., Vol. 3, 1912.

2. ALLEN, W. F.: Distribution of the lymphatics in the head, and
in the dorsal, pectoral, and ventral fine of Scorpenichthys
marmoratus. Proc. Wash. Acad. of Science, Vol. 7%, 1906.

3. -— Distribution of the subcutaneous vessels in the tail
region of Lepisosteus. Amer. Journ. of Anat., Vol. 8, 1908;
Anat, Record, Vol. 2, 1908.

4. Distribution of the lymphatics in the tail region of

Scorpenichthys marmoratus. Amer. Journ. of Anat., Vol.
11, 1910-1911.

5. ARNOLD, J.: Experimentelle Untersuchungen ueber die Ent-
wickelung der Blutcapillaren. Arch. fiir path. Anat., Bd. 53,
1871 and Bd. 54, 1872.

Ueber die Beziechung der Blut- und Lymphgefisse zu
den Saftkanilen. Arch. fiir path. Anat., Bd. 63, 1875.

v. v. Bazr: Entwickelungsgeschichte der Tiere. I. Teil, 1828.
II. Teil, 1837.

Digitized by Microsoft®

Florence R. Sabin.

. Bartyer, WAtTer A.: On the origin of the mesenteric sac and
thoracic duct in the embryo pig. Amer. Journ. of Anat.,
Vol. 8, 1908.
9, BaranskKI, J.: Die Entwicklung der hinteren Lymphherzen der
Unke (Bombinator). Bull. de Académie des Sciences de
Cracovie. Classe des Sciences Mathématiques et Naturelles.
Série B. Sciences Naturelles, 1911.
. Barrets, P.: Das Lymphgefiassystem. Handbuch der Anatomie
4 des Menschen. Von Bardeleben, 1909.

. Baum, H.: Koénnen Lymphgefasse direkt in Venen einmiinden?
Anat. Anz., Bd. 39, 1911.

2. Brcuat: Anatomie Générale, 1818.

. BoERHAVE: Quoted from Bichat., Anat. Gén., S. 596.

. Bonnet, R.: Grundriss der Entwicklungsgeschichte. Berlin,
1891; Zweite Auflage, 1912.

5. Bremer, H. L.: ‘Che development of the aorta and aortic arches
in rabbits. Amer. Journ. of Anat., Vol. 13, 1912.

. Brescuet, G.: Le Systéme Lymphatique. Paris, 1836.

7. Bupee, ALBREcHT: Neue Mittheilungen ueber die Lymphge-
fasse der Leber. Konigl. Sachs. Gesellsch. d. Wiss. zu Leip-
zig. Math.-Phys. Classe. Berichte 27-30, 1875-80, S. 161.
Die Lymphwurzeln der Knochen. Arch. f. mik. Anat.
13, 1877. i

Ueber ein Kanalsvstem in Mesoderm von Hiihnerembry-
onen. Arch. fiir Anat. und Phys., Anat. Abth., 1880.
Ueber das dem zweiten Blutkreislaufs entsprechende
Lymphgefassystem bei Hitihnerembryonen. Centralbl. f. d.
med. Wissensch., 1881.

Ueber Lymphherzen bei Hiihnerembyronen. Arch. fiir
Anat. und Phys., Anat. Abth., 1882.

Untersuchungen tiber die Entwickelung des Lymph-
systems beim Htihnerembryo. Arch. fiir Anat. u. Phys.,
Anat. Abth., 1887.

. CRUIKSHANK, W.: The anatomy of the absorbing vessels of the
human body. London, 1790.

. CLarK, A. H.: On the fate of the jugular lymph sacs, and the
development of the lymph channels in the neck of the pig.
Amer. Journ. of Anat., Vol. 14, 1912.

5. Crank, E, R.: Observations on living growing lymphatics in

the tail of the frog larva. Anat. Record, Vol. 3, 1909.

Digitized by Microsoft®

26.

27.

28.

29,

30.

31.

33.

34.

35.

36.

36a.

The Origin and Development of the Lymphatic System. 85

An examination of the methods used in the study of
the development of the lymphatic system. Anat. Record,
Vol. 5, 1911.

