Skip to main content

Full text of "Inflammation, an introduction to the study of pathology, being the reprint (revised and enlarged) of an article in Professor Allbutt's "System of Medicine.""

See other formats






















THE development of a science resembles strikingly that 
of the individual intelligence. Your child, once he 
begins to perceive, passes into the stage of perpetual 
" why ? " He is not satisfied with the recognition of 
phenomena, but would know from his elders how they 
come about. It is deserving of note that at first those 
explanations satisfy him that appeal' to the imagination 
rather than to the reason. He is not prepared to appre- 
ciate and to accept mathematical proof ; he lives in 
fairyland. Gradually his powers of reasoning become 
developed, but this under the guidance of his elders and 
teachers, and for long, while still seeking explanations, 
he accepts those given to him. There are those who 
never pass beyond this stage of deference to authority. 
But some, at least, as manhood approaches, begin to think 
for themselves ; begin to see that the explanations given 
by their elders are not all of them adequate, nor all satis- 
factory ; and so is reached the stage of youthful Sturm 
und Drang, during which period the brighter spirits, 
heedless of authority, in the exuberance of youthful 
imagination, proceed impatiently to weave and elaborate 
explanations of their own. It may be that, despite 
their little experience and the fewness of the facts on 



which they base themselves, they light upon truths of 
wide scope. But too often it happens at this period that 
inadequate hypotheses are light-heartedly imagined and 
given to the world. During this period hypothesis after 
hypothesis may be enunciated only to be found wanting. 
As manhood is entered into, the individual learns that 
it is not essential to discover an immediate explanation 
for everything, and now begins patiently to accumulate 
data in the hope that wider experience and fuller com- 
mand of facts will eventually afford material for the 
right solution of a given problem. Thus at last, with 
the necessary experience, maturity and ripe powers of 
reasoning are attained. 

Pathology has passed through all these stages : 
through its infancy, in which medicine men and priests 
and augurs explained disease by fables ; through a long 
childhood of deference to Galenistic authority ; through 
a stormy youth of wild theorising and advocacy of 
system after system iatro-chemical, iatro-mechanical, 
Brunonian, and many another ; through a period of 
reaction to the same, led by Virchow (extending roughly 
during the fifty years between 1845 and 1895), and 
now it may be said to have reached the age of maturity. 
There are those conscientious workers who continue to 
maintain that it is immature ; who hold that we have 
not passed the stage of working hypotheses as distinct 
from established theories ; who lay down that it is 
our duty still sedulously to accumulate facts without 
venturing to draw broad conclusions from the same. 
This, we think, is a mistaken opinion. There are, it is 
true, many territories which remain imperfectly explored ; 


but there are others which have been worked over so 
abundantly and have yielded so vast a harvest of facts, 
that unless we proceed to marshal these facts into order, 
and to classify them in due relationship one to the other, 
the danger is imminent that the accumulation becomes 
a very chaos ; that workers in one part of the territory, 
ignorant of the experience and material gained by those 
in another, will not merely waste their energy but work 
at cross purposes. In short, pathology will become a 
scientific Babel. The danger, indeed, is already with us ; 
we find sundry surgeons enunciating a pathology that is 
at variance with that of the physicians. Already in 
some countries, through a divorce between general patho- 
logy and bacteriology, there is a lack of eagerness on 
the part of the workers in the former to accept the 
data afforded by the latter ; and so huge has become our 
subject that no one man can have an equal familiarity 
with the recent advances in neuropathology, haematology, 
teratology, and ante-natal pathology, immunity and the 
study of infection. 

This being so, it is urgent that some of those in- 
terested in medicine should give their time and energy 
not so much to the development of any one particular 
field, as to collecting and arranging the main data bearing 
upon the causes and development of morbid conditions, 
and the establishment of the deductions which may 
reasonably be gained therefrom. If with each passing 
year it is becoming less and less possible for any one 
individual to have a familiarity with every branch of 
pathology, the time has come to provide a common basis 
of fact and theory which is acceptable to and accepted 


by the workers in various branches, thereby ensuring an 
orderly development of the science of medicine in its 
many departments, a development which, starting from a 
common base, shall be harmonious. 

Such common basis is more particularly afforded by 
the data gained through researches into the process of 
inflammation. It cannot be too strongly emphasised 
that a knowledge of the inflammatory process is the 
foundation of all pathology. We see that ultimately all 
disease is due to the disturbance of the relationship and 
equilibrium between the organism and its surroundings ; 
from the point of view of the organism every such dis- 
turbance causes injury, provided that the local resources 
are not sufficient to counterbalance it forthwith. If we 
accept the definition laid down at the beginning of this 
article, viz. that (briefly) inflammation is the local 
reaction to injury, it follows that we must first gain a 
knowledge of what constitutes the inflammatory process 
if we are to understand disease of all orders. Even if 
certain diseases are not local conditions but general, 
affecting to a greater or less extent the whole organism, 
nevertheless such general conditions are but the summa- 
tion of a series of local disturbances, and must have 
originated from one or more local disturbance ; to master 
the general we must first understand the particular and 
the local. 

Thus, if our pathology is to be not a mere catalogue 
raisonntf of names of morbid states, with precise de- 
scriptions of what those names indicate, but is to be a 
science or, in other words, an endeavour to bring into 
order and relationship the phenomena of disease, and, 


recognising the relationship between phenomena, to 
deduce the laws which underlie and determine the 
individual cases of diseases then ; obviously, the study 
of the inflammatory process is the natural starting-point 
for a right understanding of that science and what it 
can teach us. 

These considerations governed my treatment of the 
subject when, ten years ago, I was invited by Professor 
Allbutt to write the article on Inflammation for his new 
System of Medicine. I strove then to bring together all 
the data known to me bearing upon the subject of the 
reaction to local injury, and, cutting myself free from 
all the schools and' established doctrines, I endeavoured 
conscientiously to select those facts which could not be 
gainsaid, and to draw from them the deductions which 
seemed most rational. There was here no attempt to 
bring forward anything that was new or that had not 
been already recognised by individual workers. At most 
it might be said that many of the facts brought forward, 
as also the method of approaching the subject, were new 
to the text-books. It has been a source of profound 
gratification that this article has from so many sources 
been accepted as authoritative, and that, when nine years 
later I came to revise it in the light of the abundant 
observations which had been published in the meantime, 
while there was much that might be added in amplifica- 
tion, there was little to correct. The teaching which in 
1896 was to a certain extent novel is now in 1906 
-widely accepted. The revised article is here printed in 
book form in the hope that it may prove serviceable 
to the practitioner wishing to keep abreast of recent 


developments, and to the medical student who cannot 
be expected to purchase the large volumes of Professor 
Allbutt's System. And printing it thus I have to 
express my appreciation of the favour extended to me 
by Messrs. Macmillau and Co. in departing from custom 
and publishing as a separate entity what is portion of 
one of their larger works. 

In republishing it thus I have taken the opportunity 
to render it more complete and seasonable. The present 
great interest shown in Sir A. E. Wright's brilliant 
researches upon opsonins demands that his observations 
obtain fuller description (though not fuller appreciation) 
than was accorded by me a year ago. Professor Bier's 
treatment of inflammation by induced hyperaemia has 
likewise aroused widespread attention. The essentials 
of that treatment are here described, and I have been 
led to add what it is trusted will be a useful chapter on 
the application of the principles laid down in this work 
to the treatment of inflammatory states. It has seemed 
useful also to reintroduce in a modified form the para- 
graphs upon the hsemal leucocytes which, present in the 
first, were omitted in the second edition of the article in 
Professor Allbutt's System, being there more properly 
treated in a new article upon the blood; and doing this, 
to call attention to the advances gained by Schridde 
through the employment of methods which permit the 
differentiation of the various forms of leucocytes within 
the tissues. 

, MONTREAL, December 1906. 










NON- VASCULAR AREAS . . . ... .23 


























LEUCOCYTES . . . . . . ~ . .85 


PHAGOCYTOSIS . . . . . . . . .89 






SUMMARY . . . . . . . .121 


















DEGENERATIVE CHANGES . . . . . . .174 







INFLAMMATION . ..... 190 












STATE 227 





KEFERENCES . . .... 241 

INDEX 251 





IT is usual to begin the description of a morbid process 
by defining that process. In the case of inflammation, 
however, we have to deal with a process so complex, so 
modified by modifications of the many factors involved, 
and so variable in its manifestations according to the 
variety of its causes and the region of incidence, that the 
attempt to define it has proved a pitfall to pathologist 
after pathologist ; moreover, to advance a definition of 
the process at the beginning of this article in terms 
differing to any considerable extent from those employed 
by previous writers, would demand a criticism of the 
many previous attempts ; and in order that the definition 
put forward be duly supported, would necessitate an 
essay covering the whole field about to be traversed. 

Use of the Name. Yet, in the meantime, inasmuch 
as divergent views are held of the limitations of the use 
of the name inflammation, a few words of introduction are 

Two courses are before us : either to employ the name 
strictly in accordance with the primitive definition, and 
thus only to include as cases of inflammation those states 
in which there are present redness, swelling, heat, and pain, 
rigidly excluding all cases in which these cardinal signs 
and symptoms are not present ; or, on the other hand, 



PT. I 

departing from tradition, to include as inflammations all 
those morbid processes which seem to have a cause and 
, progress inseparable from and merging into the cause and 
progress of the state characterised by the classical 
symptoms. The first course is impossible ; it is as 
though one were to declare that red phosphorus is not 
phosphorus because in externals generally it does not 
agree with the definition of the yellow form made years 
before the allo tropic modification was discovered. We 
are now well agreed that of the classical signs, one, 
two, or three may be unrecognisable, and in fact absent ; 
and yet the condition of inflammation be undoubtedly 
present. 1 

The second is the only possible course, that, namely, 
which associates all those states which under suitable 
conditions may result in the production of the four classical 
states, and, moreover, originate from a common cause. 
Holding this view, it will in the meantime be well for 
me, in order to afford a starting-point for the description 
and discussion of the subject, to select from the many 
definitions one which is based not on symptomatology, 
but upon etiology, and indicates a common origin for all 
cases of inflammation. I would first note that which in 
this country has received the most cordial support, the 
definition given by the late Sir J. Burdon- Sanderson (1) 
in his well-known article in Holmes' System of Surgery : 
" The process of inflammation is the succession of changes 
which occurs in a living tissue when it is injured, provided 

1 A course allied to this lias found favour of late years among some 
surgical pathologists, who, with Hiiter (2), would limit the use of the 
term to those cases and those only in which the classical signs, or 
the majority thereof, are present and associated with suppuration, they 
urge with Zahn (3) that inflammation only occurs when pyogenic micro- 
frganisms are present, and state that when a wound heals aseptically it 
heals without inflammation. This modified course is equally impossible ; 
pyogenesis must not be confounded with inflammation. 


that the injury is not of such a degree as at once to 
destroy, its structure and vitality." This definition 
includes__too much. The haemorrhage that occurs in the 
liver when it is injured, and the changes that there occur 
in the extra vasated red corpuscles, are scarcely to be 
classed among inflammatory phenomena ; the atrophic 
changes which occur in the retina, when through injury 
it becomes detached, are due mainly to malposition and 
disuse rather than to the primary trauma. Grawitz 
(4) has put forward a definition based upon the same 
fundamental conception, but so worded as to exclude 
largely, if not wholly, the above objections ; for, as regards 
the second, it may justly be urged that atrophic changes 
are not reactive, but due to lack of reaction. Inflamma- 
tion, he holds, is the reaction of irritated and damaged 
tissues which still retain vitality. Either of these 
definitions has this great advantage, that, stating the 
cause, it clearly recognises inflammation as a process and 
not as a state. The manifestations of this process under 
favourable conditions where the region injured is a 
loose and vascular tissue, and where the injury is 
sufficiently severe or extensive are redness, swelling, 
heat, and pain : redness from the congestion of the 
vessels; swelling from the exudation of fluid and corpuscles 
from the congested vessels ; heat from the increased 
amount of blood in the region, and pain from the 
irritation of the terminations of the nerves in the region. 
To these four signs may be added a fifth, disturbance 
of function, brought* about by this departure from the 
normal condition of the region. Under unfavourable 
conditions where the region injured is dense or less 
vascular, or where the injury is less severe one or all of 
these signs may seem wanting; nevertheless a minute 
examination of the tissues will show the same succession 
of changes as in the former case. 



ACCEPTING, then, this working definition, in order to 
arrive at a due comprehension of the succession of changes 
which we take to constitute the inflammatory process, it 
will be well with Metchnikoff 1 (5) to institute a series 
of observations upon the reaction to injury exhibited 
throughout the animal kingdom from the lowest forms 
upwards to man. By this means we shall be enabled 
to determine what factors in the reactive process are 
from their constancy of primary importance ; what are 
common and essential, and what are superadded in the 
higher animals. 

TheResponseto Injury among the Protozoa. Begin- 
ning our study with the lowest and simplest forms of life 
forms so lowly that they have been regarded both as 
animals and as plants we find even here phenomena 
accompanying the reaction to injury which throw light 
upon the inflammatory process as seen in the higher animals. 
Taking as an example the amoeba, we find, in the first 

1 The succeeding paragraphs are of necessity very largely an epitome 
of sundry portions of Metchnikoff's most pregnant work upon the 
comparative pathology of inflammation. By comparing them with the 
work in question, it will, however, be seen that they depart from it 
in several points ; more especially in dwelling upon the extracellular 
excretory activities of the wandering cells, and in bringing more 
prominently forward the response to injury on the part of the fixed cells. 



place, that the nucleus plays an important part in this 
reaction. If, as Metchnikoff has shown, one of the larger 
amoebae be cut in two, the region of injury becomes 
rapidly indistinguishable the protoplasm of each moiety 
closes up, leaving no mark or scar : but of the two parts 
that which retains the nucleus grows and proliferates ; 
the other disintegrates in a longer or shorter time. As 
I have shown elsewhere (6), from every consideration 
we are forced to regard the nucleus as the dominant 
constituent of the cell, controlling its higher activities/ 
the nucleus is essential for continued growth and fonX 
reproduction. Or injury may induce changes in the 
protoplasm of the entire amoeba : thus Miss Greenwood 
(7) points out that, without necessarily bringing about 
death, the interrupted current or an aqueous solution of 
thymol leads to a process of exudation or extrusion of- 
clear hyaline spheres, or of spheres holding crystals 
and granules, from the surface of the organism a 
process resembling that occasionally seen in the cells of 
an inflammatory area in higher animals. Nor is this 
all ; apart from changes in the structure of these 
unicellular animals, differences may be seen in the 
behaviour of amoabae towards foreign bodies. According 
to Le Dantec (8), amgehre ingest noji = ioilajjiig_joreign u>^ 
substances indifferently, provided they be sufficiently small. 
It is, however, doubtful whether absolute inert particles 
are ingested. More recent studies suggest that there 
must be adherent to, or contained in the particles some- 
thing that affects the surface of the amoeba, or otherwise 
that a change in the physical -condition of the surface 
layer of the cell precedes the -formation of a pseudo- 
podium. Around each particle so ingested a vacuole is 
formed, and the fluid in this becomes increasingly acid, 
and at the same time digestive. Krukenberg (9), Reinke 
(10), and Miss Greenwood have conclusively proved 


these and similar food vacuoles in the amoeba and other 
Protozoa to contain a pepsin or digestive ferment, which, 
as Le Dantec has shown by very delicate tests, exerts its 
action in an acid medium (the general protoplasm of the 
cell-body being alkaline) ; this digestive process leads to 
the solution of food-stuffs, preparing them to be taken up 
by the protoplasm of the organism. Such portions of the 
foreign substances as are incapable of digestion become 
sooner or later extruded. It, is by this formation of 
digestive vacuoles that the amoeba acts upon and destroys 
bacteria, diatoms, and other microbes ingested by it. There 
are, however, microbic forms around which it would seem 
that no proper vacuolation is developed, or if developed, 
the acid digestive fluid is neutralised by substances dis- 
charged from the parasites ; where this is the case, instead 
of destruction there is continuance of vitality and actual 
multiplication of the invading or parasitic form, leading 
to the eventual death of the amoeba. Metchuikoff has 
observed this chain of events in one of the amoebae 
which ingests and becomes the host of a minute rounded 
form, the Microsphrera. Phenomena of like nature may 
be observed among the ciliate and flagellate infusoria. 
Here it is worthy of note that bacteria, which, as we shall 
see, are the main causes of inflammation in higher animals, 
are in these lowly forms an important, if not essential, 
source of nutrition. So far it has not been found 
possible to gain pure cultures of the amcebge : in other 
words they are unable to obtain adequate nourishment 
from the various media of the laboratory. It has been 
found, however, by Frosch, Mouton, and others (11), that 
if certain of the commoner forms be isolated, placed in a 
suspension of a pure culture of the bacillus coli, or certain 
other species of bacteria, and then be " sown " upon the 
surface of a tube of sterilised agar of broth, active growth 
and multiplication ensue, the bacilli being taken up and 


used as food. While these phenomena may primarily be 
regarded as the method employed by the Protozoa for 
the assimilation of food-stuffs, they also are clearly the 
means whereby the Protozoa defend themselves against 
living organisms which have gained entrance into them, 
and thus form the reaction to possible injury ; for when 
in certain cases the means of defence are overcome, 
the parasitic organisms gain the upper hand and lead 
to death. 

Chemiotaxis. There is yet another reaction to 
injurious influences exhibited by the Protozoa into which 
it is necessary to enter at some length. This is exhibited 
by the amoeba, but can be and has been most fully 
investigated in the myxomycetes inulticellular forms 
which can with equal propriety be classed as animals or 
plants, although usually they are included among the 
latter. These organisms form large plasmodia (masses of 
protoplasm, that is), in which, under ordinary conditions, 
the nuclei are the only indication of the individual cells 
which by their fusion have formed the masses. They 
are to be met with in leaf mould, and on the surface of 
moist decaying wood, over which they creep with an 
amoeboid movement ; and inasmuch as they may attain 
great size some species attaining twelve inches or more 
in length they form admirable material for biological 

Twenty-five years ago StahlQ 2), investigating one of 
these myxomycetes (the ^thaRum septicum, an organism 
found in tan-pits), showed that if placed upon a 
moistened surface close to a drop of mtoionj^fjaaJi.-^^; 
the plasmodium moved actively towards and into the 
infusion ; if placed similarly near to a solution of glucose 
(0*5 per cent) it moved with equal rapidity away, and so 
also in the case of solutions of various salts. These 
observations of Stahl were (if we except Engelmann's 


observations in 1881 upon the tendency of sundry 
bacj^iia to remove from regions poor in oxygen to those 
where oxygen is present in abundance) the first of a 
series of observations upon the attraction and repulsion 
of plants and portions of plants by chemical substances. 
To this property Pfeffer (13), who has made the fullest 
series of studies upon it, has given the name of chemiotaxis, 
in place of Stahl's narrower " trophotropism " ; and one 
speaks of a positive or a negative chemiotaxis according 
to the attraction or repulsion exerted. If, as Metchnikoff 
has pointed out, the advancing edge of one of these 
plasmodia (of Physarurn) be injured by cauterisation, the 
region of injury dies ; the protoplasmic currents, which 
had been advancing, reverse themselves abruptly, and 
within an hour the plasmodium has moved away, leaving 
the debris of the destroyed region behind. These experi- 
ments are so simple, and the results obtained seem so 
natural, that it may be asked whether it be worth while 
to attach a name to this property of living matter. Yet 
the name is in itself an aid to bearing these properties in 
mind ; and, as. will be pointed out later, the recognition 
of them is of material help in solving certain of the 
difficulties that present themselves in the study of 
inflammation in the higher animals. 

It may be well to note here sundry investigations 
which throw light upon the essential nature of chemio- 
taxis and, indeed, upon the motility of living matter. 1 In 
the first place it has to be recognised that this property 
of moving towards or from other substances is not a 
specific attribute of living protoplasm ; it exists as 
between substances that are not living: in fact, using 
inert matter, chemiotactic phenomena and even the 

1 A fuller account of many of the data bearing upon the physical basis of 
chemiotaxis is to be found in Wells' Chemical Pathology, Philadelphia, 
W. B. Saunders Co., 1907, page 208. 


throwing out of pseudopodia and the engulfing of 
particles can be reproduced with striking fidelity. If, as 
Bernstein (14) has shown, a drop of mercury be placed in 
a dilute nitric acid solution and now a crystal of potassium 
bichromate be" let fall in its neighbourhood, so soon as, 
with solution of the salt, the yellow area of diffusion 
impinges upon the quicksilver, this exhibits active 
movement towards the crystal and flows around it as 
though to ingest it. If the crystal be moved away, 
the mercury follows. Or again, as Rhumbler (15) has 
demonstrated, a drop of clove oil let fall into a freshly 
made mixture of glycerin and alcohol exhibits active 
amoeboid movement. It may be recalled that alcohol and 
glycerin do not easily mix, and that thus there are in the 
mixture regions in which now the alcohol and now the 
glycerin is the more concentrated. The clove oil can 
mix with the alcohol but not with the glycerin. If close 
to the clove oil there be dropped a little strong alcohol, 
the oil globule moves actively towards it with pseudo- 
podial motion. Again, a drop of chloroform in water 
acts in a most life-like manner, flowing up to some 
substances presented to it and flowing away from others. 
Let, for example, a fine thread of shellac be made to 
touch it, and the edge of the chloroform droplet extends 
along it, just as the cytoplasm of a leucocyte flows along 
a chain of anthrax bacilli, and with this the shellac 
softens and becomes coiled up within the drop until in 
this way a thread of shellac six times the length of the 
diameter of the drop becomes included within it. 

The explanation of these phenomena is, briefly : 
modification of surface tension. Physicists recognise a 
force cohesion force acting between the particles of a 
fluid and attracting the one to the other. This force 
differs in intensity for different fluids. Thus in the 
centre of a drop of one fluid suspended within another 


the forces acting upon the particles are equal in all 
directions. At the surface the cohesion force acting 
on the inner aspect of the particles may be widely 
different from that acting upon them from the fluid of 
suspension without ; they are, in fact, subject to unequal 
pressure. It is this force and this difference in force 
that constitute the surface tension, and that further leads 
the suspended fluid to assume the shape of a sphere, v 
Now let the cohesion force of the outer fluid be modi- 
fied locally, so that it approximates to that of the 
suspended drop : it follows that in the region of this 
modification the surface tension of the drop will become 
less than that on the opposite side of the drop. 
The result must be what happens when a thin-walled 
distended rubber ball is weakened on one side : namely, 
the droplet gives towards that side, and its contents 
stream towards the region of concentration of the intro-. 
duced salt or fluid. When the cohesion force of the 
suspending fluid is locally reduced, then on the contrary 
the droplet flows away. Thus it would seem that 
pseudopodial activity is at basis dependent upon the 
surface tension of 'the ectosarc of the amoeba or 

As to what in living matter pre-eminently induces 
alterations in surface tension, the researches of Hardy, 
Loeb, and Brailsford Eobertson would seem to afford a 

The extreme sensibility of protoplasm to changes in 
its environment appears to depend upon the amphoteric 
nature of the constituent molecules of living matter, i.e. 
upon their capacity to combine with both acids and bases, 
to form what toeb (16) has termed ion proteins. As 
Hardy (17) expresses it, " their electrical characters are 
conferred upon them by the nature of the reaction, acid 
or alkaline, of the fluid. If the latter is alkaline the 


particles become electro-negative and vice versa." As 
Brailsford Robertson points out (IS), when the diffusion 
of ions in the surrounding medium is uneven, as these 
act upon the protoplasm causing in that a dissociation of 
the ions attached to the protein molecules, the different 
rates at which positive and negative ions travel will 
bring about that at one side of the cell there will be a 
greater or less accumulation of one or other type of ion, 
with, as a result, increased or diminished surface tension. 
In this curiously unstable eo^uilibrium^of the constituent 
molecules of living matter, and in the ease with which they 
dissociate and enter into combination with surrounding 
ions, would seem, therefore, to lie the explanation of 
alteration of surface tension, motility and chemiotaxis. 

In the myxomycetes another point can be made out. 
Stahl observed that the plasmodium of Fidiyo, which at 
first moves away from a two per cent solution of common 
salt, will after a time (more especially if it has suffered 
from lack of water) adapt itself to the solution, advancing 
its pseudopodia or protoplasmic processes into it. With 
other myxomycetes the same adaptation has been observed. 
That is to say, by use or adaptation a negative may be 
transformed into a positive chemiotaxis. To this change 
I shall have occasion to revert. 

A similar, and suggestive, series of adaptations has 
been noted by Musgrave and Clegg (19) in their studies 
upon amoebae. They note that these in their natural 
environment the pathogenic forms in the large 
intestine, for example are selective, feeding upon only 
one of the many surrounding species of bacteria. Away 
from that natural environment they grow abundantly 
upon the surface of agar tubes in association with this 
one form. By the gradual addition of pure cultures of 
another bacillus, with which, at first, growth is not 
possible (the amcebas not taking them up, and starving as 


a consequence), the amoebae first become accustomed to 
this foreign form, then take up occasional individuals, 
until eventually they will feed upon and grow in a pure 
culture of the second form, in the total absence of 
members of the first species. Nay, more, these observers 
noted that one particular amoeba, isolated from Manila 
tap -water, and grown in conjunction with a species of 
bacteria present in that water, set up abscess-formation 
when inoculated with it into the liver of animals of the 
laboratory : from such an abscess growths of the amoeba 
could be obtained in association with the microbes in 
question. But if passage were made through two more 
animals of the same species, setting up in them liver 
abscesses, it was found impossible to gain growths in 
like manner. The amoebae were there, but growth in the 
tissues had changed their habits. They had clearly 
accustomed themselves to a wholly different form of food 
presumably to cell-products of the organism ; a point 
which throws light upon the frequent occurrence of 
tropical abscesses of the liver, in which amoebae are 
present without associated bacteria. 

The Response to Injury among the Metazoa. 
Passing from the Protozoa to the Metazoa. we reach 
immediately (or almost immediately) a series of beings in 
which the division of labour amon^r the cells has Jed to 
the development <>{' three cell-layers an outer ectoderm, 
an inner endoderm, and an intermediate layer of mesoderm. 
Even in the very lowest forms among the Metazoa it__is 
noticeable that of_these_three layers Jihere is_ one, the 
mSQ_derm, whose cells have the especial_f uncti on of 
reacting when any irritant or injurious agent gains 
entrance into the organism. Taking what are perhaps the 
simplest forms in which to observe the relationship and 
properties of these layers, Metchnikoff has studied these 
results of injury in the larval forms of Astropecten and 




other echinoderms. At one well-recognisable stage these 
larvae resemble little more than the gastrula stage of the 
embryologist ; the endoderm or hypoblast appears as a 
cul-de-sac an invagination of the ectoderm or epiblast 
while the mesoderm is represented by amoeboid cells, 
budded off from the endoderm, lying or floating in the 


FIG. 1. 1. Larva of Astropecten, modified from Metchnikoff to show end, endoderm ; 
ect, ectoderm; mes, mesodermal wandering cells; pi, plasmodinm of mesodermal 
wandering cells formed around foreign body which has punctured the endoderm. 
2. The Plasmodium, from the same larva, highly magnified ; nucl, nuclei of the 
fused wandering cells ; /, spicules of foreign matter around which the cells have 

semiliquid substance filling the general body -cavity. 1 
The_ectodern^ is so delicate that_anyL sharp aiihafamfip. can 

1 As indicating the earliest stage of these wandering cells it is note- 
worthy, as MacBride (20) has pointed out, that in the gastrula of Echinus 
esculentus two forms of wandering mesenchyme cells are present in the 
body-cavity : one obviously stellate, and attached to other similar cells 
and to the body -wall by long processes; the other, rounded forms 
(amabocytes), which, while appearing free, are connected with neighbour- 
ing cells and the body-wall by means of excessively fine threads, along 
which they appear to travel. The cells at first are therefore only 
relatively free. 


readily penetrate into the body-cavity ; and, when this 
happens, it is noticeable that the wandering mesodermal 
cells make their way to the foreign bodyT attach them- 
selves to it, and fuse into plasmodia][ ^nasaes. thus forming 
a wall, as it were, around the invajiing substance, and 
cutting it off from the_gejieral body jsyjstem. Sere, then, 
in an organism pos^ssinP' np.ij-.her np.rvrms nor vascular 
system, the reaction to injury, when that injury has not 
been sufficiently intense to cause destruction of the outer 
layer of cells, is simply and solely confined to the wander- 
ing cells of the body ; there is no effusion of fluid ; there 
is not necessarily phagocytosis on the part of these cells ; 
any digestive and destructive action on their part any 
attempt in this way to remove the foreign body must 
then be by excretion, ly extracellular action. At the 
same time, this fusion of the cells and formation of a 
plasmodium around foreign substances of greater diameter 
than the individual mesodermal cells may be looked upon 
as a mechanism whereby the equivalent of intracellular 
digestion is gained. But, as among these low forms cases 
occur in which, without the formation of plasmodia, the 
cells perform their destructive action upon bodies of 
larger size than themselves, we do not lack examples of 
what must be considered as excretory destructive powers 
on their part. That these cells in the echinoderms are 
also capable of destroying minute foreign bodies by 
intracellular action, that is, by phagocytosis, has been 
demonstrated in the larger transparent larval form known 
as Bipinnaria asterigera ; on introducing bacteria under 
its ectoderm the mesodermal cells are seen to approach, 
and by their long pseudopodia to adhere to and ingest the 
still living motile bacteria, which are rapidly digested. 

Besides this reaction to injury on the part of the 
mesodermal cells, a further response is exhibited to a 
remarkable degree among the lower Metazoa I refer to 


the great power of regeneration of lost parts, of cell- 
proliferation leading to the reproduction of destroyed 
regions. This power "is best seen in the classical 
example of the Hydra, which may be cut into many 
pieces, each (save the tentacles) capable of growing, so 
that in a relatively short time it becomes a fully formed 
individual. It is interesting to note in relation to the 
frequent tendency towards hyperplasia and excessive 
growth following upon injury in the higher animals, that 
among low forms, such as Hydra and Cerianthus, the 
same tendency is yet more strongly marked. Thus, as 
J. Loeb (21) points out, if an incision be made in the 
stem of a Hydra, a whole new oral pole, provided with 
tentacles, will branch out from the region of cell-destruc- 
tion. In the actinian Cerianthus the process is not 
quite so extensive ; yet from the lower lip of a lateral 
incision a set of tentacles develops in all respects similar 
to those around the mouth. 

Ascending to the Worms, we find that the protective 
agency devolves upon mesodermal cells suspended in the 
perivisceral fluid, and again forming the peritoneal endo- 
thelium. We arrive, that is to say, at a stage in which 
a lymphatic system may be said to be present : for the 
spaces in which the free corpuscles lie are strictly 
homologous to the lymph - containing spaces of the 
vertebrate organism, and these corpuscles may be 
regarded as lymph - corpuscles ; the peritoneal endo- 
thelium corresponds with the mesodermal peritoneal 
endothelium of vertebrata. It is the fixed as well as 
the wandering cells that take part in the process. 

