Extracted from the 
TRANSACTIONS OF THE COLLEGE OF PHYSICIANS OF 
PHILADELPHIA. 
THIRD SERIES, VOLUME VIII. THE ANATOMICAL BEARINGS OF THE SEROUS 
COVERING OF THE VISCERA. 
By 
OSCAR H. ALLIS3LD., 
SURGEON TO THE PRESBYTERIAN HOSPITAL, AND TO THE JEFFERSON 
MEDICAL COLLEGE HOSPITAL. 
[Read November 4, 1885.] 
In my remarks I shall deal only with the anatomical 
difficulties that perplex the student, and shall be con- 
tent if I can present him with so clear a description of 
them, that he will remember them, because he under- 
stands them. I shall make few allusions to the works 
of embryologists, simply stating at the outset, that all 
that is known about the subject is due to their tireless 
researches. 
At a very early stage of development, the intestinal 
canal consists simply of a straight tube, extending the 
whole length of the body of the embryo, and lying in 
front of that which subsequently becomes the vertebral 
column. It is closed at both ends, but communicates 
with the umbilical vesicle V, Fig. 10, which is external 
to the body. It is covered throughout its entire length 
by a continuous serous membrane, since at this stage of 
existence there is no horizontal partition, the diaphragm, 
to divide the body into two sections. At this period, 
the intestine antedates all the viscera, save the heart, 
and a cross section of any part of the body of the embryo, 214 
ALLIS, 
whether through the thoracic or abdominal portion, will 
represent a tube (the body walls ) enclosing a tube (the 
alimentary canal). 
But this simple anatomical expression does not long 
continue. Soon at certain points of the intestine a bud- 
ding process is manifest, and the lungs, liver, and pan- 
creas, make their appearance, whose development with 
Fig. 1. 
In this sketch of the anatomy of the embryo, I have represented both tubes 
as complete, and the serous membrane as continuous. It forms a parietal layer 
to every part of the body of the embryo, and a covering to the viscera, which 
push forward into it. The cavity, which at first extends the whole length of the 
embryo, is called the pleuro-peritoneal, and the serous membrane the pleuro- 
peritoneal. 
that of the diaphragm converts the simple expression in 
Fig. 1, into a labyrinth, every path of which the student 
feels, ends in fog, or leads to impenetrable darkness. 
At a very early date the diaphragm makes its appear- 
ance. Nothing positive of its development has as yet 
been established. Being a muscular organ, it is likely 
that it springs from the body plates, and grows from 
the periphery toward the centre. Its appearance, it 
would seem, must be prior to, or coincident with that, of 
the liver, which can be clearly defined by the middle of 
the third day. Although all the various processes of de- 
velopment are moving on together, it will be simpler to 
speak of some of them as complete in advance of others. 
Thus, by the formation of the diaphragm, we may look 
upon the pleuro-peritoneal cavity, already described, SEROUS COVERING OF THE VISCERA. 
215 
as separated into two distinct apartments, an upper, 
thoracic, or pleural; a lower, peritoneal, or abdominal. 
It has already been stated, that in the early hours of 
development there are no lungs, and when these make 
their appearance, it is as buds or offshoots from the 
oesophagus, Fig. 10, L. A simple diagram, suggested 
by Balfour and Foster, will illustrate the steps of this 
process. At the site of the oesophagus where the lungs 
are to be developed, a cross section would show an elon- 
gation, Fig. 2 (A), later a double constriction (B), and 
Fig. 2. 
still later the lungs as independent structures (C), con- 
nected at first by separate tubes with the oesophagus; 
these soon fuse, and a single permanent trachea is 
formed. 
In the demonstration of the pleural membrane, only 
the lungs and heart concern us, since these alone receive 
a covering from it, and by it are enclosed in separate 
apartments. In the description of this process, it will 
be simpler to regard the thoracic boundaries as complete 
in advance of their contents, and to represent the heart 
and lungs by three dots resting behind the serous mem- 
brane, Fig. 3. Thus we have the simple expression of 216 
ALLIS, 
a single cavity, lined by a single and continuous mem- 
brane. Little by little these structures grow forward 
into the pleuritic cavity, and, as they do so, push the 
lining membrane before them. Though destined to 
Fig. 3. 
occupy nearly the whole of the thoracic space, they do 
not advance with equal rapidity. The heart exhibits 
functional activity in advance of every other viscus, 
and its rapid development and great importance give 
it at first much greater space than that required for the 
lungs, whose function in the higher orders is never 
called into action prior to birth. In Fig. 4 I have rep- 
Fig. 4. 
resented the lungs, and the heart with its pericardium, 
pushing forward and filling up the cavity of the chest, 
and just as I have represented the chest-walls as devel- 
oped in advance of their contents, so I have represented, 
diagrammatically, the pericardium, in advance of the 
heart. Note that the viscera have not as yet separate 
compartments; that the serous lining in Fig. 3, can still SEROUS COVERING OF THE VISCERA. 