Ciark, ELrzanor Linton: Injection and reconstruction of the
jugular lymph sac in the chick. Anat. Record, Vol. 6, 1912.

General observations on early superficial lymphatics in

. chick embryos. Anat. Record, Vol. 6, 1912.

Criarx, Eviot R., and Etzanor L.: Observations on the devel-
opment of the earliest lymphatics in the region of the pos-
terior lymph heart in living chick embryos. Anat. Record,
Vol. 6, 1912.

Cuark, Exsor R.: Further observations on living growing lym-
phatics: their relation to the mesenchyme cells. Amer.
Journ. of Anat., Vol. 13, 1912.

DrascH, O.: Die Bildung der Somatopleura und der Gefasse
beim Hiihnchen. Anat. Anz. 9, 1894; S. 567.

. DzturzyNnski, A.: Untersuchungen ueber die Regeneration der

Blut und Lymphgefiasse im Schwanze von Froschlarven.
Bull. de Académie des Sciences de Cracovie. Classe des
Sciences Mathématiques et Naturelles. Série B. Sciences
Naturelles, 1911.

Evans, H. M.: On the development of the aorta, cardinal and
umbilical veins, and the other blood vessels of vertebrate
embryos from capillaries. Anat. Record, Vol. 3, 1909.

On the earliest blood vessels in the anterior limb buds

of birds and their relation to the primary subclavian artery.

Amer. Journ. of Anat., Vol. 9, 1909.

Die Entwicklung des Blutgefassystems. Handbuch
der Entwicklungsgeschichte des Menschen. Keibel und Mall.
Bd. 2. Leipzig, 1911.

Favaro, G.: Ricerche anatomo-embriologiche intorno alla circo-
latione caudale ed ai cuori linfatici posteriori degli Anfibi
con particolare riguardo agli Urodeli. Atti del l’Acad. Sc.
veneto-trentino-istriana. Cl. 1, 1906.

Ricerche interno all morfologia ed all sviluppo dei

vasi, seni e cuori caudali nei Ciclostomi e nei Pesci. Atti

del Reale Instituto Veneto di sc. lett. ed arti. Venezia. T.

65, 1906.

Digitized by Microsoft®

86

38.

39.

40.

41,

42,

43.

44,

46.

47,

48.

49,

49

Florence R. Sabin.

. GoETTE: Entwickelungsgeschichte der Unke als Grundlage einer
vergleichenden Morphologie der Wirbelthiere. Leipzig,
1874.

GoLDFINGER, GizeLLa: Ueber die Entwickelung der Lymph-
sicke in den hinteren Extremititen des Frosches. Bull. de
PAcadémie des Sciences de Cracovie. Classe des Sciences
Mathématiques et Naturelles, 1907.

GuLLUND: The development of lymphatic glands. Journ. of
Path. and Bact., Vol. 11, 1894.

Haun, H.: Experimentelle Studien iiber die Entstehung des
- Blutes und der ersten Gefiisse beim Hiihnchen. Archiv fiir
Entwicklungsmechanik der Organismen. Bd. 27, 1909.
Hatter: Elementa physiologiz corporis humani.. T. 1. 1757-

63, s. 110.

Hetty, Konrap: Hemolymphdriisen. Ergebnisse der An-
atomie und Entwickelungsgeschichte. Bd. 12, 1902.

Hever, Grorce J.: The development of the lymphatics in the
small intestine of the pig. Amer. Journ. of Anat., Vol. 9,
1909.

His, W.: Ueber die Wurzeln der Lymphgefisse in den Hauten
des Koérpers und iiber die Theorien der Lymphbildung. Zeit-
schr. f. wissensch. Zodlogie, XII, 1863.

Ueber das Epithel der Lymphgefasswurzeln und ueber

die von Recklinghausenschen Saftcanalchen. Zeitsch. f.

wissensch. Zoologie, XIII, 1863.

Untersuchungen ueber die erste Anlage des Wirbeltier-

leibes. Leipzig, 1868.