Among*tne Annelids the process of reaction to injury 
may be well followed in the earthworm by studying the 
sequence of changes that occurs around the gregarines 
which infest the male genital organs. While these 
parasites are active they by their movements prevent 



the adhesion of the wandering cells ; but so soon as they 
pass into the resting stage antecedent to spore-formation, 
the cells form a thick mass around them. The parasite 
on its part forms a thick cyst- wall ; nevertheless it may 
not infrequently be observed that, despite this protection, 
the parasite changes its appearance under the action of 
the surrounding plasmodium, and in fact is killed. While 
this is happening^ no change could be detected by 
Metchnikoff in the neighbouring blood-vessels ; these 
appear to remain completely inactive : no exudation is 
noticeable nor any recognisable change in volume. The 
nature of the injury inflicted without doubt influences 
the character of the reactive process. Thus causing 
direct injury, either by passing a thread through the 
body, or by cautery, Messing (22) found that in the 
earthworm and leech regenerative changes are both more 
pronounced and of more rapid development than is the 
accumulation of wandering cells. In six hours the 
injured epithelium might show definite signs of regenera- 
tion, while at this period but few mesodermal cells had 

While among the Worms a well-developed and closed 
vascular system is not infrequently present, in other 
animal forms, which in most respects present a much 
more complex and advanced development, namely, in the 
Molluscs, Arthropods, and Tunicates, this is not the case. 
In these the blood pours from the tubular heart sooner 
or later into the lacunae of the general body-cavity ; and 
whether veins and capillaries (i.e. finer vessels lined with 
endothelium) be absent (as is most usual), or present (as 
in the Cephalopods), the blood is sucked back from the 
body-cavity into the heart. This incomplete circulation, 
interesting as it is in connexion with the development of 
the vertebrate circulation, is interesting also because its 
incompleteness in these large and widespread classes of 


animals prevents reaction to injury from being associated 
with vascular changes. The blood in these animals, 
circulating through the ramifications of the body-cavity, 
is evidently a mesodermal fluid, if it may be so termed. 
Its corpuscles are clearly mesodermal ; and without going 
into full details as to the properties of these corpuscles, it 
may be said that they represent an interesting series of 
stages in the subdivision of labour. For example, as 
Hardy (23) has shown us, in a low form of crustacean 
like Daphnia (the water-flea) but one form of cell is 
present, whereas in the highly developed Astacus (the 
crayfish) there are three distinct forms of leucocytes (no 
red corpuscles being present^ each of which appears to 
have distinct functions. The one form in Daphnia has 
the property of taking up fat globules and food particles 
from the alimentary tract, foreign particles such as 
granules of carmine or Indian ink, and the spores of 
parasites (Monospora, Metchnikoff) ; it is granulated, 
containing minute spherules which stain with basic aniline 
dyes (basophil granules), and in certain circumstances it 
may be seen to explode with lightning-like rapidity. 
In the higher Astacus there are in the circulating haemal 
fluid two varieties of cells : one is extraordinarily 
explosive ; when removed from the body-cavity it gives 
off fine blebs or vesicles of its substance with such 
rapidity that, unless the greatest care be taken, nothing 
is seen of the- cell save its nucleus ; this form is 
phagocytic : the jother form is far more stable, and is 
loaded with large spherules which have a great affinity 
for acid dyes they are eosinophilous may be actively 
extruded, and undergo decomposition ; these cells never 
act as phagocytes. A third form, with basophil 
granules, is rarely found in the blood, and then only as 
the result of special stimuli ; but it is present in 
considerable numbers in the peculiar tissue which forms 


a sheath around certain of the arteries Haeckel's 
" Zellgewebe " ; this form is phagocytic, and can be seen 
to contain globules of ingested fat. 

Metchnikoff (24) demonstrated, in his most remark- 
able study upon a disease of Daphnia caused by the entry 
of the "spores of a yeast-like organism (the Monospora) 
into its body -cavity, that its one form of leucocyte 
can be seen to react swiftly towards the spores ; the 
cells approach them, form a plasmodium around, and 
eventually digest and destroy them. If, on the other 
hand, in consequence of their great numbers or the 
relative paucity of the leucocytes, certain of the spores 
be not attacked and develop uninterruptedly into mature 
torulae, the leucocytes show no tendency to approach 
them in fact, their neighbourhood leads to the explosion 
of the leucocytes and the torulas, multiplying, lead to 
the death of the organism. Often, again, brown eschars 
may be recognised upon the transparent carapace of a 
Daphnia, due to injuries by other individuals ; beneath 
these gears are to be found masses of leucocytes which 
remain in the region of injury until the cells of the 
tissue have proliferated and there is complete union and 

In addition, then, to the immediate reparative and 
protective reaction of the leucocytes, there is exhibited 
among the higher invertebrata a later reaction in the 
shape of proliferation of the fixed cells ; nay, at times 
the proliferative and regenerative process may be the 
more pronounced. This proliferation can be very great ; 
and cells of all forms, whether of hypo-, meso-, or 
epiblastic origin, and tissues so highly developed as the 
muscular and nervous, may participate in it. In illustra- 
tion of the ample power of tissue -reproduction after 
injury possessed by these animals, I need but mention 
the trite examples of the reproduction of the hinder 



FIG. 2. A, A Daphniu (water-flea) invaded by the parasitic yeast, Moiwsjwm. The 
deep shading, more particularly below the intestinal canal (a), is due to the 
mass of spores which have penetrated into the body-cavity. B, Anal end of an 
infected Daphnia. The elongated spores are to be seen in the lumen of the gut 
(ft), in the walls of the same, and others which have penetrated the walls and have 
entered the body-cavity are seen with leucocytes attached (c, cf). C, Stages in 
the development of Monospora : 1, Individual elongate oval torula ; 2, a torula 
budding; 3, do., later stage; 4, much elongated torula prior to" spore-formation ; 
5, torula with contained spore. D, Stages in the destruction of the spores : 6, 
spore with two leucocytes attached ; 7, 8, later stages of erosion and digestion. 


segments of divided worms, and in crustaceans the 
restoration of injured and cast-off claws and appendages. 

Many more instances might be given to show that 
the reaction to injury remains essentially a reaction on 
the part of the wandering and fixed mesoblastic cells of 
the organism, followed in sundry cases by proliferation 
of the fixed epi-, meso-, and hypoblastic cells, and by 
repair where these have been destroyed. Although 
these arthropods, molluscs, and tunicates have a vascular 
system, yet, since this_is_0en, changes in it, did they 
occur, could scarcely modify the inflammatory process. 

The Response to Injury among the Vertebrata. 
If now we pass to the vertebrates, the picture presented 
is far more complex ; not only do these present a highly 
developed nervous system, but the blood is enclosed in 
a complete vascular system. It is but just to call 
attention again to the fact that many authorities deny 
that up to this point we are justified in speaking of 
inflammation, urging that inflammation can only be 
said to be present when there is " flaming," i.e. increased 
redness and heat, due to increased blood-supply, that 
is to say, to vascular changes. While this position is 
etymologically correct we cannot accept it, because, as 
will be pointed out in the next chapter but one, we find 
that where this " flaming " is present, the process, never- 
theless, in essence, is identical with that in the lower 
animals devoid of a complete vascular system. 



LET us begin with the succession of changes that occurs 
in the simplest case, namely, in a non-vascular area, in 
one of the lowest vertebrate forms for instance, in the 
embryonic Axolotl ten to fifteen days old ; let us curarise 
it, and apply a minute crystal of silver nitrate to the 
side of its flattened transparent tail-fin, washing away 
the remains of the crystal with salt solution ; or again, 
we may pass into the tail a small needle filled with 
finely powdered carmine. By either procedure a certain 
number of cells is destroyed. The neighbourhood of the 
injury now becomes swollen, and the surrounding cells 
tumefied, vacuolated, and less refractile. This is the 
first stage that of injury and modification of the 
surrounding tissue. In a little time a few wandering 
cells (leucocytes) approach the injured region ; by the 
next day these are present in fair numbers, and can be 
seen to have taken up the particles of carmine or debris 
of the destroyed tissue. This is the second stage that v 
of immigration of leucocytes. There are no vessels in 
the transparent fin of these young axolotls, no dilatation 
of those nearest to the fin, and no diapedesis. All the 
leucocytes that pass to the part are pre-existing wander- 
ing cells of the connective tissue, a point of some little 



importance in connexion with the origin of certain of the 
pus-cells in the suppurative process of higher animals. 
The third stage is that of repair, of proliferation of the 
injured epithelium, return of the fixed cells of the tissue 
to their previous state, and emigration of the wandering 

A very similar progress of events occurs if the ex- 
periment be repeated upon the tail-fin of the young 
r Newt. The same rapid alteration in the large branched 
\Jr connective-tissue cells (which become vacuolated as their 
long processes are drawn in and shortened), and the 
same immigration of motile cells from the surrounding 
connective tissue are to be seen ; but here we now find 
the earliest evidence of vascular participation, for, accord- 
ing to Metchnikoff, complete arrest of the circulation 
may occur in the nearest vascular loop. By the next 
day the parts have returned to the normal condition. 

If from these cases we pass to mild inflammatory 
disturbances affecting the non- vascular regions of animals 
far higher in the scale, we again discover a like process 
of events. For this purpose the cornea affords the 
tissue of election ; in health it is absolutely non- 
vascular, perfectly transparent, and so thin that it can 
readily be examined microscopically. The cornea of 
mammalia, and indeed of vertebrates in general, is 
formed of fibres which run in layers parallel to the 
surface. These fibres, while roughly arranged side by 
side and parallel to one another in any given layer, are 
placed at an angle to the fibres of the layers above and 
below. Although free from blood-vessels the cornea is 
far from being devoid of channels along which lymph 
freely passes. Between the several layers there exist 
spaces in which lie the flattened connective-tissue cells 
of the organ ; and, by means of numerous fine channels, 
these spaces around the cells are connected with similar 


spaces lying in front, behind, and at the sides. Through 
this rich anastomosis of channels there is a free flow of 
lymph. These channels are really continuations of the 
body-cavity of the animal; they represent, and in fact 
play the same part as the single body-cavity of such a 
simple form as the larva of Astropecten, while the cells 
lying in the spaces are mesoblastic cells which have 
become fixed. 

Few studies are better calculated to impress the 
investigator with a sense of the depth of the well at 
the bottom of which truth lies, than a research into 
the abundant literature dealing with observations upon 
the stages of the inflammatory process as it occurs _in 
the cornea, and with the deductions therefrom. The 
adherents to successive forms of inflammatory belief 
have found in experiments upon this simple tissue 
ample support for their particular creeds. Selecting 
from the many observations those which have stood 
the test of time, I will begin with the simplest, and pass 
on to those dealing with an increasing intensity of the 
inflammatory process. 

(i.) If, as Senftleben (25) first pointed out, the centre 
of the cornea of a rabbit be washed with a strong 
solution of zinc chloride, then, in favourable cases, 
although the epithelial covering be gravely injured, there 
may be no actual rupture of the outer layers of the 
tissue. Such a cornea removed twenty-four hours later 
may show no sign of migration of leucocytes no sign, 
again, of congestion of the vessels at the periphery. 
The only indications of injury and reaction may be the 
destruction of the corneal corpuscles immediately beneath 
the cauterised area, and the appearance of a zone sur- 
rounding this in which the corneal corpuscles appear 
enlarged, distinct, and tumefied. The process may 
continue and advance insensibly to repair without the 


intervention of leucocytes ; the hypertrophying cells of 
the " granular " zone eventually undergoing karyokinesis, 
and thus by multiplication replacing the corpuscles 
destroyed. Here, then, necrosis and new grovith of the 
fixed cells of the tissue are the only recognisable factors in 
the process of repair of injury. It must be confessed that 
the conditions permitting this simplest form of reaction 
are of rare occurrence ; it is worthy of attention that 
they can exist. 

(ii.) By a slight modification of the preceding condi- 
tions another factor may be brought into play. If, after 
cauterisation in the manner above described, a break be 
made into the cauterised surface ; or if, again, without 
cauterisation, a little of the corneal tissue be removed, 
then in a few hours- a small whitish opacity is to be 
noticed within the corneal tissue in the immediate 
neighbourhood of tjie__hrfiak in the continuity, and upon 
examination this opacity is found to be due_to__a_massing 
ofLsniflll round cells. As there is at this moment no 
sign of proliferation of the connective-tissue cells of the 
cornea, these newly collected cells can only be leucocytes ; 
and further examination of their properties proves them 
to be such : there is, however, no evidence of dilatation 
of the peripheral vessels, no indication of diapedesis 
through their walls. The leucocytes, therefore, can only 
have entered into the wound from the cornea itself and 
from the conjunctiva and the lacrimal fluid bathing it. 1 
In this experiment the inflammatory process is repre- 
sented by destruction of tissue and immigration of 
leucocytes, followed by repair ; neither the vascular nor 
the nervous system play any part in it. We are forced 
to the conclusion that the leucocytes have massed them- 

1 It is to be noted that the lacrimal fluid contains even in perfect 
health, occasional leucocytes which have found their way into it through 
the tissues. 


selves in the injured area purely on their own initiative ; 
and that there must be an attraction, a chemiotaxis or 
chemiotropism, leading them actively to approach the 
region of cell-destruction. 

(iii.) Or we may proceed a step farther. A fairly 
severe aseptic injury can be produced by cauterising the 
centre of the cornea. In thus treating the pigeon's 
cornea Goecke (26) noted that the wa.nde,rjng of cells 
towards the damaged area is first visible twelve Jiours 
after the injury^ and Then proceeds from the periphery. 
Obviously the wandering cells are white blood-corpuscles, 
and pass from the peripheral vessels. In twenty-four 
hours the process and the accumulation of round cells 
reach their climax. Some of the new-comers break up, 
others, according to Goecke, show signs of division. 
But soon these foreign cells commence to wander away, 
and at the end of thirty-six hours scarce any are left. 

Turning to the fixed cells of the part, it is deserving 
of note that, before ever a leucocyte has reached the 
injured area, the corneal corpuscles, bordering upon the 
area of cauterisation, show evident signs of enlargement 
and growth. Onjbhe second day there are indications of 
active proliferation ; and these newly formed corneal 
corpuscles behave exactly like certain white blood- 
corpuscles, from which they are indistinguishable. The 
vexed question of the relative part played by the wander- 
ing white corpuscles of the blood and wandering young 
connective-tissue cells will be touched upon later. It is, 
however, well to impress upon the reader that, at a 
certain stage, what we may term histogenous and haemato- 
genous wandering cells are wholly indistinguishable by 
our present methods of study. The fight has been par- 
ticularly bitter regarding these cells in connexion with 
this very subject of experimental keratitis. 

(iv.) The observations made upon these three more 


simple cases help us materially to understand the series 
of events which occur in more intense inflammation of 
the cornea, such as that produced by injuring the surface 
and causing the entrance into the injured region of a 
small quantity of a pure culture of the pyococcus aureus. 
This may be accomplished by injecting the culture into 
the centre of the healthy cornea by means of the needle 
of a Pravaz syringe (Jacobs) (27). The micrococci so 
introduced grow rapidly, the growth so extending along 
the lymph-spaces that a branched mass of the microbes 
is produced, having the spot of inoculation as centre. 
Around the growth as it extends may be seen a sharply 
marked area in which the corneal corpuscles show 
evidences of degeneration ; the nuclei stain faintly, and 
the corpuscles, speaking generally, have a shrunken 
appearance. Here, again, the first effect of a microbic, 
as of a simple chemical injury, is to bring about degenera- 
tion of the fixed cells of the tissue. Within eighteen 
hours the zone of proliferating cocci and cell-degeneration 
is well marked ; and now the second stage begins to be 
clearly manifest, namely, the determination of leucocytes 
to the seat of injury. Within twenty-four hours there 
is a dense packing of these corpuscles around the central 
degenerated area, and great numbers of leucocytes may 
be seen converging along the lymph -spaces from the 
periphery of the cornea. This is the second stage of the 
process, the first stage of obvious reaction to the injury 
inflicted by the invading micro-organisms. If, as by 
Cohnheim 1 (28) in his original experiments upon the 
injury to the cornea, more careful examination be made 
into the stages of the deter mination of leucocytes, it can 
be seen that this determination is closely related to 

1 There can be no question that Cohnheim in his experiments induced 
not a simple keratitis but one which in the absence of aseptic precautions 
rapidly became infective and suppurative. 


changes set up in the veins at the periphery of the 
cornea ; they become more prominent, the region has a 
congested appearance, the smaller as well as the larger 
vessels are dilated, and there is abundant evidence that 
the leucocytes are passing out from the contained blood 
into the surrounding lymph-spaces. Indeed the. accumu- 
lation of leucocytes shows itself first at the periphery of 
the cornea near the vessels, and gradually approaches the 
region of injury. Into the mechanism of this diapedesis, 
and into a fuller description of the changes that take 
place in the blood-current in these distended vessels, I 
shall enter later when discussing the changes in highly 
vascular regions. Suffice it to say here that no distinc- 
tion can be made out between the behaviour of the leuco- 
cytes in the previous experiment, when they entered the 
wounded area from the external surface, and in this 
where the majority find their entrance from the blood ; 
as in the previous case the part played was evidently 
active, so must it be here also. We cannot arrive at any 
other conclusion than that some attractive force leads to 
their determination towards the inflammatory focus. It 
is the polymorphonuclear leucocytes which at first most 
actively migrate. As Councilman (29) points out, in 
experimental pyococcic inflammation, as early as fifteen 
minutes after inoculation of the centre of the cornea a 
greater number than usual is seen in the conjunctiva. A 
more granular, more sluggishly amoeboid form follows, 
most numerous in eighteen to twenty-four hours, while 
lymphocytes are not visible until the fourth day, and 
then do not so much pass out of the vessels as from the 
sheath of lymphoid tissue surrounding them. We can 
easily show, by repeating the experiment, that many of 
these leucocytes take up and contain numerous cocci, 
while other cocci remain free in the tissue-spaces. Many 
of the leucocytes degenerate and present a broken-down 



PT. 1 

appearance ; and, as at the same time an increasing area 
of the corneal tissue becomes disintegrated, an ulcer 
appears. According to the virulence of the culture and 
the reaction on the part of the organism, the process may 
now extend, a larger and larger portion of the corneal 
tissue becoming affected ; or, on the other hand, there may 


FIG. 3. Mild grade of keratitis ; commencing regeneration after forty-eight hours. 1. 
Peripheral zone of corneal corpuscles, showing enlargement with nuclear multi- 
plication. 2. Zone of degenerating granular corneal cells (b). 3. More central area 
of cells (c) destroyed and broken up by the action of the caustic, a, Processes 
from proliferating corneal cells, with nuclei, the products of direct nuclear 
division, advancing into region of irritation and degeneration. After SENFTLEBEN. 

be an arrest of the progress, the massing of the leucocytes 
preventing, as a barrier, the further extension of the 
micrococci into the lymph-spaces ; * while at the same 
time there is an advance of newly formed capillary vessels 
into the previously non- vascular tissue. It is to be noticed 
that the blood-vessels at the periphery of the cornea are 

1 Into the details of this action I shall enter more fully later. 

CIT. Ill 


prominent and dilated, and from them fine new vessels 
with very delicate walls pass towards the injured region. 
At the same time many of the corneal corpuscles, outside 
the area of destruction, can by appropriate staining be 
seen undergoing mitosis and proliferating. Thus the 
active repair of the tissue is initiated. The regenerative 
process is best observed in young animals by the use of 
caustics without causing a break in the continuity of the 
surface. Twelve hours after injury indications of cell- 
growth may be made out ; in twenty-four hours mitoses 
are numerous, but occur only in the zone immediately 
surrounding the necrosed area. The corneal corpuscles 
become larger, their protoplasm more abundant ; the 
cell-bodies stain more deeply ; the nuclei become more 
rounded and also more deeply stained ; the cells give off 
long processes always towards, never away from, the 
necrosed area and with division of the nuclei the young 
nuclei make their way into the processes, and so into 
the dead tissue. In this way new typical corpuscles are 
formed in the course of the branches, which further show 
abundant cell - inclusions ; in other words, the growing 
tissue-cells feed upon the migrated leucocytes. 



this study of inflammation, as it occurs in a region 
primarily devoid of blood-vessels, let us now pass on to 
the more complicated process of inflammation in vascular 
areas ; and, as in the previous case we considered an 
ascending or advancing series of reactive changes, so 
here let us begin with the slightest injury associated 
with the mildest reaction, and pass onward to states 
in which the inflammatory manifestations are more and 
more pronounced. 

(i.) If an incision be made with a perfectly aseptic 
instrument into the skin, also rendered aseptic, and be 
so made as to divide the dermis and tissues immediately 
below, without at the same time injuring any large 
vessel, it is the common experience of modern surgeons 
that repair takes place with the minimal amount of 
change recognisable as inflammatory. Eepair takes place 
indeed so rapidly that, if the divided structures have 
come or have been brought into immediate contact, there 
may be firm adhesion at the end of twenty-four hours. 
This is primary union, or union by first intention, which, 
rare in the old days, commonly occurs in this era of 
aseptic surgery. The full sequence of events in these 
cases cannot, it is true, be well determined by continuous 



microscopic examination ; but if the rabbit or dog be 
employed, and tissues, wounded in the manner described, 
be removed and examined at successive short intervals, 
we see .that the changes which occur are mainly, nay, 
almost entirely, related to the pre-existing cells of the 
part. The section divides a certain number of capillaries ; 
but in the very act of division the divided walls are 
apparently brought together ; and, partly by this means, 
partly by contraction, the lumina of these minute vessels 
become occluded, and the haemorrhage into the wound is 
altogether inconsiderable. Within an hour after the 
operation it is evident to the naked eye of a careful 
observer that the immediate neighbourhood of the wound 
is reddened and tumefied, but only slightly ; and, associ- 
ated with this, there is a feeble exudation between the 
opposed surfaces. But the exudation is not great, and 
even within this first hour after the infliction of the 
wound there may be development of fibrin, and coagula- 
tion of the exudate, leading to a provisional cicatrix 
cementing together the opposed surfaces. In this exuda- 
tion, and in the tissues in the immediate neighbourhood, 
the leucocytes that have undergone diapedesis may be 
few and far between, and may scarcely attract attention. 
The reaction, then, on the part of the vessels and of the 
leucocytes is of the slightest. At times the dilatation of 
the vessels is more considerable, and with this there is a 
fair amount of oozing into the wound of a thin serum, 
which has little tendency to coagulate. Thus it has not 
been an infrequent experience of surgeons that if an 
extensive skin-flap be sewn up completely the amount of 
serum accumulating between the surfaces of the wound 
during the next few hours causes not a little tension and 
discomfort. If one or two of the stitches are cut and 
this serum allowed to drain away the parts unite forth- 
with. It has become the practice to make allowance for 



this serous oozing by not completely closing one end of 
such a wound, and inserting there temporarily a " drain " 
of sterilised gauze, so that this fluid escapes immediately 
and primary union is facilitated. 

Study of sections in these cases shows that the main 
part is played by the pre-existing cells of the part ; of 
these a certain number (not so many as might a priori 
be expected) are destroyed immediately, and show all 
the signs of disintegration ; a number relatively large 
have been injured only, their nuclei remaining intact, 
though their processes or some portions of the cell-bodies 
have been cut through. It is difficult to determine these 
injuries in the small cells of the cutaneous tissues ; they 
are better seen in the peritoneum when slight inflamma- 
tory changes have there been induced. It can, however, 
be made out that the cells in the immediate neighbour- 
hood of the wound become enlarged, and, without show- 
ing signs of division, prolong themselves (that is to say, 
send out prolongations) into the region of the provisional 
fibrinous cicatrix. In this way, before the end of the 
second day, there may be a more or less complete re- 
placement of the primary unorganised cementing sub- 
stance by organised growing tissue, formed, in the first 
place, by the interlacing of processes from the neighbour- 
ing cells; in the second, and later, by a multiplication of 
these cells, together with a development of new capil- 
laries, few in number, which branch off from the slightly 
congested vessels in the neighbourhood. Thus in this 
case the process of repair is characteristically associated 
with hypertrophy and the new growth of the fixed cells 
of the tissue ; while vascular changes, exudation, arid 
leucocytosis are relatively little marked. I have, how- 
ever, never come across a case in which they have been 
entirely absent, save when the section has been truly 
extra-vascular that is to say, when it has not penetrated 


into the vascular region of the skin, and has affected 
only the epidermis and outermost layers of the dermis. 
In such cases the response to injury may show itself 
purely as a proliferation of the epithelial cells. As I 
have said, observations of this nature labour under the 
disadvantage that they must, of necessity be discontinuous. 
I bring them in at this point, inasmuch as they represent 
the mildest condition of the inflammatory reaction. I 
have not personally observed this series of changes in 
tissues which permit of continued study under aseptic 
conditions ; neither am I acquainted with any observa- 
tions wholly fulfilling these conditions made, that is to 
say, upon transparent vascular tissues subjected to the 
mildest aseptic injury and examined continuously under 
the microscope. 

(ii.) The response to injury in the cases just mentioned 
was of the slightest. Let us now pass on to cases in 
which it becomes more pronounced ; and in order to 
continue the comparative study of inflammation I would 
first describe the series of events in a highly vascular and 
transparent region in a low vertebrate animal, namely, 
in the tadpole's tail. If this be injured, either by the 
application of a caustic or by the introduction of a 
foreign inert body into its substance, a definite advance 
upon what was recognisable in the case of the axolotl, 
for example, is to be made out. Here the tail is very 
vascular, the wandering cells in the connective tissue are 
very few in number, while the blood is fairly rich in 
leucocytes, which are small relatively to the size of the 
vessels. The results of injury are a congestion of the 
vessels, noticeable within fifteen minutes, and a well- 
marked determination of leucocytes to the injured region. 
These cells, in the main, pass out from the vessels ; the 
few leucocytes pre-existing in the tissue appear to play a 
very small part. Compared with the axolotl experiment 


this observation is of considerable interest. Instead of 
a slight reaction slowly developing there is a rapid reaction ; 
instead of a slight accumulation of leucocytes there is a 
most pronounced accumulation. If there be any meaning 
in the determination of leucocytes to the region of injury, 
then evidently the active participation of the vessels of 
that region in the reactive process is fraught with benefit 
it is a further important factor developed with the 
development and advance of the organism. 

The fuller details of this vascular interference in the 
inflammatory process have been followed by many 
observers, among whom first and foremost was Cohn- 
heim (28), and to this end the frog has supplied the most 
convenient means in regions at once vascular and fairly 
transparent, such as the web of the hind feet, the tongue, 
and the mesentery. Other observers passing higher in 
the scale oT^vertebrates have employed the mesentery of 
the cat, dog, and other mammalia. Suffice it to say that, 
with slight modifications due to local conditions in the 
tissue examined rather than to the animal selected, 
the process has been found to present the same features 
throughout the whole of the adult vertebrata, from the 
reptilia upwards. For general examination, perhaps, the 
best and simplest method of observing the succession 
of changes that follow injury of a vascular area is to be 
found in Coats' modification of Cohnheim's original 
experiment upon the frog's web (30). In order to 
reproduce as nearly as possible the conditions of an early 
wound, instead of employing a caustic or chemical 
irritant, a small portion of the cutaneous surface is nipped 
off the section being just deep enough to pass through 
the cutaneous layers without causing haemorrhage. For 
the experiment to proceed satisfactorily, it is necessary 
that the frog be curarised after having been pithed. 
The web of a small frog is so thin that the changes 


occurring in and around the vessels of the part can 
readily be followed even with a high power of the 

The first change noticeable in the immediate neighbour- 
hood of the injured membrane is a dilatation of the vessels, 
first of the arterioles (although this may be very slight), 
and then of the veins ; and in this first phase there is 
a very evident acceleration of the blood-flow. At this 
early period the capillaries show little evidence of 
dilatation, but in the course of an hour expansion is 
readily distinguishable, and sundry capillary channels, 
previously invisible, become occupied by blood and show 
themselves. This first stage lasts for an hour, or in some 
cases perhaps two, and is followed by a phase of slowing 
of the blood-current. While previously a well-marked 
axial stream of corpuscles had been evident, with a 
peripheral zone of plasma devoid of corpuscles, the 
former now broadens out, the latter becomes less and less, 
and as it narrows increasing numbers of the clearer 
rounded haemal leucocytes are to be seen in it travel- 
ling at a slower rate than the more axial stream, 
and every now and then stopping beside the walls of 
the vessels, and after a short stoppage passing on 
again. The leucocytes conduct themselves as if they 
have become "sticky." 1 As pointed out by Wells (31), 
it" is eminently probable that the diffusible noxa causes 
modification of the surface tension of the endothelium, 
and it may also be of the leucocytes, and that the 
stickiness is due to this cause. We see the first signs^of 
margination. Lister demonstrated long years ago that 

1 Even so low down in the scale as Daplmia this same peculiarity is 
noticeable : there in health, as Mr. Hardy (23) has pointed out, the 
leucocytes move freely ; but, if the slightest injury be inflicted upon the 
carapace, the leucocytes, previously unadhesive, soon show the tendency 
to adhere to the walls of the body-cavity beneath the region of injury and 



PT. I 

the red corpuscles also manifest an increased viscidity, 
which again is a factor in the slowing of the blood 

As the current becomes yet slower all distinction 
between axial and peripheral streams is lost ; the cor- 


FIG. 4. Inflamed mysentery of frog to show inai-gi nation of leucocjtes in the dilated 
capillaries (a) ; migration of leucocytes (b) ; escape of red corpuscles (c) ; accumu- 
lation of leucocytes outside the capillaries (d). After RIBBERT. 

puscles, closely packed together, fill the whole lumen ; 
the leucocytes in increasing number approach the vessel- 
walls ; they adhere more firmly, and so long as a current 
is recognisable the action of the - stream leads them to 
assume a pear-shaped appearance, the rounded ends 
pointing in the direction of the current. 

As the stream slows gradually the corpuscles may at 
last move in a series of jerks synchronous with the beats 


of the heart ; or frequently in the veins and capillaries 
the mass of blood may be seen moving slowly first in 
one direction, then in the other. Frequently one or 
other of these stages is followed by complete stagnation, 
or stasis, of the blood in the vessels of the injured area 
I say frequently, for at other times little or no 
absolute arrest is seen in the vessels. Accompanying 
this stage, although observers employing other and 
chemical methods of inflicting injury have in general 
omitted to call attention to the fact, there is already a 
considerable oozing or exudation of clear fluid from the 
wound ; there is, that is to say, an outpouring of lymph, 
and this apparently from the distended vessels. Now, 
with the slowing of the stream the leucocytes, accumulated 
next to the walls of the small veins and within the 
capillaries, pass from the interior to the exterior of these 
vessels ; and, if the process be studied carefully with a 
higher power, it can be seen that this mode of passage is 
of an active, or apparently active nature. 1 A series of 
leucocytes can be distinguished, some of which are 
rounded or flattened, in immediate contact with the wall 
of the vein ; others possess a prolongation passing into 
the wall ; in others, again (or in the former if they be 
watched in the fresh specimen), the prolongation enlarges 
on the outer side of the small vessel while the portion of 
the leucocyte within the vessel becomes smaller. The 
final phase of this act of migration is that the whole 
leucocyte passes through, and is found in the lymph- 
spaces around the vessel-wall. This process of migration 
may be so general that in the course of five or six hours 
all the small veins of the region show a crowd of 
leucocytes situated along their outer surface. With these 

1 The process can be fully made out if at this stage the wounded 
region be removed, fixed immediately in weak osmic acid, and prepared 
for examination by the higher powers of the microscope. 


a greater or smaller number of red corpuscles may also 
make their escape. 

In this modification of Cohnheim's experiment a 
further stage is to be recognised. While at first the 
fluid exuded from the cut surface was clear and relatively 
free from cells and cell-debris, now, as the inflammatory 
process continues, an increasing number of leucocytes is 
contained in the exudate. The leucocytes do not remain 
in the immediate neighbourhood of the vessels, but many 
of them pass on to the injured surface ; still, it would 
seem, by active amoeboid movement. Thus at the end 
of six hours this surface may be covered by a serum or 
fluid containing large numbers of these leucocytes. Here 
then we have the first step towards the formation of a scab 
or provisional protective covering to the wound. 

Further observations cannot well be carried out in 
the pithed and curarised frog; but if an unpithed, nori- 
curarised animal be taken, and the observations upon the 
earlier stage be neglected, it can be made out that if 
irritant matter do not find entry into the wound the 
process may be arrested at this point; the leucocytes 
upon the surface may break down, and with their 
breaking down and the formation of fibrin a soft scab be 
formed : the stasis of the blood in the distended vessels 
may be followed by a re-establishment of the current 
and slow return of the vessels to their former calibre, 
while beneath the thin, soft scab the epithelial cells 
rapidly proliferate. Within twenty -four hours there 
may be abundant evidence of this new growth of the 
epithelium tending to encroach upon and cover the 
wound ; and not merely of new growth, but also of dis- 
placement and movements of the healthy cells at the 
edge of trie skin wound towards and over the bared 
surface. At the same time the region becomes less and 
less populated with leucocytes, so that not to enter 


fully at this point into the reparative process within 
sixty hours the region may show little sign of the injury 
and consequent inflammation. 

(iii.) On the other hand, if irritants of a microbic 
nature enter the wound the process may extend, as in 
inflammation of the cornea. More especially if the water 
in which the frog is kept become foul, there is a tendency 
in the inflammatory processes to spread, and in the. 
cells, both fixed and migrated, of the central area to 
break down, leading to the formation of a spreading 
ulcer. The steps of this sequence of affairs it is difficult 
to follow by continuous microscopic examination, partly 
on account of the increased opacity of the region, partly 
because the process extends over days rather than hours. 
Here, therefore, I merely mention this possible extension 
of the change with its main naked-eye appearances. 

It is not possible by continuous observation to make 
out the steps of this more extensive inflammation 
characterised by excessive emigration of leucocytes and 
destruction of these together with the fixed cells of the 
tissue the pyogenetic inflammation. Several observers, 
however, have followed its successive stages by means of 
examination of affected tissues at successive intervals 
after the infliction of injury. 