217 
be traced as a single continuous structure in Fig. 4. 
A little later the pericardium meets and unites over the 
heart, while at the same time it unites with the sternum 
in front, and the diaphragm below, thus effectually iso- 
lating the heart and lungs, and creating three distinct 
serous cavities, A, A, B, Fig. 5. Were we now to 
Fxg. 5. 
regard these processes-z. e., the development of the 
chest-walls, the diaphragm, the heart, lungs, and peri- 
cardium-as all advancing together, it would be easy to 
conceive how the pericardium by its relations to sur- 
rounding parts-i. e., the diaphragm and sternum-be- 
comes the principal factor in constructing the partitions 
of the chest cavity. 
At the outset the oesophagus was, in common with 
the entire intestine (Fig. 1), covered with a serous mem- 
brane. But for this circumstance and the one already 
mentioned, that the lungs sprout from the oesophagus, 
neither the lungs nor heart would have a serous cover- 
ing. How, then, does the oesophagus lose its covering? 
Simply enough-i. e., by the advancement of the heart 
and lungs. These structures push the thin, delicately 
attached serous membrane before them, stripping it off from 
the oesophagus, trachea, and large vessels, and appro- 
priating it to themselves. Hence a space just in front 218 
ALLIS, 
of the vertebra is formed that has no serous envelope, a 
space between the roots of the lungs and back of the 
heart (Fig. 5, D). This space is called the posterior medi- 
astinum, and the only remark that I wish to make in 
regard to it is to contrast its formation with that of the 
anterior mediastinum (Fig. 5, C ), which latter is formed 
by the fusing of the serous covering of the pericardium 
with that of the sternum and diaphragm. If this view 
is correct, then in cases of so-called congenital absence 
of the pericardium there can be no anterior medias- 
tinal space, and no separate cavities for the heart and 
lungs, while the presence of these chambers and the isola- 
tion of their contents, must prove the statement of the 
so-called congenital absence of the pericardium to have 
been an error. 
I will now pass to the abdominal portion, and take 
up the stomach and duodenum. It is from the duode- 
num that the liver and pancreas are formed, and it is by 
changes that take place, as development advances in 
these structures, that the foramen of Winslow, the pos- 
terior or lesser cavity of the peritoneum, and the four 
Fig. 6. 
layers of the great omentum are formed. Again, let me 
refer to the embryological fact, that in early life the 
intestine lies as a straight tube along the vertebral 
column, covered by the peritoneum. In order to repre- 
sent the bowel as straight, I am forced to represent the 
liver and pancreas (see dots, Fig. 6) as widely removed SEROUS COVERING OF THE VISCERA- 
219 
from the diaphragm. The liver, however, lies imme- 
diately beneath the diaphragm to the right of the median 
line, and, as it increases rapidly in size and weight, 
draws the lighter and less stable stomach and duodenum 
with it. Thus the liver, stomach, and duodenum, instead 
of continuing to look directly forward, with a layer of 
peritoneum on either side of them, gradually, so to 
speak, turn upon their right side, so that the liver lies 
in the right hypochondrium, and that which was at first 
the left side of the stomach becomes its anterior surface 
Fig. 7. 
(Fig. 7). For a time we may suppose all these struc- 
tures to remain free and unattached, and that the hand 
can be passed freely beneath the duodenum, pancreas, 
stomach, and liver; but soon the transverse duodenum 
fuses with the serous membrane upon which it lies, and 
now a small cavity is formed behind these viscera, which 
communicates freely with the general peritoneal cavity 
along the free (right) border of the liver and descending 
duodenum. The opening is large at first, extending as 
it does from the level of the pancreas (Fig. 7) to the 
diaphragm, but as the liver and duodenum increase in 
size, and approach each other, and fuse more with sur- 
rounding parts, the opening is narrowed to a foramen, 
which bears the name of Winslow. The space behind 
the stomach and liver now is limited above by the dia- 220 
ALLIS, 
phragm, below by the transverse duodenum, upon the 
left by the original mesenteric attachment, and at its 
free border closed, except at the foramen of Winslow. 