Lecithoblast und Angioblast der Wirbeltiere. Abhandl.
der mathematisch-physischen Klasse der Konig]. Sachs.
Gesellsch. d. Wissenschaften. Bd. 26, 1900.

Hoyer, H.: Ein Beitrag zur Histologie bindegewebiger Gebilde.
Arch. fiir Anat. und Phys. und wiss. Medicin, 1865.

Untersuchungen ueber das Lymphgefissystem der

Froschlarven. Bull. de Académie des Sciences de Cracovie.

Classe des Sciences Mathématiques et Naturelles, 1908.

Zur Entwicklung der Lymphgefisse bei Wirbeltieren.

Verhandl. d. VIIT Internat. Zool. Kongr. zu. Graz. (vom.

15-20, 8, 1910), 1912.

a.

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 87

A9b.

Und Udaziela, 8., Untersuchungen ueber das Lymph-
gefassystem von Salamanderlarven. Morph. Jahrb., Bd. 44,
1912.

50. Hunter, W.: Medical Commentaries. Deutsch. von. Kiihne als
medicinisch-chirurgische Beobachtungen. Leipzig, 1785.
Bd. 2.

51. Huntineton, G. 8., and McCrurz, C. F. W.: The development
of the main lymph channels of the cat in their relations to
the venous system. Anat. Record, Vol. 1, 1906-1908.

The anatomy and development of the jugular lymph
sacs in the domestic cat. Anat. Record, Vol. 2, 1908.

53. Huntineton, G. 8.: The genetic interpretation of the develop-
ment of the mammalian lymphatic system. Anat. Record,
Vol. 2, 1908.

54. Huntineton, G. 8., and McCuurs, C. F. W.: The anatomy
‘and development of the jugular lymph sacs in the domestic
cat. Amer. Journ. of Anat., Vol. 10, 1910.

55. Hunrineron, G. S.: The phylogenetic relations of the lym-
phatic and blood vascular system in vertebrates, p. 1, and
The genetic principles of the development of the systemic
lymphatic vessels in the mammalian embryo, p. 339. Anat.
Record, Vol. 4, 1910.

52.

56. Ueber die Histogenese des lymphatischen Systems beim
Saéugerembryo. Verhandl. der Anat. Gesellsch., 1910.

57. The development of the lymphatic system in reptiles.
Anat. Record, Vol. 5, 1911.

58. The anatomy and development of the systemic lym-

phatic vessels in the domestic cat. Memoires of the Wistar
Institute of Anatomy and Biology, No. 1, 1911.

59. Jotty, J.: Recherches sur les ganglions lymphatiques des
oiseaux. Arch d’Anat. microsc., T. 11, 1910.

60. Jourpain, M. S.: Sur V’existence d’une circulation lymphatique
chez les Pleurometes. Comptes Rendus de l’Académie des

. Sciences, 90, 1880, p. 1430. .

61. JourpaIn, M. L.: Sur le systéme lymphatique des tétards et
grenouilles. Comptes Rendus de VAcadémie des Sciences,
96, 1883.

62. Josstrrow, G.: Zur Lehre von dem Lymphgefissystem der
Froschlarve, des Frosches und der Eidechse. Ref. in *
Schwalbe’s Jahresber., 1905, III, 8. 353.

Digitized by Microsoft®

88 Florence BR. Sabin.

63. Josstrrow, S. M.: Ueber die Bedeutung des Cisterna chyli fir
die Lymphbewegung. Russki Wratsch. Jahrg. II. N. 29,
8. 1909-1910.

64. Josstvow, G.: Sur le systéme lymphatique du tétard, de la
grenouille et du lézard. 1 Taf. Mém. l’Acad. Imp. des Sc.
St. Petersbourg. Sér. 8. T. 15. N. 11, p. 20.

65. Der Anfang des Ductus thoracicus und seine Hrweiter-
ung. Trav. Soc. Méd. scientif. et. hyg. Charkoff. 8. 1-9.

66. Kampmeter, O. F.: The value of the injection method in the
study of lymphatic development. Anat. Record, Vol. 6, 1912.

66a The development of the thoracic duct in the pig.
Amer. Journ. of Anat., Vol. 13, 1912.

