The Relations of the Anatomy 
of the Perineum to the 
Mechanics of its 
Laceration. 
BY 
EDWARD REYNOLDS, M.D., 
BOSTON, MASS. 
REPRINT FROM VOL. XVL 
<$pnecofogicat Srattsadions. 
1891. THE RELATIONS 
OF THE 
ANATOMY OF THE PERINEUM 
TO THE 
MECHANICS OF ITS LACERATION. 
BY 
EDWARD REYNOLDS, M.D., 
BOSTON, MASS. 
REPRINTED FROM THE 
TRANSACTIONS OF THE AMERICAN GYNECOLOGICAL SOCIETY, 
SEPTEMBER, 1891. 
PHILADELPHIA: 
WM. J. DORNAN, PRINTER. 
1891. THE RELATIONS OF THE ANATOMY OF THE 
PERINEUM TO THE MECHANICS OF 
ITS LACERATION. 
By Edward Reynolds, M.D., 
Boston. 
In the midst of all the study which has been lavished upon 
the repair of the lacerated perineum, the profession has, I 
think, been too easily satisfied with the statement that its pro- 
duction is due to the strain which is put upon the pelvic 
fascia during the expulsion of the head, and too little occupied 
with the study of the exact way in which this tension is 
exerted. It is to the result of some personal observations on 
this point that I wish to call your attention to-day; and, on 
account of the necessity for brevity which is imposed upon us 
by the conditions of our meeting, I wish to confine myself to 
but two points : first, an attempt to classify the various forms 
of tear which are constantly met with, into an intelligible 
system ; and, secondly, an effort to explain the occurrence of 
these variations on anatomical grounds. Incomplete as my 
conclusions still are, I feel justified in presenting them, on the 
ground that they are the result of more than five years of 
constant study in a clinic in which nearly one thousand women 
are delivered annually by a succession of previously inexperi- 
enced men, with the natural result of a very large proportion 
of lacerations. 
At the beginning of my observations I was impressed with 
the belief that though the crescentic intra-vaginal tear first 
described by Dr. Emmet in the Transactions of this Society 4 
MECHANICS OF PERINEAL LACERATION. 
for 1883, did frequently occur, a longitudinal tear in the 
median line was much more common. At first I found a 
large proportion of these longitudinal tears; but, with increas- 
ing experience and better methods of observation, they became 
steadily more and more rare, until during the last two years I 
have failed to find a single instance in which I could persuade 
myself that this form of laceration was present; and I now 
believe that those of my earlier cases in which I thought it 
occurred, were mere instances of mistaken observation, due to 
the use of imperfect methods, and to an insufficient under- 
standing of the alterations of shape which result from the 
muscular retraction of the distended tissues; and that, if the 
case is properly studied, such tears will be found to be at least 
extremely rare, and perhaps non-existent. 
The method which I have of late pursued in the study of 
these cases is no more than that which is ordinarily adopted 
in secondary operations-i. e., first, to evert the posterior wall 
of the vagina through the vulva by the insertion of two 
fingers into the rectum, and then, after carefully sponging 
the tissues, to draw them together with tenacula, first in one 
direction and then in another, until the position in which they 
unite with the most natural appearance and with the least 
degree of tension is found ; but, simple as it is, it requires for 
success a degree of carefulness and an amount of time which 
I am inclined to think is seldom given to the subject during 
the busy moments which ordinarily succeed delivery. 
After studying carefully, in this way, a very considerable 
number of cases, I am inclined to think that all lacerations 
of the vagina and perineum can be conveniently classified as 
mere modifications of one type-form which is in itself not 
infrequent. 
This type-form is essentially that which was originally 
described by Dr. Emmet, and is illustrated diagrammatically 
in Figs. 1, 2, and 3. It is made up of three essential por- 
tions : first, two parallel, longitudinal tears of the vaginal 
walls, united at their lower extremities by a transverse tear, EDWARD REYNOLDS. 
5 
which is the second and most essential portion of the typical 
laceration. The plane of this tear is approximately at right 
angles to the axis of the vagina, and its situation is just within 
the constricting band formed by the superficial muscles. To 
these two portions are added, in the typical laceration, a 
median tear of the superficial tissues-i. e., of those which lie 
below the plane of the transverse tear. This type-form in its 
entirety is, perhaps, of more common occurrence than any 
other single variety; but, in the majority of instances, it is 
modified by the diminution or disappearance of some one or 
more of its component parts. 