Although these viscera have changed their position, and 
lie, as it were, upon their sides, yet they still, in the 
main, retain the same relation to their serous covering 
-i. e., they lie between its two layers (Fig. 7), one layer 
being in front and another behind. The two that con- 
nect the liver with the stomach are called the gastro- 
hepatic omentum; the two that connect the stomach 
and pancreas, the gastro-pancreatic omentum ; the two 
that connect the stomach and spleen, the gastro-splenic 
omentum. Let us now suppose the stomach, finding 
Ftg. 8. 
A. Incorrect; the great omentum encloses the transverse colon, and forms its 
mesocolon. B. Correct; the great omentum, sends its two posterior layers over 
the transverse colon to the pancreas and duodenum. C, D. Successive steps in 
the formation of the great omentum. 
itself hemmed in within too narrow boundaries, to begin 
a struggle for more freedom. It cannot free itself from 
the diaphragm and liver, nor from its original mesen- 
teric attachment; but finding the two layers that pass SEROUS COVERING OF THE VISCERA. 
221 
from its greater curvature, to the spleen and pancreas, 
more yielding, tugs away at these until both layers 
stretch and hang down into the abdominal cavity 
(Fig. 8, C, D, B). This constitutes the greater omen- 
tum, which is nothing more than a stretching or bulging 
forward of the two layers (C ) of the gastro-pancreatic 
omentum. As they increase in depth they may not 
inaptly be compared to two sacs (Fig. 9), one within 
Fig. 9. 
the other, the anterior layers of which, if followed 
upward, would enclose the stomach and proceed up upon 
the liver, and whose posterior layers will embrace the 
pancreas and duodenum. The space within the inner 
sac communicates, of course, with the space behind the 
stomach and liver, and may be regarded as merely an 
annex. Should these sacs fuse, it is plain the single 
membrane thus formed ought to be considered as con- 
sisting of four layers, just as the great omentum is 
described in the text-books. 
I have been particular to describe the formation of 
the great omentum without mentioning the transverse 
colon, and this is the way it should be understood.1 
1 First accurately described by J. Muller, 1830: Cruveilhier and See, Splanch- 
nologie, p. 537. 222 
ALLIS, 
Upon this head most of the text-books are in error-not 
only in their diagrams but in the text. I have examined 
about a dozen text-books, and of these, Quain, Holden 
(1868), Cruveilhier, and See, are the only ones correct. 
The prevailing error has been occasioned by observing 
that the omentum is, in the adult, usually attached to 
the transverse colon, but while attached it does not sur- 
round it as in Fig. 8 (A), or form its mesocolon, but 
passes in front and above it (B) to the pancreas and 
duodenum, from which it in fact sprung (C, D). When 
correctly represented, as in B, the colon, as Professor 
Chapman states, gets upon its anterior and upper sur- 
face five layers of peritoneum-i. e., its own mesocolic 
layer and four from the great omentum. In the text- 
books the same portion of the colon (A) gets only 
three layers. In other words, the upper surface of the 
transverse colon has five serous layers-its lower sur- 
face but one. 
With the foramen of Winslow and the great omen- 
tum the chief difficulties of the peritoneum disappear, 
though some points about the colon are not without 
Fig. 10. 
interest. Of the early embryonic condition of the intes- 
tine, that depicted in Fig. 10 is very frequently given. 
I employ it to call attention to the growth of the bowel SEROUS COVERING OF THE VISCERA. 
223 
and its extending forward, thus elongating its mesen- 
teric attachment; soon a twist takes place of the lower 
part of the intestine upon the upper, by which, that which 
is to constitute the csecal end of the colon comes to lie 
in the right iliac fossa, and that which is to constitute 
the sigmoid flexure to lie in the left iliac fossa, and the 
entire colon to form three sides of a square where it be- 
comes attached to the periphery of the abdominal cavity 
(Fig. 11). This twist of the colon to the right, like that 
Fig. 11. 
of the stomach, is probably due to the change of position 
of the liver-i. e., to the right, and to the early fixation of 
the duodenum. At first the colon has, in every part, 
quite an extensive mesocolon, but by degrees this 
shortens until its ascending and descending portions 
are so closely bound in their respective lumbar fossse 
that they lose their serous covering, except in front; not 
so with the transverse colon; its mesocolon is usually 
long enough to permit the colon to reach the body wall 
in front, and to lie just beneath the great border of 
the stomach. This length of the transverse mesocolon 
serves many purposes ; it forms a roof to the entire 
bowel below the duodenum; it also gives great freedom 
of motion to the transverse colon, so that it can accom- 
modate itself to the varied changes in the conditions of 224 
ALLIS, 
the surrounding viscera. In cases of stricture of the 
rectum and consequent extreme distention of the large 
intestine with gas, the middle of the transverse colon 
will often descend to the pelvis, giving the entire 
colon an appearance not unlike the capital letter M. 