67. KorLLiker: Sur le developpement des tissues chez les batraciens.
Annales des Sciences Naturelles, [IIe Serie, 1846.

68. Ueber die Blutkérperchen eines menschlichen Embryo
und die Entwicklung der Blutkorperchen bei Saugetieren.
Zeitschrift f. rationelle Medizin. Henle und Pfeufer. Bd.
IV, 1846.

69. Entwicklungsgeschichte des Menschen und der héheren
Tiere. Zweite Auflage. Leipzig 1876, 8. 158.

70. Grundriss der Entwickelungsgeschichte des Menschen
und der hoheren Tiere. 2 Aufl. Leipzig 1884.

71. Histologische Studien an Batrachierlarven. Zeitschr.
f. wissensch, Zoologie. 43, 1886.

72. Kune: Studien tiber die Entwickelung der Lymphdriisen beim
(Menschen. Upsala Likareforenings Féranslinger. 1903,
and Arch. f. mikr. Anat. u. Entwicklungsgesch. Bd. 63,
1902.

73. Kwowrr, H. McE.: Effects of early removal of the heart and
arrest of circulation on the development of frog embryos.
Anat. Record, Vol. 1, 1907.

74, The origin and development of the anterior lymph
hearts and the subcutaneous lymph sacs in the frog. Anat.
Record, Vol. 2, 1908. ‘

75. Lancer, C.: Ueber das Lymphgefissystem des Frosches. Die
Lymphgefasse im Schwanze der Batrachierlarven. Sitzungs-
ber. d. k. Akad. d. Wissensch. Bd. 58, 1868.

76. Lewis, F. T.: The development of the lymphatic system in

rabbits. Amer. Journ. of Anat., Vol. 5, 1906.

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 89

ee The first lymph glands in rabbit and human embryos.
Anat. Record, Vol. 3, 1909.
78. On the cervical veins and lymphatics in four human

embryos, with an interpretation of anomalies of the sub-
clavian and jugular veins in the adult. Amer. Journ. of
Anat., Vol. 9, 1909.

79. MacCattum, W. G.: Die Beziehung der Lymphgefisse zum
Bindgewebe. Arch. fiir Anat. und Phys. Anat. Abth.,
1902.

80. Matt, F. P.: Development of the ventral abdominal walls in
man. Journ. of Morphology, Vol. 14, No. 2, 1898.

81. The ccelom. Reference Handbook of the Medical Sci-
ences. William Wood & Co., New York. Second Edition.
1900-1904.

82. -— Development of the internal mammary and deep epi-
gastric arteries in man. John Hopkins Hospital Bulletin,
1898.

83. On the origin of the lymphatics in the liver. Proceed-
ings of the Association of American Anatomists, 1900.

84. Development of the connective tissues from the con-
nective syncytium. Amer. Journ. of Anat., Vol. 1, 1901-
1902, p. 354.

85. On the development of fhe blood vessels of the brain in
the human embryo. Amer. Journ. of Anat., Vol. 4, 1905.

86. A study of the structural unit of the liver. Amer.
Journ. of Anat., Vol. 5, 1906.

87. On the development of the human heart. Amer. Journ.

of Anat., Vol. 13, 1912, p. 258.

8a. Marcus, H,: Beitrige zur Kenntniss der Gymnophionen. Mor-
phologisches Jahrb. Bd. 38, 1908.

88. Mayer, J.: Ueber die Neubildung von Blutgefassen in plas-
tischen Exsudaten seréser Membranen und in Hautwunden.
Annal. d. Charité, IV, p. 41-140.

89. Mayer, Sigmunp: Ueber die blutleeren Gefasse im Schwanze
der Batrachierlarven. Sitzungsberichte, der math. natur-
wissenschaftlichen Classe der Kaiserlichen Akademie der
Wissenschaften in Wien. Bd. 91, III Abth., 1885, S. 204.

89a. MeyEr, F.: Systema amphibiorum lymphaticum disquisition-
ibus novis examinatio. Diss. inaug, Berolini, 1845.
Quoted from Hoyer (49b).