Although the variety which is found is extremely great, I 
wish, at this point in my paper, to briefly call your attention 
to a few illustrations of the more common types. I am for- 
tunate in being able to use for this purpose some reproductions 
of a set of diagrams which were drawn from life by one of my 
house officers at the Boston Lying-in Hospital, entirely inde- 
pendently of any suggestion from me, and which are conse- 
quently free from any individual bias which I might personally 
have entertained. Fig. 2 represents one of the most complete 
instances of the type-form which I have ever seen; Fig. 3, 
less symmetrical, is a more ordinary instance. Fig. 4 shows 
diagrammatically a somewhat less common tear in which only 
the transverse crescentic laceration has been produced. With 
Figs. 5 and 6 we pass to the form which is perhaps second in 
frequency, in which one of the lateral vaginal tears has failed 
to occur, and the corresponding lateral half of the crescent is 
but imperfectly developed. This form is one which is, I 
think, commonly but erroneously considered a median longi- 
tudinal tear, because the muscular retraction of the tissues, in 
the swollen state which follows parturition, not infrequently 
distorts the tear into a shape from which its true characteris- 
tics can only be elucidated by very careful observation. Yet, 
even in these instances, the methods of repair which result 
from the approximation of the edges of the torn surfaces in 
such a way as to restore them to the situation from which they 6 
MECHANICS OF PERINEAL LACERATION. 
are here represented as having been torn apart, has seemed to 
me to yield invariably much better results than can be attained 
by their division into two symmetrical longitudinal halves, 
and their approximation from side to side. 
Fig. 7 represents an instance of a very rare form of tear, 
of which I have seen only four cases. lu these tears oue of 
the lateral lacerations is alone represented, and is prolonged 
downward through the external tissues as a lateral tear well 
to one side of the anus. 
In each of these four cases this form of tear was produced 
by an unusually rapid delivery occurring at an early stage of 
labor without previous preparation to the soft parts. Three 
of the deliveries were operative; one was the birth of a five- 
pound child by a powerful spontaneous labor at seven months. 
In each instance the tear began in the fornix as a continuation 
of a lateral laceration of the cervix. 
Another not uncommon tear of which I have no illustra- 
tion is the transverse rent with both lateral prolongations, 
but without the superficial tear. 
Although many lacerations present at first sight a very 
irregular appearance, I believe that with a little study all the 
forms which present themselves can be shown to be but sub- 
modifications of these five common variations from the type 
form. 
If this attempt at classification of these lacerations is suffi- 
ciently natural to be of value, and if my belief in the sub- 
stantial uniformity in the situation and anatomical relations 
of all perineal tears is well founded, this uniformity must 
depend on some definite anatomical arrangement which de- 
termines the situation either of the weakest point of the 
tissues, or of the place of greatest tension along these par- 
ticular lines. Such an explanation seems to me to be readily 
apparent from an examination of the arrangements of the 
fascia and muscles which form the pelvic floor; but I think 
that it is less likely to be gained by a review of our well- 
established knowledge of the anatomical entities in the non- EDWARD REYNOLDS. 
7 
parturient state than by a consideration of their functional 
behavior during parturition. 
In this connection it is of course unnecessary for me to 
burden the Society with a rehearsal of the ordinary details of 
pelvic anatomy; but I may perhaps be pardoned for briefly 
enumerating the special points upon which this explanation 
rests. The pelvic floor, when considered in connection with 
parturition, seems to me to be functionally divisible into two 
strata, a superior or deep, and an inferior or superficial layer, 
and I think this division a natural one, because I believe that 
the mechanical action of these two strata during dilatation can 
be shown to be not only wholly distinct, but even antagonistic 
to each other; and that it is this antagonism which results 
in the production of lacerations, whenever the elasticity of 
the tissues is deficient. 
The upper or superior of the two strata into which I 
wish to ask you to divide the pelvic floor, is composed of the 
two layers of the superior pelvic fascia, with the enclosed 
three divisions of the coccygeus muscle, and is further rein- 
forced anteriorly by the addition of the deep perineal fascia 
(the triangular ligament of the male) with the enclosed con- 
strictor urethrae muscles. This sheet of tissue, taken as a 
whole, forms the floor of the abdominal cavity in the true 
pelvis, and is broken only by the apertures through which 
the tubular viscera pass. It is, however, much stronger at 
the sides of the pelvis than in its median portion, where it 
is of somewhat feeble construction, even when uninterrupted 
by the viscera (Fig. 8). It is firmly attached to the lateral 
edges of the vagina at the point at which this passage pierces 
it, and prolongations of its fibres pass behind the vagina to 
form a great part of the upper portion of the so-called peri- 
neal body. Its line of force is represented, diagrammatically, 
in blue, in Fig. 9. 