WHILE, as shown by Councilman (32), Grawitz and de 
Bary (33), Steinhaus, Leber, and others (34), a sup- 
purative inflammation may, under certain conditions, be 
brought about experimentally by the action of chemical 
irritants, such as mercury and turpentine ; yet under 
ordinary pathogenetic conditions suppuration is induced 
by the growth of micro-organisms within the tissues. 
Hence it is better to study the conditions as induced by 
the inoculation of pus -producing microbes into one or 
other tissue. A very full series of observations upon 
the development of abscesses through the agency of the 
Staphylococcus pyogenes aureus has been made by 
Hohnfeldt (35). He inoculated small quantities of pure 
cultures of the microbe subcutaneously into rabbits. 
Four hours after inoculation the vessels of the region 
were found densely filled with corpuscles, and in them 
a commencing margination of the white corpuscles was 
discernible. Leucocytes were present within the tissue 
in numbers greater than normal ; although, compared 
with later stages, they were infrequent. They were of 
two kinds the mononuclear in the majority, the poly- 
morphonuclear l in lesser numbers ; both forms were 

1 This term while precise is painfully sesquipedalian. For myself I 



congregated mainly around the line of entrance of the 
injecting needle. Many of the connective-tissue cells 
were so swollen as to be rounded rather than flattened. 
The injected cocci, lying in the lymph-spaces, were 
scattered through the tissue ; in part free, in part already 
ingested by cells, not only by the leucocytes, but also by 
connective-tissue cells, the number within leucocytes 
being not inconsiderable. Preparations made at the end 
of ten hours showed the same conditions, but more 
distinctly. There was ample evidence of migration of 
the leucocytes, margination in the congested vessels, 
various stages of passage through the vascular walls, and 
large collections of the cells in the perivascular lymph- 
spaces ; from these they spread into the spaces between 
the bundles of connective-tissue fibrils. The cocci lay 
in the lymph-spaces in increased numbers, and the 
massing of leucocytes corresponded in position to the 
accumulation of microbes. In these regions the leuco- 
cytes were mainly poly morphonu clear, but in the 
boundary zone away from the cocci the mononuclear 
form predominated. At the end of twenty hours there 
was further accentuation of these conditions. As yet 
an abscess proper had not formed, but there were 
enormous numbers of leucocytes and also of micrococci ; 
the fibrillse of connective tissue were widely separated by 
the collections of leucocytes, which clustered round and 
hid the connective-tissue cells. With the completion of 
forty -eight hours a well-defined abscess had formed, 
separated sharply from the surrounding healthy tissue. 
The centre of the abscess was seen to consist of densely- 
packed leucocytes mingled with large growths of cocci. 
These leucocytes were almost entirely polymorpho- 

am willing to see it replaced by the earlier polynuclear, provided it be kept 
in mind that so used polynuclear is an abbreviation and not the synonym 
for multimtclear. 


nuclear ; and in this central area the nuclei of some 
showed fragmentation. Such accumulations of living and 
dead leucocytes in fluid matrix constitute pus. Neither 
leucocytes nor connective- tissue cells showed the slightest 
indication of mitosis. In the central area all traces of 
the previous capillaries had disappeared ; in the peri- 
pheral zone they were easily recognisable, being fully 
injected and showing a marginal disposition of their 
leucocytes, many of which could be seen (in osmic acid 
preparations) fixed in the process of migration. The 
majority of the cocci lay in these leucocytes. Even 
where the colonies of the microbes were thickest there 
the majority were intracellular. Passing towards the 
periphery the number of cocci became smaller and 
smaller. At the periphery they could be seen not 
only to be intracellular, but also free in the lymph- 
spaces ; and Hohnfeldt, with other observers, saw them 
definitely grouped within the endothelial cells of the 
peripheral vessels. Thus it may be noted that at 
this stage the proliferating microbes extended into the 
healthy tissues outside the abscess. In the centre of the 
abscess the original tissue had wholly disappeared ; 
nearer the periphery light streaks and bundles' of the 
disintegrating fi brill a3 could be recognised between the 
leucocytes. Not till about the tenth day did new growth 
of tissue begin to show itself. During the preceding six 
days there had been more breaking down of the poly- 
morphonuclear cells, characterised by fragmentation of 
the nuclei and by fatty degeneration of the cell- 
substance. But by the tenth day the periphery had 
begun to assume the appearance of granulation tissue ; 
it contained numerous capillaries and new-formed 
connective tissue with characteristic epithelioid cells or 
fibroblasts possessing large, oval, pale-staining nuclei. 
In these cells, as in the connective-tissue cells of the 


surrounding healthy tissue, the numerous steps of 
indirect cell-division could be made out. In this granu- 
lation tissue cocci were absent and leucocytes were in- 
frequent. In the soft, cheesy, central area masses of 
cocci were still present. Whether these were living or 
dead, Hohnfeldt did not determine ; he inferred (what 
has since been proved by several observers to be an 
unsafe inference) that inasmuch as they stained well 
with aniline dyes they were alive. 

Thus, to sum up Hohnfeldt's observations, the pro- 
cesses occurring in a suppurative inflammation that ends 
in healing are the following : 

1. Primary multiplication of the pyogenic organisms 
with no immediate reaction. 

2. Congestion of the region of invasion, with margina- 
tion of the leucocytes ; signs of degenerative changes in 
the tissue cells of the region. 

3. Collection, in the region, of mononuclear leuco- 
cytes ; then immigration of polymorphonuclear leuco- 
cytes ; multiplication of the cocci. 

4. Ingestion of large numbers of the microbes by the 
polymorphonuclear leucocytes and other cells, including 
the endothelial cells of the vessel-walls. 

5. Increasing immigration of leucocytes until the 
tissue becomes densely packed. This is accompanied by 
a yet greater proliferation of the microbes, which extend 
(i.e. are carried by lymph-streams or by cells) into the 
region outside the developing abscess. 

6. Coincident destruction of the tissue of the affected 

7. Degeneration of the leucocytes within what is now 
the sharply defined abscess. 

8. Eventual proliferation of the connective tissue at 
the periphery of the abscess ; formation* of fibroblasts 
in the highly vascular surrounding zone ; absorption, 


cicatrisation, and encapsulation of the debris of the 
leucocytes and micrococci. J 

There are not a few points in connexion with these 
observations of Hohnfeldt that deserve discussion ; very 
possibly he has misinterpreted certain of the appearances. 
On the whole, however, he draws a full and accurate 
picture of the successive stages of suppurative inflamma- 
tion, and I may defer discussion to a later review of the 
action of the leucocytes and of the formation of fibrous 
tissue respectively. 

However, before leaving this general description of 
the series of anatomical changes induced by injury, there 
is another phase of the inflammatory process set up by 
pathogenic micro-organisms which must not be passed 
over I refer to those cases in which, instead of ending 
in repair, there is extension and generalised disease. The 
stages preceding extension vary witli the nature of the 
microbe ; thus, in some cases, the reaction to the invasion 
of the microbe is mainly leucocytic (as with inoculations 
of the x micrococci of suppuration), in others it is mainly 
exudative or serous, the congestion of the vessels being 
followed by abundant exudation of serum into the tissues. 
This is the case in inoculation of animals such as 
rabbits, guinea-pigs, and fowls with cultures of micro- 
organisms which are peculiarly virulent in their behaviour 
towards these animals. Such a serous or exudative 
inflammation is, for instance, well seen if the vibrio 
Metchnikovi be inoculated into the pectoral muscles of 
a fowl. Within twelve hours, it may be, the seat of 
inoculation becomes greatly swollen, and on section is 
found reddened and congested ; while from it drains 
an abundance of relatively clear, faintly reddish serum 
containing but few leucocytes. 

In such a case as this the micro-organisms appear to 
pass with ease from the centre of infection into the 

OH. V 


surrounding tissues, and thence into the lymphatics and 
general circulation, whence they may be obtained within 
twenty-four hours. Where there has been a well-marked 
abscess - formation in the region of invasion there, as 
already indicated, it is true that the microbes may be 
found outside the abscess at a fairly early period ; but in 
the main, proliferation is limited to the abscess, and the 
blood remains free and sterile. Under certain conditions 
of greater virulence of the pyogenic microbes it is found 
that as the abscess extends it becomes ill-defined there 
is no sharp demarcation between the collected leucocytes 
and the surrounding tissue ; the columns of leucocytes 
spread indefinitely from the centre, and numerous micro- 
cocci are intermingled with them. Where this is the 
case there is a marked tendency for the microbes to find 
their way into the general circulation from this irregular 
peripheral extension along the lymphatic spaces, and to 
set up a condition of septicaemia as in the more serous 
inflammation described above. 

Septicaemia, or the passage of micro-organisms into 
the blood, 1 with all the results of such a passage the 
condition which sundry French observers have described 
as inflammation of the blood we shall not discuss here. 
In septicaemia we pass beyond the local response to 
injury, we deal with a state of general systemic dis- 
turbance. Nevertheless certain phases of the septicaemic 
condition throw light upon the inflammatory process. 

In the first place, it is of interest to note that when 
the infective micro-organisms and their products are 
within the vessels they fail to induce the cardinal symp- 
toms of inflammation. They do not lead to exudation 

1 This is the sense in which the term Septicaemia is employed by patho- 
logists and bacteriologists ; by some surgical writers the word is employed 
as synonymous with Toxaemia ; to avoid confusion there is now a tendency 
to use the more precise term Bactericemia to indicate the presence of 
microbes in the blood. 


of fluid from the blood or to widespread diapedesis of 
leucocytes. The stimulus, whatever it be, which leads 
to these phenomena at the point of invasion is no longer 
called into activity when the noxa is within the circu- 
latory apparatus. This is the reverse of what might be 
expected were the inflammatory process primarily due to 
a modification of the endothelium of the vessel -walls 
by the irritant, a modification passively permitting the 
exudation and passage outwards of the leucocytes. 

This statement that infective micro-organisms and their 
products circulating within the blood fail to induce inflam- 
matory changes, would seem to need modification when the 
development of metastatic abscesses is taken into account 
of secondary abscesses, that is, in tissues at a distance, 
and not in direct continuity with the original focus. 
But a study of the mode of production of these abscesses 
shows that the statement still holds. Such abscesses 
originate round emboli of pyogenic micro-organisms 
in the capillaries. Sundry cocci are arrested in a 
capillary, proliferate, and fill the vessel. It must not be 
thought that the plugging or embolism is an immediate 
process in these cases that the bacteria are present in 
the blood in such numbers that a cluster of them, passing 
into a capillary, blocks it. Experience shows that, with 
a bacterisemia of this extent, death ensues so swiftly that 
there is no time for abscess -formation. Either the 
bacteria are contained in a small mass of infected blood- 
clot, which is the immediate cause of the blockage (as, 
for example, in abscess of the kidney secondary to 
infective endocarditis, and in lung abscesses following 
upon suppurative thrombosis of the lateral sinus), or, in 
other cases, isolated bacteria passing with the blood into 
a capillary are taken up by a phagocytic endothelial cell, 
and, instead of being destroyed, multiply, lead to the 
death of that and surrounding cells, and form a colony 


occluding the capillary. It is only when a minute 
vessel is thus occluded that the abscess-process begins, 
that is to say, when by this occlusion the vessel has 
become extravascular ; and while it is true that, primarily, 
the arrest of pathogenic microbes within the capillaries 
is often associated with a small accumulation of intra- 
vascular leucocytes and with degenerative changes in 
the vascular endothelium, the metastatic abscess, as such 
forms not by accumulation of leucocytes in the occluded 
vessel, but around it ; the leucocytes migrating from 
surrounding capillaries. 

The Production of Suppurative Inflammation in 
Serous Cavities. Certain important modifications of 
the process above described as occurring in abscess 
formation are to be observed when inflammation affects 
the highly vascular walls of serous cavities, such as 
the abdominal, the pleural, or the peritoneal. From the 
very vascularity of the affected layers there is, in the 
first place, a tendency to a relatively abundant exudate, 
and in the second, this exudate pouring into a large sac 
is apt to bring about the extension of the irritant over 
a wide surface. Thirdly, not only do we find leucocytic 
migration, but, in addition, the endothelium covering the 
serosa now plays an important role. If a moderately 
severe inflammation be set up by introducing a culture 
of the Bacillus coli into the abdominal cavity of the 
guinea-pig, then, as pointed out by Beattie (36), if 
samples of the inflammatory exudate be withdrawn at 
intervals, a definite increase in the polynuclears is recog- 
nisable in from one hour to two and a half hours, and in 
non-fatal cases the number becomes greater and greater, 
until from thirty to thirty-six hours have elapsed. In 
such non-fatal cases the number of these cells returns to 
the normal in about seven days. At from the eighteenth 
to the twenty -fourth hour a cell of a wholly different 



type makes its appearance. Unlike the polynuclear 
leucocyte, the nucleus of this form is characteristically 
not lobate, but round or oval round in the smaller 
cells of this type, and then deep staining with but little 
surrounding protoplasm ; oval and pale staining in the 
larger cells, and then surrounded by a large cell body, 
constituting the so-called " hyaline cell." As Eanvier 
showed (37), these cells are of endothelial origin, derived 
from the actively proliferating serous endothelium. From 
this period onwards these cells increase in number, until 
during the third day in non-fatal cases they greatly out- 
number the polynuclears, and may be still present up to 
the eighty - fourth hour, when the latter have almost 
completely disappeared. Reference will be made to 
these cells and their properties in the next chapter. 



IN the second place, through this study of advancing 
inflammation it is of interest to trace the very close 
relationship that exists between inflammation and fever. 
Besides the local changes here described, local injury is 
accompanied by systemic disturbances. These may be 
slight or grave. Take, for instance, progressive abscess- 
formation, or follow the development of a malignant 
carbuncle in man. At first the reaction is purely local, 
but very soon, long before any of the micro-organisms 
are capable of detection in the blood, there is a raised 
temperature and a slight febrile state, the fever becoming 
more and more evident as the local process becomes 
more and more extensive, until with the detection of the 
microbe in the blood the most severe fever, with con- 
stitutional disturbance, sets in. Local inflammation, 
then, without any other possible explanation than either 
the nervous irritation to which it may give rise, or the 
passage into the general circulation of the soluble 
products of bacterial growth and tissue - destruction, or 
both, may lead to the development of the febrile state. 
How large a share is played by these two possible factors 
it is difficult to say. That bacterial products injected 
into the circulation lead to the rapid production of the 



febrile state rests on ample evidence ; but whether these 
act directly by inducing increased cellular activity, or 
indirectly by stimulating the cerebral centres, we cannot 
absolutely say. As yet we have little accurate know- 
ledge of the part played by the nervous system in the 
development of the febrile state. This, however, may 
safely be declared, that the more we study the continued 
fevers of bacterial origin, the more do we discover that 
these commence by a local inflammatory disturbance. 
The continued fevers of bacterial origin are, in general 
the continuance, or rather the extension, of a primarily 
localised inflammatory lesion. This it may be noted is 
not necessarily the case with fevers of protozoal origin, 
caused, like malaria, by the introduction of the specific 
organisms into the circulating blood through the agency 
of puncture of subcutaneous capillaries by insects that 
" stiflg-" The sting of such insects may, it is true, cause 
a local inflammation, but this bears no direct relationship 
to the multiplication of the infective microbe in the blood. 
In this conception of inflammation as essentially ja 
local process we find ourselves at variance with Eibbert 
"(38), whose views have been somewhat widely accepted 
among German pathologists. Ribbert holds that " in- 
flammation includes both local and general processes " 
(loo. cit. p. 11). He calls attention to the fever which 
may supervene in the course of inflammation, the active 
new formation of leucocytes, mainly from the bone 
marrow, and discharge of the same into the blood, whereby 
leucocytosis is set up, and to the increased development 
of bactericidal and anti-toxic substances in regions distant 
from the area of primary injury ; also to the various 
degenerative processes in distant organs cloudy swelling, 
fatty degeneration, etc. evidently due to the diffusion of 
toxic substances either from the inflamed area or from 
organised bodies, such as bacteria, which have escaped 


from the zone of active inflammation. But at the same 
time he admits (1) that these general processes can be 
induced experimentally without there being any local 
disturbance, as also that (2) mild grades of inflammation 
may lead to no general involvement of the system. He 
acknowledges thus that the two are not inevitably and 
essentially united. 

The logical outcome of such an inclusion of the 
general body disturbances into our conception of inflam- 
mation leads to a lack of anything like a boundary 
between the condition of inflammation and that series of 
general disturbances which for want of a better term we 
speak of as (general) infection. According to Ribbert, 
inflammation must be made to include infection. 

We see no " must " in the case ; on the contrary, and 
we gravely doubt the wisdom of such teaching. One 
might as well banish wholly the idea of inflammation as 
a distinct entity. It appears to us both more sensible 
and more in accordance with tradition to hold that (1) 
inflammation is the local reaction to injury, (2) that this 
may or may not be accompanied or followed by systemic 
reaction, and (3) that this systemic reaction to injury 
covers so wide a field, including not only the discussion 
of fever, leucocytosis, anti- toxic formation, and the 
principles of immunity, but also (as Ribbert neglects to 
note) that of pain and shock, that advisedly it should be 
considered^ apart and then in several chapters. That in 
studying the phases of the local reaction we must 
repeatedly advert to the systemic reaction is one thing ; 
it is another thing to regard systemic reaction as an 
essential part of the process. 



THE main facts gathered thus far concerning the 
inflammatory process, and the conclusions to be drawn 
therefrom, may now be placed in order before I discuss 
in detail the various factors in the process. They are 

1. Injury, when it is not so widespread and severe as 
to lead to the death of the individual, is followed by a 
reaction on the part of the organism. This reaction may 
be (a) ^ocslised. (b) generalised ; it is the first of these 
that constitutes iE^mmation. 

2. In unicellular organisms the continued vitality of 
the individual after injury, and in multicellular organisms 
the vitality of the individual cells, is dependent primarily 
upon the persist enceLflf the^nucleus ; if this be destroyed 
or removed, the rest of the cell is incapable of complete 
restitution and continued growth. 

3. In unicellular organisms the reactive process is 
twofold, and consists of (a) destruction or removal of the 
irritant; destruction being brought about by a process of 
intracellular digestion, removal by extrusion of the 
irritant ; (b) new growth of the organism. 

4. This response to injury on the part of unicellular 
organisms is essentially reparative. 

5. In multicellular organisms, with division of labour 
among the constituent cells of the individual, there is a 



separation of functions ; the twofold local reaction to 
local injury is yet more clearly marked ; but 

(a) The destruction or removal of the irritant is in 
the main accomplished by the wandering cells of meso^. 
bias tic origin. 

(b) The new growth to replace the tissue destroyed 
by the irritant proceeds in the main from the fixed cells 
of the tissue. 

6. Ascending the scale of multicellular organisms, a 
division of labour and differentiation of function is 
discoverable among the wandering mesoblastic cells. 
Whereas in the lower forms of the Metazoa one form of 
leucocyte alone is present, in the higher forms two or 
more varieties can be distinguished which possess 
different properties and act differently towards irritants 
introduced into the system. 

7. According to the nature of the irritant causing 
the injury, the leucocytes are actively attracted in 
greater or less numbers to the region of injury, surround 
the irritant, and remove or destroy it by means very 
similar to those employed by unicellular organisms. 
Where the irritant is present in the form of discrete 
particles, some at least of the leucocytes may incorporate 
the particles, and remove them or destroy them by a 
process of digestion. Others of the leucocytes in the 
higher Metazoa do not act thus as phagocytes ; never- 
theless they are equally attracted to the focus of 
inflammation, and .presumably, tend to counteract the 
irritant by some other (extracellular) means. 

8. While to the wandering cells appears to be allotted 
the main duty of removing deleterious and irritant 
matters, certain of the fixed cells of the organism, notably 
the endothelial cells of the vessels, also exert these 

9. Among the very large number of Metazoan forms 


in which no complete vascular system is present, this 
attraction of the leucocytes to the region of injury is at 
first the sole recognisable response to injury, proliferation 
of the fixed cells occurring in the neighbourhood of the 
injury at a later period. The relation between leucocytic 
migration and tissue proliferation is, however, a variable/ 
quantity ; there may be active proliferation in response? 
to injury with little evidence of wandering in of leuco-^ 

10. Among the higher Metazoa, in which there is 
a well-developed vascular system, the determination of 
leucocytes to the region of irritation still continues, nay, 
more, is markedly aided by the participation of the vessels 
in the inflammatory process. 

11. The vascular phenomena in inflammation may be 
regarded as serving two main purposes (a) the pouring 
out of increased fluid into the injured area; (b) the afflux v ; 
and migration of leucocytes. 

12. Even in the highest Metazoa, possessing fully 
developed vascular systems, the response to injury in a 
non-vascular area, such as the cornea, may be associated 
with no change in the surrounding vascular areas, but 
purely with a determination to the injured area of leuco- 
cytes already free in the surrounding tissues. 

13. The second phase of the inflammatory process, 
that of tissue -repair, but very rarely occurs without 
evidence of previous migration of leucocytes and exudation 
from the congested vessels ; nevertheless with certain 
mild grades of injury it can occur. 

14. A comparative study leads inevitably to the con- 
clusion that the determination of leucocytes to the region 
of injury is the most constant and most characteristic 
early response to injury recognisable throughout the 
Metazoa, and that it must be regarded as the most 
important factor in the first stage of the inflammatory 


process. The vascular phenomena noticeable in the 
higher Metazoa must be regarded as a second and highly 
important adjuvant factor of later development. New 
tissue -formation is the prominent characteristic of the 
later stages of the process. 

15. As among the Protozoa, so in the Metazoa, the 
response to injury is consistently in the direction of 
repair of injury. 

This general survey of the response to injury through- 
out the animal kingdom demonstrates most clearly that 
the same broad principles, the same methods of defence 
and repair on the part of the organism, are called into 
activity from the lowliest forms to the highest ; that, in 
fact, no line can be drawn to separate one set of phenomena 
as truly inflammatory from another set which, while also 
a response to injury, is non-inflammatory. Although it 
is jtrue that the term inflammation implies a reddening 
and congestion of the vessels, we find upon closer examina- 
tion that reddening and congestion are not the funda- 
mental but superadded features in the process of repair 
of injury features superadded as the organism advances 
in its place in the animal kingdom. Thus if we are to 
comprehend the process satisfactorily, we must pass 
beyond the narrower acceptation of the term. 

Having thus sketched broadly the general phenomena 
of the inflammatory process, it will be well now to describe 
in fuller detail the factors of this process among the 
higher vertebrata, and to bring together the more import- 
ant results of the study of the respective functions of 
the wandering cells, the vessels, the fixed cells, and the 
systemun inflaminaiion. 





As I have already shown, there is more than one form of 
leucocyte in the mammalian organism, and these several 
forms evidently possess different attributes, and act 
differently in the reaction to injury. Inasmuch as they 
have been variously classified so variously, in fact, that 
it is often far from easy to collate the various descrip- 
tions and to discuss the forms distinguished by one 
observer in the terms of another it is necessary to give 
the chief classifications of them and their relations. 

The first to discriminate between the forms of white 
corpuscles in the blood was Wharton Jones (39) so long 
ago as 1846. He drew a distinction between 

A. Granule cells Finely and coarsely granular. 

B. " Nucleated " cells Non-granular. 

These observations were confirmed and advanced by 
Max Schultze (40), who made out the following forms : 

1. Small round cells with round nucleus and little clear proto- 


2. Larger cells with round nucleus and more clear protoplasm. 

3. Cells with finely granular protoplasm, and one, two, or more 


4. Cells with coarse granules in the protoplasm. 



The distinctions drawn were, so far, purely morpho- 
logical ; and very little notice was taken of these 
varieties for a long period until Ehrlich (41), in a 
notable series of papers extending from 1878 to 1887, 
drew attention to the fact that the wandering cells of 
the organism react diversely towards the different 
aniline dyes, and possess diverse tinctorial affinities 
indicating chemical differences in the nature of certain 
constituents of the cell bodies. The granules of the 
previous observers were found to be variously affected 
by the dyes employed ; they were shown not to be fatty, 
but as Ehrlich put it of the nature of a glandular 
excretion ; l and comparing the effects of the two groups 
of aniline colours that in which the dye is associated 
with the acid constituent of the salt and that wherein 
the dye forms the base (the " acid " and " basic " aniline 
dyes respectively) he made out the existence of five 
forms of granulation associated with as many varieties of 
wandering cells. His table of cells, according to their 
granulation, is as follows : 

a. granulation Eosinophil Cells frequent in horse's blood, 
present constantly in small numbers in human blood ; 
numerous in medulla of bones of rabbits, dogs, guinea- 
pigs, etc. Stain deeply with acid aniline dyes. Granules 
large and coarse. 

fl. granulation Amphophil Cells frequent in rabbits and 
guinea-pigs in blood ; present also in medulla of bones. 
Stain both with acid and basic dyes. Granules fine. 

y. granulation Basophil Large cells found in the con- 
nective tissue, from the frog upwards, " Mastzellen " ; in 
blood of man only in certain cases of leukaemia. Stain 
only with basic dyes. Granules coarse. 

8. granulation Fine Basophil The " mononuclear " leucocyte 
of human blood. Granulation fine. Stain with basic dyes. 

1 J. Weiss has studied the micro-chemical reactions of the eosinophilous 
granules, and concludes that they are of albuminoid nature ; since they 
were found not to be digested in gastric juice, he would ally them with 
the nucleins. 


. granulation Neutrophil The most frequent leucocyte 
of human blood, polymorphonuclear or " poly nuclear." 
Stain only in neutral dyes not in acid or basic. 

Ehrlich and his pupils and Kieder (42) have done 
much to throw light upon the relative numbers of the 
leucocytes possessing these different granulations in 
different diseases, but at first they accomplished little 
in discovering -the origin of the various forms, their 
functions, or their relationships. We owe the first 
satisfactory studies upon the properties to the long- 
continued and wonderful series of researches upon 
Phagocytosis initiated by Metchnikoff, who made out 
that the different wandering cells of the body act 
differently towards microbic and other foreign particles 
introduced into the organisms. Thus he was led to 
draw a distinction between 

1. Lymphocytes immature leucocytes. 

2. Large hyaline cells, mononuclear, phagocytic, " macrophages." 

3. Smaller neutrophil cells, polynuclear, " microphages." 

4. Eosinophil leucocytes very rarely phagocytic. 

English observers, notably Professor Sherrington (43), 
the late Dr. Kanthack (44) and Mr. Hardy (45), and 
more recently, following these, Dr. Durham (46), Pro- 
fessor Muir, and Dr. Beattie (36), have made notable 
advances in the determination of the function of the 
wandering cells in inflammatory and other conditions. 
Of American observers, Dr. A. E. Taylor (47) deserves 
special mention. Kanthack and Hardy have materially 
simplified the classification given by Ehrlich by dividing 
the leucocytes of the blood into : 

1. Coarsely granular),. , ., 

2. Finely granular | 0x yP hl1 cells - Dining with acid dyes. 

3. Coarsely granular \ ,, 

4. Finely granular j Bas P M cells - ^aming with basic dyes. 

5. Lymphocytes. 

6. Hyaline cells. 




Their coarsely granular oxyphil cells are the eosino- 
phil cells of most writers ; their finely granular are the 
neutrophil and amphophil of Ehrlich. They prove 
conclusively that Ehrlich's neutral stain is in no sense to 
be regarded as such, but must be considered as a weak 
acid dye. 


Kanthack and Hardy. 




Finely granular 
Coarsely granular 

f Neutrophil cell.\ 
\ Amphophil cell. / 



Finely granular 

Basophil cell with 


5 granulations. 

Coarsely granular Mastzellen(7gran- 

Coarsely gran- 

basophil. illations). 

ular basophil. 

Lymphocyte. Lymphocyte. 



Hyaline cell. 


/Hyaline cell. 

In connexion with this subject of inflammation yet 
another classification must be entered into. The above 
forms are all to be found in the blood-stream in the 
inflamed tissue. We have to recognise also other 
wandering corpuscles which are not necessarily derived 
from the blood. It may be questioned whether it is 
right to speak of wandering cells of tissue origin as 
leucocytes, that term having first been employed as 
synonymous with white blood corpuscles, but on the old 
presumption that practically all the wandering cells seen 
in inflammation are derived from the blood, it has been 
customary to speak of all orders of these cells by this 
name, and certainly their characters are such as to 
render it difficult to distinguish them one from the 


other save with care. Thus as a matter of custom and 
of practical utility we elect to speak of all these various 
cells indifferently as leucocytes. 1 

Thus further we may divide the leucocytes concerned 
in the inflammatory process into the three following 
groups : 

1. Hcematogenous, viz. those derived from the blood- 
stream ; this includes polymorphonuclear (neutrophil 
and amphophil polynuclear) and eosinophil (coarsely 
granular oxyphil). 

2. The histo-hcematogenous include the lymphocytes 
and hyaline cells and other cell forms apparently derived 
from the lymphocyte. All these may either have passed 
into the inflammatory focus from the blood -stream, or 
may be the result of proliferation of cells already within 
the tissue. 

3. The histogenous cells originate locally as the result 
of local tissue proliferation, and not by passage into the 
parts from the blood-stream. 

The recent observations of Schridde (48), to which 
I shall refer later, indicate that improved methods of 
staining are capable of showing that there are distinc- 
tions to be made out between the lymphocytes of the 
blood and those of the perivascular sheaths, lymph 
glands, lymph follicles, and tissues in general. If this 
be so this class will largely disappear, and we shall be 
able to make a sharp separation between haematogenous 
lymphocytes on the one hand and histogenous on the 
other ; all that will be left in the histo-hsematogenous 
class being the hyaline cells of endothelial origin. 

1 Some modern workers like Schridde are attempting to confine the 
term leucocyte purely to the granular cells of the blood, and make a 
distinction between leucocytes and lymphocytes, which appears to me to 
be indefensible. 



WE need here say little regarding these, they are so fully 
discussed in all text-books of medicine and treatises upon 
the blood. We need but recall that the polymorpho- 
nuclear neutrophil cells are those which, passing from 
the blood, predominate in all cases of acute inflammation 
leading towards suppuration and abscess formation ; that 
in human blood, as in most mammals, the eosinophiles 
while very characteristic are generally rare, though they 
may be noticeably increased in number in certain diseases, 
notably where intestinal parasites, trichina, etc., are present 
in the organism ; the mast cells are a still rarer form. 
While such cells may occasionally be found in the blood, 
we are personally doubtful as to whether those seen in 
inflammatory foci have wandered as such out of the blood- 
vessels. Their granules are large and stain deeply with 
ordinary basic dyes such as methylene blue, and they must 
not be mistaken for small clusters of cocci. 


Here, as above noted, we have first to consider the 
lymphocyte and its modifications. It is now accepted by 
practically all workers that the lymphocyte seen within 
the tissues, and the small cell with round, deeply staining 
nucleus so large as almost completely to fill the cell, and 


having a scanty ring of surrounding cytoplasm, may be 
derived both from the blood-stream and by proliferation 
of the lymphoid cells normally present to a greater or 
lesser extent, more particularly in the sheaths of smaller 
veins. Until recently it has been regarded as impossible 
to make a distinction between these two groups of cells. 
Now Schridde (by his modification of Altmann's stain) 
makes out very definitely that granulation can be made 
out in these cells, and that the granules of the haemal 
lymphocytes are of a wholly different type to those seen 
in the tissue lymphocytes and the cells of lymph follicles, 
and, basing himself with Ehrlich upon the fact that 
the granules of the different varieties of cells are constant 
features of the same and present constant characters, he 
concludes that here we have to deal with two distinct 
cell varieties. The granules of haemal lymphocytes he 
declares are very small, and of about the size of those 
seen in neutrophil leucocytes, and by his method they 
take on a brownish red tint. Those of the peri-vascular 
lymphocytes are larger, few in number, and of a 
pronounced red colour. 