Digitized by Microsoft®

90 Florence R. Sabin.

90. Maxrmow, A.: Untersuchungen ueber Blut und Bindegewebe.
Archiv fiir mikroskopische Anatomie und Entwickelungsge-
schichte, Bd. 73, 1909.

91. McCuurz, C. F. W.: The development of the thoracic and right
lymphatic ducts in the domestic cat (Felis domestica).
Anat, Anz., Bd. 32, 1908.

92. McCuure, C. F. W., and Sitvester, C. F.: A comparative study
of the lymphatico-venous systems in adult mammals. Anat.
Record, Vol. 3, 1909.

93. McCuurg, C. F. W.: The extra-intimal theory and the develop-
ment of the mesenteric lymphatics in the domestic cat (Felis
domestica). Verhandl. der Anat. Gesellsch., 1910.

A few remarks relative to Mr. Kampmeier’s paper on
the value of the injection method in the study of lymphatic
development. Anat. Record, Vol. 6, 1912.

95. Mrtrzer, S. J.: The distribution of solutions in cardiectomized
frogs. Journ. of Exper. Med., Vol. 13, 1911.

96. MierzesEwskI, Lapistas: Beitrag zur Entwicklung des Lymph-
gefassystems der Vogel. Bull. de l’Acad. des Sciences de
Cracovie. Classe des sciences mathematiques et naturelles,
1909. :

97. Mitter, A. M.: The development of the jugular lymph sac in
birds. Amer. Journ.‘of Anat., Vol. 12, 1912.

Histogenesis and morphogenesis of the thoracic duct in
the chick; development of blood cells and their passage to
the blood stream via the thoracic duct. Amer. Journ. of
Anat., Vol. 15, 1913.

98. Minot: Die Entwicklung des Blutes. Handbuch der Entwick-
lungsgeschichte des Menschen. JKeibel und Mall. Bd. II,
Leipzig 1911. :

99. Mouuter, S.: Ueber den Bau der kapillaren Milzvenen (Milz-
sinus). Arch. f. mikr. Anat., Bd. 76, 1911.

100. Nucx, A.: Adenographia curiosa. Leide, 1691.

101. Panprer: Beitrige zur Entwickelungsgeschichte des Hiihnchens
im Ei, Wiirzburg, 1817.

102. Pannizza, B.: Osservazionni antropo-zootomico fissiologiche.
Pavia, 1830.

103. — Sopra il sistema linfatico dei rettili, ricerche zootomiche.
Pavia, 1833.

94.

97a.

Digitized by Microsoft®

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

115.

117.

The Origin and Development of the Lymphatic System. 91

Pensa, A.: Della struttura e dello sviluppo dei gangli linfatici
degli uccelli (Ancer domesticus). Ric. f. n. Lab. di Anat.
norm. d. R. Univ. di Roma. Vol. 12, 1906-07. Tav. 15.
Fig. 2.

Studio sulla morfologia e sulla topografia della cis-
terna chili e del ductus thoracicus nell’uomo ed in altri mam-
miferi. Ric. f. n. Lab. di Anat. norm. d. R. Univ. di Roma.
Vol. 14, 1908-09. Tav. 2.

PuiatNer, EH. A.: Hinige Beobachtungen ueber die Bildung der
Capillargefasse. Miiller’s Archiv f. Anat. Phys. und wis-
sensch. Medicin, 1844, S. 525-526.

Potinsxi, W.: Untersuchungen ueber die Entwicklung der sub-
cutanen Lymphgefisse der Saéuger, in sonderheit des Rindes.
Bull. de PAcad. des Sciences de Cracovie. Classe des
Sciences Math. et Naturelles, Serie B, 1910.

Prevost ET Lesert: Mémoire sur la formation des organes de
la circulation et du sang dans les batraciens. Ann. des
sciences nat. 3 Serie, 1844. T. I, II et III.

RaDWAnskA, ‘M.: Die vorderen Lymphherzen der Frosche. Bull.
intern. Ac. Se. Cracovie, 1906.

Ranvier, L.: Morphologie du systéme lymphatique. De l’ori-
gine des lymphatiques dans la peau de la grenouille. C. R.
Ac. des sc. T. 120, 1895, p. 182.

