Conditions Underlying the Infection 
of Wounds. 
By William H. Welch, M.D. CONDITIONS UNDERLYING THE INFECTION OF 
WOUNDS : 
Including a Discussion of Disinfection with Reference to Treatment of Wounds, 
of the Relation of Bacteria to Suppuration, of the Resistance of Tissues to 
the Multiplication of Bacteria, and of the Effects of Antiseptic Agents on 
Wounds. 
BY WILLIAM H. WELCH>/m.D., 
Professor of Pathology, Johns Hopkins University. 
That the presence of certain kinds of bacteria is an essential condi- 
tion of wound infection is a fact so well established, and so generally- 
recognized, that no discussion as to this point is likely to arise in this 
assembly. The practical results of the application of this doctrine to 
the management of wounds are the most eloquent testimony to its life- 
saving truth. The recognition of this truth, even before its complete 
demonstration, and the introduction of methods of wound treatment 
based upon it, will remain the immortal merit of Lister, even although 
every detail of his treatment be replaced by measures found by riper 
experience to be better suited to the purpose. 
The simple conception which was the basis of early antiseptic proced- 
ures in surgery—that a wound to which bacteria gain access becomes 
infected in the same way as a sterilized infusion of meat undergoes 
putrefaction when a single suitable germ enters—has been greatly 
modified. Furthermore, it is found that the traumatic infections pre- 
sent their own peculiar problems, which must be studied by them- 
selves, and cannot be solved by analogies drawn from observations of 
other specific infections, such as anthrax and the septiciemias of the 
lower animals. The study of wound infections involves the considera- 
tion of many varying and often complicated factors, relating both to 
the agents of infection and to the individual exposed to infection. 
Without consuming time with any historical review, I shall proceed 
at once to indicate some of the questions which are of especial import- 
ance, and which may be profitably considered in this discussion. 
What are the microorganisms concerned in the infection of wounds, 
and how do they act ? 
How are we to explain the great differences in the effects produced 
by the pyogenic bacteria, their apparent harmlessness under some con- 
ditions, their fatal virulence in others ? CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
Wliat are the ways by which bacteria gain access to a wound ? 
How often are bacteria to be found in wounds treated antiseptically 
or aseptically ? What are the characters of these bacteria and whence 
do they come ? 
What are the best means of surgical disinfection ? 
No attempt can be made on this occasion to treat exhaustively a 
single one of these questions. My aim will be to present a brief sur- 
vey of our knowledge, as well as of the defects in our knowledge, 
concerning some of the more important questions ; to offer some results 
of personal observation and experiment, and to indicate in conjunction 
with the co-referee the lines which this discussion may follow. It does 
not seem necessary, indeed it would be presumptuous before this audi- 
ence, to dwell in detail upon established facts which are the common 
property of all who are interested in the subject. 
So far as personal observations and experiments are concerned, I 
may say that these have been made in the Johns Hopkins Hospital 
and Pathological Laboratory, and that I am indebted to several of my 
colleagues and co-workers there for work which they have done either 
independently or under my supervision. I take this opportunity of 
acknowledging especially the assistance of Drs. Halsted, Abbott, 
Ghriskev, Robb, and Howard. 
This paper will relate only to the ordinary inflammatory and sup- 
purative traumatic affections, and not to specific wound infections, 
like tetanus, diphtheria, and hospital gangrene. 
It would be wearisome on this occasion to consider or even enum- 
erate all the bacteria which have been found in suppurative and other 
traumatic inflammatory processes, so that in this connection I shall 
touch only upon a few points drawn chiefly from our own experience. 
We have found the staphylococcus pyogenes aureus far more fre- 
quently than any other species of bacteria in furuncles, abscesses, 
osteomyelitis, and other forms of suppuration. The prevalence in 
Strassburg, according to Levy, of the staphylococcus pyogenes albus 
in all kinds of suppurative inflammation (excepting furuncles), raises 
the question whether there may not be differences in different regions 
as regards the relative frequency of the various kinds of pyogenic 
cocci. 
We have, however, met a white staphylococcus in a certain class of 
cases with great frequency, viz., in small stitch abscesses, and in the 
slighter grades of inflammatory disturbance of wounds treated anti- 
septically or aseptically. This coccus has been often seen by Bos- 
sowski and others in the same class of wounds, and has been hitherto 
identified with the staphylococcus pyogenes albus of Rosenbach, but CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
3 
we have some hesitation in so regarding the micrococcus which we 
have found in the cases mentioned. The staphylococcus pyogenes 
albus is generally described as not differing essentially from the aureus, 
except by the absence of the yellow color in the cultures. This coccus, 
however, differs from the staphylococcus pyogenes aureus in much 
greater slowmess of liquefaction of gelatin and of coagulation of milk, 
and in far less virulence when inoculated into the circulation of rab- 
bits. It may be an attenuated form of the staphylococcus pyogenes 
albus. Until this is settled I propose to call it the staphylococcus 
epidermidis albus, as it is an almost constant inhabitant of the epider- 
mis, as will be apparent when I come to speak of the bacteria in aseptic 
wounds. I shall only add here that this coccus may be present in 
graver suppurative inflammations, but then it has been nearly always 
associated with some other pyogenic organism, or has assumed the 
form of the typical staphylococcus pyogenes albus. 
It is, perhaps, not out of place to remark that the diagnosis of the 
staphylococcus pyogenes albus in cultures from abscesses and other 
suppurations should not be made too hastily. It may require several 
days before colonies of the aureus assume a distinct yellow color. 
Cultures upon potatoes are particularly favorable for the speedy devel- 
opment of this color. There may be marked differences in the rapidity 
with which different colonies of the aureus from the same source and 
in the same plate or roll cultures turn yellow. 
The efforts to differentiate into distinct species the pathogenic strep- 
tococci have thus far met with little success, so that the weight of 
opinion favors the view that the streptococci of erysipelas, of phleg- 
monous inflammations, of septicaemia, of puerperal fever, and of 
various forms of angina, belong to one and the same species. In con- 
formity with the results of most investigators I have been unable to 
determine any decisive differences between the streptococcus erysipe- 
latos and the streptococcus pyogenes. Yon Lingelsheim has recently 
divided the streptococci into two groups : one non-pathogenic, called 
the streptococcus brevis, and the other pathogenic, the streptococcus 
longus. Among the points of distinction which he gives there is one 
which I have not found constant, viz., the invisible growth of all 
pathogenic streptococci upon potato. The behavior is usually as stated 
by Yon Lingelsheim, but I have isolated from a case of phlegmonous 
cellulitis, and from one of septicaemia,, a streptococcus presenting a 
distinct grayish-white growth upon potato. Yon Lingelsheim made 
no use of one of our most valuable differentiating media, viz., steril- 
ized milk colored with litmus. In this medium the common strepto- 
coccus pyogenes produces a very firm coagulum with separation of 4 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
nearly colorless serum, as has been pointed out by E. Frankel. I have 
from different sources isolated pathogenic streptococci which do not 
coagulate milk, although they change the color of the litmus to a lilac 
pink, indicating the production of a certain amount of acid. 
While we must admit that all inflammatory affections caused by the 
pyogenic staphylococci may be produced also by streptococci, and, with 
the probable exception of erysipelas, the reverse also holds true, never- 
theless we are not prepared to accept the statements of Levy and some 
other recent writers that the character of the microorganisms, whether 
staphylococci or streptococci, has no influence upon the symptoma- 
tology of the disease. In conformity with most investigators, we have 
found the staphylococci prone to form circumscribed areas of suppura- 
tion, whereas the pyogenic streptococci have a tendency to produce 
spreading phlegmons with lymphangitis and a wide zone of surround- 
ing redness and inflammatory oedema. It is the custom in our gyneco- 
logical wards to isolate the operative cases in which streptococci are 
found at the time of the operation in the seat of the disease, or subse- 
quently in the wound. 
Of other pyogenic cocci we have found the staphylococcus pyogenes 
citreus twice in abscesses, the staphylococcus cereus albus three times 
in combination with other bacteria in suppurating wounds, the diplo- 
coccus pneumoniae once in a fibrinous peritonitis, and the micrococcus 
tetragenus once, together with the staphylococcus aureus in an abscess 
of the lung. 