Plasma-Cells. There is another form of mononuclear 
cell which must, we think, be included in this series, 
although, as to its properties and origin, there has been 
and still exists a very active controversy. This is the 
almost notorious plasma-cell. Unna (49) first named 
these in 1891, or rather misnamed them, using a term 
previously employed by Waldeyer for another form of 
cell. Unfortunately his description was not wholly 
adequate, and various later observers have evidently 
included and described under this term cells of more 
than one order, so that confusion is worse confounded. 
The cell which most closely conforms to Unna's descrip- 
tion, and which, therefore, I regard as the plasma-cell 
proper, has the following attributes : it has a relatively 


small, round or oval, not indented nucleus, coarsely 
granular, rich in chromatiri and, further, staining darkly ; 
this nucleus is situated excentrically. The cell-body 
stains deeply with Unna's methylene blue ; the shape 
within the tissues is liable to considerable variation 
often rounded or oval the cells may be polygonal or 
even drawn out into a spindle ; they are obscurely 
amoeboid. As to the possession of phagocytic properties, 
opinion is somewhat .divided. The general agreement is 
that they are not seen to take up bacteria ; they may, 
however, contain inert particles of foreign matter such 
as vermilion or coal-pigment. As they grow larger and 
older, the cytoplasm tends to become somewhat vacuolated, 
and the nucleus to be less deeply stained. As to their 
eventual fate, nothing definite can be said at present. 
Cells having these characters are normally present in 
lymph-glands, spleen, and bone-marrow, and here every 
transition can be made out between them and the 
ordinary lymphocyte, which similarly, it may be noted, 
has a small, round, deeply-staining nucleus (even more 
deeply staining so as to appear homogeneous). Accord- 
ing to von Marschalko (50), they may be noted in any 
inflammation within twenty-four hours. More recent 
observers state that they do not appear in quantity 
within the zone of inflammation until several days have 
elapsed. Now, as noted by Justi (51) and Councilman 
(29), there is normally around the veins a lymphoid 
sheath, and in inflammatory conditions this sheath 
becomes noticeably augmented. The cells here multiply 
by direct division, and, in addition, here and there 
mitoses are visible (Justi). That lymphocytes are 
endowed with faint amoeboid activity (Wlassow and 
Sepp and A. Wolff (52)), and that, further, they migrate 
from the vessels, has now been pointed out by several 
observers (Councilman, Almkvist (53)). It is quite 


possible, therefore, that the increase in the perivascular 
lymphoid sheath is in part due to migration from the 
vessels, and, therefore, that some at least of the hsemato- 
genous lymphocytes_develop into plasma-cells^ Though, 
admitting this, it seems to me evident that the majority 
of the plasma -cells are derived from the proliferating 
perivascular tissue, and thus are not immediately haemato- 
genous. It is on these grounds that I am led to include 
them here in order to emphasise the fact that not all 
the leucocytes seen in the field of inflammation are of 
vascular derivation. 

I have, I admit, here collated and selected out of a 
large number of very divergent descriptions those state- 
ments regarding the plasma-cell which conform with the 
conclusions I, personally, have reached regarding them. 
I am fully prepared to find that not all who have studied 
these cells will agree with what is here laid down, but 
to discuss the divergent views would occupy too much 
space. The reader, however, will find references to the 
leading papers bearing upon these cells given in the 
bibliography at the end of this study. The views here 
given in the main agree with those of von Marschalko, 
Paltauf (54), Justi, Mallory, and Councilman. In their 
properties the cells here indicated conform largely to 
Maximo w's " poly blasts " (55), cells which Maximo w 
regards as distinct from the plasma-cell. 

Maximow's very full and elaborate studies have 
deservedly attracted wide attention. The cells which 
he terms " poly blasts " are mononucleated amoeboid 
cells, having phagocytic properties, and, according to 
him, these may have more than one origin : (1) the 
majority he derives from migrated lymphocytes ; but 
(2) some he regards as of endothelial origin, derived 
from endothelial cells of perivascular lymph tissue, 
serous membranes, etc., which may have passed through 


I>T. II 

the stage of being clasmatocytes (vide page 80); and 
(3) a small number he would derive from wandering 
cells, pre-existing in the tissues. He holds that all 
these cells may develop into large, somewhat actively 
amoeboid cells, as suggested by their irregular shape, 

^ VJ 

^Q 1 * 



FIG. 5. From a case of acute interstitial nephritis (man), drawn by camera lucida, 
i* a Zeiss, ocular -2. Showing plasma-cells (1) in the interstitial tissue between the 
tubules. The epithelium of the middle tubule (2) is degenerated ; the tubule con- 
tains polymorphonuclear leucocytes (3). After COUNCILMAN. 

characterised by possessing a very definite centrosome 
situated close to the nucleus. I am inclined to think 
that Maximow's selection of tissues and of methods 
has led him to overlook the distinction between young 
lymphocytes and cells of endothelial origin. I cannot 
but think that had he studied the peritoneum, for 
example, he would have reached other conclusions. 


The views here laid down are largely supported, 
though at the same time considerably extended by 
Schridde, using his new stains, which, it may be 
added, have the great advantage that by them it is 
possible to differentiate tinctorially the different forms of 
leucocytes present in the inflamed tissues. Schridde also 
sees every grade of development from the perivascular 
lymphocyte to the plasma-cell, though no relation between 
haemal leucocytes and this form of cell. He, however, 
goes farther and finds three types of plasma-cells. The 
commonest of these is that with neutrophil granulations 
or weakly oxyphil. Rarer are those with acidophil or 
oxyphil granules (in which the oxyphilic character is 
somewhat less pronounced than is the case with the 
ordinary eosinophils) ; while the rarest of all are those 
with metachromatic basophil granules corresponding to 
the mast-cells of the blood. We have thus, according to 
Schridde, two parallel series of cells one derived in the 
main (in the adult) from the bone-marrow, thence gaining 
entrance into the blood ; the other derived from the 
perivascular and lymph-follicular lymphocytes. These 
results need confirmation; I would only here say that, 
having seen Dr. Schridde's preparations and knowing 
his excellent technique, I am strongly inclined to accept 
his findings. Using a modified technique Beckton of 
Cambridge (56) confirms Schridde's observations so far as 
regards the existence of granules in lymphocytes and 
plasma-cells and their similarity. 

Hyaline Cells derived from the Vascular Endo- 
thelium. We shall discuss these in connexion with 
histogenous leucocytes in general. Here we would only 
note that there is a certain amount of evidence that 
vascular endothelium can give origin to clear hyaline 
cells with little or no granulation. 



THE part played by these is most clearly seen in the 
study of inflammation of the peritoneum. Many years 
ago Cornil and Ranvier (37) called attention to the very 
active proliferation of the endothelial cells of the great 
omentum in inflammations set up by the injection of 
dilute solutions of silver nitrate. They pointed out that 
these cells, flat and lamellar under ordinary conditions, 
become greatly enlarged with swollen nuclei which, within 
a few hours, undergo direct division, so that many cells 
come to possess two or three nuclei. Indirect division 
(mitosis) is only seen, according to Cornil and Toupet, in 
forty-eight hours. Some more recent observers have 
recognised it much earlier, Beattie (36) as early as the 
tenth hour. So swollen are the cells that they project 
prominently from the surface, being often pear-shaped and 
attached merely by a pedicle. As a result of the active 
proliferation, here and there large, semi-detached clumps 
of these cells may be made out. Cornil and Ranvier 
did not study the exudate ; this has been done by more 
recent observers, notably by Durham (46) and Beattie. 
The former, employing non-lethal intraperitoneal injections 
of several microbes, has described the same changes in 
the omentum as those described by Cornil and Ranvier, 
and has noted, more particularly, that these swollen cells 



give origin, by direct division, to large, clear, mononuclear 
cells or leucocytes. These may be derived from the 
peritoneum generally, but the omentum is especially 
active in this respect. He identifies these cells with 
Metchnikoff s " macrophages," and Beattie has made a 
full study of the same. They are, according to him, the 
most characteristic form of leucocyte seen in peritoneal 
inflammation. After non-fatal peritoneal injection of 
the B. coli in the guinea-pig, he first observed an 
increase in the polymorphonuclear cells of the peritoneal 
fluid, which begins three hours after the injection and 
reaches its maximum in from six to thirty hours. 
Mononuclear leucocytes are first seen to increase in 
number about the eighth hour. On an average these 
are in numbers equal to or even greater than are the 
polymorphonuclears. From now onwards (in non-fatal 
as distinguished from fatal cases) they definitely pre- 
ponderate. As during the next two days resolution 
proceeds, the polymorphonuclears become fewer and 
fewer until the few cells present in the exudate are 
almost entirely mononuclear. These mononuclear cells 
(the hyaline cells of the late Prof. Kanthack and Mr. 
Hardy, and of other observers) vary in size. According 
to Beattie there is every transition from small cells 
resembling lymphocytes with a round or kidney-shaped 
nucleus rich in chromatin, and with scanty protoplasm 
up to cells four or five times as large, having a rounded 
or kidney-shaped nucleus which does not stain nearly as 
deeply as that of the smaller forms, but shows deeply- 
staining nodes of chromatin in the nuclear network. 
These larger cells have abundant cytoplasm, often 
showing extensive, fine vacuolation. In these respects 
the cells are identical with the swollen endothelial cells 
of the omentum. Nor are they merely passive agents 
cells cast off in a moribund condition from the inflamed 


peritoneum and undergoing degeneration. They are 
phagocytic ; they take up bacteria, though, in this 
respect, they are not, with certain exceptions, so active 
as are the polymorphonuclears. Thus, if tubercle bacilli 
be injected, they are found, in a few hours, almost 
exclusively within the mononuclear cells and not in the 
polymorphonuclears. This is in harmony with our 
general experience that polymorphonuclear leucocytes, 
while not negatively chemiotactic to tubercle bacilli, are 
not potently attracted thereto, so that these more 
leisurely mononuclears have the opportunity to take 
them up. What is most characteristic is that these 
mononuclear cells are active cellular phagocytes ; they 
take up other cells of the exudate polymorphonuclears, 
eosinophils, and red blood-corpuscles. A single large 
mononuclear cell of a peritonitic exudate may often be 
seen containing three or four other cells or the remains 
of the same. 

There can, therefore, be no doubt that, in peritoneal 
inflammation at least, an important series of free cells or 
leucocytes is of endothelial or tissue origin ; the question 
is whether all these mononuclear cells are of like 
development. While clearly the larger forms are indis- 
tinguishable from the swollen cells of the peritoneal 
endothelium, and while the time of their appearance in 
the exudate coincides closely with the onset of active 
proliferation of that endothelium, we have the apparently 
contradictory statement that every transition is also to 
be observed between them and small lymphocytes ; while 
other observers studying lymphocytes proper are equally 
convinced that these cells never develop into the large 
hyaline cells above described. Now lymphocytes proper 
may be found in the peritoneal exudate. These 
apparently contradictory results may be harmonised by 
pointing out that, in the first place, in the very earliest 


stages, most cells are indistinguishable, and that, in, the 
second place, a distinction can frequently be made out 
between the lymphocyte proper and the embryonic cell 
of endothelial origin. The former has a deeply-staining, 
rounded nucleus ; the latter a nucleus which, while also 
deeply staining, tends to be indented. There is, however, 
another difficulty ; if the omentum be examined during 
the period of peritoneal inflammation, its vessels are seen 
to contain mononuclear cells which, like these " hyaline " 
cells of the exudate, are devoid or almost devoid of 
granulation, and these in tra vascular cells are not to be 
distinguished from the medium-sized hyaline cells of the 
exudate. Why cannot some of the mononuclear cells 
of the exudate be haematogenous ? There is, apparently, 
no reason why we should deny this hsematogenous origin 
of some at least of these cells, nor would it seem that, 
making this admission, we land ourselves in further 
difficulties. On the contrary, the evidence accumulating 
of late years appears to point in one direction, namely, 
that the mononuclear cells seen within the vessels during 
the process of inflammation are similarly of endothelial 
origin. Metchnikoff, from 1883 onwards, was the first 
to suggest that the large mononuclear leucocytes seen in 
vessels and lymph-glands are identical in their properties 
and characters with the endothelial cells of these tissues, 
and he directed attention to their peculiar properties of 
ingesting other leucocytes, and named them " Macrophages" 
(a barbarous but a convenient term). Buffer's (57) studies 
of the lymph-glands wholly confirmed these observations. 
The fullest studies upon the histology of these cells are 
by Mallory (58), who described, with great detail, the 
large cells with slightly -staining, grooved, or indented 
nucleus, present more particularly in the lymph-glands in 
typhoid fever, and possessing, in short, all the characters 
of Metchnikoff's macrophages. He describes them as 


originating from the proliferation of the endothelial 
cells of lymph -spaces, lymphatic vessels, and blood 
vessels. He has observed their mitosis, and has demon- 
strated the migration of such proliferated endothelial 
cells into the adjoining connective tissue (Fig. 6). 
These observations were confirmed by Councilman in his 
studies upon keratitis (29). We have, indeed, to 
recognise the very active part played by the vascular 
endothelium in inflammation, its pronounced proliferative 
and phagocytic properties. As pointed out by Abbott, 


FIG. 6. Large cells of endothelial origin (macrophages of Metclmikoft') from centre 
of lymph-gland in typhoid fever which have actively taken up lymphocytes (?.) and 
red blood-corpuscles (r. b. c.). MALLOEY. 

Nicholson, and myself (59), within ten minutes after 
inoculating the B. coli into the circulation, the endothelial 
cells of the liver capillaries are swollen and have taken 
up numerous bacilli (Fig. 7). Before our observations, 
Werigo (60) had demonstrated the process whereby 
hepatic endothelial cells give off pseudopodia, by which 
anthrax bacilli are seized and subsequently englobed. 
Eecently Behring and Much (61) have called attention 
to the great phagocytic powers of the endocardial 
endothelium. As Beattie points out, in omental in- 
flammation the cells lining the veins become swollen 


and then detached from the wall, and various stages of 
the separation can be made out. Study of the sinuses 
of the spleen in typhoid (Mallory) shows the process 
very well ; the lining cells swell to three or four times 
their original size and become separated, so that a sinus 
may become distended with these free cells. As they 
swell, their cytoplasm becomes vacuolated, and poly- 
morphonuclears and other cells are seen to be ingested. 
Without entering into a discussion of the finer details 
of nuclear staining (vide Beattie's article) I would point 
out that the evidence now accumulating tends to the 
conclusion that Ehrlich's mononucleated non- granular 

FIG. 7. Endothelial cell of capillary of rabbit's liver, fifteen minutes after 
intravenous inoculation of culture of B. coli, showing bacilli ingested and under- 
going disintegration. (The whole of the nucleus is not here in focus.) 


blood -cells (41) are of endothelial origin. 1 Contrary, 
therefore, to Gulland, Saxer, and Uskow (62), I am of 
opinion that there is far from being a common origin for 
all the white blood-corpuscles ; one group at least is of 
endothelial origin, just as another is of lymphoblastic. 
And thus, recognising that the former are capable of 
migration, it must be admitted that cells of this order 
seen in exudates may be of both local and hsematogenous 

A word may here be said regarding the fate of these 
large mononuclear cells. Evidence seems to show that 

1 With these I do not include the neutrophilic and other myelocytes 
seen in the blood in certain disorders : they, it need scarcely be said, are 
cells of a different order, clearly derived from the bone-marrow. 


they do not, to any large extent, wander back into the 
blood-vessels. Various stages of degeneration of these 
cells may be recognised in the later stages of peritoneal 
inflammation ; they tend to be extensively vacuolated, 
and thus the conclusion is that they in the main eventu- 
ally disintegrate locally. While this appears to be true 
in connexion with bacterial inflammations, where more 
inert substances are taken up they clearly are capable of 
wandering in considerable numbers back into the circu- 
lation. Thus Metchnikoff points out that, if washed 
nucleated red corpuscles of the frog be injected into the 
peritoneum of a warm-blooded animal, large mononuclear 
cells containing the easily recognisable remains of the 
corpuscles can be detected after several days in the 
mesenteric glands, the vessels of the liver, and the spleen. 
Cells of the same order, which have not become free, 
evidently, according to Kanvier, take an active part in 
the formation of peritoneal adhesions. Muscatello, 
Graser, and Borst (63) unite in the opinion that the 
epdothelium of serous surfaces can form connective 
tissue, though of late von Brunn (64) and Monckeberg 
have strongly contested this view, at least as regards 
serous endothelial cells. Their objections are not, to 
me, wholly convincing. More recently Baumgarten (65) 
has brought forward what appears to be definite evidence 
that vascular endothelium under mild grades of irritation 
gives rise to fibroblasts and connective tissue. 

Epithelial Leucocytes. Mononuclear cells of endo- 
thelial origin seen in inflammation of solid tissues, 
noticeably in tubercle formation, are, from their general 
appearance, often spoken of as epithelioid cells. But if 
we employ the term epithelium, as is usual in English- 
speaking countries, to denote membranes derived from 
the primitive epiblast or hypoblast, this usage is apt to 
cause confusion. There are, however, true epithelioid 


cells ; I refer to the large phagocytic cells seen in 
pneumonic conditions, clearly derived from the flattened 
epithelium lining the air -sacs. With Pratt (66) we 
must recognise, in the pneumonic exudate, two forms of 
large mononuclear cells : (1) large cells, round or oval, 
with a crescentic, vesicular nucleus placed towards the 
periphery and apt to contain cell-inclusions, and (2) a 
cell present in the early stages of acute lobular pneumonia, 
larger than the polymorphonuclears, but not so large as 
the form just mentioned, but, like that, having a large, 
vesicular, indented nucleus and a very slightly granular 
protoplasm. These also are phagocytic, and Pratt 
expresses himself as doubtful regarding their origin. 
From his description, and since they are also found in 
the capillaries and in the tissues, it would seem clear 
that these are the endothelial mononuclear leucocytes men- 
tioned above, whereas the former are of epithelial origin. 
Free Cells Leucocytes derived from other 
Tissues. In the experimental inflammations of the 
cornea it can be recognised that the tissue-cells of the 
boundary zone in which there is no actual destruction 
undergo active proliferation. . The corneal corpuscles, 
for example, as already indicated, become swollen, send 
out processes towards the point of injury, their nuclei 
stain more deeply, undergo division, and now two young 
cells of embryonic type are present where one had been 
before (vide Fig. 3, p. 30). These young cells are of the 
hyaline or mononuclear type, and, what is more, like the 
cells of endothelial origin, actively take up polymorpho- 
nuclear and other cells. Nor are the connective-tissue 
corpuscles alone in giving rise to cells of this order. 
Highly complex tissue -cells, like muscle fibres, when 
injured, are apt to exhibit direct divi7ion and multiplica- 
tion of their nuclei; these nuclei are apt to separate 
from the main mass, carrying with them a small body 


of surrounding cytoplasm. This process in connexion 
with the degenerating tadpole's tail has been well 
described by Metchnikoff, Barfurth, Griffiths, and others 
(67). When so separated, these cells are not to be 
distinguished from hyaline mononuclear leucocytes and, 
like them, actively digest cell-debris, and so forth. 

Clasmatocytes. In certain fine connective -tissue 
membranes, both in warm- and cold-blooded animals, 
there are to be seen considerable numbers of large, 
apparently wandering cells, elongated or much branched, 
with a rounded or elongated nucleus which Eanvier 
(68), on account of their tendency to cut off or leave 
behind them in their wanderings small remnants of 
protoplasm, has termed clasmatocytes. Marchand (69) 
has included them among his leucocytoid cells. Accord- 
ing to Eanvier, these originate as leucocytes i.e. are 
derived from the blood which have become more or less 
stationary in the tissues, and, on one hand, in their 
proliferation may give rise to cells of the exudate, and, 
on the other hand, may develop into true connective- 
tissue cells. While agreeing with Eanvier in this latter 
respect, Marchand denies that they originate from 
ordinary leucocytes, and regards them as identical with 
what Saxer (70) has termed "primitive wandering cells," 
which can give rise to leucocytes and nucleated red 
corpuscles. He thinks that by repeated division they 
develop lymphocytes, and that they gain entrance into 
the circulation as a form of mononuclear leucocyte, and, 
as already stated, that they may take part in the 
formation of connective tissue. Maximow concurs in 
holding that they may form fibroblasts and mononuclear 
" poly blasts." But now, as Lubarsch (71) points out, an 
identical form of cell is to be found in keratitis, with the 
same remarkable long processes, which are apt to be cut 
off and lie isolated in the tissues. If we then can 


determine the nature of these cells, we may arrive at 
some conclusion between these divergent views, but here, 
I must confess, there are difficulties. Lubarsch regards 
the characteristic corneal cells with spear-like processes 
as derived from mononuclear leucocytes which wander 
into the part. Grawitz (72), on the other hand, holds 
that these cells are modified corneal tissue -corpuscles. 
Councilman, and before him Senftleben, are very clear in 
their descriptions of the way in which, in certain grades 
of keratitis, the peripheral zone of corneal corpuscles 

FIG. 8. Clasmatocytes. After (1) RANVIER and (2) MAXIMOW. 

enlarge, send out processes into the necrosed area, and 
proliferate. This process we have just described. Their 
statements, thus, are in agreement with that of Grawitz. 

It will be seen, reading over the description of these 
various forms of cells which we regard as histogenous, as 
distinguished from hsematogenous, that certain features 
are common to all of them : the tendency to attain to 
large size ; the amoeboid character ; the rounded, or, at 
most, kidney -shaped nucleus ; the tendency to take up 
other cells ; the liability, on the one hand, in prolifera- 
tion, to give off wandering cells of the mononuclear type, 
on the other hand, to develop into fixed tissue - cells. 



Taking all these facts together, it will be seen that the 
clasmatocytes group themselves with the histogenous 
group of wandering cells, and as such I must classify 
them. In short, much, if not all, of the controversy 
regarding these different forms 'of cells becomes assuaged 
if we regard the cells of the mononuclear type with rela- 
tively abundant protoplasm the so-called hyaline cells 
as of endothelial and tissue origin, whether they occur 
within the blood or in the inflammatory focus. 

To those who have not followed the copious literature 
of the last few years these questions regarding the 
nature of the different forms of cells seen in inflamma- 
tion may appear to be small and unimportant, and the 
controversy regarding them to be very petty. But much 
depends upon the proper solution of this question, more 
particularly as regards the nature of the new -formed 
tissue which develops in areas of inflammation, and the 
difficulty in arriving at correct conclusions is very great. 
Any one who has studied inflammatory tissues will find 
out that, at first, classification of the different forms of 
cells seen seems to be almost hopeless. As a matter of 
fact, the youngest forms of newly developed cells are 
often so exactly alike that no means at present at our 
disposal serve to distinguish them; hence, for example, 
the controversy whether the large mononuclear cells give 
origin to lymphocytes. It is only by carefully following 
the successive stages seen in the proliferation of any one 
form of cell that we can arrive at any conclusion, and 
then by a combination of methods, for the employment 
of only one or two methods of staining is apt to mislead. 

Let me once more repeat that I am fully prepared to 
find that others do not agree with the conclusions here 
reached ; nay, more, I would add that I am personally 
fully prepared to modify my views regarding these cells in 
the light of further research. In the meantime, to sum 

CH. X 



up, let me say that the completed table of the leucocytes 
concerned in inflammation would appear to be the 
following l : 

a 2 


w SP 


POL YMORPHONUCLE A R (poly nuclear, 

finely granular oxyphil, neutro- 
phil, and ainphopliil cell). 
EOSINOPHIL (coarsely granular oxy- 
phil, macroxycyte). 

LYMPHOCYTE (? of two types). 

PLASMA-CELL (? histogenous). 

nuclear leucocyte, hyaline cell 
(in part), "epithelioid cell" (in 

EPITHELIAL wandering cell (large 
hyaline cell, in part). 

CONNECTIVE TISSUE wandering cell 
(including CLASMATOCYTE). 

MUSCLE AND OTHER tissue wander- 
ing cells. 

Originating in adult mammals 
from the bone-marrow, and 
migrating from the blood 
into the inflammatory area. 

Originating from lymphoid 
tissue and from vascular 
and other endothelia re- 
spectively ; present in in- 
flamed area either by migra- 
tion from blood or as result 
of local proliferation. 


^Originating locally as result 
of local tissue proliferation. 

In the previous edition the Mast- Cell was placed 
" very doubtfully " among the hsematogenous leucocytes. 
But whether seen in the blood or in the tissues in chronic 
inflammation this cell is degenerate in appearance, nor is 
there any evidence that it migrates from the blood into 
the tissues. Whether it is a degenerate plasma-cell or 
clasrnatocyte, and so histo-haematogenous or a degenerate 
histogenous cell is still undetermined. 

The genetic relationships I would suggest are indi- 
cated in the following table : 

1 It must be clearly borne in mind that all the forms encountered in the 
blood in various morbid states are not included in this table. A concise 
description of these other forms, with their presumed relationship, is to be 
found in the article, by R. C. Cabot (73), in Osier and McCrae's System of 





6 d 

O -^ 02 

O H 




H f\ <1 

CO hi 

| 2 ^ 

g & 
^ j^ P 





i-^ 02 O 

P2 <1 -i 

^ a c ' " " ~ 2 
^ c ^ 

! 1 i 




i i H 

K ^ "^ 








t-^ ^ ,^^ 



a ^ 

^x-3 1 1 

r, | 

i Js 


r/} ^ 

^-^ ^ ^ ^^ 

fa 02 



S flH 


h-3 -2 w 

PQ ^>r2 

o s 

li" ( ' E " 


03 * 

hp| ^j EH 

S s 


* 1 


S I ' c 

s >s ^ K 

fc ^ 

w g 


^ s 

"~~^^ O 0^^ 

\ W |o 



W <J 



^ ^^ *^ 

fa 1 "1 

fc fa 





M I H 

- H 



S g 


P ^ W'o 

/ o o ^^ 9 J 


E ' 


JJ / 

/ S W rt H r^ 

/ S rC Srt * ^ *H 

5 S 


1 / 

(^ ^ ^ 


H S / 



g s 


^ ^^ ^0 

^^3 ^ 

ft EH 

sj ^ 




g ^ ^| S g 

02 O 



^ O wC H H-5 

P t u 


bn ^> 


- w 


^ i i ^^ 



S >/ 

^3 ,-. - &3 



o \ 

:-, "^ W H 4 (_ 



\v O I-C ^"'f-H O O 



>, O *" P ^ Oi O 


>T ^J ^ -M C H ^ 



H ** jj ctf 3 2 

tf " 


S "jS- S 1 ^ 









11 B 

^ w 







AND here, though in so doing I to some extent 
anticipate and refer to matters to be discussed in some 
detail later, it is most convenient to sum up briefly the 
facts determined regarding the parts played by these 
various forms in inflammation. 

Polymorphonuclear Leucocyte. In acute inflammation, when the 
irritant is not too intense, this is the form most often attracted to 
the focus of irritation, migrating most rapidly and in the greatest 
numbers. It is the characteristic pus cell ; is actively phagocytic, 
particularly for bacteria ; secretes bodies of the nature of enzymes 
notably, as shown by Leber, Fr. Miiller, and Jochmann and others, 
of a proteolytic nature, and, either while active or in the process of 
dissolution, liberates antitoxic and antibacterial substances. It may 
(1) wander back into the lymphatics or blood-vessels ; (2) undergo 
dissolution and disintegration in situ ; or (3) be ingested by the 
proliferating tissue-cells and mononuclear leucocytes. It has nothing 
to do with tissue formation (74). 

Eosinophil Leucocyte. This is also actively attracted, and that 
at an early period, towards the inflammatory focus, migrating from 
the surrounding tissues and also from the vessels, but it is never 
the preponderating cell present. In peritoneal inflammations it 
is found in great numbers in the omental vessels, and in sub-acute 
and chronic inflammations of certain tissues, such as the skin, may 
also be relatively abundant. Very rarely is it seen to ingest 
bacteria, so that, to all intents and purposes, it is non-phagocytic. 
Kanthack and Hardy (44) and Hardy and Wesbrook (75) have 



ascribed a secretory activity to these cells, associated with the 
reduction in number and apparent discharge of the coarse granules, 
but later observers have failed to confirm their observations. The 
mode of action, therefore, of these cells remains unsettled. Opie 
has noted (76) that, during the course of various pyogenic in- 
fections, this form of cell disappears wholly or in part from the 
peripheral circulation, and after injection of a variety of bacteria 
into the peritoneal cavity of guinea-pigs he. noted a similar 
disappearance ; but now, examining the mesentery, during the 
height of the infection, the eosinophils were fcruud collected in the 
blood-vessels, actively migrating into the peritoneal cavity, where 
they were collected in considerable numbers upon the surface of the 
omentum. They clearly, therefore, have a part in the inflammatory 
process, but what that part is must still be considered undecided ; 
while there are indications that they proliferate in situ, all observers 
agree that they take no part in the formation of new tissue. 

Mast-cell. This is seen particularly in sub-acute inflammations 
(the coarse basophil granules, staining intensely with carbol-thionin 
and basic aniline dyes, and taking on a metachromatic nuance, 
must not t>e mistaken for clusters of small cocci). The nucleus 
stains feebly, according to the usual descriptions. Professor Miller 
of the University of Missouri has shown me his preparations, and 
permits me to refer to them. They definitely show that what has 
been taken for a feebly staining nucleus is a court or area where 
once the nucleus had been. The actual nucleus undergoes dis- 
integration and discharge. This cell is sluggishly amoeboid, and 
may detach portions of its cytoplasmic processes with their contained 
granules along the track taken by its pseudopodia. Nothing has 
been made out regarding the activities of this form of cell, save 
that Maximow shows that in the course of the inflammatory pro- 
cess it undergoes atrophy and disintegration (77, 78, 79). Professor 
Miller's studies confirm these views, and indicate further a complete 
want of relationship between the relative abundance of haemal and 
tissue mast-cells in inflammatory states. Schridde indicates that 
this type of cell within the tissues is a rare variety of plasma-cell. 
Cabot calls attention to their frequent association with eosinophilia. 

Lymphocyte. This is only slightly amoeboid, and does not 
migrate very actively from the vessels to the injured area in acute 
inflammations. It is, however, capable of this migration, and, in 
some chronic inflammations, is the preponderating cell ("small round 
cell "). It is found more particularly in masses around the vessels, 
and this as a result of (1) migration ; (2) proliferation of the 
pre-existing lymphoid tissue. It is not phagocytic for bacteria, but 
may ingest particles of inert matter. According to Bergell (80) 


lymphocytes obtained from a tuberculous focus possess a definite 
fat splitting enzyme or enzymes, which is still active at a heat 
sufficient to destroy the cells. The small lymphocyte, with a 
scarcely noticeable zone of cytoplasm, may give origin to the 
middle-sized lymphocyte, with more cytoplasm and deeper staining 
round or oval nucleus, and to the plasma-cell. There is no 
evidence that these plasma-cells are actively wandering, or that 
they return to the blood ; they occur especially in connexion with 
the formative and chronic inflammations, but whether they actively 
participate in the formation of new fibrous tissue is still a matter 
of debate. 

Histogenous Wandering Cells. These appear in the inflammatory 
area in general at a definitely later period than do the polymorpho- 
nuclears and the eosinophils, and in certain forms of inflammation 
(e.g. of serous membranes) may become the preponderating type, 
exhibiting well-marked phagocytic properties, being not so active 
as regards most species of bacteria as are the polymorphonuclear 
leucocytes, but peculiarly active as regards other cells and cell- 
debris. They are sluggishly motile, and though those of vascular 
endothelial origin may migrate from the vessels into the tissues, 
they are rarely recognised passing back into the vascular system 
from the inflammatory area ; they are apt to become vacuolated. 
They multiply both by direct division and mitosis. They may 
(1) undergo local necrosis and disintegration ; (2) may form giant- 
cells, either by plasmodial fusion or by direct division of the 
nuclei ; or (3) may be active factors in the formation of new tissue. 
Certain forms (e.g. those derived from corneal corpuscles, and the 
clasmatocytes) give off frequently long pseudopodial processes. 

A word should here be said concerning sundry cells 
of later development, appearing as a result of inflammation 
giant-cells, Ranvier's cells, and Gluge's corpuscles. 
Of these the last are evidently leucocytes of the hyaline . 
type which have taken up the fatty products of tissue- 
degeneration ; the second colossal cells breaking down 
with great ease are of endothelial origin. Giant-cells 
would seem to be of more than one variety ; some appear 
to be due to aberrant cell-growth, wherein the nuclei 
undergo division without the protoplasm of the cell- 
body following suit. The characteristic giant-cells of 
tuberculosis and chronic inflammation may now be said 


with fair certainty to be plasmodia, in all respects 
comparable to the masses of fused cells seen to form in 
the lower animals around foreign bodies, and by Kanthack 
and Hardy around masses of bacteria in the lymph 
of frogs outside the body. The observations of Borrel 
(81), Duenschmann (82), and Faber strongly support 
this opinion. Hektoen's careful studies tend also thereto 
(83). But while this is the case, it has to be admitted 
that in some giant cells of endothelial origin developing 
in the course of sub-acute inflammation direct division is 
to be seen affecting the nuclei, as demonstrated by Duval 
and White in their studies upon experimental glanders 
(84). There is indeed no reason to be adduced why 
both cell fusion and amitosis should not manifest them- 
selves in the course of development of an individual 
giant cell. 



IN the case of most pathogenic micro-organisms, after 
inoculation into the organism, a very considerable pro- 
portion are to be discovered, sooner or later, within 
wandering cells which have collected in the region of 
inoculation. I have already mentioned more than one 
case of this nature in discussing the comparative pathology 
of inflammation. Evidently under certain conditions one 
of the functions of certain of the leucocytes is to attack 
and incorporate bacteria. The leucocytes having these 
properties in mammals are more especially the polymorpho- 
nuclear and the non-granular or faintly granular mono- 
nuclear cells of endothelial and tissue origin ; the former, 
as already indicated, migrating from the blood, coming 
more rapidly into action and being most active from a 
very early period in tissues abundantly vascularised ; the 
latter, in the main of local development, becoming active 
at a later period, and then acting as phagocytes towards 
other cells as well as bacteria. It is, for instance, the 
polymorphonuclear cell which is found in overwhelming 
numbers in an extending subcutaneous abscess, and these 
are seen to contain great numbers of the micrococci. 