Structure des ganglions lymphatiques du pore. C. R.

Ac. des sc. T. 121, 1895, p. 800.

Etude morphologique des capillaires lymphatiques des

mammiféres. C. R. Ac. des sc. T. 121, 1895, p. 856.

Développement des vaisseaux lymphatiques. C. R. Ac.

des se. T. 121, 1895, p. 1105.

Aberration et régression des lymphatiques en voie de

développement. C. R. Ac. des sc. T. 122, 1896, p. 578.

Des lymphatiques de la villosité intestinale chez la rat

et le lapin. C. R. Ac. des se. T. 123, 1896, p. 923.

La théorie de la confluence des lymphatiques et la

forphologie du systéme lymphatique de la grenouille. C.

R. Ac. des se. T. 123, 1896, p. 970.

La théorie de la confluence des lymphatiques et la

développement des ganglions lymphatiques. C. R. Ac. des

sce. ‘I. 123, 1896, p. 1038.

Digitized by Microsoft®

92

118.

126.

127.

128.

129.

130.

131.

182.

Florence R. Sabin.

Morphologie et développement des vaisseaux lymphatiques
chez les mammiféres. Archives d’Anatomie microscopique.
T. 1, 1897, p. 69 et p. 137.

. Von RECKLINGHAUSEN: Die Lymphgefiisse und ihre Beziehung

zum Bindegewebe. Berlin, 1862.
Das Lymphsystem. Stricker’s Handbuch der Lehre von
den Geweben. Leipzig, 1871-72.

. ReicHert, Kk. B.: Studien des physiologischen Instituts zu Bres-

lau. Leipzig, 1858.

Entwicklungsleben im Wirbelthierreich. Berlin, 1840
und 1857.

Entwicklung des Meerschweinchens. Abh. der Berliner
Akad., 1862.

. Remax, R.: Untersuchungen ueber die Entwickelung der Wir-

beltiere. Berlin, 1850-55.

Ueber blutleere Gefaisse (Lymphgefasse) im Schwanze
der Froschlarve. Miiller’s Archiv f. Anat. Phys. und wiss.
Medicin. 1850. §S. 102 und S. 182.

Roucer, CHarLes: Mémoire sur les développement, la structure
et. les propriétés physiologiques des capillaires sanguins et
lymphatiques. Arch de Phys., V, 1873.

Rickert: Ueber die Abstammung der blutigen Gefassanlagen
beim Huhn, und ueber die Entstehung des Randsinus beim
Huhn und bei Torpedo. Sitzungsber. d. k. bayer. Akad.
d. Wissensch. zu ‘Miinchen, 1902.

Rickert, J.; und Motiigr, 8.: Die erste Entstehung der Ge-
fasse und des Blutes bei Wirbeltieren. Handbuch der ver-
gleichenden und experimentellen Entwickelungslehre der
Wirbeltiere. Hertwig, Bd. I. T. 1, 1906.

Sasin, F. R.: On the origin of the lymphatic system from the
veins and the development of the lymph hearts and thoracic
duct in the pig. Amer. Journ. of Anat., Vol. 1, 1901-02.

On the development of the superficial lymphatics in

the skin of the pig. Amer. Journ. of Anat., Vol. 3, 190-4.

The development of the lymphatic nodes in the pig

and their relation to the lymph hearts. Amer. Journ. of

Anat., Vol. 4, 1905.

Further evidence on the origin of the lymphatic endo-

thelium of the blood vascular system. Anat. Record, Vol. 2,

1908.

Digitized by Microsoft®

The Origin and Development of the Lymphatic System. 93

133. The lymphatic system in human embryo, with a con-

sideration of the morphology of the system as a whole.

Amer. Journ. of Anat., Vol. 9, 1909.

Die Entwicklung des Lymphgefissystems. Handbuch

der Entwicklungsgeschichte des Menschen. Keibel und Mall,

Vol. 2, Leipzig, 1911. Manual of Human Embryology,

Keibel und Mall. J. B. Lippincott Co., 1912.