The list of bacilli which may be concerned in suppurative and other 
inflammatory affections is much longer than was formerly supposed, 
and is likely to be still further extended. Those which we have 
chanced to encounter are : the bacillus pyogenes fcetidus, twice in 
abscesses containing bad-smelling pus ; the proteus Zenkeri, once in 
pure culture in an ovarian abscess with purulent salpingitis ; a bacillus 
apparently identical with the bacillus emphysematis maligni of Wiek- 
lein, once in an emphysematous cellulitis of the arm ; a bacillus appar- 
ently identical with the bacillus enteritidis of Gartner, in the brain and 
other organs of an infant three months old with cerebral abscess and 
meningitis following operation for imperforate anus ; the typhoid 
bacillus, once in pure culture in an osteomyelitis of the ribs following 
typhoid fever ; a hitherto undescribed bacillus, rapidly liquefying 
gelatin and characterized by rapid solution or peptonization of milk, 
without coagulation, once in peritonitis ; bacilli which could not be 
cultivated in our ordinary culture media, in three cases of suppurative 
inflammation ; and the bacillus coli communis, fifteen times. CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
5 
As these bacillary infections are far less common than those caused 
by micrococci, I shall take some other occasion to describe them, and 
here shall ask your attention only to one group, viz., those associated 
with the bacillus coli communis, as they have been relatively common 
in our experience and present some points of especial interest. 
The first observation of the bacillus coli communis in connection 
with wound infection was made by Tavel, in 1889. This has been fol- 
lowed by a few isolated observations of this organism either in the 
unchanged organs of the body or in suppurations, until recently A. 
Frankel reports its presence in nine out of thirty-one cases of peritonitis. 
I first came across this bacillus in the organs of the body in 1890, in a 
case of multiple fat-necrosis with pancreatitis, which I reported to the 
Association of American Physicians. As in this case diphtheritic coli- 
tis existed, it seemed probable that the lesion of the intestine opened 
the way for the entrance into the circulation of this inhabitant of the 
healthy intestinal canal. This view subsequent experience has con- 
firmed, for I have already isolated in pure culture from the internal 
organs the bacillus coli communis in twenty-five autopsies where some 
distinct lesion of the intestinal mucous membrane existed, such as 
ulceration, diphtheritis, hemorrhage, traumatic injury, and I have 
almost uniformly failed to find it outside of the intestine when no 
demonstrable lesion of the mucous membrane existed. I am therefore 
prepared to say that this bacillus is an extremely frequent invader in 
intestinal disease, although I have no evidence to offer that it does 
any harm except under certain especial conditions. The autopsies 
were in nearly every case made within a few hours after death, 
once in less than one hour. Moreover, the colon bacillus does not 
invade the blood and organs in the process of post-mortem decom- 
position. 
The cases in which we have found the colon bacillus under circum- 
stances pointing to its pathogenic action have been as follows : per- 
forative peritonitis, four cases; peritonitis secondary to intestinal 
disease without perforation, two cases ; circumscribed abscess, three 
cases ; and laparotomy wounds, six cases. 
Its presence, several times in pure culture, in laparotomy wounds 
treated aseptically, although apparently not a source of serious trouble, 
was not a matter of indifference. It was generally accompanied with 
moderate fever and a thin, brownish, slightly purulent discharge, of 
somewhat offensive, but not putrefactive odor. The smooth and rapid 
healing of the wound was interfered with. In some of the cases there 
was evidence of intestinal disorder ; in others this was not apparent, 
and infection from without could not be excluded. 6 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
For the purposes of the present discussion, perhaps the chief interest 
of our observations concerning the colon bacillus is that they furnish 
an illustration of the possible predisposition to infection afforded by 
intestinal lesions, and also give an example of the much-disputed auto- 
infection. 
Of suppurative inflammations which we have examined bacteriologi- 
cally with negative result, may be mentioned abscesses in which the 
bacteria were presumably dead, several hepatic abscesses caused by 
the amoeba dysenteriae, and many cases of suppurating buboes and of 
pyosalpinx, doubtless caused by the gonococcus. 
From the foregoing brief summary, mainly of our own limited ex- 
perience, it is clear that the bacteria which may be concerned in surgi- 
cal infections are many, but the pyogenic cocci of Ogston, Rosenbach, 
and Passet far outrank in frequency, and therefore in importance, all 
other bacteria combined. 
In view of the more practical matters to be considered, we cannot 
tarry long over the theories of suppuration, interesting as just this 
part of the subject is at the present time. 
There are very good reasons to believe that the process of suppura- 
tion serves a useful purpose, and is one of the most important and 
efficient weapons employed by Nature in combating invading micro- 
organisms. Man is not subject to those forms of septicaemia, common 
in many lower animals, in which enormous numbers of bacteria are 
present in the blood ; but, on the other hand, he is particularly sub- 
ject to suppurations and other localized inflammations. The same 
microorganisms, for instance the diplococcus pneumoniae and the bacil- 
lus anthracis, which in many of the lower animals invade the blood 
in large number, cause in man localized inflammations. In the 
experimental septicaemias of lower animals there is an unmistakable 
relation between the extent of the local reaction at the point of inocu- 
lation on the one hand, and the duration of life and number of organ- 
isms present in the blood on the other. The more extensive and 
intense the local reaction, the fewer, as a rule, are the bacteria found 
in the blood and the longer the duration of life. There are two varie- 
ties of the swine-plague bacillus, the one kills rabbits in twelve to 
eighteen hours with trifling local reaction and a large number of bacilli 
in the blood, the other produces extensive purulent infiltration around 
the site of subcutaneous inoculation and permits the animals to live 
for several days, and at the autopsy very few bacteria are found in 
the blood. In a susceptible animal the virulent anthrax bacillus pro- 
duces an acute septicaemia, the attenuated bacillus a local abscess. 
There is even experimental evidence that if inoculation of a suscepti- CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
ble animal with the virulent anthrax bacillus be speedily followed by 
inoculation at the same point with a pyogenic organism so that sup- 
puration soon ensues, the development of anthrax septicaemia may be 
warded off and the animal recover. It seems reasonable to infer, 
when the same organism produces in one animal a rapidly fatal septi- 
caemia without local reaction and in another species only a local 
abscess, that the latter animal is protected by the barrier of suppura- 
tion. As is well known, the fatal infections from wounds received at 
post-mortem examinations are not generally those where marked local 
reaction with suppuration appears at the seat of the often trifling 
injury. 
Exactly how abscess formation checks the invasion of bacteria we 
do not know. That bacteria may die, often in a short time, in pus 
both within and outside of the body, has been demonstrated. The 
leading theories are these : leucocytes and other cells act as phagocytes, 
taking into their bodies the bacteria and killing them;—the wall of 
leucocytes and other cells at the margin of an abscess acts as an 
obstacle to the passage of the bacteria into the surrounding tissues ;— 
pus contains chemical substances injurious to bacteria or antidotal to 
their toxic products;—the bacteria starve in pus, not being able to 
assimilate such concentrated food. Something can be said in favor of 
each of these theories, more in favor of some than of others, but none 
is proven and we cannot stop to discuss them here. 
That pyogenic bacteria set up suppuration by chemical substances 
produced by them, seems to be proven. The zone of necrosis around 
these bacteria which can be demonstrated as the first effect of their 
lodgment in the tissues points to such chemical action. Moreover, from 
cultures of pyogenic cocci several chemical products have been ob- 
tained capable of causing suppuration. The most interesting observa- 
tions here are the recent ones of Buchner and his pupils, who have 
found the proteid constituents of many bacteria capable of producing 
suppuration. These bacterio-proteins of Buchner possess in a remark- 
able degree the property of positive chemotaxis, that wonderful quality 
by which certain chemical substances influence the leucocytes to mi- 
grate from the vessels and move actively toward them. A flood of 
unexpected light seems destined to come from the studies of chemo- 
taxis to illumine many dark problems in pathology. 
Since we know that the chemical products and constituents of cer- 
tain bacteria are the direct agencies in the production of suppuration, 
it is less surprising to find that various other chemical substances are 
likewise able to cause suppuration. But important as this fact is for 
the theory of suppuration, the demonstration that these other chemical 8 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
irritants, such as turpentine, nitrate of silver, etc., can cause suppura- 
tion has no especial bearing upon the ordinary suppurations in human 
beings. 
We come now to the consideration of a division of our subject beset 
with difficulties at every turn, but one no less interesting to the sur- 
geon than to the bacteriologist. 