I am careful to repeat that the colon has its own 
transverse mesocolon, and that it does not depend 
upon the great omentum for it. The latter is asserted 
by some authors, much to their discredit. As the colon 
twists upon itself, and comes to lie across the duode- 
num, its mesocolon (Fig. 11) furnishes a fixed point for 
the commencement of the mesentery of the small intes- 
tines, which begins at the left border of the second 
lumbar vertebra and proceeds downward to the right 
iliac fossa. This base of the mesentery is scarcely 
longer than six inches, while its periphery, encircling 
the bowel, is nearly eight times the length of the body. 
The coiling of the small intestine is greatly facilitated 
by the looseness of its connection with its serous 
envelope. This is especially noticeable in foetuses and 
young infants. In all such cases the entire small intes- 
tine can be drawn from its peritoneal sheath, leaving 
the mesentery unbroken save at the opening through 
which the gut is extracted. By the disposition of the 
small intestine within the borders of the colon a mutual 
benefit accrues. The small intestine is placed beneath 
the abdominal muscles, where every exertion will excite 
peristaltic motion, and therefore aid in carrying its 
contents to the colon. By the same muscles the same 
pressure through the small intestines is exerted upon 
the colon, which cannot escape its influence, and as its 
contents cannot revert to the small intestine they are 
carried forward to their destination. SEROUS COVERING OF THE VISCERA. 
225 
It is interesting to examine the various portions of 
the alimentary canal with reference to their relation to 
the serous covering. Thus, from the oesophagus, it has 
been stripped oft* and appropriated by the lungs and 
heart. The latter had need .of it as organs of ceaseless 
activity; while the oesophagus, being but a mere con- 
duit for articles of alimentation, was better off without 
such a covering; but with the stomach it is far other- 
wise: this is more than a receptacle'for food; it must 
have capacity for liquids, taken often with great ra- 
pidity and in great quantity. To be equal to all'the 
emergencies that the stomach may be called upon to 
meet, the great omentum is formed (Fig. 8, B, C, D). 
This permits large bloodvessels to course along both 
curvatures of the stomach, a circumstance only present 
in this organ, and make a vascular network about it, 
which, in case of thirst, will abstract water from the 
stomach with wonderful rapidity. The disappearance 
of covering from the transverse duodenum, the forma- 
tion of the mesentery and the mesocolon, have already 
been touched upon. 
It remains briefly to speak of the portal circulation. 
The blood to the digestive tract comes from the abdom- 
inal aoita from three principal vessels, viz., the coeliac 
axis, the superior and inferior mesenteric arteries. Now. 
the aorta lies behind the peritoneum in its entire course, 
and consequently receives no covering from it; not so 
with the three large branches to the digestive tract. 
These, as well as all the blood from the digestive tract 
to the liver, lie throughout their entire course between 
two layers of serous membrane. The coeliac axis 
(Fig. 12) supplies the liver, stomach, pancreas, spleen, 
and duodenum, but none of its vessels pass of 226 
ALLIS, SEROUS COVERING OF THE VISCERA. 
the stomach to reach their destination. A peculiarity 
in the venous blood from the digestive tract is, that it, 
like the arteries, passes behind the stomach. If the 
mesenteric vein had passed beneath the duodenum, it 
would have been obliterated; if it had passed in front 
of the pancreas, the weight of the omentum would have 
dragged upon it; had it gone in front of the stomach, 
Fig. 12. 
Diagrammatic. Introduced to show the course of the venous blood of the 
digestive apparatus to the liver. 
then a full stomach or acts at vomiting would have 
interfered with its function; in its course to reach the 
portal vein it passes between the pancreas and duo- 
denum, and under the stomach ; the portal vein thus 
comes to lie anterior to the foramen of Winslow; but 
neither it nor the hepatic arteries have any hand in 
creating this foramen, as stated in at least one promi- 
nent text-book.1 
1 For a description of the formation of the tunica vaginalis and descent ot the 
testes, as related to the peritoneum, see Annals of Anatomy and Surgery, Nov. 