The conditions lea'ding' to this phagocytosis have been 
worked out in remarkable detail by Metchnikoff. He 
has amply demonstrated that the microbes can be taken 


up in a living condition. Thus, if the Vibrio Metchnikovi 
(a form closely allied to the cholera spirillum) be 
inoculated into the anterior chamber of the eye of an 
immunised animal, within a very few hours phagocytes 
are seen filled with the small, slightly curved vibrios. 
If now one such cell be isolated, placed in a drop of 
broth upon a cover-slip, made into a hanging-drop 
preparation and examined under the microscope, it is 

FIG. 'J. Resolution of acute infectious disease (relapsing fever) spleen pulp of monkey 
(a Macacus), showing (a) microphage, multiuuclear, with incepted spirocha-tes ; 
(6) solitary, and (c) forming dense tangle, (d rf) nuclei of splenic tissue (Zeiss, ^\, 
ocular 4 ; x 1515 diam.). METCHNIKOFF. 

seen that the broth causes the death of the leucocyte ; 
while with time, and favourable temperature, the microbes 
proliferate rapidly and completely fill the corpuscle until 
it disintegrates, whereupon they proceed to multiply in 
the surrounding fluid. This seizing and incorporation of 
microbes does not then necessarily lead to their death. 
In certain cases of acute disease there may be abundant 
phagocytosis, and the disease progress nevertheless ; the 
phagocytes being destroyed by the products of the 
incorporated organisms. This is the case in mouse 
septicaemia, in swine erysipelas, and (as has been shown 




by Gabritchewsky (85) iu diphtheria. As Eoux remarks : 
" Us out fait de leur mieux en englobant les microbes, 

FIG. 10. Two giant-cells, seen under high magnification ( x 1515 diain.), from a rodent, 
the spermophile, inoculated with tuberculosis, to show stages in thte destruction 
of the bacilli, a, Unaltered bacillus ; b, bacillus staining badly, and with greatly 
thickened capsule; c, bacillus granular and breaking up; d e, "shadows." 

mais ceux-ci se sont adaptes au milieu inte'rieur des 
cellules, et ils ont triomphes" (86). 

In other less acute diseases, such as gonorrhoea, and 
in chronic maladies, such as in tuberculosis, leprosy, and 


glanders, the bacilli may in certain stages be found 
within the cells and rarely free in the lymph-spaces, they 
appear to be almost parasitic, after the manner of the 
microsphsera previously referred to as infesting the 
amoeba. In these cases it would seem as though the 
toxic properties of the microbes and the antagonising 
powers of the cells were nearly balanced. In tuberculosis, 
for instance, it is not unusual to find in the giant-cells 
some bacilli which evidently are undergoing degenerative 
changes, staining poorly and irregularly, or but faintly 
traceable as unstained, translucent shadows, while else- 
where they are apparently proliferating despite their 
intracellular position. 1 

And this equality, or almost equality of the resisting 
powers of cells and microbes may explain the chronic 
nature of the diseases above mentioned. Nevertheless, 
in general, it may be stated that there is some relationship 
to be recognised between the amount of phagocytosis and 
the virulence of the microbe ; with a given bacterial 
species the more virulent the microbes, the less the 
proportion of them taken up by the cells, the longer 
the time before the phagocytes come into action, As 
in the unicellular organism, so in the wandering cells of 
Metazoa the mode of destruction of the included microbes 
can, under suitable conditions, be seen to be digestive. 
Several observers have seen the anthrax bacillus, in frogs 
and other animals, wholly or in part surrounded by a 
vacuole developed within the leucocyte ; and, as an 
evident result, the portion so surrounded has been seen 

1 It is, however, unsafe to declare in all cases that because a micro- 
organism continues to stain well therefore it was living at the moment 
the preparation was taken and fixed by heat. Thus in pneumonia after 
the crisis a fair number of diplococci may be found within the leucocytes 
of the expectorated contents of the alveoli, and these may stain perfectly 
well ; yet it may be impossible to gain .a single growth of the diplococcus 
from the same material. 


to become swollen and fainter when stained, until it has 
undergone a veritable digestion and dissolution. By the 
use of appropriate reagents the vacuoles in general are 
seen to be faintly acid in contrast to the more alkaline 
surrounding cytoplasm. 

As with the lower organisms, so with the wandering 
cells of the higher, there is an evident attraction, or 
chemiotaxis, whereby these cells pass towards the microbes 
and their products. The chemiotactic properties of the 
wandering cells have been especially studied by 

Fio. 11. Phagocytes, macrophage and microphage, to show stages of digestion and 
destruction of bacilli, from spleen and eye respectively of white rat with anthrax. 
In 1, part of the bacillus is unaffected, but a vacuole has formed around the other 
part, which further has now lost the power of tak ng the stain. In 2, various 
stages are seen, the bacilli passing through the granular, badly staining, to the 
vacuolated, unstained stage, until finally but faint " shadows " are observable 
(Zeiss, TV oc. 3). METCHNIKOFF. 

Pekelharing (87), Leber (34), Massart and Bordet (88), 
and Gabritchewsky (85). Of the results obtained by 
these observers the most important are that leucocytes 
are variously attracted towards various substances. Thus 
Leber found that the introduction into the system of 
finely-powdered copper and various compounds of mercury 
caused an abundant collection of the wandering cells 
around the particles, while powdered gold, silver, and iron 
exerted no such attraction. Gabritchewsky and Buchner 
(89) showed that the products of bacterial growth in 
general possessed chemiotactic properties yet more 
powerful than simple chemical compounds. While the 


degree of positive chemiotaxis is found to vary within 
wide limits, the examples brought forward of negative 
chemiotaxis exerted by bacterial products have so far 
been very few so few as to support the contention of 
the late Prof. Kanthack (90), that it is very doubtful 
whether any microbes by their products actually repel 
the leucocytes, though they are capable of causing the 
rapid destruction of the attracted leucocytes, and so of 
rendering the area around the microbes relatively free 
from wandering cells. 

This chemiotaxis would also seem in general to be in 
the inverse ratio of the virulence of the microbes. I say 
in general, for with chemiotaxis as with phagocytosis 
there appear to be exceptions to any uniform law ; and 
cases can be brought forward of diphtheria, for example 
in which the leucocytes, instead of being repelled, are 
attracted in great numbers to the region of inoculation 
of a most virulent bacillus. Indeed it may be doubted 
whether it is the bacterial toxins as such that, diffused 
into the surrounding medium, constitute the force 
attracting the leucocytes towards the bacteria, We now 
have come to distinguish two orders of pathogenic 
microbes, those which during active growth give off 
diffusible toxins, and those elaborating toxins which 
during active existence are not diffused, but remain 
intracellular (endotoxins) (91), so that the medium in 
which they grow has little or no toxic effects, although 
if the bacteria be collected and frozen at the lowest of 
temperatures and then crushed, their freed body-juices 
are found to be extraordinarily toxic. [MacFadyen and 
Rowland (92).] The diphtheria bacillus belongs to the 
first order, forms like the B. typhosus and the cholera 
spirillum to the second. It is only with the death of 
the bacteria of this latter order that the toxins are 
liberated to any extent. Now as regards chemiotaxis 


and phagocytosis we can draw no line of demarcation 
between the two orders. Or, more accurately, up to the 
present there has been no adequate study of the inter- 
action between leucocytes and bacteria of the exotoxic 
and endotoxic groups respectively. It is suggestive that 
the " type " producer of ejcotoxins, the (jjpht.hp.ria, barillas, 
induces active chemidtaxis and phagocytosis, followed 
by disintegration of the leucocytes, while the tubercle 
bacillus which characteristically gives rise to ejdntny.ins 
is weakly chemio tactic, gives rise to a leisurely phago- 
cytosis, and persists for long within the leucocytes with- 
out setting up pronounced reaction. The law would 
seem to be not that the extent of positive chemiotaxis is 
in inverse ratio to the virulence of the different species 
of bacteria as towards the human being or other larger 
animal, but that with a given species of microbe the greater 
the virulence of the strain the less the positive chemiotaxis. 
It has still to be determined whether it is the toxin as 
such, or some other diffusible substance produced coincid- 
ently, that attracts the leucocytes towards the bacteria. 

A very good study of the action of bacteria of different 
degrees of virulence can be made by repeating an experi- 
ment of MetchnikofF. The rabbit is an animal susceptible 
to the growth within its tissues of the bacillus of anthrax. 
As is well known, there are various means whereby the 
virulence of this microbe can be diminished ; so that if 
cultures of the " attenuated " bacillus be inoculated into 
susceptible animals, these, instead of causing a fatal 
disease, induce but a transient local inflammatory 
disturbance, accompanied by fever, and followed by 
complete recovery. If now a small quantity of a 
virulent culture of the bacillus be inoculated into the one 
ear of a rabbit, and an equal quantity of an attenuated 
culture into the other, the results are very instructive. 
Within twenty-four hours it can be noticed that an acute 


inflammation has been induced in both ears ; in both the 
vessels round the seat of inoculation are greatly con- 
gested, but whereas at the seat of inoculation of the 
virulent organism there is a serous inflammation so 
intense that the skin is raised and separated from the 
subjacent tissues by a clear, transparent, reddish fluid 
which also infiltrates the deeper tissues, in the other ear 
there is not nearly the same amount of swelling and 
serous exudation ; the region of inoculation is more 
opaque and solid. Upon more minute examination the 
serous fluid in the first ear is found to contain relatively 
very few leucocytes ; the firmer mass in the second is 
composed of a huge aggregation of leucocytes. 

Before proceeding farther it will be well to sum up 
the phagocytic hypothesis of inflammation as upheld by 
Metchnikoff and those who see in this phenomenon the 
all-important factor in inflammation and the repair of 
injury (as also in the production of immunity), in order 
that, having put clearly forward the tenets of those 
upholding the hypothesis, I may the more readily state 
wherein lies the strength and wherein the weakness of 
the doctrine. 

This hypothesis may be summed up in the following 
theses : 

1. That certain of the leucocytes present in the blood 
and lymph, notably the polymorphormclears (microphages) 
and the large mononuclear hyaline cells (macrophages), 
are capable under certain conditions of taking up bacteria 
which have gained entry into the system. 

2. That in addition to these, the splenic corpuscles, 
the cells forming the endothelium of capillaries, and 
other fixed cells of mesoblastic origin, possess the same 
property, although they exert it to a less extent. 

3. That these phagocytes seize upon and destroy 
living and active microbes under certain conditions. 


4. That the more virulent the microbe the less the 
tendency for the leucocytes above mentioned, and for the 
other fixed cells, to take up the bacteria. The less 
virulent the microbe the more extensive the phagocytosis. 

5. That in addition to this power on the part of 
certain cells (the phagocytes) to take up and destroy 
certain bacteria, another factor has to be called in to 
explain why the wandering cells of the body migrate 
towards the focus or foci where the micro-organisms have 
gained an entry into the body. This factor is the 
" chemiotaxis " exerted by the products of bacterial 
growth, and by some other substances, such, for example, 
as the products of death of tissue and of wandering cells, 
and experimentally also certain chemical irritants as, for 
example, turpentine and mercury. In the case of the 
virulent microbes the leucocytes are not attracted to the 
focus of infection. There is a " negative " chemiotaxis, 
and thus, in the absence of phagocytosis, the proliferation 
of the microbes takes place without hindrance ; whereas 
the less virulent microbes and their products attract the 
leucocytes, they exert a positive chemiotaxis, so that 
there is a migration of leucocytes through the capillary 
and venous walls to the focus of infection, and the 
leucocytes taking up the microbes tend to arrest the 
infective process. 

6. That the leucocytes may become accustomed and 
eventually attracted to substances from which at first they 
were repelled, and thus a negative may be transformed 
into a positive chemiotaxis. 

7. That the cells, having once acquired positive 
chemiotactic properties in relation, to the products of any 
specific microbe, retain and transmit these properties 
through a series of cell-generations, the length of which 
varies according to the microbe, the extent of the primary 
reaction, and the idiosyncrasies of the individual. 



8. That, consequently, the cure of zymotic or mycotic 
disease, whether localised or general, and immunity also, 
are mainly brought about by the activity of special 
cells (the phagocytes), and are primarily dependent upon 
the attraction existing between these cells and the 
products of bacterial metabolism. 

9. The process of inflammation is essentially the 
endeavour on the part of the organism to promote the 
migration of leucocytes, and to aid the inclusion and 
destruction of the irritant. " The essential and primordial 
element of a typical inflammation is a reaction of phagocytes 
against the irritant (agent nuisible)." Or, more fully, 
" inflammation is to be regarded, on the whole, as a 
phagocytic reaction of the organism against irritants, a 
reaction which at times is accomplished by the wandering 
cells alone, at times with the aid of the vascular (fixed) 
phagocytes, or with that of the nervous system." 

These are the main headings, if we may so term them, 
of Metchnikoffs hypothesis (for, to my knowledge, he 
has never formulated it clause by clause). Yet other 
data must be added to make it complete, data drawn 
from Metchnikoffs more recent studies (93), which we 
shall discuss in fuller detail later : 

10. The destruction of microbic irritants by the cells 
is a process of digestion : the cells elaborate ferments 
(cytases) which differ in their properties ; that- elaborated 
by the microphages (polymorphonuclears) is termed by 
him microcytase ; that by the macrophages (hyaline 
mononuclears), macrpcytase. These ferments are the 
main factor in the destruction of bacteria. 

11. These ferments are formed and act within the 
living, active cell; they are endx)-enzymes. But with 
the death and disintegration of the phagocytes (phagolysis) 
they become liberated, and, through their liberation, the 
body-fluids become bactericidal. Since cells which are 


capable of acting as phagocytes liberate these cytases, the 
process is to be included p^g^ytir 

12. It must be granted that when extracellular the 
cytases alone are incapable of causing bacteriolysis ; 
bacterial destruction is brought about by the interaction 
of the cytase and another product of cellular activity- 
the " fixateur " (immune body or intermediary body). 
While the entases are not specific, acting indifferently on 
various microbes^ the fixateurs are specific, a different 
fixateur being developed in the process of immunisation 
against each specific microbe. These fixateurs are 
actively secreted from, the living cell, and are present in 
the humours and exudates of the immunised animal. 
Metchnikoff has relatively little to say regarding these, 
but admits their existence, and further believes that they 
are developed from the same order of cells as are the 
cytases ; hence he holds that they also are to be con- 
sidered as results of phagocytic activity. 

I have laid down these later theses somewhat in 
advance in order to present the doctrine as a whole, and 
that I may the more easily take up the objections that 
have been raised, some of which have been well met by 
Metchnikoff and his fellow-workers, though others still, 
in my opinion, have not been fully controverted. 

In the terms of this hypothesis, then, phagocytosis is 
the all-important factor in the inflammatory process, the 
vascular, exudative, nervous, and other phenomena being 
auxiliary means whereby the phagocytic properties of the 
wandering and fixed mesodermal cells may be brought 
more fully into action : the determination of leucocytes 
that I have described is almost entirely to be attributed 
to an endeavour on the part of these cells to take up 
and destroy the irritant. To what extent is this doctrine 
to be accepted ? 

It must in the first place be laid down without the 


slightest hesitation that phagocytosis is a factor in inflam- 
matory processes ; no antagonist of the doctrine nowadays 
is prepared to deny this. Nay, more, each succeeding 
year we recognise more fully that it is an important 
factor. There is not a single bacterial disease that affects 
man in which, to my knowledge, it has not been shown 
that the causative microbes are liable to be taken up by 
the cells, if not in man himself, at least in other warm- 
blooded animals. And, by repeating a very simple 
experiment devised by Leishinan (94), this phagocytic 
activity of the human white blood corpuscles can very 
easily be demonstrated. It is only necessary to clean 
the finger, take a drop of blood, mix it with an equal 
quantity of a suspension of any pathogenic organism, 
drop the mixture on a slide, place over it a cover-slip, 
place the preparation in an incubator at blood-heat for a 
quarter of an hour, then rapidly removing the cover-slip, 
fix the film and stain it by any good bacillary stain. 
Ordinary polymorphonuclear leucocytes present in the 
film can then be seen to have taken up, in this short 
time arid under these not wholly favourable conditions, 
abundant bacteria. There may be twenty or more present 
in a single cell. 

The Resistance Period. It is when a hypothesis 
has been applied with the object of attaining practical 
ends, and is found to afford the results expected, that its 
Value is proved, and we would point out that one of the 
most important recent advances in abdominal surgery 
has been developed from a basis of this hypothesis of 
phagocytosis. As already indicated, the injection of 
various microbic and chemical irritants into the abdominal 
cavity is followed by a pronounced leucocytosis, follow- 
ing upon which, when particulate bodies are present, 
there is seen to be abundant ingestion by the cells. 
Taking these facts into consideration, Issaeff (95) 


determined that, if a peritoneal leucocytosis be induced 
by intra-abdominal injection of saline solutions, serums, 
etc., and then (twenty-four hours later) various species of 
spirilla be inoculated also into the abdominal cavity, 
the resisting power of the animal to these pathogenetic 
microbes is very greatly augmented. Dr. Durham (46) 
confirmed and showed that this was true, not only of the 
spirilla, but of the B. typhosus and other microbes. He 
found that the leucocytosis and the increased resistance 
lasted for some four or five days. Von Mikulicz and 
other surgeons have made use of this " IssaefF resistance 
period." A day prior to performing any serious opera- 
tion, they inject nucleins, saline solutions, or serum, 
harmless in themselves, but capable of setting up an 
abundant leucocytosis. They have found that, thereby, 
they materially diminish the dangers of subsequent infec- 
tion ; the leucocytes, already in the peritoneal cavity in 
great numbers, take up and destroy bacteria which have 
chanced to gain entrance, before they have time to 

One has but to read Metchnikoff's fascinating work 
on L' Immunity dans Us maladies infectieuses (93), to be 
impressed with the extraordinary diversity and number 
of the researches in connexion with various pathogenic 
organisms whereby the phenomena of phagocytosis have 
been established and confirmed. Yet, already in the 
course of discussion, there are one or two respects in 
which we are not wholly in agreement with the general 
terms in which the hypothesis is set forth. (1) We 
doubt, for instance, the actual negative chemiotaxis the 
actual repulsion from bacterial irritants. At most, with 
Prof. Kanthack, we are prepared to see an absence of 
positive attraction. This, however, is a minor point and 
does not seriously affect the main question. (2) We 
have difficulty also in recognising that , /the extent of 


chemiotaxis and of phagocytosis is in absolute inverse 
ratio to the virulence of the microbes. This obtains, we 
admit, in connexion with any one species ; the more 
virulent the members of that species, the less the phago- 
cytosis. But, comparing different species, it does not 
fully hold and, if virulence be dependent on toxic 
properties, then, when many of the most virulent bacteria 
do not diffuse out their toxins, we have to suppose that 
it is not the toxins but some other and diffusible sub- 
stance, elaborated pari passu with the toxins, which is 
the cause of the attraction. Here again, while disagree- 
ing, we do not think that this renders the hypothesis 
in its essence untenable. Granting this, phagocytosis 
might still be the all- important factor. (3) What 
deserves emphasis is that, by dwelling upon one function 
of the cells that of phagocytosis the hypothesis is 
incomplete. It is not permissible, in the first place, to 
expand the term phagocyte so as to include cells which 
do not eat, and if we agree that bacteria are destroyed 
and their toxins neutralised, not alone by intracellular 
but also by extracellular activities, substances being 
discharged into the lymph and blood plasma which 
arrest bacterial growth, neutralise the toxins, or lead to 
dissolution of the bacteria, then our hypothesis passes 
from being one of phagocytosis pure and simple into a 
" cellulo- humoral" one. Now we have to admit this, 
and, for a fuller comprehension of inflammation, have to 
appreciate duly all the factors that go to make up the 
very complicated picture. While we wholly agree with 
Metchnikoff that the later development of his hypothesis 
is a natural expansion, and while we understand the 
natural inclination to speak of it still as the phagocytic 
hypothesis, we find that the position is not clearly under- 
stood by every one, and the objection not infrequently 
raised that much that is now included in the hypothesis 


is not phagocytosis at all, is difficult to controvert in a 
few words. It may further be urged with some justice 
that the extracellular and excretory activities of the 
body cells are minimised, and certain other factors in 
the process largely neglected. 

This notwithstanding, Metchnikoff's views have been 
purposely given this prominence. It is through him 
and through them, and through the active controversy 
and critical research initiated by these remarkable 
studies, that we have reached our present standpoint 
regarding, not merely inflammation, but the wider sub- 
jects of infection and immunity. And what is more, 
they are in our opinion in the main incontrovertible. 

Now, in order to arrive at what we personally regard 
as a wider and sounder view, it may be well to detail 
briefly and to criticise the more important of these 
researches directed against Metchnikoff's hypothesis. 



THE first strong attack and that when Metchnikoff 
upheld an uncompromising phagocytosis and nothing 
else was originated by Nuttall (96). So long ago as 
1874, Traube and Gschleiden (97) had called attention 
to the antiseptic properties of the blood, and in his 
presidential address at the London meeting of the Inter- 
national Medical Congress in 1881 Lord Lister had 
noted that a drop of putrid blood, teeming with microbes, 
if diluted with a little water, when added to pure blood 
that had been received into a sterile vessel, might leave 
that blood unchanged for days, the blood remaining 
sweet, and having, evidently, the properties of arresting 
the growth of the organisms of putrefaction. In 1887 
von Fodor (98) of Buda-Pesth more definitely called 
attention to these bactericidal properties, but it was 
Nuttall, then a Johns Hopkins student working under 
Flu'gge, whose exact studies called particular attention 
to the properties of the humours of the organism. In 
an attempt to repeat Metchnikoff 's researches upon the 
destruction of the anthrax bacillus, this observer noticed 
that if he placed a fine canula containing a fresh culture 
of attenuated anthrax bacilli in the tissue of a rabbit's 
ear, there resulted in sixteen hours a rich cellular 
exudation ; but phagocytosis appeared not to reach its 



maximum for twenty-two hours, and even then half of 
the bacilli lay free and not taken up by the cells ; and he 
found, further, that the free bacilli showed involution 
and degeneration to the same extent as did the ingested. 
This led him to study the effect of blood-serum, defibrin- 
ated blood, and lymph upon the bacilli, and he discovered 
that these fluids had a remarkably rapid action, destroy- 
ing great numbers within a very few hours. Nuttall's 
very full research appeared to show conclusively that 
the bacteria -destroying power resided largely in the 
serum, and that in inflammation the exuded fluid rather 
than the leucocytes brought about the destruction of the 
microbes. Further, Nuttall found that the bactericidal 
substance was destroyed by being subjected for one hour 
to a temperature of 55 C., so that blood-serum outside 
the body, thus heated, lost all its powers of arresting 
bacterial growth. A definite, if variable, quantity of 
bactericidal matter was present in a given quantity of 
the heated serum, capable of destroying a certain number 
of bacteria, so that, if more than that number were 
added, the excess survived and, in the course of a few 
hours, showed active multiplication. These observations 
were confirmed and extended by Nissen (99), Behring 
(100), and Buchner (102), and a most valuable series of 
contributions have been made by Hankin (101), Buchner, 
Vaughan(103),Tizzoni and Cattani (104),Behring,Ehrlich 
and his pupils (105), and, others, upon the nature and 
properties of the substances to be derived from the 
blood -serum of animals either naturally immune to 
certain diseases, or rendered immune by one or other 
procedure. What is more, it has been recognised that 
two orders of substances are recognisable ; one capable 
of destroying pathogenic microbes, the other not de- 
stroying them, but rendering their products inert. 
We cannot here give these important observations in 


detail (93), as they bear more particularly upon the 
subject of immunity. It was shown from several sides 
that, in the blood-serum and removed body-fluids as also 
under certain conditions within the tissues, there occurs 
active destruction of bacteria without the bacteria being 
ingested by the cells. The experiments were so clear, so 
easy to repeat and confirm, that we can well remember 
the time when, to believe in phagocytosis, even as an 
auxiliary process in the arrest and cure of disease, was 
to be regarded as to be lacking somewhat in common 
sense. The destruction of bacteria in the economy was 
held to be almost entirely in some cases entirely 
brought about by the action of the humours of the body. 
But this humoral hypothesis pure and simple was 
soon seen to be inadequate. Its upholders could not 
say how and whence the bactericidal substance was 
derived, and the more sera were studied whether 
gained from healthy animals, from individual cases of 
the disease, from animals naturally immune to certain 
diseases, or those rendered immune by inoculation of this 
or that, microbe the more impossible it became to 
recognise any sure relationship between the bactericidal 
power of the serum and the extent of resistance displayed 
by the individual affording that serum. It was noted 
also that the different humours of the body differed in 
bactericidal properties in a way that could not be 
adequately explained in the terms of this hypothesis. 



IF the serum and if the blood-plasma contain bactericidal 
substances, these must in all likelihood be developed by 
certain cells, and thus at bottom the humoral conception 
must be cellular ; and the very fact of the great increase 
in the bactericidal properties of the blood immediately 
on its withdrawal from the body, must suggest that in 
the changes which occur in the extravascular blood there 
is a liberation and solution of bactericidal substances. 
Now the first and foremost of these changes is the 
breaking down of the leucocytes as the blood begins to 
clot. It may, therefore, be that this breaking down of 
the leucocytes, with liberation of their contents, is capable 
of explaining the increased bactericidal action of defibrin- 
ated blood and blood-serum. This, let me repeat, is not 
phagocytosis, at most it is a function of cells which can 
act as phagocytes. If correct, it seems that that group 
of wandering cells which more particularly exhibit 
phagocytic properties is also able to bring about the 
destruction of bacteria and the neutralisation of toxins 
by other means. As a matter of fact evidence has 
steadily accumulated showing that more particularly the 
wandering cells of the body and also, probably, certain 
orders of fixed cells possess these other means. Many 
years ago Eibbert (107), in his studies upon the fate of 



spores of sundry species of mould (aspergillus and mucor) 
inoculated into the anterior chamber of the rabbit's eye, 
had found that two stages of reaction were to be made 
out ; at first the spores and developing hyphal filaments 
became surrounded by dense clusters of leucocytes which 
remained in apposition to but did not ingest the micro- 
organisms. They appeared to bring about a weakening 
and lowering of vitality on the part of the spores and 
filaments, so that, after a time, other cells wandering 
into the part could manifest their phagocytic action and 
take them up. Eibbert, it is true, attributed the lower- 
ing of vitality to the walling-in (" Wallbildung ") of the 
leucocytes and consequent lack of nutrition. The fact 
remains that he demonstrated a preparatory extracellular 
action upon the micro-organisms by the leucocytes. 

That the leucocytes contain bactericidal substances 
was first demonstrated by Hankin (101), who obtained 
from the lymphatic glands and spleens of animals 
immune to anthrax (dogs and cats), a proteid of the 
nature of a globulin, identical with Halliburton's cell- 
globulin ft, and having a bacteria-killing power similar 
to that possessed by blood-serum. In later observations 
upon the rat he showed that there was a relationship 
between the amount and activity of these " defensive 
proteids " and the power of resistance of the animal to 
the disease. Thus Hankin showed that in animals 
possessing the power of destroying bacilli, the organs 
containing the largest collections of leucocytes yielded 
notable quantities of a bacteria -destroying substance. 
These defensive proteids are now more commonly known 
as alexins, this name having been given to them by 
Buchner, who, beginning as a strong supporter of the 
humoral hypothesis, ended in assuming a position which 
more nearly approached that of Metchnikoff. It was 
shown by Denys and Havet (108), of Lou vain, that 


exudates rich iu leucocytes have a more intense bacteri- 
cidal power than has the blood -serum of the same 
animals, and that the blood and exudations of the dog 
freed from leucocytes, either by infiltration or by centri- 
fugal action, lose their bactericidal power, regaining it 
when the leucocytes are re-introduced. Buchner (109) 
fully confirmed these observations. He showed that, if 
sterilised emulsions of the gluten of wheat be injected 
into the pleural cavity of a dog or rabbit, its presence 
leads to the pouring out of an aseptic exudation peculiarly 
rich in leucocytes, and this exudation is much more 
bactericidal than is the blood or the serum of the animal. 
He noted further that, like the alexins of blood-serum, 
this bactericidal substance of inflammatory exudates is 
destroyed by heating to 55 C. Several other observers 
-Bail (110), Schattenfroh (111), Van der Velde (112), 
Jacob (113), Lowit (114), and Bordet (115) have 
obtained similar results, employing various methods. Thus 
it is now generally accepted that the alexins or bacteri- 
cidal substances of the serum, as of exudates, are derived 
very largely from the leucocytes, and that when, therefore, 
bacteria undergo destruction in the humours of the body 
without the intervention of cells, that destruction is 
indirectly due to the leucocytes. 

Yet another line of work independently points in the 
same direction. Vaughan (103) separated from blood - 
serum nuclein, a body which so far had been found 
exclusively in connexion with the nuclei of cells. His 
work was independently confirmed and extended by 
Kossel, working with nucleic acid, the main constituent 
of nuclein. This nuclein is either itself bactericidal or 
has a bactericidal substance in association with it. The 
presence of such a body in the serum is best explained 
by the disintegration of nucleated cells, i.e. of leucocytes. 
Clearly then leucocytes of certain orders contain 


bactericidal substances ; can we recognise not merely a 
liberation of anti-bacterial substances by disintegrating 
leucocytes, but a functional active secretion of the same 
such as has been held by Kanthack and Hardy, and 
by Buchner and his school ? Metchnikoff largely denies 
this ; the microcytase and the macrocytase, the digestive 
ferments of the two main orders of leucocytes, are, 
according to him, endo-enzymes, acting with the cell-body 
and only discharged with the dissolution of the cell. At 
a very early period it was shown by Nuttall that the 
blood-serum removed from the body acts far more rapidly 
and energetically than does the blood-plasma and lymph 
within the body. The disparity of action between the 
two is remarkable. Thus Lubarsch (116) has shown 
that in order to kill a rabbit by intravenous inoculation 
of anthrax bacilli into the circulating blood, at least 
16,000 virulent bacilli must be introduced. In other 
words, the whole circulating blood is able to destroy only 
some 16,000 bacilli at a time. On the other hand, a 
single cubic centimetre of the blood-serum derived from 
such a rabbit can, outside the body, kill an even greater 
number of bacilli in a few minutes. The normal blood- 
plasma, therefore, does not appear to be actively bacteri- 
cidal, and this is confirmed by certain striking observations 
by Gengou (117). If rabbits' blood be obtained with 
every precaution against admixture with the body-fluids, 
be received into paraffined test-tubes, and be centrifugalised 
immediately so as to remove all the cells, a serum or 
plasma is obtained which will remain liquid for several 
days. So obtained, there has been a minimal breaking 
down of the leucocytes. This plasma has practically no 
bactericidal properties, but, on the contrary, serum from 
the same animal, obtained from blood allowed to clot, is 
actively bactericidal. In this process of clotting there is, 
as we know, an extensive destruction of leucocytes. 


But Gengou's experiments merely demonstrate that in 
the normal animal not subjected to bacterial invasion the 
leucocytes circulating in the blood do not actively dis- 
charge bacteriolytic substances into the plasma : they do 
not controvert the possibility that, under bacterial stimulus, 
the cells of the body actively secrete these bacterial 
substances. It is, we confess, difficult either to prove or 
disprove that this is the case. If bacteria or their pro- 
ducts be injected into the body-fluids, and these are found 
to become increasingly bacteriolytic, it may well be urged 
that the result is wholly due to resulting disintegration 
of the leucocytes. It must be noted further that other 
observers repeating Gengou's work (though often not his 
identical methods) have not gained such clear results 

Pfeiffer's Phenomenon. A few years ago proof posi- 
tive of such exudation, not by leucocytes, it is true, but 
by endothelial cells, seemed to have been discovered by 
Pfeiffer of Berlin (119). If a few drops of a pure culture 
of the cholera spirillum be injected into the peritoneal 
cavity of a guinea-pig, vaccinated against cholera, and, a 
few minutes later some of the peritoneal fluid be ex- 
amined, it is seen that very few leucocytes are present, 
and, this notwithstanding, the spirilla, instead of having 
their usual curved " comma "-like form, have become 
globular, the majority staining badly and being dead. 
Outside the body the blood-serum of a highly immune 
guinea-pig has similar properties. If heated to 55 C. 
for an hour these properties are lost. There is, therefore, 
present in the peritoneal fluid, as in the blood-serum, a 
body of the nature of an alexin which clearly is, to some 
extent, responsible for this remarkable alteration in the 

It is impossible to deal in detail with the important 
controversy that has raged regarding the meaning of 




this Pfeiffer's phenomenon. Suffice it to say that Pfeiffer 
regarded the substance causing the phenomenon as an 
excretion from the peritoneal endothelium ; that Metch- 
nikoff demonstrated that it was best explained by what 

1. 3. 