How are we to explain the extraordinary variability in the effects 
produced by the pyogenic cocci ? The specific infectious agents with 
which we first became familiar, the bacteria of the natural and of the 
experimental septicaemias of lower animals, are nearly constant in the 
effects which they produce when inoculated in small quantity into 
susceptible animals. Not so with the pyogenic cocci. Here we meet 
the most puzzling differences. We find the same coccus in the most 
insignificant epidermal pustule which we find in a dangerous phlegmon, 
in osteomyelitis, in pyaemia, in septicaemia, in acute ulcerative endocar- 
ditis. The manifold varieties of puerperal fever, which we now know 
to be a typical wound infection, may be caused by apparently the 
same streptococcus, producing a mild endometritis, a pelvic abscess, 
localized or general peritonitis, pyaemia or the most virulent and 
rapidly fatal septicaemia. 
If we seek an explanation of these things by experimentation upon 
animals, wTe find that in order to get any positive effects at all it is 
often necessary to introduce enormous quantities of our cultures, con- 
taining vastly more bacteria than we can suppose to be concerned in 
the primary infection in human beings. If we inoculate smaller quan- 
tities, these can often be disposed of by the animal without any mani- 
fest symptoms. 
It must be confessed that we stand here before problems many of 
which still await solution, but it will not do to pass them by on this 
occasion without some discussion, although time will permit me to 
touch upon only some of the salient points. 
It is well to bear in mind that the inflammatory infections of wounds 
do not represent specific morbid entities in the same sense as do anthrax 
and typhoid fever, for example. The latter diseases are caused by a 
single specific germ and no other, and present a definite and character- 
istic clinical and anatomical picture, whereas the former correspond to 
the reaction of the body toward a great variety of noxae and offer 
manifold variations in symptoms and lesions. 
The quantity of a culture of the staphylococcus aureus required to 
produce suppuration is not the same for all tissues and all parts of the 
body. A mere trace, a fraction of a drop, of a dilute suspension of 
the yellow staphylococcus in salt solution when injected into the rab- CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
9 
bit’s eye will set up a suppurative inflammation. It takes a larger 
amount of such a suspension to produce an abscess in the loose sub- 
cutaneous tissue of the back than in the dense tissue of the ear of a 
rabbit or in muscular tissue. Enormous quantities, five to ten cubic 
centimeters and more, of the suspension can be injected often without 
effect into the normal peritoneal cavity of rabbits and dogs, as has 
been shown by the well-known experiments of Grawitz and his pupils, 
confirmed by several other investigators. 
It is probable, as claimed by Grawitz, that this difference in the 
behavior of different tissues depends in large part upon the rapidity 
with which the injected material, more particularly the toxic sub- 
stances, are absorbed. But we must also reckon with a predisposition 
apart from this in the normal tissues, for the introduction even of very 
large numbers of staphylococci into the circulation of rabbits is fol- 
lowed by the formation of abscesses only in certain situations, mainly 
the kidneys, heart and certain muscles, although we must suppose that 
the cocci have been carried to all parts of the body. 
That there are variations in the virulence of different cultures of the 
pyogenic cocci is admitted by most writers and may be inferred from 
the discrepant results of various experimenters as regards the amount 
of the culture required to produce suppuration. These discrepancies 
cannot be explained wholly by different conditions of the experiments. 
We have found no especial difference in the result, for example, 
whether we introduced into the peritoneal cavity of dogs cultures of 
the staphylococcus aureus according to Grawitz’s method, or according 
to Pawlowsky’s method, or by a small laparotomy wound. By one as 
well as by another method we were able to inject 5 c.c. and often more 
of a suspension in salt solution of the yellow staphylococcus without 
any appreciable effect. 
In order to test this matter of varying virulence, I have injected into 
the ear veins of rabbits bouillon cultures, forty-eight hours old, of the 
staphylococcus aureus obtained from many different sources. Con- 
siderable variations in virulence were found to exist both in the origi- 
nal cultures and in the successive generations of the same culture. The 
most virulent cultures were one from a beginning furuncle and one 
from a fresh peritonitis. The injection of 0.1 c.c. of these cultures 
caused the death of the rabbit in twenty-four hours with a large 
number of staphylococci in the blood and many necrotic foci in 
the kidneys. After the second generation the virulence was les- 
sened. From a case of hypertrophic cirrhosis of the liver with- 
out suppurative complications a moderate number of colonies of 
the staphylococcus pyogenes aureus were found in cultures made from 10 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
the liver. 1.5 c.c. of bouillon cultures, forty-eight hours old, of this 
staphylococcus were injected without any effect into the ear-veins of 
two rabbits, the animals being still alive eight months after the injec- 
tion. These extremes are met only exceptionally. In the great ma- 
jority of cases we have found the injection of 0.2 to 0.3 c.c. of bouillon 
cultures sufficient to kill the animal in four to seven days with the 
usual abscesses in the kidneys and heart and occasionally elsewhere. 
As a rule there was no material change in the degree of virulence for 
many successive generations, but sometimes without any apparent 
cause there would occur a marked weakening of the virulence either 
after a few generations or after many. 
Our conclusions from these experiments are that while variations in 
virulence occur and may explain in large part the varying results of 
different experimenters operating upon animals under similar condi- 
tions, they are not sufficient in degree or in frequency to afford ade- 
quate explanation of the great differences in the effects of the pyogenic 
cocci upon human beings, or at least can do so only in part. 
How far can we apply to human beings the experiments showing the 
great tolerance of animals toward the pyogenic cocci ? The experi- 
mental evidence upon this point is naturally scanty, but such as we 
possess indicates that man is not equally tolerant, that his tissues re- 
spond more readily by suppuration to inoculation of pyogenic cocci. 
The experiments upon this point of Garre, Bumm, Bockhart, Fehleisen, 
Waterhouse—and these names should not be mentioned without ex- 
pressing the debt of gratitude which science owes to them—are too 
well known to necessitate here any statement of their results. But 
while these experiments appear to indicate less immunity on the part 
of human beings toward the ordinary staphylococci of suppuration—a 
matter, however, wdiich cannot be considered absolutely certain—they 
have also demonstrated that the dilference, if it exists, is only one of 
degree, and that large numbers of the pyogenic cocci may be inocu- 
lated into human tissues without effect. Waterhouse, for example, 
injected under the skin of the abdomen and of the scrotum 0.25 c.c. 
of a suspension of the staphylococcus aureus, made by mixing one 
loopful from an agar culture with 5 c.c. of water, and the result was 
completely negative. 
It is true that undue emphasis should not he laid upon the negative 
results, for there are equally trustworthy experiments with positive 
result, showing that even small quantities of suspensions of the pus 
organisms may cause suppuration, both in animals and in human 
beings. CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
11 
That there exist bacteria capable of producing abscesses regularly, 
even when inoculated in very small quantity into animals, is proven by 
the discovery by Dr. Bolton of a hitherto undescribed bacillus in gar- 
den earth. This bacillus, for cultures of which I am indebted to Dr. 
Bolton, who will hereafter publish its description, produces with cer- 
tainty localized abscesses when inoculated in small amount into the 
subcutaneous tissues of rats, mice, and rabbits. When injected into 
the circulation it causes multiple abscesses in the kidneys, joints, bones, 
and other situations. 
Inasmuch as it is by their toxic products that the pyogenic bacteria 
do injury, it is not surprising to find that it makes a great difference 
in the result whether these bacteria enter the tissues already equipped 
with a reserve force of this poisonous material, or must begin the fight 
unarmed. There is abundant experimental evidence to show that even 
in very small number the pus-producing micrococci will cause suppura- 
tion if they are accompanied, or soon preceded or followed, in sufficient 
amount and concentration by various toxic substances which are 
present in pure cultures of the cocci, and the extent of the inflamma- 
tion bears a relation to the quality and quantity of these substances. 
The experiments and researches of many investigators, such as Hankin, 
Brieger and Frankel, Buchner, Dirckinck-Holmfeld, Grawitz, Leber, 
Dubler, Kronacher, Ribbert, have shed light upon this subject. It 
would seem as if the issues of the battle between the invading micro- 
cocci and the tissues depend often upon the first blow ; and if the 
invader can strike this with the aid of powerful weapons which he has 
forged before he enters, the victory, for the time at least, is with him. 
This matter of accompanying toxic substances is probably of great 
importance in our understanding of the potentialities of the living 
agents of wound infection as they occur under natural conditions. 