A critical study of the evidence presented in several

recent articles on the development of the lymphatic system.

Anat. Record, Vol. V, 1911.

On the origin of the abdominal lymphatics in mammals

from the vena cava and the renal veins. Anat. Record, Vol.

6, 1912. :

Der Ursprung und die Entwickelung des Lympfge-
fassystems. Ergebnisse der Anatomie und Entwickelungs-
geschichte, 1918.

137. Sata, L.: Sullo sviluppo dei cuori limfatici e dei dotti toracici
nell’ embrione di pollo. Ricerche Lab. di Anat. Norm. d.
r. Univ. di Roma. Vol. 7, 1899-1900.

138. Saxer, Fr.: Ueber die Entwickelung und den Bau der normalen
Lymphdriisen und die Entstehung der roten und weissen
Blutkérperchen. Anat. Hefte. Bd. 6, 1896.

139. Scuwann: Structur und Wachstum der Thiere und Pflanzen.
Berlin, 1839.

140. Stnvester, CHARLES F.: On the presence of permanent com-
munications between the lymphatic and venous systems at
the level of the renal veins in adult South American
monkeys. Amer. Journ. of Anat., Vol. 12, 1912.

141.:SmrrH, Heten W.: On the development of the superficial veins
in the body wall in the pig. Amer. Journ. of Anat., Vol. 9,
1909.

142. Srannius: Ueber Lymphherzen der Vogel. Miiller’s Archiv
fiir Anat., Phys. und wiss. Medicin, 1843.

143. STRICKER, S.: Studien ueber den Bau und das Leben der capil-
laren Blutgefisse. Sitzungsbericht der Math.-Naturwiss.
Klasse der k. Akad., Wien. Bd. 52. Abth. 2, 1865, 8. 379.

Untersuchungen ueber die Kontractilitat der Kapillaren.

Med. Jahrb., Wien, 1878.

134.

136.

136a.

144.

Digitized by Microsoft®

94 Florence R. Sabin.

145. Stromsen, F. A.: A contribution to the anatomy and develop-
ment of the posterior lymphatic heart of turtles. Publication
132 of the Carnegie Institution of Washington, 1910.

146. On the relations between the mesenchymal spaces and
the development of the posterior lymph hearts of turtles.
Anat. Record, Vol. 5, 1911.

146a. On the development of the prevertebral (thoracic)

duct in turtles as indicated by a study of injected and unin-
jected embryos. Anat. Record, Vol. 6, 1912.

147. Szity, A.: Ueber das Entstehen eines fibrillaren Stiitzgewebes
im Embryo und .dessen Verhaltniss zum Glaskérperfrage.
Anat. Hefte. Bd. 35, 1908.

148. THoma: Untersuchungen ueber die Histogenese und Histome-
chanik des Gefassystems. Stutgart, 1893.

149. Titney, F.: The development of the veins and lymphatics in
Tragulus meminna, erxleben. Amer. Journ. of Anat., Vol.
13, 1912.

150. VircHow: Die Cellularpathologie. Berlin, 1858.

151. Wietixy, W.: EHinige Beitrage zur Anatomie und Physiologie
der Lymphherzen. Ref. in Hoffmann-Schwalbe’s Jahresber.,
1884.

152. Ueber die Anwesenheit vielzihliger Lymphherzen bei
den Froschlarven. Zool. Anz. Bd. 9, 1886.
153. Weitere Untersuchungen ueber die Lymphherzen und

Lymphgefisse einiger Amphibien. Ref. in Hoffmann-
Schwalbe’s Jahresb., 1889.

154. Wotrr: Theoria Generationes (Original 1759). Ausgabe von
1764.

155. Wysorzky, N. Tu.: Materialien zur embryonalen Entwick-
lungsgeschichte der Gefasse. Journ. f. norm. u. path.
Histologie u. klin. Medicin. Herausgegeben von M. Rudneff.
St. Petersburg, 1875. Schwalbe und Hoffmann, Jahres-
berichte. 1876. S. 145.

Digitized by Microsoft®

Digitized by Microsoft®

Digitized by Microsoft®

Digitized by Microsoft®