FIG. 12. Pfeiffer's Phenomenon. Effect of human blood-serum upon spirillum cholerse 
(1, 2) ami bacillus typhosus (3, 4) respectively. 

In the two upper figures the blood-serum has been heated to destroy the intermediate 
bodies (immune bodies, opsonins). Leucocytes have been added, together with a 
suspension of the bacteria, and the mixture placed at 37 C. for fifteen minutes. 
It is seen that the polymorphonuclear leucocytes take up the unaltered bacteria. 
In the two lower figures are shown the effects of the unheated blood-serum. In 
other respects ,the treatment is the same. Under the action of the serum the 
bacteria swell up and become spherical, losing their power of taking the stain. 
There is equally active phagocytosis. In the leucocytes are to be seen some un- 
altered bacteria. These have been ingested before the serum has acted upon them. 

he has termed " phagolysis "- by the rapid clumping of 
the leucocytes of the peritoneal cavity upon the omentum, 
followed by their dissolution and liberation of bactericidal 
substances ; he has shown further that the phenomenon 
does not develop in connexion with body-fluids, such as 


the aqueous humour, containing few or no leucocytes ; 
and that it does not occur if phagolysis be prevented ; 
while Bordet (120) demonstrated that the reaction needed 
the presence of two substances an alexin or microcytase 
derived from the broken - down leucocytes, which is 
destroyed by heating to 55 C., and a second substance, 
not present in normal animals but present in the serum 
and peritoneal fluid of vaccinated animals, which is only 
destroyed at a temperature of 68 to 70 C. This 
substance belongs to the group of bodies which have 
been termed fixateurs, amboceptors, immune bodies, 
intermediary bodies, etc. In short, says Metchnikoff, 
it is the breaking down of the peritoneal leucocytes 
that is the main factor in bringing about Pfeiffer's 

But where does the second substance, the intermediary 
body, which also is essential for the development of 
Pfeiffer's phenomena, come from ? It is present, not 
only locally, but in most of the body humours. To be 
brief, there is no evidence that this substance is, under 
normal conditions, purely intracellular. The evidence, 
so far as it goes, is all to the contrary. As with the 
allied group of substances known as the agglutinins and 
the whole group of immune bodies produced in the 
development of immunity, these fixateurs or intermediary 
bodies are widely distributed in the organism, even in 
the excreta of immunised animals. They appear to be 
actively produced, and that over relatively long periods, 
and so we must conclude that they are products of living 
cells. This, at least, is the view accepted by all who 
have studied them, and, though he has singularly little 
to say regarding them, MetchnikofPs view partakes of 
this opinion (vide 93, p. 816). He regards them as also, 
in the main, produced by those cells which, under other 
conditions, act as phagocytes. That is, to say the least, 



doubtful. As Sir A. E. Wright (121) points out, the 
organisms which exhibit Pfeiffer's phenomenon under 
favourable conditions notably the spirillum of cholera 
and the B. typhosus may be taken up by the peritoneal 
leucocytes very rapidly, before the phenomenon has 
manifested itself. When so taken up, they do not swell 
up and become globular, though others, ingested later, 
have a globular appearance. If the phagocytes themselves 
produce the intermediary substance, why should not the 
bacteria undergo like alteration ivithin the bodies of the 
phagocytes ? It is only outside these cells that the 
change takes place. 

Wright's Phenomenon. Where, indeed, these inter- 
mediary bodies are produced is, at the present time, an 
open question. That they play an important part is most 
evident, but, what is more, the recent observations of 
Wright and Douglas and other co-workers (121) show 
that, in connexion with the majority of pathogenic 
bacteria, phagocytosis occurs with difficulty, if at all, in 
the absence from the body-fluids of certain of these inter- 
mediary bodies. These particular ones they have termed 
opsonins. 1 Wright and his fellow-workers confirmed the 
observation of Nuttall, Stern, and others that blood-serum 
has no bactericidal effect upon the pyococcus aureus. 
They were thus led to investigate to what extent active 
phagocytosis might explain the destruction of pyococci 
gaining entrance into the organism, and found that an 
active phagocytosis occurred, but only under certain 
conditions. If, along the lines of Irishman's experiment, 
the corpuscles are, by centrifugalisation, separated from 
the blood, and now a mixture be made of the washed 
white blood corpuscles, the serum, and a suspension of 
the pyococci, and this mixture be kept for a quarter of 
an hour at the temperature of the blood, active phago- 

1 From opsono (or obsono), " I cater for." 


cytosis occurs ; the polymorphonuclear leucocytes are 
seen to contain, each of them, numerous cocci. But if 
now, before making this mixture, the serum be heated to 
65 C. and then added, there is little or no phagocytosis. 
This points to the existence of something in the unheated 
serum which favours the process of ingestion of the 
bacteria, and, as a matter of fact, they show that a 
component of the active serum has a preparatory_action 
upon the suspended bacteria, so that the bacteria subjected 
to an unheated serum for a time and then placed in the 
heated inactivated serum are taken up more readily than 
those placed direct into an inactivated serum. They 
show, further, that in the process of rendering an indi- 
vidual more refractory to pyococcic infection, his serum 
comes to contain increasing quantities of this ppsqnin or 
intermediary substance. So that, taking the washed 
white blood corpuscles of an ordinary individual, adding 
these to a mixture of the serum of the immunised indivi- 
dual with a suspension of pyococci, the leucocytes of this 
normal individual take up a much greater number of the 
cocci in a given time than do washed white blood corpuscles 
of the immunised individual suspended in thecentrifugalised 
serum of the normal individual, to which a -culture of 
pyococci has been added. What is true of the staphy- 
lococci obtains also with the long series of other bacteria, 
the only exception noted being in connexion with 
members of the diphtheria group. These opsonins are. 
found to be distinct from alexins or bactericidal sub- 
stances, apparently distinct also from the ordinary inter- 
mediary bodies, for their activity is greatly lowered by 
subjection to a temperature of 55 C. But, as with the 
intermediary bodies, it seems improbable that they are 
secreted by the phagocytic cells. Sir A. E. Wright and 
Douglas found that the blood-serum of cases with abscesses 
might be from six to ten times more active in promoting 


phagocytosis than the serum extracted from the pus of 
the abscesses. Are they secreted by the phagocytic cells 
or by leucocytes of any order ? It would not seem 
possible to give a definite answer to the question at the 
present time. Wright and Eeid (122) pointed out 
that the exudates produced by a given bacterium may 
contain no opsonins of any order, an observation which 
has been confirmed by Opie (123). If leucocytes 
secreted these substances, then we should expect that 
with collection of leucocytes, the containing fluid 
should on the contrary afford opsonins in the greatest 

Aggressins. But here other considerations have to 
be taken into account, which would seem adequate to 
explain the absence of opsonins in these circumstances. 
As pointed out some years ago by Ainley Walker (124), 
as again by Welch (125), if the cells of the organism 
produce antibodies of various orders when stimulated by 
microbes and their products, it is at least probable that 
the resistance of these microbes to the action of the body 
humours is to be explained by a reciprocal production of 
antibodies by them. And, as a matter of fact, Bail 
(126) has demonstrated the existence of bodies of this 
nature. Injecting typhoid bacilli, for example, into the 
pleural cavity of animals of the laboratory and so setting 
up a serous inflammation, he collected some of the fluid, 
centrifugalised, to remove the bacteria and cells and 
gained a fluid that was apparently inert, that could 
be inoculated into other animals without causing any 
symptoms. But now if he inoculated this fluid into 
those animals along with a sublethal dose of the typhoid 
bacilli, that dose forthwith became lethal. Instead of 
the bacteria being destroyed, they underwent active 
proliferation. Clearly there is a something in the 
inoculated exudate that has paralysed the protective 


agencies of the body, something presumably supplied by 
the bacteria which had grown in the fluid, and this 
something is not of the nature of a toxin, for the fluid 
is not toxic. To this Bail has given the name of 

Hence where an exudate is found devoid of opsonins, 
it is permissible to surmise either that the aggressius 
given off by the bacteria have neutralised the opsonins 
present in that exudate, or that the leucocytes of that 
exudate which under normal conditions should have 
afforded opsonins have been paralysed by the aggressins, 
and opsonin production has been inhibited. 

FIG. 13. Granular (oxyphil) leucocytes attacking and coming into apposition to chain 
of />'. ramosus. From a Ziegler's chamber preparation, peritoneum of guinea-pig. 

These researches of Sir A. E. Wright have established 
beyond doubt that besides the actual engulfing and 
digestion of bacteria by the leucocytes another factor has 
to be taken into account (127). 

It is interesting to note that the late Prof. Kanthack 
and Mr. Hardy (4-4), whose observations I quoted in a 
previous edition of this article as affording direct evidence 
that the leucocytes secreted bactericidal substances, were 
the first to demonstrate clearly the compound -nature 
of the bacteriolysis. Adding a suspension of anthrax 
bacilli to frogs' lymph and studying what happened, by 
prolonged examination under the microscope, they 
observed that cells possessing granules staining with 
eosin (amphophil cells) attacked the bacilli, coming into 


contact with them, and, in this process, the granules 
were seen to be discharged. These cells did not actively 
ingest the bacilli, but they, while remaining extracellular, 
showed evidence of degeneration. Only later and by 
another form of leucocyte was there, in the frog's lymph, 
active phagocytosis. In other words, according to them, 
one set of leucocytes secreted the substances acting upon 
the bacilli, and prepared these to be ingested by another 
order of these cells. 

It must be admitted that these observers introduced 
not a little confusion by speaking of certain cells as 
eosinophilous, which, as a matter of fact, correspond in 
most respects to the ueutrophil and amphophil poly- 
morphonuclears of the mammal, to cells which in higher 
animals are definitely phagocytic. Mesnil and others 
have been unable to confirm these observations ; but 
then they have not repeated Kanthack and Hardy's 
experiments according to the lines laid down by them ; 
they have attempted other methods. It must be 
admitted that in the frog the experiment does not 
always succeed. I have, however, made and seen pre- 
parations in which this loss of granulation and coincident 
degeneration of the bacilli could not be denied. We 
have not been able to convince ourselves that in the 
mammal the coarsely granular eosinophil possesses a like 
excretory function, although Kanthack and Hardy con- 
vinced themselves that this was the case. In his last 
confirmatory article upon this subject Hardy (45) calls 
attention to the fact that after contact with bacilli the 
coarsely granular oxyphil cells of the frog are diminished 
in size, and, what is more, that as the cells crawl over 
chains of bacilli they leave behind them a slime. Bacilli 
coated with such a slime never grow. Beckton (56) and 
others have of late called renewed attention to the 
similarity in size and staining properties between the 



granules of eosinoplrils and lymphocytes and the granules 
of known secretory cells. The observations of Durham 
and others indicate a little -understood alteration in 
bacteria preparatory to ingestion, which, indeed, would 
suggest that the leucocytes of mammals do afford some 
preparatory secretion. If, for example, a relatively 
abundant suspension of actively motile but not highly 
virulent bacilli be introduced into the peritoneal cavity 
of an untreated guinea-pig and a drop of the peritoneal 

FIG. 14. Leucocytes from peritoneal cavity of guinea-pig into which a microbe of low 
virulence (Sanarelli's " Versailles " vibrio) has been inoculated, to show "hedge- 
hog " appearance around the mononuclear hyaline cells (a). The eosinophils (b) 
and the lymphocytes (c) were not affected (the polymorphonuclear not shown 
to a less extent than the hyalines). After DURHAM. 

exudate be examined, it is noted at a certain period that 
while many of the bacilli are moving freely in the fluid, and 
may impinge with impunity upon the lymphocytes and 
eosinophil cells present, those that come into contact with 
the hyaline mononuclears become arrested and motionless, 
and the individual leucocytes may be seen taking on a 
" hedgehog " appearance, the immobilised bacteria adhering 
by one end to the leucocyte. This adhesion and immobilisa- 
tion, prior to ingestion, cannot, I think, be explained except 
by presupposing the existence of a zone of diffusible 
material discharged from these cells and having a prepara- 
tory action upon the bacilli. Putting all the facts together, 


it would seem that, under certain conditions, some pre- 
paratory or intermediary substances are separated from 
leucocytes of one or other order which aid in the process 
of phagocytosis, lut the evidence is very far from con- 
clusive that all the preparatory substances are of leucocytic 


I AM thus led to modify MetchnikofFs conception of the 
mode of action of the leucocytes to some extent, and to 
believe that the following more nearly represents the 
state of our knowledge. 

1. Phagocytosis the process of irigestion and diges- 
tion on the part of the cells of the organism is the 
factor most generally involved in the destruction of 
pathogenic organisms within the system ; both fixed cells, 
like those of the endothelium, and certain orders of free 
cells leucocytes can manifest this property. 

2. Where the cells are able to take up living 
bacteria, these bacteria have, in most cases, to be subjected 
to an extracellular action by substances present in the 
surrounding medium (opsonins) prior to digestion. 

3. Bacteria may also undergo destruction without 
phagocytosis taking place. Where this is the case, the 
bacteriolytic substance (cytase) is liberated into the 
medium upon the death and disintegration of cells that 
are potentially phagocytes. It, however, cannot act 
without the intervention of a second intermediary body 
(fixateur) present in the medium. 

4. Whether leucocytes and other cells of the body 
can actively secrete the bacteriolytic substance when 
stimulated, must be left an open question ; it must, 



however, be recognised that certain leucocytes secrete 
and discharge substances which, if not directly bacterio- 
lytic, are preparatory to and essential for the destruction 
of the bacteria. 

I have so far utilised Metchnikoff's terminology, 
using the term " cytase " for the intracellular bacteriolytic 
substances and, thereby, tacitly accepting the view that 
these bodies are ferments (the termination " -ase " being 
the conventional method to signify an enzyme). This is 
the view of the French school of bacteriologists generally. 
How far is this accurate ? Other workers are by no 
means convinced that this is their nature. Pfeiffer and 
Ehrlich l regard the intermediary body or amboceptor as 
the ferment ; and there is much to be said in favour of 
this view, but, so long as the nature of enzyme action is 
not clearly understood, so long as it is currently accepted 
that, in organic fermentations, only two factors are 
involved, for so long must there be confusion regarding 
what is the true ferment or enzyme. As a matter of 
fact, it would seem that in all organic enzyme action for 
the development of the complete cycle there are at least 
three factors requisite, and we have to decide which of 
these three is the ferment. The realisation that three 
factors are and must be involved immediately aids our 
comprehension of the facts here brought forward, and 
renders it easy to grasp the otherwise most puzzling 
order of affairs presented in the co-existence and mutual 

1 "Da unter dem Einfiuss ties Addiments (intermediary body) Erschei- 
iinngen auftreten, die man mit Pfeiffer als der Verdauung analog ansehen 
muss, so werden wir nicht fehlgehen, wenn vvir dern Addiment den 
Charakter eiues Verdanungsfermentes vindizieren." Ehrlich mid Morgen- 
roth, Uber Haemolysin. 1. Mitteil. Gesammelte Arbeiten von Ehrlich, 
p. 12. It is, however, but right to state that in more recent communica- 
tions Ehrlich appeal's studiously to avoid any definite comparison between 
the bodies concerned in the production of immunity and the digestive 
and other ferments. I gather that at present he maintains an open mind 
in the matter. 

CH. xv SUMMAEY 123 

action of the ainboceptors and complements, of fixateurs 
and cytases, opsonins, lysins, et hoc genus omne. 

The ferments of the body should not be looked upon 
as a special group of chemical compounds, but, on the 
contrary, enzyme action should be viewed as a property 
of a series of substances possibly of very different chemical 
composition. This view was seriously discussed by 
Woodhead (128) many years ago. The study of the 
catalytic action of finely divided platinum, of manganese 
dioxide, etc., supports this view. I am thus inclined to 

FIG. 15. Diagram to illustrate the conception regarding ferment action and its 
application to the side-chain hypothesis of immunity. FS, molecule of ferment- 
escible substance with side-chains ; FR, molecule of fermentator ; F, ferment ; 
FI F,i, etc., the successive stages in the action of the ferment. 

consider that the ferments, or bodies presenting enzyme 
action, may be regarded as a certain order of unsatisfied 
molecules, which enter into combination with certain 
other molecules in the surrounding medium for which 
they have affinity ; a chemical action takes place 
resulting in the enzyme becoming attached to certain 
atom groups of the fermentescible molecule. It is not 
even necessary to suppose that as the result of this 
chemical action the atom groups are immediately de- 
tached. So soon as this union takes place, the enzyme 
molecule becomes satisfied, and, were there no other 
substances present to disturb the equilibrium, further 


action would cease. For ferment action to continue 
there must be some third substance present, which we 
may term the fermentator or complement, having an 
affinity for the atom-group represented by the ferment 
plus side -chain, or atom -group, of the fermentescible 
substance. As the result of its presence, a second union 
takes place, and what had been the unsatisfied atom- 
group of the fermentescible substance of the above com- 
pound, becomes now attached to and combined with the 
fermentator. The ferment becomes free, once more 
unsatisfied, and ready to act again on the fermentescible 
substance. We may go farther and suppose that the 
atom-group or side-chain of the fermentator, thus modified 
by combination with the atom-group from the ferment- 
escible substance, may either remain attached to the 
fermentator (complement) or become liberated as a 
separate entity. 1 

According to this conception of ferment action, which 
is the ferment the intermediary body (fixateur) or 
the lysin, complement or cytase ? Clearly, I think, the 
former ; the cytase can only be regarded as the 
fermentator, the substance essential for the final act in 
the process. And this view is supported by Wright's 
phenomenon, in which the bacterial bodies must have 
the intermediary body (opsonin) in association with them 
before they can be acted upon by the cells. Possibly it 
is a matter of convention which we shall term the 
enzyme or ferment, whether the intermediary body or 
the third substance essential for the completion of the 
process. Yet it seems more just that the active factor 
in the process should receive the name ; just as, with 
Pawlow and Bayliss and Starling (129), and in opposi- 
tion to Metchnikoff and Delezenne (130), I would lay 

1 I have expanded this conception of enzyme action in my Principles 
of Pathology, vol. i. p. 58. 

CH. xv SUMMAKY 125 

down that in tryptic digestion the enterokinase is the 
true ferment, and would regard the trypsinogen as the 
fermentator. And thus, to sum up, I find myself in 
harmony with Pfeiffer and Ehrlich in regarding the 
amboceptor or intermediary body as the ferment, and, in 
place of speaking of the cytase of the phagocytes, would 
speak of the cytolysin or the bacteriolysin. 

To complete this section I will here add other con- 
clusions deduced from a study of the later stages of 
inflammation and discussed in a later chapter ("Upon 
the part played by the fixed cells in the Inflammatory 
Process "), viz. : 

5. In the later stages of inflammation the growing 
fibroblasts may often be seen to contain leucocytes in 
process of digestion. Presumably, therefore, a certain 
number subserve nutrition. 

6. Others are, in certain cases, recognisable in the 
lymph-spaces outside the inflammatory focus, containing 
the debris of dead tissue. Emigration can therefore 
occur as well as immigration. 

7. The process of development of wandering into 
fixed cells has been observed ; cells of what are here 
termed the hsematogenous type have never been observed 
to become thus converted ; the histogenous alone would 
seem able to undergo the change. 

8. The contrary process of development of wandering 
cells from degenerating tissue (muscle -fibres) has also 
been recorded by more than one observer. 



WHENEVER injury to the tissues leads to vascular dilata- 
tion there is an increased effusion of fluid from the 
blood. The extent of this effusion varies greatly ; it 
varies with the tissue affected, the state of the organism, 
and the quality and nature of the irritant. Dense tissue 
permits of little exudation, while loose vascular tissue, 
under the action of an irritant of no great intensity, may 
undergo great exudative swelling. There is, for instance, 
a peculiar liability in serous and cutaneous surfaces (or 
more truly in subserous and dermal layers), when 
inflamed, to manifest abundant exudation. Their 
vascularity and the slight external resistance would 
appear to explain this liability. There is a like tendency 
to abundant exudation from mucous surfaces, though 
here to a large extent, so long as the covering epithelium 
is not destroyed, the exudation is governed i.e. partakes 
of the nature of an excretion, abundant mucus from the 
cells being also discharged, and is not extreme. With 
destruction of the covering epithelium it may become 
profuse ; choleraic diarrhoea is perhaps the most pro- 
nounced case of exudation in the human organism. That 
some general state of the organism is a* factor concerned 



is seen when virulent anthrax bacilli are inoculated 
subcutaneously -into an ordinary rabbit and into one 
that has been rendered immune : in the former the 
exudation is of a serous nature, in the latter little fluid 
is exuded from the vessels a clear indication that, in 
the development of the immune state, not merely the 
leucocytes, but also the capillary walls, at the least, 
become altered in their behaviour to the toxins. The 
effect of the quality of the irritant is observable upon 
comparison of the results of inoculation of various 
microbes. Some cause little exudation of fluid. These 
are in general of low pathogenic quality, but not 
always ; certain virulent microbes (such as those of 
tetanus) lead, when inoculated, to relatively little 
effusion of fluid from the vessels. On the other hand, it 
may be stated definitely that where in a moderately 
dense tissue the injection of a pure culture of a micro- 
organism leads to well-marked exudation, the micro- 
organism is of high virulence. 

Can any meaning be ascribed to this effusion, or, to 
express the same idea in words which shall not offend 
those who fear the semblance of teleological ascriptions, 
has the increased pouring out of fluid into the tissues as 
the result of irritation been of proved benefit to the 
species, so that those individuals have survived who have 
manifested this reaction and have conveyed it to their 
offspring ? Is it an attempt at increased nutrition in the 
injured region ? It has been suggested, in accordance 
with Virchow's conception of inflammation, that the 
injury, stimulating the surrounding fixed cells, leads to 
increased local metabolism ; and that the exudation is a 
means of bringing to the region the increased nourish- 
ment demanded by the increased cellular activity. But 
inasmuch as exudation is most marked in those cases 
where 4here is most profound and rapid cell destruction, 


and again at the early stage of the inflammatory reaction, 
when evidences of growth and proliferation of the fixed 
cells of the region may be, and most often are, wholly 
wanting, this view can scarcely be upheld. Yet at a 
later period of the process, and again in chronic inflamma- 
tion, the overgrowth of the connective-tissue cells would 
appear to stand in some relationship to the over-nutrition 
caused by the continued dilatation of the vessels and the 
pouring out of excessive lymph into the tissues. It must, 
however, be admitted that the exact relationship subsist- 
ing between nutrition and cell overgrowth is still a 
matter of debate among pathologists. 

That the exudation exerts a " flushing-out " action is 
very evident in many cases. Thus the inflammation 
induced by plunging an animal's leg into hot water is 
accompanied by great increase in the amount of lymph 
obtainable from the efferent lymphatics of the part. 
Experimentally it may be shown (Samuel (131) and 
others) that seven to eight times as much lymph may 
drain away from an inflamed as from an uninflamed 
region. It is shown also by the presence of streptococci 
in the lymph channels outside the area of acute in- 
flammation in erysipelas, by the frequent implication of 
the nearest lymph-glands in suppurative disturbances, 
and by the appearance of lesions, due to the direct action 
of bacterial products, in organs far removed from the 
focus of bacterial proliferation in such diseases as diphtheria 
and tetanus, wherein, as a rule, the bacteria remain 
strictly localised. It is clear that the exudation into an 
inflamed area may be much in excess oj' the normal 
transudation ; that it can accomplish a removal of irritant 
matters. It is clear also, from more than one of the 
examples given above, that a process which may be 
beneficial to the region of injury may be harmful to the 
system as a whole. 


Study of the " flushing-out " effects of the inflammatory 
exudate affords, in fact, a good object-lesson as to how 
far the reaction to injury is to be regarded as purposive. 
Yet we are forced to see that there is a certain amount 
of adaptation. Where the irritant can be conveyed to 
the exterior an abundant exudative inflammation generally 
occurs an abundant flushing ; where it can be conveyed 
into one of the body cavities the same holds good ; but 
here a mechanism is often called into action (deposit of 
fibrin) whereby the exudate with its contained irritants 
is held within the serous cavity for days and weeks after 
all signs of active inflammation have subsided. The 
organism, that is to say, would seem to restrain its 
drainage to the general lymphatic system. Where the 
irritant is merely the product of tissue-change the profuse 
exudate is rapidly conveyed away ; where, on the other 
hand, the injury is of bacterial origin, the passage of lymph 
from the focus of inflammation is, generally speaking, 
not nearly so free ; it is of thicker consistency and drains 
away slowly. In short, as I have already indicated, 
where the microbe is not too virulent a cellular rather 
than a serous inflammation is produced ; and in place of 
abundant flushing an increased antibacterial and antitoxic 
action of the exuded lymph comes into play. 

But besides the mere " flushing-out " the exudation 
has often another effect, namely, dilution of the irritant 
and reduction of its injurious properties, so that it acts 
with lessened force upon the tissues, and permits the 
wandering cells to be attracted to the region where they 
may exert their functions. Where a comparatively mild . 
physical irritant leads to abundant exudation the flushing- 
out action appears to be in the ascendant, where microbic 
irritants cause great local inflammatory oedema, judging 
from the less extensive lymph flow from the region, the 
diluent action must be regarded as the more important. 


130 INFLAMMATION i-r. n 

I have already pointed out that a relation may be traced 
between the intensity of bacterial irritation and the 
extent of the exudation. In short, there may be a great 
exudation under two apparently opposed conditions : in 
the presence of comparatively mild physical irritants, and 
in that of severe bacterial irritants. In the former case 
it more especially subserves removal, in the latter dilution 
of the poison. 

That in general the exudate exercises a beneficial 
effect upon the process of an acute or subacute inflamma- 
tion, and that in general this is inadequate rather than 
excessive has been brought home to us of recent years by 
the remarkable success which has followed Bier's method 
of. treatment (132), when properly carried out. This 
treatment consists in the production of a moderate grade 
of increased hyperrcmia in the affected part whether (1) 
by the application of compressing bandages above the 
part, so that the free return of venous blood is obstructed, 
(2) by cupping, so that the lessened pressure causes 
determination of blood to the part, or (3) by the local 
application of heat as from electric or hot air baths, lead- 
ing to increased vascular dilatation. All these means, 
indeed, bring about dilatation of the capillaries, thinning 
of their walls, and increased pouring out of fluid into the 
tissues. The swelling and redness of the part increase, 
but the patient benefits. Care, it is true, has to be taken 
and each individual case treated upon its merits ; there 
must also be alternation of periods of treatment and rest. 
If properly carried out, as the part becomes swollen the 
sense of local pain disappears, and it is replaced by a 
feeling of general well-being, the inflammation ceases 
to extend, and actual absorption takes place of the 
inflammatory products. 

How does this treatment bring about its good effects ? 
Since corresponding results are gained whether we slow 


the circulation by compression or increase it by the local 
application of heat, it seems scarce likely that increased 
passage of leucocytes into the part is the essential process. 
On the other hand all these methods favour the increased 
passage of fluid exudate into the affected tissue. It is to 
this that the good effects are to be ascribed, more 
particularly to dilution of the irritant and conveyance to 
the part of an increased amount of antibodies. We can 
to some extent harmonise the good effects gained by Bier 
by his treatment with those coincidentally gained by Sir 
A. E. Wright in his opsonic treatment, 1 by supposing that 
increase in the exudate means increased amount of 
opsonins brought to bear upon the bacteria setting up 
the inflammation whereby these are prepared for ingestion 
by the leucocytes present in the inflamed area. Cer- 
tainly the exudate contains opsonins ; these pass, that 
is, out of the blood-plasma into the tissues. 

1 Wright's treatment cannot be used for very acute cases, but has been 
found valuable in subacute and recurrent conditions. It is based upon 
the inoculation, subcutaneous, of accurately graduated amounts of dead 
cultures of the microbe that has caused the specific form of inflammation. 
The results of such inoculations are that the "opsouic index" or relative 
amount of opsonins in the blood-serum (and so in the body fluids) becomes 
definitely increased, and with this the capacity of the organism to with- 
stand the growth of the specific microbe. 



THE fundamental characteristic of inflammatory exudation 
as compared with ordinary lymph is its richness in pro- 
teins. Whether we regard lymph as a filtrate pure and 
simple from the blood, or follow Heidenhain in regarding 
it as the result of a selective filtration, it is highly 
probable that in inflammation the exudate approaches in 
its composition more nearly to the blood -plasma than 
does ordinary lymph. The dilatation of the capillaries, 
the consequent thinning of the endothelial layer with, 
it may be, the opening of some lacuna between the 
individual cells, and the direct action of the irritant 
upon these cells, may all be expected to aid the transuda- 
tion. In this way the amount of protein matter in the 
lymph may be increased. But equally important must 
be the addition of proteins due to the breaking down of 
leucocytes and tissue -cells. I have already discussed 
this destruction of the cells, and need not here give the 
evidence of its occurrence. 

This increase in the solids, mostly proteids, of the 
exudate, has been well shown by Samuel, by measuring 
the flow of lymph from the main lymph-channel of the 
dog's foot. In one such experiment he found that there 
was discharged in successive periods of three hours : 

1. Untreated foot: 4'0 lymph containing 4 to 
5 per cent of solids. 



2. Foot subjected to venous obstruction: 28*5 
of lymph containing 2 to 3 per cent of solids. 

3. Foot inflamed: 28'5 of lymph containing 7 
per cent of solids. 

The figures in the third case of inflammation do 
not represent the whole exudate. So thick was the 
lymph that it tended to clot and obstruct the canula, 
and there was, in addition, much oedema and swelling of 
the foot. But obviously, as Ainley Walker (133) points 
out, from twenty to thirty times more proteid matter 
may drain away from an inflamed than from a healthy 

In addition to the proteins the inflammatory lymph 
may contain other substances worthy of more than pass- 
ing note. Of these the more important are ferments, 
the results of proteolysis (notably fibrin and its pre- 
cursors, nucleo- albumins and albumoses), and in many 
cases mucin, together with bactericidal substances, and, 
where bacteria are present, the products of their growth. 
Various extractives have been noted. Exudates rich in 
cells and disintegrated tissue-products pus, for example, 
may contain glycogen, fats, and, as Klotz working in 
my laboratory at McGill University has recently shown 
(134), a very definite amount of soaps. 

The presence and amount of these substances depend 
largely upon the intensity and character of the inflam- 
mation. Thus the total quantity of proteins, and the 
proportion of fibrin, albumin, and globulin present, vary 
within wide limits. The following table l of observations 
made by Halliburton (135) shows well this variation in 
proteins, and the difference existing between inflammatory 
exudations and dropsical effusions : 

1 These figures are thoroughly in accord with those of other analyses 
by Reuss, Hofmann, Mehu, and Letulle (136). 




Percentage Quantity of 

Sp. Gr. 







Acute pleurisy, Case 1 






Case 2 






,, Case 3 






Hydrothorax . . "\ 
Average of three cases / 






Exudates, as a rule, have an acidity above 10 with 
decinormal soda solution, using phenol-phthalein as an 
indicator, transudates under 10. The difference is too 
slight to be relied upon. Eeuss, Hofmann, and others 
have shown that the amounts of solids and extractives 
vary very slightly ; it is the albuminous matters which 
mainly determine the variations in the specific gravity. 
The factors determining the amount of albumin are 
many. Thus, as a general rule, more albumin will 
be found in a pleural exudate than in a peritoneal 
exudate. Some observers, however, place peritoneal 
exudate first in order. The pericardial and the sub- 
cutaneous come next. It has been noted, further, 
that the specific gravity and amount of albumin are 
somewhat higher in right-sided pleurisies than in left 
(Berheim and Brunting). The amount of albumin in 
the blood is also a factor. Thus in ansemia or 
hydraemia the specific gravity of exudates is lowered. 
Other factors are also given ; among them the not 
infrequent combination of mechanical effusion or 
obstruction with active inflammation. The most im- 
portant of all would appear to be the extent of irrita- 
tion of the affected part. Thus, in four cases of 
pleurisy, Euneberg compared the amount of albumin 
found in the pleural fluid with that present in acute 
blisters produced on the patients by cantharides ; the 


average of the former was 5*43, of the latter 6 '2 per 
cent ; without exception, the percentage in the blister- 
fluid was found the higher. It may be laid dawn (Miller 
(137)) that fluids with the specific gravity of 1018 or 
higher, with at least 4 per cent of albumin, are of inflam- 
matory origin ; of from 1010 to 1015, with albumin u] 
to 3 per cent, are due to venous stasis ; of less than 101 
with albumin under 1 per cent, are due to hydrsemil 

The Cells of the Exudate. Much study has, of late 
years, been devoted to cytodiagnosis, to the diagnosis of 
inflammatory and other conditions by a study of the 
cells present in the removed fluids. It cannot, however, 
be said that, for our present purposes, much has been 
elicited beyond this, that abundant polymorphonuclears 
indicate an active inflammation ; a preponderance of 
lymphocytes, either a tuberculous infection or sub-acute 
inflammation of other nature of some little duration, 
though it has also been noted that in the early stages 
of a tuberculous inflammation (e.g. in a rapidly developing 
tuberculous pleurisy) there may also be an abundance of 

Fibrin. Between the amount of fibrin present in 
exudations and the amount of peptones there is an 
inverse ratio. Peptones are especially developed in 
connexion with suppurative inflammation ; and the more 
an inflammation tends to be suppurative the greater is 
the breaking down of the fibrin, as also of fixed and 
wandering cells, and the more evident the production of 
peptones, or more correctly of albumoses, until in chronic 
abscess - formation of fair extent these pass into the 
general circulation, and are excreted and recognisable in 
the urine. 