The inferior layer of the pelvic floor is composed, anteriorly, 
of the bulbo-cavernosus, and the transversus perinei muscles, 
with the two layers of the superficial fascia which enclose 8 
MECHANICS OF PERINEAL LACERATION. 
them; and posteriorly, of the sphincter ani and the fibres 
which bind this muscle to the coccyx. It must be remem- 
bered that anteriorly, as is stated by Ranney, the superficial 
perineal fascia is a somewhat firm sheet of non-elastic tissue, 
which is especially strong at the points where it is reflected 
over the transverse muscle; but that in its posterior portion 
it is so slightly developed that the sacral half of this lower 
layer is in reality composed of but little more than the 
yielding sphincter ani muscle, the peri-rectal fat, and the 
external skin. It is drawn in yellow in Fig. 9. 
To arrive at the conclusion as to the etiology of perineal 
lacerations, to which I wish to lead you, it is now necessary 
to examine separately the action of these two layers during 
parturition. 
When the head passes the cervix and meets the resistance 
of the pelvic floor, it must first dilate the vagina and the 
opening in the superior layer, by pushing to each side the 
stronger lateral portions of the superior fascia, and by putting 
upon the stretch the weaker central part, at the same time 
stretching the whole structure downward. This process is 
represented diagrammatically in Figs. 10 and 11. In the 
former, which is supposed to be a section through the supe- 
rior pelvic fascia and vagina, at right angles to the axis of 
the passage, the stronger lateral portions of the fascia are 
shown as ending somewhat abruptly in their attachment to 
the vaginal walls. According to a well-known phenomenon 
of mechanics, it is probable that if the supporting sling which 
is so produced is to tear, the separation will occur at one of 
the two points where the stronger tissues join the weaker, i.e., 
at A or B, and would then naturally result in the production 
of the first portion of the typical tear; i.e., the two longitu- 
dinal lateral prolongations of the crescent. 
As the head advances, either by dilatation of this structure 
or by laceration of the vagina, it follows from the obliquity 
of the vaginal axis to the fascia, that the posterior portion is 
put more and more upon the stretch by the downward and EDWARD REYNOLDS. 
9 
forward progress of the presenting part; while its anterior 
half is, upon the contrary, carried up and forward and 
crowded against the pubes by the same process, as is rep- 
resented in Fig. 11, in which the posterior half of this seg- 
ment is represented in blue and the anterior half in red. If 
then the anterior half of this sheet of tissue is relaxed and 
the posterior half is stretched, by the process of dilatation, it 
is evident that the action of the whole may be considered as 
limited to that of the posterior segment, i. e., to a crescentic 
hood of fascia, which is eminently well fitted to resist that 
downward and forward movement of the fourchette with 
which we all familiar during the stage of expulsion. If now 
this posterior segment could slip backward over the head, its 
tension would be relaxed and its resistance to the process of 
dilatation annulled, and it is the analysis of the force which 
prevents this, which is the next step in my argument. 
The merest glance at the external surface of the perineum 
during the advance of the fourchette at the end of labor, will 
demonstrate to the most careless observer that the elasticity 
of the sphincter ani is so great that the restraining powers of 
the superficial layer of the perineal floor are limited almost 
entirely to the action of its anterior half. If, now, it is re- 
membered that the efficient anterior segment of the superficial 
fascia is connected to the efficient posterior segment of the 
superior layer only by the comparatively feeble indifferent 
connective tissues of the perineal body, as is indicated dia- 
grammatically in Fig. 9, I think it will be conceded that the 
point of least tensile strength is probably situated at this 
point, and that the diagrammatic view ofthe restraining forces 
which is given in Fig. 12 is not unfair. 
In this figure, the passages are supposed to be already partly 
dilated and the perineum is bulgiug slightly before the head, 
the escape of which is still restrained by a hood of tissue, 
whose action may be summarized as that of a Y-shaped band, 
formed by the union of the posterior segment of the superior 
fascia with the V-shaped superficial fascia. If, when the 10 
MECHANICS OF PERINEAL LACERATION. 
head is in this situation, the tissues of the perineal body are 
supposed to yield along the black line A in Fig. 12, there 
will be an immediate gain in the length of the restraining 
band, which gain is exactly represented in Fig. 13, in which 
the dotted black line is an exact reproduction of the position 
of the colored lines in Fig. 12, while the colored lines are 
allowed to sink downward and forward by just the amount 
which is gained by such a separation, which is, of course, the 
transverse portion of the typical tear. 