Here we have to do, not with pure cultures of the pyogenic cocci, still 
less with those that have been washed in sterilized salt solution or 
water, but with pus-producing organisms which have come from all 
sorts of sources, which have been engaged in very different activities, 
which have been growing under various conditions or have been long 
dormant, and which are mixed with many kinds of bacteria. There 
is proof that under some of these conditions the infectious material 
may possess a degree of virulence with which we are not familiar in 
our artificial cultures. When a strong, healthy man (I have now in 
mind such a case in a medical student) dies in a few days from septi- 
caemia caused by the inoculation of a mere scratch on the finger with 
fluid from a puerperal peritonitis, it is the quality of the infectious 
material which brought about the fatal result, and not any especial 12 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
predisposition of the individual. When we find, as has been done, 
that the peritoneal fluid contains in pure culture the streptococcus 
pyogenes, and that the same organism is present in pure culture in 
the patient accidentally infected with this fluid, the observation is just 
as convincing and clear in its interpretation as if the experiment had 
been made intentionally upon an animal. No number of experiments 
showing that millions of apparently the same species of micrococcus 
can be injected in artificial cultures into the peritoneal cavity of dogs 
and rabbits can do away with the force of such an observation. No 
surgeon, no obstetrician, however strong may be his trust in the de- 
fensive powers of the animal tissues and fluid, would willingly permit 
the smallest particle of that peritoneal fluid to come into contact with 
a fresh wound or the uterus after childbirth. 
The differences in virulence which have been found to exist between 
inflammatory exudates from various sources containing pyogenic bac- 
teria are much greater than those observed in the cultures of the same 
bacteria on artificial media. To explain this, some assume that the 
differences in virulence pertain to the pyogenic cocci as such, that they 
are specifically endowed with different biological attributes ; while 
others think that the varying virulence relates not so much to the 
bacteria as to the character of the toxic substances with which they 
are associated. In favor of the latter view, which is the one advo- 
cated by Bumm, Fehleisen, Chantemesse, and others, may be cited 
such observations as that of Bumm, who found that the injection into 
the peritoneal cavity of a rabbit of a quarter of a drop of fluid from a 
case of acute septic puerperal peritonitis quickly killed the animal with 
peritonitis, whereas the streptococci artificially cultivated from the 
same fluid were much less virulent ; or the experiment of Fehleisen, 
showing that a minimal quantity of an artificial culture of the staphy- 
lococcus aureus added to a little of the clear serum obtained from the 
germ-free zone of inflammatory oedema around a spreading cellulitis 
was capable of producing extensive abscesses, whereas the mixture of 
the same organism with water had no such effect. The question is 
still an open one, and very likely both factors are concerned. 
We have thus far confined our attention to the microorganisms of 
infection, and have said nothing concerning the conditions predispos- 
ing to infection in the exposed individual. Everybody believes in the 
doctrine of predisposition, some more than others. The tendency at 
the present time is certainly not to minimize its importance. Predis- 
position is a term to conjure with. It is often made to explain in a 
vague sort of way things which we do not understand. Nevertheless 
it is a very real thing, and we cannot pass it by here without notice, CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
13 
even if this must necessarily be very brief. Every surgeon knows 
that wounds in some persons do much better than in others, and that 
some kinds of wounds are much more prone to suppurate than others. 
The interesting studies upon immunity which have shed so much 
light upon the nature of predisposition toward some infectious diseases 
have not as yet cleared up the question of immunity against the pyo- 
genic cocci. We know that the healthy tissues can dispose of a cer- 
tain number of these cocci under ordinary circumstances, but how 
they do it we do not know. 
An experiment of Roger is suggestive. He found that the strepto- 
cocci of erysipelas grew as well in the blood-serum of rabbits ren- 
dered immune as in the blood-serum of other animals, but that in the 
former they lost their virulence. As we now know that the protective 
influence of the blood-serum of immune animals consists quite as much 
in the power to destroy the poisons produced by bacteria as in the power 
to kill the bacteria directly—already demonstrated examples of this 
are diphtheria, tetanus, and septicaemia produced by the diplococcus 
pneumoniae—it is not unreasonable to suppose, in the light of Roger’s 
experiments, that this antidotal capacity of the blood and animal fluids 
may be one of the means employed by Nature in disposing of the pyo- 
genic cocci. That these cocci are not directly killed by the extra- 
vascular blood-serum of rabbits, dogs, swine, and human beings, or if 
at all in very small number, we know. On the other hand, this serum 
is an excellent medium for their growth. But if the toxic products of 
the bacteria are destroyed by the fluids of the body, the bacteria can 
do no harm, and are at the mercy of the tissues, as has been shown by 
the recent brilliant researches of G. and F. Klemperer on the diplo- 
coccus of pneumonia. When we have a clearer insight into the nature 
of immunity against the pyogenic bacteria, our understanding of the 
conditions underlying the infection of wounds will be greatly advanced. 
I cannot undertake to discuss all the general conditions of the body, 
such as diabetes, syphilis, alcoholism, anaemia, obesity, typhoid and 
other fevers, Bright’s disease, etc., which have been regarded as pre- 
disposing causes of infection with the pyogenic cocci, some of these 
diseases upon most conclusive clinical evidence. Gartner has recently 
brought forward evidence derived from experiments on animals show- 
ing that general anaemia and hydraemia render easier the infection with 
small quantities of the staphylococcus aureus, and Ribbert has demon- 
strated that the presence of toxic products of the same microorganism 
in the circulating blood favors the development of foci of suppuration, 
a fact which evidently bears upon the pathology of pyaemia and of 
some cases of furunculosis, as well as upon the importance of evacuat- 14 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
ing pus. An instance has already been given of the predisposition to 
infection afforded by intestinal lesions. In dismissing thus hastily the 
matter of general predisposition to suppuration, it will not be under- 
stood that these few words are a measure of the importance of the sub- 
ject. I believe that the surgeon cannot be too thorough in the exami- 
nation, before contemplated operations, of all the important organs 
and functions of the body, and that, wherever possible, he should 
endeavor to put the patient into the best possible condition of health 
before undertaking a severe operation. 
Of very immediate practical interest to the surgeon is a knowledge 
of the various conditions in and about a wound which favor the lodg- 
ment and development of pyogenic bacteria. In a general way it may 
be said that anything which interferes with the integrity of the living 
tissues in a wound is a predisposing cause of suppuration, in case suit- 
able microorganisms gain entrance. Experiments have shown that the 
necroses produced by chemical irritants, such as carbolic acid and cor- 
rosive sublimate, favor the multiplication of the microorganisms of 
suppuration. Dr. Halsted has shown that the irrigation of fresh 
wounds by a solution of corrosive sublimate as weak as 1 to 10,000 is 
followed by a distinct line of superficial necrosis demonstrable under 
the microscope. 
We are not so well informed as to the influence exerted by blood in 
a wound. On the one hand, Von Bergmann and most modern sur- 
geons lay the greatest stress upon prompt and careful haemostasis in 
surgical operations ; on the other hand, Schede has revived and more 
fully developed an old method of treatment, by which a certain class 
of wounds are permitted to fill with a blood-blot, and he and other 
surgeons have obtained rapid and aseptic healing by this method. Is 
this blood-clot, as such, a source of danger in the same sense as dead 
tissue is ? As has already been mentioned, fresh blood-serum does not 
possess any such germicidal power over the pyogenic cocci as it does 
over the typhoid bacillus and many other bacteria. Not being able to 
find any direct experimental evidence upon the point, I have, with the 
assistance of Dr. Howard, made a large number of experiments upon 
dogs. The operations were done with strict antiseptic or aseptic pre- 
cautions. In most of the experiments a cavity was chiselled out of 
bone, and this and the rest of the wound were allowed to fill with a 
coagulum of blood, the conditions pertaining to similar operations in 
human beings being observed. The blood-clot, after its formation, 
was inoculated with a culture of the staphylococcus aureus, either by 
injecting into it a few drops of a fresh bouillon culture or by inserting 
a platinum loop carrying a bit of the growth on agar. The outcome CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
15 
of the experiments was that the so-called organization of the blood- 
clot went on as it does in human beings, and the wounds did not sup- 
purate. The staphylococci survived, at least for many days, in the 
clot, but they did not appear to multiply. This result is in conformity 
with the experience of Grawitz, who found that the aureus lives for a 
long time, although it does not multiply, in blood-coagula outside of 
the body, and that no development of this organism takes place in a 
solidified mixture of equal parts of nutrient gelatin and blood. 
It is not my province to consider the extent of application of the 
method advocated by Schede to human surgery, nor the advantages of 
retaining the material employed by Nature in filling up cavities and 
pockets in fresh wounds as against the insertion of sterilized extran- 
eous material and the obliteration of the dead spaces by deep stitches, 
which, without great care, are liable to strangulate tissue, produce 
undue tension, and interfere with the circulation. 
Where the healing of the wound by the blood-clot method is not 
directly purposed, undoubtedly it is important for its aseptic course to 
check the oozing of blood and to prevent the unintended formation 
and retention of blood-coagula whose presence is not arranged for by 
the surgeon in the management of the wound. That loss of blood, as 
such, predisposes to suppuration has already been mentioned. 