Into the discussion of the mode of formation of fibrin 
I need not enter here, intimately connected as the 


subject is with the inflammatory process. The greater 
text-books of Physiology enter exhaustively into the 
matter. Suffice it to say that, as in the blood, a direct 
relationship is made out between the breaking down of 
leucocytes and the development of this substance in 
inflammatory exudations. 

It is in connexion with inflammation affecting serous 
and epithelial surfaces l that fibrin is more clearly 
recognisable, forming, it may be, thick coatings of the 
badly named " inflammatory lymph " over the inflamed 
surfaces. This deposit is in all respects comparable to 
the formation of thrombi in the blood-vessels (138). 
Here, as there, the deposit occurs only when the 
endothelium has undergone destruction and the roughened 
sub-endothelial tissues are exposed And here also the 
fibrin may be deposited either in filamentous or homo- 
geneous and hyaline form according to circumstances. 
How far the blood-platelets are involved in the production 
of inflammatory fibrin is a matter deserving further 

The above statement gives the general consensus of 
opinion among pathologists of the present time regarding 
the mode of origin of inflammatory fibrin. And yet 
this view has been vigorously opposed by Neumann 
(139), at least as regards pseudo-membranous inflam- 
mation. According to him, careful examination of 
fibrinous inflammation of the serosa and of diphtheritic 
inflammation of the mucous membranes shows that the 
deposit of fibrin does not lie upon, but under, the 
endothelium or epithelium. The statements of other 
observers (Marchand, Orth, Ziegler (140)) that endothelial 
cells can be seen here and there still remaining under 
the fibrin, he strongly controverts ; the cells so seen 

1 Of epithelial surfaces, more especially those covered by a single cell 
layer, as notably the pulmonary alveoli. 


are, he declares, swollen connective-tissue cells, and the 
hyaline glistening bands of fibrin are seen to be directly 
continuous with the connective-tissue fibrils ; the bands 
of fibrin, in short, are derived from, and are a modification 
of, the swollen and altered connective - tissue fibrillse. 
This change Neumann terms " fibrinoid degeneration " of 
the connective tissue. He admits that, in acute cases 
free from connective - tissue proliferation, the fibrin is 
more probably the result of exudation. Now it is quite 
true that there are cases in which one cannot say 
certainly what is the nature of the cells seen under the 
fibrin ; yet there are other cases in which there cannot 
be a doubt as to their endothelial nature. As Lubarsch 
(71) points out, this can be proved experimentally by 
causing a minimal acute local peritonitis, by simply 
pulling out a loop of intestine, pinching it with sterile 
forceps, and replacing it. If the animal be killed, in 
thirty-six hours small spots may be found here and 
there, which are covered with a finely granular, scarcely 
visible deposit of fibrin ; in such local deposits a 
continuous layer of endothelial cells can be made out 
passing well under the fibrin, which is pure exudate. 
Neumann's layer of endothelium covering the false 
membrane certainly exists in several cases, but it is 
clearly of secondary development. Gaylord (141), who 
made a full study of the subject under Orth, has shown 
clearly that after introducing fibrin or fibrin-fprmiug 
fluids into the serous cavities, the endothelium proliferates 
and covers over this foreign fibrin, sometimes before any 
signs of organisation appear in the mass. Undoubtedly 
connective tissue undergoes a change in the inflammatory 
area ; the bundles of fibrillse swell, become more hyaline, 
lose their fibrillar appearance, and form more homo- 
geneous glistening bands. But this modified connective 
tissue can generally be distinguished from fibrin proper 


if suitably stained by employing van Gieson's stain, or 

The researches of Leo Loeb (142), conducted in my 
laboratory, possibly throw some light upon this subject. 
Taking a little lobsters' blood in which coagulation has 
been delayed by the addition of a solution of adrenalin 
chloride, and placing this upon a slide, then covering 
this with a second slide and pulling the one slide over 
the other so as to exert traction, it can be seen under 
the microscope that the cells, arranging themselves in 
rows, become transformed into a system of threads, and 
here and there the threads can be seen passing through 
a cell or even through the nucleus of a cell ; the cells 
often become spindle-shaped and may either be so drawn 
out that their protoplasm forms long threads, or fine 
fibrillar threads may be seen actually passing through 
several cells. Loeb produced a similar transformation 
into fibrils by traction upon the protoplasm of exploded 
cells. It is worthy of note that, during either process, 
the cell - granules disappear, and these fibrillse have 
staining and other reactions which connect them both 
with fibrin and with connective tissue. Many of them, 
for example, stain well by Mallory's connective- tissue 
stain. In other words, these observations of Loeb favour 
the view that the conversion of the protoplasm of 
connective - tissue cells into fibrillse is the result of 
tension and traction, i.e. of physical agents, and that the 
same is true also of the development of the threads 
of fibrin. If the same process be at work in both 
conditions, there is little wonder that it is difficult to 
draw a sharp line of distinction between the intra- and 
extra-cellular process in tissues where both are occurring 
at the same time. 

Leaving out of account coagulation -necrosis as not 
occurring in direct connexion with exudates, it may be 


said that similar fibrin formation is frequently recognisable 
in connexion with primary inflammation of parenchy- 
matous tissues. Where there are abundant and disten- 
sible lymph-channels extensive clotting may be seen in 
the lymph. This is peculiarly well marked in the con- 
tagious pneumonia of cattle (contagious pleuro-pneumonia). 
In acute inflammation of various organs, by appropriate 
methods of staining, similar formations of threads of fibrin, 
often starting from cells as centres, may be observed in 
the tissue spaces. Many years ago Werneck de Aquilar 
(143) demonstrated the frequent existence of fibrin in 
acute tubercular processes. 

The beneficial effects of fibrin formation in serous 
cavities have been rendered abundantly manifest by the 
increase in abdominal surgery. No one who has followed 
any considerable number of operations for appendicitis 
can have failed to remark how, in case after case, despite 
the intricacy of the intestinal coils and their mobility, 
the strongly irritant matter produced by gangrene of the 
appendix, or oozing through perforations in it, is restricted 
within a relatively small space by the surrounding 
fibrinous adhesions which form rapidly between the 
intestinal loops. By this means alone the peritonitis 
is restricted and " regional," instead of being generalised 
from the onset. Even when inflammation (as in peri- 
carditis) affects the whole extent of a serous cavity, the 
layer of fibrin acts as a protective coat closing the 
lymphatic " stomata," hindering the free absorption of 
the morbid material by the lymph- and blood-vessels, and 
filtering bacteria out of such fluid as does find its way 
through to the tissues beneath. It is not a little 
remarkable to call to mind how case after case of 
purulent pericarditis or purulent pleurisy may be 
examined in which, despite the intense suppurative 
disturbance in the serous cavity, the tissues at the other 


side of the deposit of fibrin the myocardium or the 
lung tissue show little or no tendency to abscess 
formation. Let there be primary abscess formation or 
gangrene in the lung, and perforation of the pleura and 
pleurisy may supervene : generalised pleurisy, however 
intense, does not lead to this unless complicated by other 
disease. Let there be primary or metastatic abscess in the 
myocardium, then there may be aneurysm and rupture of 
the heart ; yet such rupture produced by extension inwards 
of a purulent pericarditis is of the utmost rarity. Let 
there be inflammation originating in the submucosa of the 
intestines, as in enteric fever, and perforation may result ; 
general peritonitis, while often due to perforation, never 
so far as I can find directly induces that event. 1 In 
all these cases the natural protective layer of the serous 
surface is removed or gravely injured at a very early 
stage ; and the layer of fibrin, replacing the serous endo- 
thelium, forms an effective barrier. I may add that the 
mucin, extruded so as to form a layer over inflamed 
mucous surfaces, presents a similar protective action. 

But here again attention should be called to the fact 
that, while we can thus recognise an action beneficial to 
the economy in the laying down of fibrin, the adaptation 
to the needs of the economy is very far from being 
perfect, and the ultimate results are even replete with 
danger. I know of no better example than is to be 
derived from a study of the great omentum in a series of 
cases of abdominal disturbance (145). Time and again 

1 Where, however, there are localised pockets of pus such perforation 
may occur, and what I have termed "exogenous perforative ulceration" 
of the intestines is much commoner than is generally suspected (144). 
Thus in 700 autopsies in all conditions, I encountered it seventeen times. 
The conditions favouring the development of , the condition are adhesions 
after general or local peritonitis : formation of a pocket of pus, or abscess, 
between them ; tension, acting with most effect on a soft walled viscus ; 
compression, anaemia, and malnutrition of the wall ; lessened vitality ; 
infection ; ulceration ; perforation. 


we find that this which, with the peristaltic action of the 
bowels, appears to be in constant movement over their 
surface, has become attached by fibrinous adhesions over 
some inflamed area, thereby acting as a plaster or pad, 
reinforcing a weak point and, by the adhesions, preventing 
generalised peritoneal infection. From this aspect alone, 
the great omentum can only be likened to a brooding 
abdominal providence. But, when these adhesions 
organise, the omentum, now firmly attached, forms a 
band or bands of most dangerous import ; now constrict- 
ing a coil of the intestine and so causing obstruction, or 
kinking the bowel, or leading to internal hernia and 
volvulus. In short, the late results of adhesions may be 
very serious. 

The fibrin so thrown out, while it may (1) be dissolved 
by the action of bacterial products, or (2) undergo 
complete absorption by the cells and fluids of the body 
with restitutio ad integrum of the affected areas, may 
also (3) form a frame -work upon which new tissue- 
growth occurs with replacement by organised connective 
tissue. This new tissue -formation in inflammation we 
shall discuss later. 

Passing now to the ferments and ferment-like bodies 
present in the exuclate, I may briefly state that these are not 
only generated and excreted by the pathogenic bacteria 
present, but are liberated by the breaking down of the 
wandering cells. Abundant evidence of the existence 
of bacterial ferments capable of acting upon proteids, 
gelatine, sugars, etc., is supplied by the study of the 
growth of these microbes outside the body. No less 
than six such enzymes are said to be produced by the B. 
pyocyarieus, for it has been shown that dead cultures of 
this organism will liquefy gelatine, coagulate milk, and 
redissolve the coagulum, invert cane sugar, split up fats, 
and decompose proteids. That ferments also originate 


from the wandering cells has been demonstrated by 
Leber (34), who, placing pieces of copper in the anterior 
chamber of the eye, thereby produced a purulent collec- 
tion devoid of microbes, and showed that the exudate 
was capable of digesting proteid matter. In this the 
leucocytes do not differ from the other cells comprising 
the organism. There have of late years been abundant 
observations by Salkowski (146), Jacoby (147), Conradi 
(148), and many others, upon the phenomena of autolysis, 
or the self-digestion of liver, muscle, and in fact most 
tissues, in which it has been shown that without bacterial 
action, but by the action of their own juices, the cells 
are able to digest and disintegrate themselves. Only in 
the inflammatory exudate the conditions favouring such 
self -digestion would seem to be markedly augmented. 
Mliller (149) has shown that pus has a strong digestive 
action upon dead tissues, and to such autolytic action is 
now ascribed, on what appear to be convincing grounds, 
the resolution of the pneumonic exudate. Flexner has 
recently demonstrated that autolysis also occurs among 
bacteria. Cultures of the micrococcus of epidemic 
cerebro-spinal meningitis are very short lived. A smear 
from such cultures thirty hours old may show abundant 
typical diplococci ; the same cultures a few hours later 
may show not a single coccus. Dying they have digested 
themselves. Eegarding the ferments, Opie (150), setting 
up non-bacterial inflammation by injections of turpentine, 
has demonstrated the existence of two proteolytic enzymes 
present in greater or lesser amount, according to the 
nature of the predominating cells of the exudate ; the one 
first described by Friedrich Mtiller is peculiar to the 
polynuclear leucocytes, and is most efficient in a weakly 
alkaline medium. The second present in the mono- 
nuclear leucocytes digests in the presence of acid ; it 
resembles the so-called autolytic enzyme present in 


almost all the tissues of the body the enzyme which 
leads removed tissues to digest themselves in the absence 
of any bacteria of putrefaction. But at the same time 
he has determined that in the serum of the exudate there 
exists an " anti-enzyme," or body arresting the activity 
of these enzymes. The conclusion is that digestion of 
fibrin and of dead cells is not apt to occur until either 
the leucocytes are present in overwhelming amounts (as 
in thick pus), or where, with arrest of the inflammation, 
the fluid exudate, and with it the anti-enzyme, becomes 
drained off. He found, for example, that a single injec- 
tion of turpentine into the pleural cavity led to a fibrinous 
exudate ; repeated injections, by causing abundant poly- 
nuclear leucocytosis, converted the fibrinous into a purulent 
exudation, with evidence of -softening and disintegration 
of the fibrin already laid down ; as also that when he had 
established a serous inflammation, complete aspiration of 
the fluid was followed by the appearance of localised 
empyema or suppuration. It is evident that in the 
absence of the antibody of the serum the enzymes left 
adherent to the fibrin over the pleural surfaces cause 
active proteolysis, and that the products are irritative, 
attracting further leucocytes to the part. At autopsy I 
have been impressed at the frequent tendency to the 
formation of localised abscesses in cases of general peri- 
tonitis that have been operated on, ' the fluid being 
drained away. At operation these cases have been 
found to be of the serous type. Now they have become 
purulent. These observations of Opie upon the existence 
of a body in the serous exudate inhibiting the proteolytic 
activity of the leucocytic products suggest that it is not 
always advisable to perform complete paracentesis, i.e. 
complete removal of serous inflammatory fluids, however 
wise it may be to remove the denser purulent collections 
as fully as possible. 



It would seem, therefore, that, more especially in 
pyogenic inflammation, the removal of dead tissue cells 
and dead leucocytes may, to a large extent, be due to 
the action of the inflammatory exudate, apart from any 
phagocytic action on the part of living active cells ; 
although this also comes often into play. 

The bactericidal substances present in the inflamma- 
tory exudate have already been considered. We have 
abundant evidence that substances capable either of 
destroying microbes or of hindering their growth are 
present therein. 

Summary. To sum up what is known concerning the 
inflammatory exudate, it may be said 

1. That the exudate varies in amount and in character 
with (a) the nature and intensity of the irritant, (b) the 
condition of the organism, (c) the region of irritation. 

2. That while it undoubtedly augments the nutrition 
of the affected region, increased nutrition at the early 
stage of an acute inflammatory process would not seem 
to be of benefit or to play any important part. At a 
later stage and in chronic inflammation the increased 
nutrition possibly aids the hyperplasia. 

3. That in many cases the exudate exerts a beneficial 
action by flushing out the injured area. But this same 
flushing-out, by distributing the microbic irritants, may 
also be harmful to the economy. 

4. That the exudate plays an important part in 
diluting the irritant. 

5. That the development of fibrin in certain inflam- 
matory exudates is associated with the breaking down of 
the wandering cells, and is of manifest benefit in so far 
as it circumscribes the inflamed area, and prevents the 
passage of morbid material outwards. Here again the 
action, by favouring the development of organised 
adhesions, may be deleterious. 


6. That the exudate may possess digestive functions, 
causing the production of albumoses and other products 
of nitrogenous disintegration ; the ferments being 
developed from the cells alone when the exudate is 
aseptic, from these and the microbes together when 
pathogenic microbes are present. 

7. That the exudate may further contain substances, 
generated by the cells, capable of hindering bacterial 
growth, and of destroying pathogenic microbes. 



THE study of the action and function of the leucocytes in 
inflammation has profoundly modified our conception of 
the inflammatory process. When the leucocytes were 
regarded as purely passive agents, and their diapedesis as 
purely secondary to modified conditions of the blood- 
current and of the vascular walls, Cohnheim's hypothesis 
was that most generally accepted ; this hypothesis 
regarded the changes in the vessels as of the first 
importance. Thus it was that for several years our 
attention was mainly concentrated upon the determina- 
tion of the various changes of the vessel -walls, and of the 
mechanism whereby these changes were brought about. 
Nowadays less attention is directed to this side of the 
inflammatory process, and it may be said that during the 
last ten years little advance has been made in determin- 
ing the mechanism of the dilatation that accompanies 
inflammation. The subject, indeed, is beset with 
difficulties. It is most difficult to observe the changes 
that occur in the cells forming the endothelium of the 
congested vessels ; we are still, for instance, far from 
being * sure whether the opinion of Arnold (151) is 



correct, namely, that the leucocytes, and, it may be, a 
large portion of the exuded plasma, find their way out 
through the dilated stomata between the endothelial 
cells ; or whether the leucocytes pass directly through 
these cells as one soap bubble may be passed through 
another. And when we come to discuss whether the 
inflammatory exudation be a filtration, or whether, on 
the other hand, it be more of the nature of an excretion, 
or what may be termed a selective filtration certain 
components of the blood-plasma being permitted to pass 
through, while others are withheld we are met with 
the difficulty that, of the extravasated leucocytes, a 
varying proportion undergo rapid destruction and 
dissolution. Thus, in analysing the inflammatory serum, 
we are not dealing simply with the extravasated fluid, 
but with a fluid which in addition contains proteid and 
other constituents derived largely from broken-down 
white corpuscles, and in part, it may be, from the 
modified cells of the inflamed area. 

Though Arnold's observations upon the altered con- 
dition of the vascular endothelium in inflammation 
appear at first very convincing, upon further study they 
seem at most to indicate that with dilatation of the 
vessels there is an increase in the size of the spaces 
between the endothelial cells. They do not, however, 
prove that these are other than virtual spaces filled with 
intercellular substance ; and indeed Arnold himself came 
eventually to the conclusion that some such substance 
was present. That viscid, gelatinous substances injected 
into the circulation may be detected passing through 
these stigmata is not a proof that the spaces are actual ; 
all it proves is that the walls are weaker in these 
regions ; it must be remembered that increased force and 
increased intravascular pressure are necessary to promote 
the passage of the injected mass along the vessels.' The 


passage of the mass through the walls may therefore be 
an " artefact." 

Kolossow (152) has demonstrated that the endothelial 
cells of the intima of vessels are not absolutely inde- 
pendent units, and that they are connected one with the 
other by numerous fine brides or bridges of cytoplasm. 
Between these bridges are the stigmata ; stomata larger 
spaces are not normally present between capillary 
endothelial cells. He holds that, normally, the cuticular 
portions of the cells are in apposition, but that with 
distension the stigmata from being potential become 
actual spaces, through which the migration of leucocytes 
and the escape of fluid may take place. 

There is this further difficulty in the assumption that 
these are actual spaces that in acute inflammation the 
exuded fluid contains a smaller quantity of proteids than 
does the blood -plasma. It is true, no doubt, that the 
stigmata are so small they may possibly act like the 
pores of a filter, and consequently may not permit the 
free passage of certain constituents of blood-plasma. To 
enter into the large subject of the nature of lymph would 
be to pass too far afield ; I can here only say that taking 
into consideration the abundant evidence we possess of 
the activity of endothelial cells influenced also, it may 
be, by loyalty to my old master Heidenhain I have not 
become convinced by the brilliant researches of Starling 
that these cells have no selective activity, governing to 
some considerable extent the quality and the quantity of 
the exudate. 

We have not a little evidence that these cells play an 
important part in the vascular phenomena of inflammation. 
To their power of taking up microbes and acting as 
phagocytes I have already referred ; into their connexion 
with the slowing of the blood-stream I shall enter later. 
Here I would point out that microscopically these cells 


can be seen to alter during the inflammatory process ; 
they become enlarged and project into the lumen of the 
smaller vessels, and in my experience this enlargement 
affects not only the cell-bodies, but also the nuclei, which 
at the same time would seem to contain more chromatin 
and to stain more intensely. In cases of chronic 
inflammation the enlargement is followed by prolifera- 
tion, notably in the arterioles and capillaries, a process 
which may lead to the ultimate occlusion of these small 
vessels. And in acute inflammation mitosis is to be seen 
occurring in these endothelial cells at an earlier period 
than in the surrounding tissues. 

A further and very important process intimately 
connected with the proliferation of the endothelium of 
the capillaries is the formation of new vessels as the 
result of continued inflammation. It . is true that 
Eindfleisch (153) and others have described this as being 
brought about by vaso-formative cells situated externally 
to the vessels ; and that others have advanced so far as 
to suggest that there are cells in the newly-forming 
granulation-tissue which become hollowed out and gain 
attachment to the pre-existing capillaries in a manner 
wholly similar to that observable in the vascular zone of 
the embryo of the chick. 

The search for the earliest signs of new capillaries is 
a matter of some difficulty. I will not peremptorily 
state that Eindfleisch mistook an arrangement of cells 
not unfrequently seen in granulation-tissue for stages in 
the development of new vessels. My own observations 
coincide with those of Arnold, Ziegler (154), and of the 
majority of those who have more recently studied the 
question, and lead me to regard the formation of new 
capillaries as originating from the endothelium of the 
vascular loops already in existence. This, I hold, must 
now be regarded as settled. 


The first step in the process is often recognisable, in 
cases of pleurisy and pericarditis, in the projection of 
loops of pre-existing capillaries beyond the line which 
indicates where the serous endothelium used to be, and 
into the fibrinous clot now adherent to the sub- 
endothelial layer. Such loops are markedly distended, 
and " point," as it were, at right angles to the denuded 
surface. A similar pointing or giving way of the wall 
along the convex margin of the loop is also to be made 
out not unfrequently in newly-developed capillaries. In 
these there is not, as might be expected, a thinning of 
the endothelium along this outer margin, but certain of 
the cells on the contrary appear large and active. At 
times a small sharp protrusion of the vessel-wall can be 
detected in the region of pointing. This is best seen in 
the capillaries that are themselves but newly formed, and 
composed of nothing but a layer of young endothelial 
cells. In this layer the protrusion can be made out to be 
in direct continuity with the endothelial cells of the 
region. At first it is solid, but in the later stages it can 
be seen to be nucleated, and to be growing by pro- 
liferation of the endothelial cells which thus jut 
outwards. Even before any further change is noticeable 
in this projection from the capillary wall it may be seen 
to be united with a similar process originating from a 
neighbouring vascular loop. Finally, it would apppear 
that the joined processes become hollowed out, and thus 
are developed into fully formed capillary loops. It seems 
impossible to make precise observations on the phenomena 
of new vascular formation in its successive stages. I can 
but state that these appear to be the steps of the 
process. By what means the new vascular projections 
join together to form loops we are ignorant. Metchnikoff 
suggests that there must be an attraction between the 
neighbouring projections achemiotaxis leading them to 


come into apposition ; that chemiotaxis is a factor in the 
formation of new vessels has been indicated by Council- 
man (29), who has pointed out that where, in keratitis, 
the lesion is exactly central these new vessels .are seen 
advancing inwards all around the periphery ; where the 
lesion is eccentric these form only on the side nearest the 
lesion. That they do join is very clear to those who 
have studied granulation-tissue, or have observed the 
vascular network connecting the previously separated 
surfaces of a wound. 

A further function of the vessel- walls is to be seen in 
the slowing of the blood-current. It is difficult, and in 
fact impossible, to explain this slowing by altered 
diameter of the arteries and veins. The alterations 
observed in the diameters of the vessels of the inflamed 
area are such as, acting alone, would lead to increased 
rate of flow. Nor, again, is the apparent amount of 
exudation, and of lymph-flow from the affected part, 
sufficient to make it probable that (as Wharton Jones 
(155) first suggested) the slowing is in the main due to 
the concentration of the blood, relative drying of the 
corpuscles, and consequent increase of friction ; while 
this may be an adjuvant we must, I think, find some 
more potent factor. What this factor is was pointed out 
long ago by Lord Lister (156), who, in 1858, noticed 
that, coincident with the slowing of the blood-strearn, the 
corpuscles move sluggishly along the vessel-wall as though 
attracted by it. He essayed to prove this by an 
experiment performed previously by Weber (157). He 
ligatured a frog's leg, then irritated a portion of the web 
by a little mustard, and found that, although the blood- 
current had ceased, there was nevertheless an accumulation 
of corpuscles in the vessels of the irritated area, the 
corpuscles gliding into the affected region and becoming 
adherent there. Eyneck (158) has shown that this 


accumulation is not due to increased adhesiveness of the 

FIG. 16. Formation of new vessels in inflammatory tissue. (1) From a Ziegler's chamber 
(i.e. one formed of two cover-slips) left in the peritoneal cavity for forty-eight days. 
The tissue formed between the cover-slips consists of uninucleated and multi- 
nucleated formative cells. It is bounded by fully-formed new capillaries, and in 
the angle between these the solid buds or processes of developing new capillaries 
are well seen. (2) Formatis'e (connective tissue) cells in direct connexion with the 
endothelium of newly-formed capillaries. From a similar preparation. ZIEGLER. 

red corpuscles, inasmuch as similar slowing and stasis 



may be induced if the blood of the frog's leg be replaced 
by milk and the web irritated. In this case there is a 
gradual slowing of the stream of milk and accumulation 
of the fatty globules in the inflamed area. If, on the 
other hand, the vascular endothelium be killed by the 
action of circulating metallic poisons, then he found that 
no stasis occurred. And in favour of these views of Lord 
Lister and Eyneck is the fact already noted, that in 
inflammation the endothelium of the vessel-walls becomes 
altered, the cells becoming enlarged. With this, as 
evidenced by the conduct of the white corpuscles, they 
become more adhesive, and this adhesiveness with the 
associated increased friction between the vascular walls 
and contents I regard as the first factor in bringing about 
the slowing of the blood- stream. Let the current once 
accelerated be rendered slower by this increased friction, 
then transudation may accentuate the accumulation of 

Summary. While there is very much yet to be 
learned concerning the part played by the blood-vessels 
in inflammation, and while our present knowledge of this 
branch of the subject can only be regarded as very 
imperfect, the following may, I think, safely be said to 
epitomise what is known at the present time : 

1. That the vascular walls, and more especially the 
endothelial cells lining the capillaries, play an active and 
not a passive part in the inflamed area. 

2. These cells have the power of throwing out 
pseudopodia and of taking up non-motile bacteria. 

3. They are larger and more prominent during 
inflammation than they are under conditions of health. 

4. From them are developed the new vascular loops 
in cases of more chronic inflammation. 

5. Their inner surface would seem to become more 
adhesive in inflammation, and by this, in the first place, 


to lead to the arrest and adhesion of the leucocytes and 
red corpuscles. 

6. Similarly they would seem to cause an increased 
resistance to the passage of the blood-current, and thus 
tend to slow the rate of blood-flow. 

7. The slowing of the stream may further be aided 
by the passage through the walls of increased amounts 
of fluid from the blood. 

8. It is impossible by analysis of the inflammatory 
exudation to determine whether this be a mere filtrate 
or be the result of a selective activity of the endothelium. 
On the whole there appears to be a combination of the 
two processes. 

Other properties of the blood-vessels in respect of 
inflammation will be better discussed in later sections in 
connexion with the discussions on the part played by the 
nerves, and on the formation of new tissue. 



BY his researches, Cohuheim (1867) forcibly attracted 
the attention of pathologists to the diapedesis l of 
leucocytes in inflammation a process which had already 
been very clearly described years before by W. Addison 
(159) (1843) and Waller (160) (1846) in England; 
and yet earlier (though without grasp of the connexion 
between the diapedesis and inflammation) by Dutrochet 
(161) in France (1828). Cohnheim recognised the 
amoeboid nature of the leucocytes, and saw that once 
outside the vessels they moved actively, but eventually 
he could not discover that their penetration of the vessel- 
walls was anything but passive, although twenty years 
previously Augustus Waller had clearly described the 
active nature of the process ; and this failure on 
Cohnheim's part to recognise the true nature of diapedesis 
confirmed him yet more strongly in the view that the 
all-important factors in the inflammatory state were the 
changes in the vessel-walls, and, it may truly be said, 
arrested his advance towards a fuller comprehension of 
the subject. 

1 It has been objected that the term "diapedesis" should strictly be 
employed to denote only the passive transit of red corpuscles out of the 
vessels. If the word was originally employed in this sense it was an 
incorrect use : the term clearly indicates an active process a "footing 
through " or jumping through. 



It must be acknowledged that there is much which 
would seem to support this view of the passivity of the 
leucocytes. No one is prepaied to attribute active 
movements to the red corpuscles, nevertheless in inflam- 
mation a certain number of these escape through the 
vessel-walls. In the inflammation affecting some organs, 
notably the lungs, the number effecting a passage is very 
considerable. If, then, the red corpuscles emerge 
passively, why should not the emergence of the white be 
passive also ? Add to this the very important observa- 
tions made by Cohnheim, that where the circulation is 
arrested by compression of the artery there diapedesis 
ceases. This, if invariably true, would seem to indicate 
that when once by changes in the vessel the leucocytes 
adhere to the wall, the further passage through that wall 
is due to the vis a tergo of the blood-pressure. 

This, however, is not a safe deduction to draw from 
the experiment referred to. When the artery of an 
inflamed area is compressed the stoppage of the blood- 
stream not only reduces the pressure, but also affects the 
quality of the blood and the conditions of the vessel- 
walls ; moreover, it must profoundly affect the vitality or 
at least the activity of the contained leucocytes. These 
considerations alone render the experiment valueless as a 
proof of the passive nature of the diapedesis. Again the 
passage outwards of red corpuscles does not occur in the 
earliest stages of reaction to irritation ; it never precedes 
the diapedesis of the leucocytes (save where there is gross 
injury), but follows it. A capillary or small vein in the 
inflamed frog's web, for example, may be seen wholly 
filled with corpuscles, the peripheral zone being quite 
annihilated, and numerous red corpuscles lying in im- 
mediate contact with the walls ; nevertheless at first 
leucocytes only emigrate. This difference must be due 
to some special property of these cells. The leucocytes 


in the blood-stream are not necessarily globular passive 
agents, but are capable of independent movement. Leber 
(34), in his long series of studies, has pointed out that if, 
with due precautions, a hooked glass tube (closed at its 
outer end where it catches into the incision in the wall) 
be inserted into a large vein no thrombosis may be set up 
around the intravascular portion, and yet, upon removal, 
a large collection of leucocytes may be found in the tube, 
attracted by a .drop of mercury placed, along with normal 
salt solution, within it. (Mercury is a substance which 
within the tissues leads to an accumulation of leucocytes.) 
Here, then, there must be active attraction and active 
movement of the leucocytes within the blood - stream. 
And Lavdowsky (163) has described very exactly what 
other observers had also noted, namely, that in inflammation 
the leucocytes in the outer zone of the blood-stream do 
not simply adhere passively to the wall, but move back- 
wards and forwards before they attach themselves and 
emigrate, as though seeking for a point of less resistance. 
At times this movement is in a direction opposite to that 
of the blood-current. Further, Councilman has called 
attention to the suggestive fact that in the process of 
migration the nucleus is always directed to the objective 
point and, with a small surrounding of cytoplasm, is 
the first part of the cell to pass through the capillary 
wall. More than one observer has seen a relationship 
between the labile, broken-up character of the nucleus 
of polymorphonuclear leucocytes and their function of 
passage through minute orifices in the capillary walls. 

If both within and without the vessels the leucocytes 
can be actively amoeboid, it is strange that they should be 
passive in the process of diapedesis which to the eye has 
so characteristically amoeboid an appearance. 

As above stated, the compression of the artery passing 
to an inflamed area is in most cases sufficient to arrest 


diapedesis in that area, and I have suggested that this 
arrest may be due to the altered environment of the 
leucocytes. Now, if an embryonic form be taken, in which 
the tissues would seem to possess greater inherent vitality 
coupled with less sensibility, the arrest does not necessarily 
occur. Thus, Metchnikoff has noted that diapedesis of 
the leucocytes can be followed in the tadpole's tail, after 
the animal has been curarised to such an extent that the 
heart has ceased to beat and the blood in the capillaries 
has been brought to a standstill. 