1883. CHAPMAN, SEROUS COVERING OF THE VISCERA. 
227 
[ After the reading of the preceding paper:-] 
Dr. Chapman said : I have listened with much interest to the paper 
of Dr. Allis. There are several facts which confirm what he has said 
about the disposition of the peritoneum. Examining the abdomen of 
an adult, it seems that the disposition of the peritoneum is as given in 
Gray, but that is not so. If the abdomen of a foetus at eight months 
or term is opened, the peritoneum will be found as described by Dr. 
Allis. 
If you take a net, making one sack, and spread it over a number of 
objects on a table, representing the liver, kidneys, and the other ab- 
dominal organs, and then invaginate the net, we shall have a repre- 
sentation of the greater and smaller sack of the peritoneum. There is 
then no difficulty in placing the net over the different objects, as the 
peritoneum is disposed over the viscera. 
In the body of the foetus, you can raise the sack and demonstrate 
that it consists of four layers. By inserting a pipe in the foramen of 
Winslow, the sack may be distended. If the sack is raised, you find 
the transverse colon covered as it is in this drawing. In the adult, 
the sack becomes adherent to the covering of the transverse colon, and 
we then have five layers, instead of four as shown in Gray. 
In almost all the mammalia (at least of several hundred that I 
have examined), I find the peritoneum disposed as here represented. 
The transitory stage through which the peritoneum passes in the 
human animal, is retained as a permanent condition in the lower 
animals. This is true, not only of the peritoneum, but of the other 
organs of the body. In the other mammals, this sack can be lifted, 
leaving the transverse colon covered with peritoneum. The only ex- 
ceptions that I have found, have been in the gorilla, the chimpanzee, 
and ourang. In all these animals, the transverse colon adheres to the 
greater sack, as it does in man. It, therefore, seems that the facts of 
comparative anatomy, and the facts of embryology thoroughly bear 
out what has been said by Dr. Allis. It is no longer a question of 
theory, but it is exactly what is represented here. 
There are one or two other points in connection with the develop- 
ment of the peritoneum that may be mentioned. The peritoneal and 
pleural cavities may be regarded as simply large lymphatic sacks. 
If milk is thrown into the peritoneal sack, in a few hours the lym- 228 
PANCOAST. SEROUS COVERING OF THE VISCERA. 
phatics of the diaphragm will be filled and the fluid may even pass to 
those of the pleura. This will serve to explain the fact why you may 
coincidently find the peritoneal cavity filled with fluid or pus, and at 
the same time the pleural cavity. The peritoneum is nothing but a 
large dilated lymphatic sack placed around the abdominal organs. 
In regard to the development of the pleura, Dr. Allis begins by 
saying that there is one general cavity. It seems to me that it would 
be better to consider the pleura as consisting of two sacks in the be- 
ginning, and that the lungs are thrust between these, thus producing 
the same thing as represented here. This will account for there being 
no communication between the two. There is one reason for not con- 
sidering the pericardium as a part of the pleura, and that is, the fact 
that the phrenic nerve runs between the pericardium and pleura. 
Another point as regards the peritoneum is one which is of interest 
to surgeons, and that is, that the viscera being pushed forward, carry- 
ing the peritoneum before them, can have no peritoneum behind. 
Hence in hernia occurring between the quadratus lumborum and latis- 
simus dorsi muscles, there will be no peritoneal covering to the bowel. 
Dr. W. H. Pancoast : I agree with Dr. Allis in his demonstra- 
tion of the development of the peritoneum. While really the diagram 
in the text-book of Gray may not be correct, yet practically it is 
correct. When these different layers become adherent, they practi- 
cally constitute one layei* and they may be demonstrated as one layer. 
I may add one interesting fact. Year after year in making this 
demonstration, I have represented the layers of the peritoneum by a 
mosquito-netting, and this has shown that the ordinarily described 
arrangement of the peritoneum is not correct. You cannot take a 
membrane like that and take the different viscera and place them in 
the proper position and demonstrate the great omentum at the same 
time. You must pin up the netting to represent the omentum. 
I might mention one point in regard to the pericardium. We find 
that this membrane is firmly attached to the diaphragm and extends 
upward, surrounding the heart and passing up over the great vessels 
and mingles with the deep fascia. In violent exercise, as in athletes, a 
full inspiration precedes the effort and this might seriously affect the 
heart, if it were not for this support passing through the cavity of the 
thorax and to the deep fascia of the neck.