These separations having taken place, the head is restrained 
only by a thin hood formed by the superficial tissue, which, 
unless it is extremely distensible, will then tear along the 
median line, both because that is the weakest point of the 
fascia, on account of the decussation here of the fibres from 
the two sides, and because of the lateral tension exerted by 
the strong bands of fascia which cover the superficial trans- 
verse muscles. This tension is excellently demonstrated in 
Savage's Anatomy of the Pelvic Organs, and has long been 
received as a sufficient reason for the median situation of the 
superficial rent. 
These explanations seem to me to be sufficient to account 
for the frequent appearance of the type-form, but even if they 
are correct to this extent, much yet remains to be said on the 
reasons for the variation of other tears from this form, and 
upon the precautions and processes which should be adopted 
for their prevention and repair. On these points my observa- 
tions are, however, as yet rather too incomplete for full and 
satisfactory publication. I can only say that those tears in 
which the longitudinal element predominates, as in Figs. 5, 
6, and 7, seem to be always the result of rapid labors with 
powerful expulsive forces; while those in which the trans- 
verse tear is extensive and the other elements of the type are 
ill-developed, as in Figs. 3 and 4, follow as a rule upon 
extremely slow second stages. 
I am inclined to believe that when the second stage has 
been extremely rapid, the lateral wings of the tear are the EDWARD REYNOLDS. 
11 
primary element in its production; and that the existence of 
even one deep longitudinal tear frequently affords sufficient 
space to permit the expulsion of the head without the produc- 
tion of a well-developed transverse rent, as in Figs. 5, 6, and 
7. On the other hand, when the tear follows a very slow 
expulsion, it is probable that the transverse rent is the primary 
element, and that its lateral prolongations, which are then 
seldom well marked, are merely secondary, and are the results 
of the momentary rapidity of progress which is permitted by 
the appearance of the transverse tear. 
When only one of the upper longitudinal tears is well 
marked, it is usually found upon the side to which the fore- 
head was originally directed. How far this last fact may be 
due to the friction of rotation remains to be determined, and 
the field for observation of the relative frequency of the 
several forms, in accordance with the varying presentations 
and with the differing positions X)f the head, may also afford 
another interesting point for observation, to which I hope to 
devote some study in the future. 
It is at once apparent, if these views are correct, that the 
essential element in the repair of all lacerations must consist 
in the insertion of such sutures as will draw the torn edge of 
the upper layer downward and those of the lower upward, 
until they meet at the central point from which they were 
originally separated; but the importance of such sutures is so 
well understood, and the choice of methods for their insertion 
has been so frequently discussed, that I have thought it best 
to leave that point untouched, and to restrict myself to-day 
entirely to the anatomical side of the subject. Fig. 1. 
Diagram of the type-form. 
Fig. 2. 
Fig. 3. Fig. 4. 
Fig. 5. 
Fig 6. 
Fig. 7. Fig. 8. 
Diagrammatic view of the superior pelvic fascia, from above. Fia. 9. 
Diagram of the restraining forces. 
The force of the upper layer of fascia is represented by the blue line, that 
of the lower layer by the yellow lines. Diagrammatic section at a right angle to the axis of the vagina. 
The superior pelvic fascia is represented in blue, the vaginal walls in red. 
The black lines at A and B show the probable position of the lateral tears. Fig. 11 
Diagram of the relations of the two segments of the superior pelvic fascia 
during full dilatation of the parturient canal. 
The position of the posterior segment is shown by the heavy blue line, that 
of the anterior segment by the heavy red line. The broken colored lines 
show the lines of tension in their respective segments. The broken black lines 
show the relations of the posterior segment to its normal position. Fig. 12, 
Diagram illustrating the position of the restraining tissues at the moment 
when the perineum first bulges before the head. 
The posterior segment of the superior fascia is drawn in blue, the anterior 
segment of the superficial fascia in yellow. The black line A shows the situ- 
ation of the probable transverse rent through the indifferent connective tissue 
of the perineal body. Fig. 13. 
Diagram illustrating the position of the restraining tissues after the occurrence 
of the transverse tear. 
The broken lines are exact reproductions of the position of the colored lines 
in Fig. 12. The position of the colored lines shows the amount of space gained 
by a separation to the extent of the black line A in Fig. 12.