In contrast with the negative results of the experiments just men- 
tioned stands a series of positive ones which illustrate the readiness and 
uniformity with which suppuration of an infected wound ensues which 
contains masses of tissue strangulated by ligature. These wounds were 
inoculated with the same cultures which yielded negative results as 
regards the infection of the blood-clot. 
We have made also a large number of experiments upon dogs by 
ligating portions of the omentum and then injecting cultures of the 
staphylococcus aureus, into the peritoneum. In most of these cases gen- 
eral peritonitis developed, in some localized peritonitis, and in some no 
peritonitis followed the inoculation. 
In order to demonstrate the influence of foreign bodies in favoring 
suppuration, we inserted into the peritoneal cavity of nine dogs pieces 
of potato presenting a growth of the staphylococcus aureus, and in every 
instance general peritonitis developed, although in no case was the 
insertion of similar pieces of sterilized potato followed by peritonitis ; 
and the injection of 1 c.c. bouillon culture of the staphylococcus aureus 
into the peritoneal cavity of twenty-three dogs was not followed in a 
single instance by peritonitis. 
There is a gratifying harmony between the views entertained by 
bacteriologists concerning the power of the living tissues to overcome 16 
CONDITIONS UNDEBLYLNG THE INFECTION OF WOUNDS. 
a certain number of pyogenic bacteria, and the tendency of the modern 
surgeon to respect these tissues more and more, not to destroy their 
vital capacities by the unnecessary application of strong chemical dis- 
infectants, not to bruise them, not to make them too tense, not to 
strangle them, not to suffer the presence in wounds of spaces and 
foreign bodies, which remove bacteria from the influence of the living 
tissues and fluids. 
From this necessarily hasty and imperfect survey of this division of 
our subject, it is apparent that, while there is no reason to doubt that 
the pyogenic cocci are specific agents of infection, the effects which 
they produce depend upon a variety of conditions, such as the source, 
the number, and the virulence of the micrococci, the accompanying 
toxic substances, the part of the body invaded, the readiness of absorp- 
tion, the presence of foreign bodies and of pathological products, the 
general state of the patient, and the condition and handling of the 
wounded tissues. 
In giving due weight to each of these factors one should not forget 
that the infectious material may exist under natural conditions, in a 
state capable of causing traumatic infections just as directly, just as 
certainly, just as independently of predisposition, as infection of a 
susceptible animal takes place with the anthrax bacillus. 
As to the various ways by which pathogenic bacteria may gain 
access to wounds, there is at the present time general agreement of 
opinion that the greatest danger is from contact with infected hands, 
instruments, and other objects. The danger of infection by contact is 
a lesson which has been learned no less by bacteriological workers in 
the laboratory than by practical surgeons and obstetricians. 
The possibility of infection from the air, insignificant as it may he in 
comparison with contact infection, cannot be ignored quite as much as 
some seem inclined to do. The staphylococcus pyogenes aureus has 
been repeatedly found in the dust floating in the air, particularly of 
surgical wards where there are suppurating cases. The streptococcus 
erysipelatos was found by Von Eiselsberg in the air of a ward contain- 
ing cases of erysipelas and the streptococci found by Prudden in the 
air of hospital wards were probably identical with this. When we had 
more confidence than we now have in the power of chemical disinfect- 
ants to destroy all bacteria which might accidentally get into a wound 
during a surgical operation, it seemed proper to disregard the air as a 
source of infection, and no less a surgeon than yon Volkmann could 
say, “ Auch auf einem Abtritte wiirde ich dreist operiren wenn die 
Hande rein waren.” A privy, by the way, is not the most dangerous 
place which he could have selected. CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
17 
Now that our trust in chemical disinfectants for the purpose named 
is shaken, and that what is called aseptic surgery is the watchword of 
the day, I believe that a surgeon who aims at the best will try to have 
the air of his operating-room as free from germs as possible, and will 
have it so constructed that the floor and walls and all that is in it can 
be readily cleaned and disinfected. He will regard the influence of 
currents of air and of commotion in the room in stirring up dust, and 
will not ignore the value of moisture in laying dust and in keeping it 
in its place. These suggestions may seem pedantic when the whole 
tendency of surgery now is to simplify technique and to throw over- 
board unnecessary ballast, but they appear to me to rest upon bacteri- 
ological facts which should not be ignored. 
An example has already been given in this paper of auto-infection 
from the intestinal canal by the bacillus coli communis. The not in- 
frequent invasion of the pyogenic cocci in typhoid fever, diphtheria, 
scarlet fever, and other diseases with lesions of the alimentary tract, 
are probably explicable partly by these lesions opening a passage for 
the bacteria into the circulation, and partly by the predisposition 
afforded by the presence in the body of toxic substances belonging to 
the primary disease. It is known that the pyogenic cocci are often 
present in the alimentary canal. For instance, in a case of perfora- 
tive peritonitis from typhoid fever, recently examined in my labora- 
tory, there were isolated from the peritoneal exudate, not in single 
but in many colonies on the roll cultures, the staphylococcus aureus 
and the streptococcus pyogenes, in addition to the typhoid bacillus, 
the colon bacillus, and an unidentified bacillus liquefying the gelatin. 
Rinne observed in liis experiments on dogs that the injection of 
sterilized putrid fluids together with staphylococci into the peritoneal 
cavity was followed by suppuration of all open wounds, which other- 
wise healed kindly, but that subcutaneous wounds were unaffected. 
The bacteria found in the suppurating open wounds, however, were 
not those injected, but were derived from the air. This is unquestion- 
ably an interesting and important observation, but he goes too far in 
supposing that such wounds, as well as other loci minoris resistentisB, 
may not become infected from pyogenic cocci in the circulation, as 
has been demonstrated by the experiments of De Wildt, Waterhouse, 
and others. 
That even under the most unfavorable conditions of general infec- 
tion a wound may heal by first intention is known to surgeons. In 
one of our cases, for example, the patient soon after an extensive 
operation for removal of a cancer of the breast, developed diphtheritic 
ulcerative dysentery ; and at the post-mortem examination were found, 18 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
in addition to the dysentery, pneumonia, abscess of the lung, and 
fresh pleurisy, with wide distribution of the staphylococcus aureus ; 
neverthless, the operation wound had remained perfectly healthy with- 
out a trace of suppuration. 
Into the burning question of auto-infection from the genital tract in 
puerperal women I cannot enter, save to say that the pyogenic cocci 
seem to be present only exceptionally in the normal tract and do 
not thrive there if intentionally put in, but that they are found in 
pathological utero-vaginal secretions, when, of course, they may be- 
come a source of puerperal infection. The question is a difficult one 
and requires further investigation. This passing mention is only in- 
tended to indicate that the subject belongs to our theme and may 
appropriately enter into the discussion. 
I must beg permission to defer for a few moments the question of 
infection from the skin of the operator and of the patient. 
That an aseptic wound is not necessarily one free from bacteria has 
been known since the early days of antiseptic surgery, the subject 
having been investigated by Ranke, Demarquay, Fischer, Schuller, 
Watson Cheyne, and others. Kiimmell found that pieces of muscle, 
adipose tissue, or connective tissue taken from fresh wounds immedi- 
ately after irrigation during the entire operation with corrosive subli- 
mate solution 1 : 10,000, contained bacteria, and the same result was 
obtained after frequent washing with sublimate solution 1 : 1000. 
The most recent investigation of the bacteria in fresh wounds 
treated antiseptically is by Bossowski, who found, of 50 cases, 10 with 
negative and 40 with positive result from the bacteriological examina- 
tion. He found the staphylococcus albus 26 times, the staphylococcus 
aureus 9 times, the streptococcus pyogenes 2 times, and other organ- 
isms, non-pathogenic, 8 times. He several times found a coccus, 
liquefying gelatin after several days, growing at first white and then 
slowly turning yellow, and incapable of producing suppuration in rab- 
bits or in wounds. This he proposes to call staphylococcus gilvus. In 
every case in which the staphylococcus aureus or the streptococcus 
pyogenes, with or without other organisms, was present, suppuration 
occurred. On the other hand the staphylococcus albus produced gener- 
ally no interference with the healing of the wound, but sometimes it 
caused a little suppuration of the drain canal or a stitch abscess. In 
the majority of cases there was absolute prima intentio. The small 
number of the colonies indicated meagre development in the wound. 
Drs. Ghriskey and Robb have made under my observation the bacte- 
riological examination of 45 laparotomy wounds treated with strict 
antiseptic precautions in the gynecological wards of the Johns Hop- CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
19 
kins Hospital. The method of treatment and the technique adopted 
for obtaining the cultures have already been published by Dr. Robb, 
in the Bulletin of the Johns Hopkins Hospital, July, 1891. 
Of these 45 cases the result was negative in 14, positive in 31, 
or nearly 69 per cent, of the cases. The organisms found were the 
staphylococcus albus 19 times, the staphylococcus aureus 5 times, the 
bacillus coli communis 6 times, the streptococcus pyogenes 3 times, 
once alone and twice in combination with the albus. Of the cases 
with the aureus in only two was there reason to suspect infection from 
without; in the others this organism was present in the seat of dis- 
ease for which the operation was performed. In the first case with 
streptococcus pyogenes the operation was performed for an ovarian 
abscess which contained streptococci in large numbers ; the other two 
cases probably became infected in some way from the first patient. 
In all of the cases presenting the staphylococcus aureus or the 
streptococcus pyogenes the wound suppurated, and, as a rule, the 
general condition of the patient was bad. 
In many of the cases where the white staphylococcus was found 
there was no disturbance in the healing of the wound. This was 
true especially when the organism was found in small number, and 
when it made its first appearance subsequent to the first dressing, 
which took place usually twenty-four to forty-eight hours after the 
operation. In many cases, however, the white coccus was the cause 
of more or less trouble, although rarely of a serious nature. It was 
found to travel down to the bottom of the wound along the side of 
the drainage-tube, and any purulent discharge atti'ibutable to its pres- 
ence was usually confined to the tract occupied by the drainage-tube, 
so that the drainage-tube distinctly favored the invasion and growth 
of the coccus. Sometimes fever without suppuration seemed attribu- 
table to the presence of this coccus. That this organism is the most 
frequent cause of the ordinary stitch abscesses has already been men- 
tioned. 
Under especially favorable circumstances this white staphylococcus 
may cause peritonitis, as is shown by a fatal case of hystero-myomec- 
tomy where at the autopsy a volvulus of the ileum was found. The 
stump and the laparotomy wound both looked healthy, but there was 
a fresh fibrino-purulent peritonitis clearly starting from that part of 
the peritoneum agglutinated to the inner edge of the laparotomy 
wound. This part corresponded to the peritoneum covering the 
twisted ileum. The twist was not so tight as to have produced gan- 
grene or even a marked hemorrhagic condition ; but it had interfered 
sufficiently with the circulation and the nutrition of the peritoneum 20 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
to have rendered this a favorable soil for the growth of an otherwise 
comparatively innocent bacterium, our white staphylococcus having 
been found in pure culture in the inflamed peritoneum. 
The efforts to find out the origin of this very common inhabitant of 
wounds, treated aseptically or antiseptically, have led us to some inter- 
esting and new observations concerning the bacteria of the skin. 
The skin may have all sorts of bacteria upon its surface, but, like 
the mouth and the intestine, in addition to these it has its own dis- 
tinctive bacterial flora. 
If the hands be thoroughly scrubbed with soap and hot water with 
a sterilized brush, or if this be followed by washing the hands in sub- 
limate solution and the mercury be precipitated by sulphide of am- 
monium, the cultures obtained from shapings of the skin so treated 
will generally be found to contain, as the prevailing organism, the 
white staphylococcus, and often this will appear in nearly or quite 
pure culture. 
But the most important point is that this coccus is very often pres- 
ent in parts of the skin deeper than can be reached by any known 
means of cutaneous disinfection save the application of heat. We 
were directed to this conclusion first by experiments on animals. 
Then the observation of the same white coccus in pure culture time 
after time in wounds where every possible antiseptic precaution had 
been taken, pointed to the same deduction. More conclusive evidence 
was afforded by the examination of skin stitches in cases where at the 
time of the operation it was proven that the silk used for stitches was 
sterile and that the surface of the skin after thorough disinfection 
was sterile. The silk sutures when removed were proven both by 
microscopical examination and by roll or plate cultures to contain with 
great regularity the white staphylococcus, often in considerable num- 
ber, often enclosed within leucocytes, and this not only where a stitch 
abscess had formed but also where there was not a trace of suppura- 
tion or visible reaction around the stitch. 
A crucial experiment is the following : The skin is thoroughly dis- 
infected, in the manner to be presently described, so that culture-tubes 
of nutrient agar or gelatin inoculated by scrapings from its surface 
remain sterile. A silk thread, sterilized by steam and proven by 
culture methods to be sterile, is passed one or more times by means of 
a sterilized instrument through the skin, is withdrawn, and at once a 
tube of melted nutrient agar is inoculated with the thread and rolled. 
This amounts to a ready means of making cultures directly from the 
deeper layers of the epidermis and the skin, and is a method applica- 
ble for many purposes to the bacteriological examination of the skin. CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
21 
By this method the presence of the white staphylococcus, often in 
pure culture, has been repeatedly demonstrated in parts of the epi- 
dermis deeper than were acted upon by any methods of disinfection 
of the surface of the skin. So far as our observations extend, and 
already they amount to a large number, this coccus may be regarded 
as a nearly, if not quite, constant inhabitant of the epidermis. It is 
now clear why I have proposed to call it the staphylococcus epidermidis 
albus. It possesses such feeble pyogenic capacity, as is shown by its 
behavior in wounds as well as by experiments on rabbits, that the 
designation staphylococcus pyogenes albus does not seem appropriate. 
Still, I am not inclined to insist too much upon this point, as 
possibly this coccus, which has hitherto been unquestionably identi- 
fied by Bossowski and others with the ordinary staphylococcus pyo- 
genes albus of Rosenbach, is an attenuated or modified form of the 
latter organism, although, as already mentioned, it presents some 
points of difference from the classical description of the white pyo- 
genic coccus. 
We can now understand how, without any flaw in the antiseptic 
technique of the surgeon, this microorganism may be present in 
wounds, and we have a satisfactory explanation of the frequent occur- 
rence of stitch abscesses, although, of course, the inference should not 
be drawn that the white staphylococcus is the only bacterium which 
may be concerned in the production of these annoying complications. 
How much practical importance attaches to the demonstration of 
this coccus in the deeper layers of the epidermis or the glandular 
appendages of the skin I am not prepared to say. The surgeon 
with good technique who does not bother himself about it is not 
likely to be severely punished by the behavior of his wounds. 
Those who put drainage-tubes and other extraneous substances into 
their wounds I think will have to consider it. Dr. Halsted, on 
the basis of researches on the bacteria of the skin and the dif- 
ficulties of complete disinfection of the skin of the patient, has 
abandoned for nearly all wounds the use of skin stitches, the edges of 
the wound being brought together with admirable coaptation by sub- 
cutaneous sutures. The results, both as regards the scar and the 
aseptic healing of the wound, have been most gratifying. Stitch 
abscesses are, of course, avoided by this procedure. 
Another coccus which we have found, although less frequently than 
the white coccus, on the surface and in the deeper layers of the epi- 
dermis, is one already referred to which corresponds to that called by 
Bossowski staphylococcus gilvus. This organism seems to bear about 
the same relation to the pyogenic yellow staphylococcus which our 22 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
epidermal white coccus does to the typical pyogenic white staphy- 
lococcus. 
The staphylococcus pyogenes aureus we have found very frequently 
on the hands of surgeons and their assistants who have to do with sup- 
purating cases of any kind. It may be present in this situation at 
least for several days after contact with surgical cases. It was met 
only exceptionally upon the hands of other persons. 
The demonstration of microorganisms in layers of the skin deeper 
than can be disinfected by present methods suggests that more 
attention than seems now to be customary should be paid to the skin 
of the patient as a source of traumatic infections. It also admonishes 
us to receive with caution the statements recently made concerning 
the elimination in suppurative diseases of staphylococci by the sweat. 
A considerable part of our work has been devoted to the subject of 
surgical antisepsis, and particularly to the disinfection of the skin. 
Dr. Abbott has conducted in my laboratory careful experiments re- 
garding corrosive sublimate as a disinfectant against the staphylococ- 
cus pyogenes aureus, in which many fallacies in previous work on the 
same subject have been pointed out. In view of the time already con- 
sumed, I must ask permission simply to refer to Dr. Abbott’s paper on 
this subject, already published in the Bulletin of the Johns Hopkins 
Hospital for April, 1891. 
The conditions for the efficient action of chemical disinfectants have 
been found to be far more complicated and less easily controlled than 
was formerly supposed, and the substitution, wherever applicable, of 
the simple and certain methods of disinfection by heat, such as have 
been long employed in bacteriological laboratories, is to be com- 
mended. Chemical disinfectants still have their place for many pur- 
poses in the operating room, but their place is not in fresh, healthy 
wounds. 
Thorough scrubbing of the skin with soap and warm water by a 
sterilized brush removes many bacteria, but not all, and it cannot be 
regarded as a satisfactory means of cutaneous disinfection. 
The fallacy in previous work on disinfection of the skin with cor- 
rosive sublimate has been that in testing its efficiency the sublimate 
was not first precipitated by sulphide of ammonium or some other 
alkaline sulphide. If this precaution, to which attention was first 
directed by Geppert, be observed, it will be found that corrosive 
sublimate accomplishes much less than is generally supposed. Our 
revision of the work relating to cutaneous disinfection with sub- 
limate has led to some curious and interesting observations, and to 
results which at first seemed paradoxical. CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
23 
By examining the hands of surgeons who are in the habit of wash- 
ing them daily in solutions of corrosive sublimate, it was found that 
the mercury becomes so intimately incorporated with the epidermis 
that its presence there can be demonstrated by means of sulphide of 
ammonium at least many days after any contact with mercurial solu- 
tions has taken place. Micrococci in the epidermis which have not 
been killed by washing in sublimate solutions, but which have been 
brought into such relation to the sublimate that even after prolonged 
washing of the skin with alcohol and water they will not grow on 
culture media until the skin is washed with sulphide of ammonium, 
may also remain a long time in the epidermis. Hence it may happen 
that prolonged scrubbing of the hands of such persons simply with 
soap and warm water may remove so many superficial bacteria that 
the cultures from the scrapings of the epidermis may show very few, 
sometimes even no, colonies ; whereas, when this is followed by 
washing in sublimate and then in sulphide of ammonium, a much 
larger number of colonies appear in the cultures. This apparently 
paradoxical result, which is obtained only from the hands of those 
who have previously washed them in sublimate solution, has no refer- 
ence to the application of the sublimate immediately after the soap 
and water, but is to be explained by the liberation, by means of sul- 
phide of ammonium, of the bacteria held in check by the mercury used, 
it may be, several days before the experiment. The same result is, of 
course, obtained if the sulphide of ammonium be applied immediately 
after the scrubbing with soap and water. These observations upon 
the persistence of mercury in the epidermis and its long-continued 
inhibition of the growth of bacteria, make it necessary in all work 
upon disinfection of the hands to first precipitate the mercury with 
sulphide of ammonium whenever the experiments are to be made upon 
hands which have been washed in sublimate solutions, even if this has 
occurred a long time previously. Exactly vrhat relation the mercury 
in the epidermis holds to the bacteria which it does not destroy, but 
whose growth in our nutrient media it prevents, we cannot say. We 
may, perhaps, think of these bacteria as enveloped in an albuminous 
combination of mercury. One thing is certain—that, when the subli- 
mate has been as thoroughly washed off from the skin as possible with 
water, or has been applied days before, the nutrient gelatin or agar is 
not rendered unfit for the growth of bacteria by the mere presence of 
the small quantity of mercury carried into it with scrapings from the 
epidermis, for the bacteria which have reached the epidermis after the 
application of the sublimate—and these are often identical with those 
inhibited by the mercury—develop as usual. It is only those bacteria 24 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
which were originally brought into contact with the sublimate in some 
such manner as that suggested which will not grow until after the 
application of sulphide of ammonium, and it is not — as has been 
usually supposed in other observations of a similar kind in disinfectant 
experiments with sublimate—the alteration of the nutrient medium by 
the presence of a trace of sublimate which inhibits the growth of the 
colonies. 
As to the practical efficiency of disinfection of the skin with solu- 
tions of corrosive sublimate, it is to be said that this agent, when 
properly applied, kills most of the bacteria upon the surface of the 
skin. The washing of the skin with alcohol immediately before the 
use of the sublimate increases its efficiency to a marked degree. If 
Fiirbringer’s method be carried out according to the strict letter of his 
directions it yields fair results, but it is not certain. If the mercury 
after employment of this method be precipitated by washing the 
hands in sulphide of ammonium, it will be found that the results are 
much less favorable than would appear by cultures made from the 
skin and under the nails, without the use of ammonium sulphide. It 
is especially the scrapings under the nails and around the matrix of 
the nails which yield positive results when ammonium sulphide is 
used, but often negative ones without this precaution. It need hardly 
be said that in our experiments all of the well-known, although often 
neglected, precautions to insure the full strength of the sublimate 
solutions were observed. 
It may be urged that it is not necessary actually to kill the bacteria 
upon the skin ; it is sufficient if they are rendered incapable of growth, 
and as most of those which are not killed by the sublimate do not 
grow upon our ordinary nutrient media, it is reasonable to infer that 
they will not grow in wounds. This line of argument certainly 
deserves consideration ; nevertheless, there is no positive proof that 
these bacteria will not grow in wounds under some conditions, and 
surely one will feel safer with a method of disinfection which actually 
kills the bacteria. 
I shall not detain you with the results of our experiments with other 
disinfectant agents. These will be published in a short time else- 
where. I shall simply state here that we have thus far obtained the 
best results in disinfection of the skin by the following method : 
1. The nails are kept short and clean. 
2. The hands are washed thoroughly for several minutes with soap 
and water, the water being as warm as can be comfortably borne, and 
being frequently changed. A brush, sterilized by steam, is used. The 
excess of soap is wrashed off with water. CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
25 
3. The hands are immersed for one to two minutes in a warm satu- 
rated solution of permanganate of potash and are rubbed over thor- 
oughly with a sterilized swab. 
4. They are then placed in a warm saturated solution of oxalic acid, 
where they remain until complete decolorization of the permanganate 
occurs. 
5. They are then washed off with sterilized salt solution or water. 
6. They may then be immersed for two minutes in sublimate solu- 
tion, 1:500. 
The bacteriological examination of skin thus treated yields almost 
uniformly negative results, the material for the cultures being taken 
from underneath and around the nails. This is the procedure now 
employed in the gynecological and surgical wards of the hospital. 
The principal conclusions of this paper may be summarized as 
follows : 
The number of different species of bacteria, particularly of bacilli, 
revealed by the systematic study of traumatic infections is much 
greater than was formerly supposed. The pyogenic staphylococci and 
streptococci, however, are by far the most common causes of suppura- 
tive affections of wounds. 
A coccus, which may appropriately be called the staphylococcus 
epidermidis albus, is a nearly, if not quite, constant inhabitant of the 
epidermis, lying both superficially and also deeper than can be reached 
by present methods of disinfection of the skin. This coccus is found 
frequently in aseptic wounds. It may be the cause of disturbances, 
usually of a relatively slight degree, in the healing of the wound, 
especially when drainage-tubes are inserted. It is the most common 
cause of stitch abscesses in wounds treated antiseptically or aseptically. 
The bacillus coli communis is a frequent invader of various organs 
of the body in cases with ulcerative or other lesions of the intestinal 
mucous membrane. In such cases its presence is usually unattended 
by evidence of pathogenic action, but this bacillus may be associated 
with inflammatory affections of wounds, with peritonitis, and with 
abscesses. 
There are many reasons for believing that the process of suppura- 
tion serves a useful purpose in combating bacteria and preventing 
their invasion of the circulating fluids and the tissues of the body. 
The pyogenic bacteria set up suppuration by means of chemical 
substances produced by them and entering into their composition. 
The studies of chemotaxis have shed much light upon the mode of 
action of these substances. 26 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
The effects produced in the animal body by the pyogenic cocci are 
determined by many factors relating to the infectious agents and to 
the individual exposed to infection. There are differences in these 
effects depending upon the species of animal, upon the tissues and 
part of the body infected, upon the readiness of absorption from the 
infected part; upon the source, the number, and the virulence of the 
organisms ; upon the nature and amount of the toxic substances 
accompaying and produced by the bacteria, upon general predisposing 
conditions of the body, and upon local conditions in a wound such as 
the presence of foreign bodies, of pathological products, of dead 
spaces, of bruised, necrotic, and strangulated tissues. 
Infectious agents, as they occur under natural conditions, may pos- 
ses greater virulence than the same bacteria in artificial cultures, and 
this probably depends upon accompanying toxic substances. 
Results of experiments on animals explain clinical experience con- 
cerning the aseptic healing of wounds by the so-called organization 
of a blood-clot. 
The tissues of a wound should be handled so as to interfere as little 
as possible with their vital capacity to overcome bacteria. 
Although the greatest danger of infection of a wound from without 
is by direct contact, nevertheless the possibility of infection from the 
air should not be disregarded. 
Auto-infection may take place by the entrance into the circulation 
and tissues of pyogenic bacteria from the alimentary and the genital 
canals, but there is no evidence that this can occur when these tracts 
are in a healthy condition. Moreover, with the requisite lesions of 
these tracts other general and local conditions of the body are im- 
portant, if not essential, factors in bringing about pyogenic or septic 
infection. 
The presence in the circulating blood and tissues of certain chemical 
products of pyogenic and of putrefactive bacteria, as well as that of 
various other injurious substances, favors the growth in wounds of 
septic and pyogenic bacteria, both of those which may be carried to 
the part by the circulating fluids and those which may enter from out- 
side of the body. 
Whenever we have been able to demonstrate the presence in wounds 
in human beings of the staphylococcus pyogenes aureus or of the 
streptococcus pyogenes the wound either was suppurating or subse- 
quently it suppurated. 
Only in the minority of cases were the aseptic wounds rvhich we ex- 
amined free from bacteria. By far the most common organism in 
these wounds pursuing an aseptic course is the staphylococcus epider- CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
27 
midis albus, which without the presence of a drainage-tube or other 
foreign body rarely causes suppuration in the wound. 
The presence of microorganisms in layers of the skin deeper 
than can be reached by existing methods of cutaneous disinfection 
points to the skin, especially to that of the patient, as a source of infec- 
tion to be carefully guarded against. 
The substitution so far as possible of subcutaneous for cutaneous 
sutures lessens the chances of infection from this source, and particu- 
larly those of stitch abscesses. 
Wherever applicable in surgical antisepsis, disinfection by heat 
should be preferred to that by chemical agents. 
Previous experiments to determine the efficacy of disinfection of 
the skin with corrosive sublimate are vitiated to a considerable extent 
by the failure to precipitate the mercury in the form of the sulphide 
before testing by culture methods its germicidal power on the skin. 
The mercury may remain for days and weeks intimately incorporated 
with the epidermis. 
Epidermal bacteria not killed by the sublimate may be brought into 
such relation with it that they will not grow in ordinary culture media 
until the mercury is precipitated as the sulphide, and such bacteria 
may remain for days and weeks in the epidermis. 
The results of Fiirbringer’s method of disinfection of the skin are 
found to be less favorable when they are tested after precipitation of 
the mercury with ammonium sulphide than without this precaution. 
The best results in cutaneous disinfection we obtained by a method 
in which permanganate of potash followed by oxalic acid plays the 
principal disinfectant role.* 
References to authors mentioned in the text, arranged according to order of 
citation. 
Levy: Archiv fiir exp. Path. u. Pharmakologie, 1891, Bd. xxix. p. 135. 
Bossowski: Wiener med. Wochenschr., 1887, Nos. 8 and 9. 
Von Lingelsheim: Zeitschrift flir Hygiene, 1891, Bd. x. p. 331. 
E. FrUnkel: Centralbl. f. Bakteriologie, 1889, Bd. vi. p. 393. 
Wicklein: Virchow’s Archiv, Bd. cxxv. p. 75. 
OUrtner; Correspondenzbl. d. allg. iirztl. Vereins von Thtiringen, 1888, Heft 9. 
Tavel: Correspondenzbl. f. Schweizer Aerzte, 1889, No. 13. 
A. FrUnkel: Wiener klin. Wochenschrift, 1891, Nos. 13-15. 
* Cultures were exhibited showing various species of bacteria mentioned in the 
paper, the white staphylococci and other bacteria isolated from the epidermis, the 
growths from skin sutures, and illustrating the results of different methods of disinfec- 
tion of the skin, and the importance of precipitation of the mercury with ammonium 
sulphide when testing the germicide action of corrosive sublimate on the skin. 28 
CONDITIONS UNDERLYING THE INFECTION OF WOUNDS. 
Welch: The Medical News, May 34,1890, p. 566. 
Ogston: Arch. f. klin. Chirurgie, Bd. xxv. 1880, and Brit. Med. Jour., 1881, p. 369. 
Roseribach: Die Mikrotirganismen der Wundinfectionskrankheiten des Menschen, Wies- 
baden, 1884. 
Posset: Untersuch. liber d. Aetiologie d. eiterigen Phlegmone des Menschen, Berlin, 
1885. 
Buchner: Berliner klin. Wochenschrift, 1890, Bd. xxvii. pp. 673 and 1084. 
Grawitz: Charite-Annalen, 1886, Jahrg. xi. Virchow’s Archiv. 1889, Bd. cxvi. 
Grawitz and De Bary: Virchow’s Archiv, 1887, Bd. cviii. 
Pawlowsky: Virchow’s Archiv, 1889, Bd. cxvii. 
Garre: Fortschr. d. Med., 1885, No. 6. 
Bunrni: Sitzungsber. d. phys,-med. Gesellsch. zu Wiirzburg, 1885, No. 1. 
Bockhart: Monatshefte f. prakt. Dermat., 1887, No. 10. 
Fehleisen: Aetiologie d. Erysipels, Berlin, 1883. 
Waterhouse: Virchow’s Archiv, 1890, Bd. cxix. p. 343. 
Hankin: British Med. Journ., July, 1890. Centralbl. f. Bakteriologie, 1891, Bd, ix. p. 
336. 
Brieger and Frdnkel: Berliner klin. Wochenschr., 1890, p. 370. 
Buchner: Loc. cit. 
Dirckmck-Holmfeld: Annales de l’lnstitut Pasteur, 1888, p. 469. 
Grawitz: Virchow’s Archiv, 1887, Bd. cx. 
Leber: Fortschr. d. Med., 1888, p. 460. 
Dubler: Ein Beitrag zur Lehre von der Eiterung, Basel, 1890. 
Kronacher: Die Aetiologie und das Wesen der acuten eitrigen Entziindung, 1891. 
Ribbert: Die path. Anat. u. die Heilung der durch den Staph, pyog. aur. hervorge* 
rufenen Erkrankungen, Bonn, 1891. 
Bumm : Mtinchener med. Wochenschrift, 1889, No. 43, and Arch, ftir GynSkologie, 1889, 
Bd. xxxiv. p. 335. 
Fehleisen: Arch. f. klin. Chir., 1886, Bd. xxxvi. p. 966. 
Chantemesse: Progr6s med., 1890, Nos. 13 and 19. 
Roger: Le Bulletin med., 1890, p. 966. 
G. and F. Klemperer: Berliner klin. Wochenschr., 1891, Nos. 34 and 35. 
Gttrtner: Ziegler’s Beitriige zur path. Ant., etc., 1891, Bd. ix. p. 376. 
Ribbert: Op. cit. 
Halsted: “The Treatment of Wounds, with Especial Reference to the Value of the 
Blood-clot in the Management of Dead Spaces.” The Johns Hopkins Hospital Reports, 
vol. ii, No. 5,1891. 
Von Bergmann: Klinisches Jahrb., Bd. i. 1889. 
Schccle: Arch. f. klin. Chir., 1886. Bd. xxxiv. p. 345. 
Grawitz: Virchow’s Archiv, 1889, Bd. cxvi. 
Von Eiselsberg: Arch, f. klin. Chir., 1887, Bd. xxxv. 
Prudden: The American Journal of the Medical Sciences, May, 1889, p. 453. 
Von Volkmann: Quoted from an article by Fritsch in Deutsche med. Wochenschrift, 
1890, No. 19, p. 397. 
Rinne: Uber den Eiterungsprocess und seine Metastasen, Berlin, 1889, p. 61. 
De Wildt: Abstract in Centralbl. ftir path. Anat., 1890, Bd. i. p. 9. 
Waterhouse: Loc. cit. 
Ranke, Demarquay, Fischer, ScliilUer, quoted from references in Watson Cheyne, Anti- 
septic Surgery, London, 1883. 
Watson Cheyne: Op. cit., and Brit. Med. Journ., Sept. 30,1884. 
KUmmell: Verhandl. d. Deutschen Gesellschaft ftir Chirurgie, XIV. Congress, 1885. 
Bossowshi: Loc. cit. 
Halsted: The Johns Hopkins Hospital Bulletins, 1889, vol. i. p. 13. 
Geppert: Berliner klin. Wochenschrift, 1889, Nos. 36 and 37. 
Filrbringer: Untersuchungen und Vorschriften tiber die Desinfection der HKnde des 
Arztes, Wiesbaden, 1888.