It* is evident, therefore, that with our present know- 
ledge we must regard the diapedesis of the leucocytes as 
an active migration, and must look upon the blood- 
pressure, the disposition of the blood-stream, and the 
altered condition of the endothelium of the dilated vessels 
as adjuvants in the process. The slowing of the blood- 
stream and fhe diminished pressure in the inflamed 
capillaries render it more easy for the leucocytes to 
accumulate close to the vessel-wall ; the dilatation of the 
vessels and consequent thinning of the walls, with the 
opening, perhaps, of larger spaces of cement substance or 
stigmata between the individual endothelial cells, render 
it more easy for the leucocytes to accomplish the passage ; 
but the movement from within the capillaries to the 
tissue- spaces outside is an active process due to amoeboid 
movement of the leucocytes themselves. The continuity 
of the vessel -wall once destroyed, other cells red 
corpuscles may be pressed passively through the walls. 

If this view be accepted, we are bound to look beyond 
Cohnheim's limit of changes in the vessel- wall for the 
stimulus which, originating in the area of irritation, acts 
upon the vessel-wall arid the leucocytes in contact with 
it, and, having first set up changes in the former, so 
reacts upon the latter that they emigrate ; or, to put it 
in other words, are attracted out of the capillaries 




towards the focus of irritation. It has already been 
shown that the movement of wandering cells in the 
tissue is due to the attraction of a diffusible product of 

FIG. 17. 1. Capillary (inflamed) of frog seen in profile, exhibiting margination of 
leucocytes, assumption of pear-shaped form, and migration through wall ; a leuco- 
cyte adherent by long process. 2. A leucocyte in process of diapedesis, showing 
the pseudopodia in the outer aspect of capillary. From mesentery of rabbit ; also 
in profile : higher magnification. LAVDOWSKY. 

bacterial growth and of tissue change, and of sundry 
organic and inorganic materials the force to which 
the name of positive chemiotaxis has been given. 
Chemiotaxis must be invoked to explain the active 
emigration of the leucocytes from the capillaries, and 


again to explain its cessation under other conditions. 
Thus, while the exposed mesentery of a frog is a tissue 
in which diapedesis can be observed with facility under 
ordinary conditions, if it be washed with a weak solution 
of quinine, the leucocytes in the vessels remain globular! 
cease to adhere to the walls, and do not emigrate. This! ' 
observation, first noted by Binz (164), has been confirmed! 
by others, among whom Disselhorst (165) made oul 
also that, if these same leucocytes be removed from thj 
vessels, they exhibit their usual amoeboid movements. 
The quinine has not paralysed them, as Binz supposed ; 
but, as Metchnikoff pointed out, it has neutralised the 
previous positive attraction, a negative or repulsive 
chemiotaxis being brought into play. It is difficult to 
see how these observations can be otherwise explained. 

The view that diapedesis is an active process gains 
further support from, and at the same time explains 
certain interesting observations made by Bouchard (166), 
Eoger (167), Charrin (168), and Buffer. These observers 
have independently shown that in sundry instances the 
results of local injection of virulent cultures are greatly 
modified if, shortly before or coincidently, the microbes 
and their products are introduced into the circulation. 
Thus, as Buffer points out (169), a drop of the culture 
of the bacillus pyocyaneus inoculated into the anterior 
chamber of the rabbit's eye leads ordinarily to a great 
migration of leucocytes to an acute purulent inflamma- 
tion. -If, however^ the toxins produced by this microbe 
have previously been injected into the circulating blood, 
no accumulation of leucocytes follows inoculation into 
the eye. Buffer also extended most suggestively certain 
observations of Boger(167). Subcutaneous or intramus- 
cular inoculation of the rabbit with the bacillus of 
symptomatic anthrax leads to the production of a local 
abscess with extensive accumulation of leucocytes. After 


simultaneous injections of fluid containing virulent bacilli 
and their products into the vein of the ear and the 
muscles of the hind leg, Buffer found the rabbit dead 
within fifteen hours, with a huge tumour in the inocu- 
lated limb. Here, upon examination, the muscle-fibres 
were found widely separated by fluid exudate, in which 
there had been great multiplication of the bacilli ; but 
leucocytes were entirely absent. In these two cases we 
havajjlfc^ore diapedesis and determination of leucocytes 
follej Hie purely local action of the toxin ; want of 
diapSSJ^and absence of leucocytes when the toxin at 
the same time circulates in the blood-stream. Wholly 
in line with these are the observations of Sidler (1*70). 
Solution of iodine, injected subcutaneously into the ear 
of a rabbit, sets up an inflammation characterised by 
extensive exudation with associated abundant migration 
of leucocytes into the part. With the injection of 0*05 
per cent iodine solution peripherally into the ear vein 
there is the same or even more extensive fluid exudate, 
but no leucocytosis. If any large proportion of the 
leucocytes which find their way to a focus of irritation 
emerge from the blood-stream, these divergent results 
are only to be explained by some hypothesis which is 
capable of reconciling the difference in the action of the 
leucocytes when they are circulating in normal and in 
toxin- or irritant-containing blood respectively. 

Now, the results in these cases are entirely consonant 
with what we know concerning the sensitiveness and 
reaction to stimuli not only of unicellular organisms, but 
also of the higher animals. Organisms, whether lowly 
or of most complex development, only perceive and 
react to alteration in their environment when the altera- 
tion exceeds a definite ratio. Thus, as Pfeiffer has 
pointed out, a motile bacterium (the " B. termo ") is 
attracted towards solutions of peptone ; if it be already 



in a peptone solution, in order for it to be attracted 
towards and move into a more concentrated solution, this 
last must be five times as strong as is the former. This 
is in conformity with the psychophysical law of Weber- 
Fechner : that sensibility increases in arithmetical ratio 
when the stimulus or excitation increases in geometrical 
ratio or, in other words, reaction is in proportion to the 
logarithm of the excitation. The only possible explana- 
tion that I can see of the above observations of Buffer, 
Eoger, and Sidler is that the passage and want of 

I passage of the leucocytes out of the vessels depends upon 
the ratio of diffusible bacterial products present in the 
blood-stream and in the tissues respectively. Where the 
products are localised at one focus in the tissues, the 
leucocytes are attracted out of the unaltered blood, and 
there is active diapedesis ; where there was already a 
solution of the bacterial products in the blood, the ratio 
of difference between the percentage amount of toxin in 
blood and tissue may be insufficient to stimulate the 
leucocytes ; no diapedesis then ensues. 

As is well shown in the experiment with symptomatic 
anthrax, the presence of the bacillus and its products in 
the circulating blood did not prevent inflammation at the 
region of local injection ; inflammation and exudation 
were abundantly manifest there was, in fact, a more 
extensive exudation than ever. The irritant that is to 
say, the toxic products of the bacilli at the point of 
injection was in no wise hindered from exerting effects 
upon the fixed cells of the vessel- walls, and promoting 
all the changes in calibre and condition of the walls and 
in the blood-stream characteristic of inflammation. But 
with vascular changes, if anything more prominent than 
in the case where local inoculation alone had been practised, 
the leucocytes stayed within the vessels. Now the only 
cause to which we can attribute this abstention of the 


cells from emigration is lack of attraction certainly not 
lack of vascular change or lack of blood-pressure. 

Summary. We are thus led to the following con- 
clusions regarding the passage of cells out of the blood- 
stream into an inflamed area : 

1. The diapedesis of the leucocytes is, as the name 
implies, an active and not a passive process ; it is due to 
active amoeboid movements on the part of the cells. 

2. The stimulus leading to diapedesis is that of positive 
chemiotaxis. It is the attraction exerted upon the 
leucocytes by the diffusible substances associated with 
the irritant. 

3. Irritants, if themselves diffusible, or the diffusible 
substances developed while the irritants are within the 
tissues, are capable of two separate actions : one direct 
upon the vessel-walls, leading to vascular changes ; the 
other through the walls upon the leucocytes, whereby 

emigration may be induced/ 

4. These two actiomffieed not (and frequently do ,not) 
manifest themselves pari passu. t^AZ^. &fy**<0& 

5. In relation to diapedesis, the dilatation of the 
vessels, the altered rate of blood -stream, the altered 
disposal of the corpuscles in the stream, and the modified 
endothelium, may all be regarded as adjuvants. 

6. The passage of red blood corpuscles from the blood- 
vessels into the inflamed area is passive, due to the blood- 
pressure and to lack of continuity of the vessel-walls. 
Such lack of continuity is favoured in many instances by 
the migration of the leucocytes through the walls. 




IF the vascular changes in inflammation were due to 
reflex influences proceeding from the central nervous 
system, and were in fact controlled by the centres in the 
brain and spinal cord (as has been held by the supporters 
of neuro-humoral hypotheses), then, in the first place, 
there should be a rapid -and almost immediate response 
on the part of the vessels of any region on the introduction 
of an irritant. But this is not by any means constantly 
observed. Thus, as Cohnheirn pointed out, if croton oil 
be rubbed upon a rabbit's ear, more than an hour may 
elapse before the first beginnings of hypersemia can be 
detected ; yet the inflammation eventually set up may 
be very intense. In the second place, section of all the 
nerves passing to any region of the body should have this 
effect, that injury in the region in question should be 
unaccompanied by the ordinary vascular reaction. But 
this is not the case. Divide all the nerves which supply 
a rabbit's ear, for example, and then injure that ear, either 
by heat, cold, or inoculation of pathogenic micro-organ- 
isms, and inflammation manifests itself with all the stages 
recognisable in an ear with intact nerve -supply. The 
vascular changes which accompany inflammation can 



occur, then, independently of any central nervous in- 

We can proceed further, and state that regions deprived 
of their nerve-supply are peculiarly prone to inflammatory 
changes. But this liability to inflammatory disturbances 
in such regions is not directly due to the destruction of 
vasomotor tracts and the cutting off of central influences 
from the vessels of the part, but is, it would seem, 
immediately connected with the loss of sensation. Divide 
the ocular branch of the fifth nerve of a rabbit, and, if 
the eye be not protected, ulceration and necrosis of the 
cornea manifest themselves in the course of a few days. 
Protect the eye, either by bringing the lids together or 
by placing a shade over it in such a way that dust and 
foreign particles are prevented from settling upon the 
surface, and no such ulcerative disturbance manifests 
itself. From this it is clear that the primary cause of 
the inflammation is not any trophic change in the region, 
but is the lack of sensation, whereby irritant substances 
are permitted to gain a lodgment upon the outer surface 
without any attempt being made to remove them. That, 
in addition, there is a lowered vitality in parts deprived 
of their nerve-supply, and that this renders those parts a 
more favourable seat for inflammatory disturbances is more 
than probable ; nevertheless, this would not seem to be the 
primary cause of the increased liability to inflammation. 

This, then, in the first place, is clearly recognisable 
that the vascular changes accompanying inflammation 
can occur independently of central nervous influences. 
Hence it follows that there must be a peripheral mechan- 
ism controlling the vessels. It remains, therefore, to 
determine the nature of this peripheral mechanism : is 
it wholly under the guidance of peripheral nerve-cells 
situated in the vessel-walls, or is it, in part at least, 
idiocellular ? In the present state of our knowledge the 


answer to this question must be guarded. The more 
carefully the in nervation of the various regions is studied, 
the more clearly is it demonstrated that throughout all 
the tissues of the body there exists a wonderfully fine 
and complicated net-work of nerve-filaments with occa- 
sional isolated ganglion-cells. Yet proof is wanting that 
this system in connexion with the vessels is sensorimotor. 
Indeed, so far as regards the heart and ventricular muscle 
(which may be looked upon as the region of the vascular 
system wherein the motile portion of the walls has 
become specially developed), the researches of Eomberg 
and His lead rather to the conclusion that the peripheral 
nervous system subserves sensation alone. Nevertheless, 
there are observations to the contrary. Mall (171), for 
instance, has demonstrated the existence of motor nerves 
in the portal vein, finding that after ligature of the 
thoracic aorta stimulation of the splanchnics causes the 
portal vein to contract until the lumen almost disappears. 
The portal vein, it is true, has functions differing from 
those of most other veins ; but this observation renders 
it possible that some veins, at least, may be modified by 
central stimuli in the course of the inflammatory process. 
At the same time, the more the activity of the various 
tissues is studied, the more fully it is seen that many 
cells retain what may be termed reminiscences of an 
earlier and more embryonic condition in which their 
functions were varied and less specialised. There is an 
inherent probability that the endothelial cells can react 
directly to stimuli, and that they are capable of idiopathic 
contraction and expansion on appropriate stimuli. We 
have seen that these cells are capable of taking up 
microbes, and thus obviously exhibit an independent 
activity similar to that observed in the amoeba or the 
wandering phagocyte. If these cells, then, are capable 
of throwing out pseudopodia, and thus of enclosing non- 


motile bacteria, are they not capable of contracting and 
expanding, as a whole, according to the stimulus of 
altered environment ? As a matter of fact, such con- 
tractility of the endothelial walls of the capillaries has 
been demonstrated by Klebs (172) and Severini (173). 
I cannot but conclude, then, that at least the endotheliuin 
of the capillaries is to some extent self-regulative or 
neuro-muscular. It is quite possible that the muscular 
coats of the smaller arteries are likewise capable of self- 
regulation, and respond directly to stimuli. In their 
studies on the sympathetic inner vation of the rabbit's 
ear, Dr. and Miss Meltzer (174) have demonstrated 
that after section of all the sympathetic fibres to 
the ear, the resultant hypersemia disappears in some 
days, the vessels contracting, and this before any signs 
of nerve-regeneration can be detected. In other words, 
the arteries regain their tonicity without the aid of 
extrinsic nervous influence. That inflammatory dilatation 
of the vessels differs from simple vasomotor dilatation 
they demonstrate by the action of adrenalin : this is 
followed by rapid contraction in the latter condition, 
but is without effect in the former. The continuance of 
pulsation in the veins of the bat's wing when the nerves 
have been cut through, suggests either that the muscle 
of the walls has independent action or the existence of 
local ganglionic centres. 

There is, indeed, something characteristic about the 
inflammatory dilatation of vessels something not 
observed in other conditions in active arterial hyper- 
semia or in venous obstruction. As Samuel (131) points 
out, if intense venous obstruction of the rabbit's ear be 
produced, the larger vessels are seen greatly engorged ; 
but puncture of the skin outside the visible vessels, or 
even of the ear with a fine needle in such a position, is 
not followed by the escape of even a drop of blood 


the capillaries remaining contracted. Puncture outside 
the visible vessels in the inflamed ear and the outpouring 
of blood is quite extensive. There is, in short, an active 
dilatation of the capillaries ; indeed, contrary to the usual 
teaching, I am inclined to the belief that the hyperaemia 
of inflammation is not a true, active, or arterial hyper- 
asmia ; still less is it a passive or venous hypenemia ; 
it is, I hold, sui generis, and best described as a capillary 

This view that the vascular phenomena of inflam- 
mation can occur independently of the central nervous 
system and of the peripheral nerves does not imply 
that the nervous system, central and peripheral, is 
without its influence upon the process; far from it. 
We have evidence, in the first place, that the state of 
the vascular walls is modified after destruction or sever- 
ance of the nerves. I do not here refer only to the 
consequent alterations in calibre of the vessels, but also 
to the changes in other properties. Thus Gergens 
(175), and to a less extent Riitimeyer (1*76), noticed 
that after destruction of the spinal cord the blood- 
vessels of the frog permit a larger quantity of fluid, 
and even particles of granular colouring matter, to 
permeate them. 

In the second place, we have evidence that the central 
nervous system exercises some direct influence upon the 
inflammatory process. From Cohnheim onwards it has 
been a matter of common observation that when all the 
nerves of a part have been severed, the stages of the 
process succeed each other with greater rapidity. It 
may be that the modified state of the capillary walls, 
noted in the preceding paragraph, is capable of account- 
ing for this fact, and that, in the absence of central 
influences, dilatation of the vessels and exudation of 
fluid lead to the cardinal symptoms of redness and 


swelling, with associated changes in the tissue, at an 
earlier period. 

Of the part played by the different sets of nerves the 
external ear of the rabbit again furnishes an excellent 
study. This part has a double nerve-supply through 
the auriculars (major and minor) from the cervical 
plexus and the sympathetic branches from the superior 
cervical ganglion : stimulation of the former leads to 
dilatation of the ear vessels (vaso- dilator action), of the 
latter to contraction of the same (vaso-constrictor). If, 
as shown by Samuel (177), the sympathetics be divided 
on the one side, and the auricular branches upon the 
other, the ear vessels of the former side become widely 
dilated (unimpeded action of vaso-dilators), and those of 
the latter markedly constricted. Under these conditions, 
if both ears be subjected to the action of water warmed 
to 54 C. there is a characteristic difference in their 
reaction. In the organ deprived of sympathetic influence 
the congestion and hypersemia become yet more pro- 
nounced : an acute inflammation sets in which proceeds 
rapidly to recovery. In the opposite ear, with its con- 
stricted vessels, no hyperaemia is set up ; but there is 
stasis, and gangrene may supervene. These results have 
been confirmed by Roger (178), who, taking a rabbit and 
dividing the sympathetic on one side, and then inoculating 
both ears with like quantities of a culture of the strepto- 
coccus of erysipelas, found that the erysipelatous process 
manifested itself much more promptly upon the paralysed 
side, and came to an end at an earlier date. The reverse 
was the case when the auriculars of the one side had 
been divided : here the process was of slower develop- 
ment than on the intact side, and of slower course, 
resulting in mutilation of the organ. Further confirma- 
tion is afforded by the researches of Meltzer and Meltzer 


The inference to be drawn from these observations is 
that section of all the nerves passing to the rabbit's ear 
permits the inflammatory process to run a more rapid 
course ; section of the sympathetics (vaso-constrictors) 
alone has the same effect ; while the uncontrolled action 
of the sympathetics after section of the auriculars (vaso- 
dilators) hinders or prevents the manifestation of the 
ordinary processes of inflammation, and by preventing 
the destruction or removal of irritant matter favours 
necrosis of the tissues. We have yet to learn whether 
these results are capable of a general application, and to 
discover how far they are borne out by clinical observa- 
tions on diverse cases of localised paralysis. So far as 
they go they afford direct evidence of the power of the 
central nervous system to modify the course of the 
inflammatory process, while they demonstrate admirably 
how potent an auxiliary is the dilatation of the vessels in 
the inflammatory process. 

Other evidence that the state of the nerve-supply of 
a region influences the manifestation of inflammation is 
afforded in sundry neuropathies. In all of these, in the 
present state of our knowledge, it is difficult to trace out 
the nervous factors associated with the lesions to which I 
refer. Our knowledge of the respective influences of 
trophic and vasomotor nerves is far too limited to permit 
us to say more than that a relation exists between the 
condition of the nerve-supply of the affected area and the 
inflammatory lesions there observable ; that in a certain 
number of cases inflammation affecting the area supplied 
by one branch of a nerve may have associated with it 
definite inflammatory disturbances in the areas supplied by 
other branches of the same nerve, and that, similarly, when 
inflammation affects a viscus, inflammatory phenomena may 
be sympathetically developed in regions innervated from 
the same area in the brain or spinal cord. I have already 


given examples in support of the first statement : the 
familiar redness, swelling, heat, and pain of the side of 
the face which may accompany toothache is an example 
in support of the second. Head and Campbell's thorough 
studies upon Herpes zoster (179) show that lesions 
affecting the nerve centres in this case the posterior 
root-ganglia may be the main factor in the development 
of an intense inflammatory process in the areas governed 
by those centres a process so far unassociated with the 
presence of bacteria, although the appearances suggest 
infection. Another example is to be found in the acute 
nephritis which at times rapidly follows the passage of a 
catheter or the impaction of a stone in the urethra. It 
is not unlikely that many of these sympathetic inflamma- 
tions are not direct, but secondary. Thus, the first 
noticeable symptom of catheter fever is suppression of the 
urine. Such suppression might be brought about either 
by reflex contraction of the renal arteries, or, contrari- 
wise, by reflex great dilatation and congestion of the 
vessels of the kidneys. If it be caused by the former, 
then the nephritis can only be regarded as secondary, 
and as due to the injury done to the organ by the stoppage 
of its blood-supply for some little time. Undoubtedly in 
many cases of catheter fever the nephritis is infective, 
but in some the condition supervenes so rapidly that it is 
difficult to believe that we have to deal with an ascending 
infection. When there is infection it develops in such 
a way that .we are led to see that the altered condition of 
the organ under various influences has favoured the 
inflammatory process. 

From the multitude of the factors involved, these 
examples, taken separately, afford at most only a great 
probability that the nervous system can directly originate 
inflammatory changes. There is, however, the clearest 
proof that the nervous system does possess this power, 


and this is afforded by the results of certain observations 
upon hypnotic effects. In some persons susceptible 
to hypnotic suggestion, the suggestion that a red-hot 
substance has been placed upon the hand will, in the 
course of a few minutes, lead to great reddening of the 
part supposed to have been burned, and this reddening 
may be followed by great local exudation and swelling 
in fact, by all the symptoms of acute inflammation, save 
this, that the exudation is simply serous, not passing on 
to anything approaching pus formation, i.e. the migration 
of leucocytes would seem to be wanting. Here, then, 
actual inflammatory reaction follows supposed injury. 
Characteristically there is a tendency for the lesion to be 
bilateral, " reflected " upon the corresponding region of 
the other side of the body. . 

It is unnecessary to do more than point out the light 
that this intervention of the central nervous system 
throws upon the subject of counter-irritation, and upon 
the modifications of the course of inflammations brought 
about by idiosyncrasy of the individual. 

From what has been said in the preceding paragraphs 
it follows that : 

1. Acute inflammation in all its stages may proceed 
regularly in the absence of all centrifugal nervous 

2. The vessels of an injured area are capable of 
reacting apart from central influences ; it may be either 
directly, or under the control of a peripheral system of 

3. The central nervous system is capable of modifying 
the process of inflammation. It would appear that 
when the vaso-dilators alone are called into action the 
successive stages of the process are accelerated. When 
the vaso-constrictors alone are acting the process is 


4. Centrifugal impulses alone, apart from any local 
injury, may originate a succession of phenomena of 
inflammation in a part. 

5. Hence, in all probability, a nervous and central 
origin must be ascribed to some, at least, of the 
sympathetic inflammations seen to occur in areas 
supplied by the other branches of a nerve supplying 
a part primarily inflamed ; and again in areas supplied 
from the same region of the brain or cord as the 
inflamed organ. 




As a consequence of irritation two opposed processes may 
be manifested in the cells of the affected area, changes 
leading to impairment and death, and changes leading 
to overgrowth and proliferation ; degeneration and 

Either of these two processes may, it is true, be 
wholly wanting. In very acute suppurative disturbances, 
destruction of tissue -cells, and the steps leading to 
destruction may be the only recognisable changes. 
Again, in the first stage of most injuries, whether of 
mechanical, chemical, or bacterial nature, degenerative 
changes are wont to take the lead. On the other hand, 
there are irritants so mild that little or no cell-destruc- 
tion results from their action ; an extreme example of 
this category of inflammations is seen in those epithelial 
overgrowths commonly known as " corns," due, as Sir 
James Paget pointed out in his lectures, -to intermittent 
pressure, and irritation of moderate intensity. 1 Other 

1 It may very well be that this is not an extreme example. Certain 
neoplasms, or tumours proper, not improbably develop as a conse- 
quence of some irritation having an intensity just sufficient to induce 



examples are to be found in the " catarrhal " inflamma- 
tions, in which there is marked initial overgrowth and 
proliferation of the cells of mucous membranes ; and in 
tuberculosis, again, in which characteristically the earliest 
effects upon the pre-existing cells, produced by the 
presence and growth of the tubercle bacilli, are those of 
enlargement and multiplication necrotic changes, as a 
rule, only appearing at a much later stage. Once more, 
in the later healing stages of injuries, cell-proliferation 
may be in the field alone. Nevertheless, in a very great 
number, if not in the majority, of inflammations, the two 
processes may be found occurring together destruction 
and degeneration being in evidence at the focus of 
irritation, and growth and proliferation towards the 
boundary zone, where the irritant is acting in a less 
concentrated form. 

Although the two processes are thus so frequently 
associated, it will be well, for the orderly review of our 
subject, to consider them separately. 

Degenerative Changes. Death of the pre-existing 
cells as an immediate consequence of injury cannot be 
regarded as one of the phenomena of the inflammatory 
process. Immediate death of the cells may be a result 
of injury, and the disintegration of the dead cells may in 
itself lead the way to all the symptoms of inflammation. 

cell-proliferation, and continued for a time sufficiently long to impress 
upon the cells of the affected tissues the habit of rapid multiplication. 
There is evidence both in animal and vegetable pathology favouring this 
relationship between inflammation and neoplastic growth. The objection 
may be raised, with considerable force, that substances which lead to cell- 
proliferation are stimuli and not irritants, and that a line should be 
drawn between inflammation proper and overgrowth the result of irrita- 
tion. I, for one, would willingly make this difference, but while it is 
easy to draw the line in certain well-marked examples, in others, as I 
shall proceed to show, cellular proliferation is so essential a part of the 
whole inflammatory process that the division becomes impossible. 


But cessation of action is not reaction, nor is failure 
response, and throughout this study inflammation has 
been considered as the reaction following injury, and the 
response to it. Thus immediate death of tissue-cells is 
resultant and not reactive, and may be eliminated from 
the category of the essential phenomena of inflammation. 
The same is to some extent true of cell-degeneration, 
but not entirely. While it is impossible now to accept 
Virchow's view, that inflammation is essentially a process 
characterised by increased nutritive changes in the cells 
of the tissues (180), it remains most probable that in 
very many cases irritation induces increased, even if 
perverted, activity of certain orders of cells. The pro- 
liferation, swelling, and more or less rapid degeneration 
of these cells cannot be wholly ascribed to the toxic 
influence of the irritant, but must in part be regarded as 
a result of over-stimulation and overwork. This is most 
noticeable in connexion with catarrhal and paren- 
chymatous inflammations. In parenchymatous nephritis, 
for example, such as that set up by cantharidin or 
infection, the cells especially affected are those whose 
functions are especially excretory ; and their degeneration 
would appear to be intimately related to the performance 
of their functions. Such degeneration, preceded or 
accompanied, as it so frequently is, by excessive prolifera- 
tion, may truly be regarded as reactive, and not as wholly 
and primarily destructive. 

Of the degenerations which affect the tissue-cells (and 
often at the same time the leucocytes) in inflammation 
there are many varieties ; in fact, according to the nature 
of the irritant, one, or other, or all the degenerations 
affecting the tissues in different pathological conditions 
may manifest themselves, save, perhaps, simple atrophy 
and pigmental degeneration (as apart from pigmental 
infiltration). Most commonly recognised are cloudy 


and fatty changes, but mucoid and hydropic changes are 
far more frequent than is generally noted. Even so 
specialised a change as amyloid degeneration has been 
observed occurring locally in chronic inflammations as, 
for example, in gummas ; while in these same chronic 
lesions hyaline degeneration in the vessel- walls is very 

There is a further form of degeneration met with in 
inflammatory disturbances which deserves more recogni- 
tion than it has generally received. This is what may 
be termed reversionary metamorphosis. To speak of it 
as reversionary degeneration is, perhaps, a misnomer, 
although, if the irritation be continued, the modified 
cells undergo atrophy and destruction. It is not seen 
in acute inflammations, but in more chronic cases. As 
I have pointed out elsewhere, functional activity and 
active growth and multiplication of the cells of the 
organism are largely incompatible. The performance 
of function necessitates katabolism and breaking down ; 
growth demands a building-up and increase of the 
cellular bioplasm. Thus, in the course of active cell- 
multiplication, the individual units lose, to a certain 
extent, the finely differential histological features associ- 
ated with the performance of special function, and revert 
to a simpler, more embryonic type. All inflammatory 
new growth, when active, presents, indeed, a reversionary 
character, the cells assuming appearances resembling 
those seen in the process of foetal development. Even 
in the simplest of all tissues, namely, white fibrous 
connective tissue, this is to be observed. But, in some 
instances, it is peculiarly well marked. Muscle-fibres, 
for example, are developed from sarcoblasts large cells, 
without a sign of striation but with abundant cytoplasm, 
tending to become multinucleated. In subacute inflam- 
mations affecting a muscle, the striation disappears, the 



nuclei multiply, and, from the common multinucleated 
mass, there may be budded off or separated, isolated 
cells resembling hyaline leucocytes with abundant 
cytoplasm ; resembling, in fact, the embryonic sarcoblasts. 
The embryonic liver shows no differentiation between 
liver-cells and bile-duct ; both are developed from in- 
different strands of cells. In the process of development 
those destined to be liver-cells enlarge, become polygonal, 
and gain abundant cytoplasm ; those destined to become 
bile-ducts remain small, multiply, and become arranged 
as an epithelium. Now, in chronic interstitial hepatitis 
(portal cirrhosis) of the progressive type, we at times 
surely recognise at the outer border of the affected 
lobules that the liver-cells become smaller and smaller. 
Their nuclei also are small, and they have little surround- 
ing cytoplasm ; they form into little worm-like strands 
of cells isolated by the newly formed connective tissue ; 
here and there such strands can. be followed in direct 
continuity with the more normal liver-cells of the lobule. 
These are not true bile-ducts, as they are sometimes 
mistakenly called : the cells are irregularly arranged and 
do not form a true epithelium ; they are more of a 
reversion to the indifferent stage. 1 

All these degenerations are inevitably associated 
with disturbances of the functions of the affected cells, 
and lead to their death if the irritation which has 
induced them be continued. But death is not the 
final stage to be considered. The ultimate fate of 
the necrosed cells varies according to the situation of the 
inflamed area, the intensity of the irritation, and the 

1 While these appearances can be made out in progressive cirrhosis, it 
should be noted that, in cirrhotic livers showing reactive hypertrophy, 
another condition giving somewhat the same general appearance is to be 
detected, namely, active budding and new formation of liver parenchyma 
proceeding from the bile-ducts (181). This is a process of another order, 
one of ascent, not of descent. 


specific character of the irritant. From a free surface 
the dead material may be freely cast off. In acute 
suppurative inflammations, whether superficial or deep 
and, in general, wherever there is an abundant deter- 
mination of leucocytes, there obtains a digestion and 
solution of the necrosed cells ; and, as I have already 
pointed out, this is associated with the liberation of 
many ferments, the development of peptones, albumoses, 
fatty acids, and other bodies, and is brought about 
largely through the extracellular action of the leucocytes 
and processes of autolysis. When there is a large area 
of cell-destruction, with well-developed encystment and 
limitation of necrosis by granulation-tissue, the solution 
of the dead material and subsequent absorption may be 
incomplete, and a fatty debris left behind, which may 
eventually become infiltrated with lime salts (calcareous 
infiltration, in the production of which, as Klotz (134) 
has recently demonstrated, the conversion of fatty acids 
into soluble soaps plays an important part). In tuber- 
culosis, despite the presence of many leucocytes in the 
immediate vicinity, the dead material of the centre of 
the tubercle undergoes very little absorption, but. remains 
as an inspissated, cheesy, and, later, calcareous mass. 
In syphilis, on the other hand, in large gummas, while 
there is similar death of the central cells and absence of 
removal, fatty metamorphosis does not occur nearly to 
the same extent, nor is there the same tendency towards 

Lastly, although very little is known about the 
subject, it must be pointed out that along with the 
tissue-cells the intercellular matrix undergoes modifica- 
tions or degenerative changes during inflammation. 
Among these, in all probability, is to be classed an 
increase in the amount of intercellular mucin, a mucoid 
degeneration. The inflammatory exudate is in. many 


cases rich in mucin, and although our knowledge of the 
changes in the matrix is scanty, the fact that the tissue- 
cells in general show little evidence of storage of mucoid 
or mucinogenous material, renders it probable that what 
mucin is formed is either excreted or elaborated between 
the cells. Connective-tissue fibrils, as part of the matrix, 
become swollen and fused into translucent b^nds or masses, 
in which the individual fibrillse are no longer distin- 
guishable : in acute inflammation the next stage is the 
dissolution and disappearance of this collagenous matter. 
In chronic disturbances they are especially prone to 
hyaline change. 



IN the lower animals, as we know, injury and actual 
removal even of a large portion of the body may be 
followed by the complete reproduction of the lost part. 
In man, however, this reproduction of lost tissue is 
reduced to its lowest point ; the higher the tissue the 
less, and the less perfect, the reproduction. Speaking 
generally, the tissues which show the greatest potentiality 
for reproduction are the least highly organised those 
composed of similar units. The " connective tissue " 
the lowest and most widely distributed retains the 
largest powers of proliferation and hyperplasia. 

In ordinary inflammation, hypertrophy and hyper- 
plasia l of the connective-tissue cells are absent at the 
focus of irritation. Here degeneration is predominant. 
It is in the peripheral zone, away from the maximum 
concentration of the irritant, that (as shown in case after 
case of Leber's long series of studies upon injury to the 
cornea) the connective-tissue cells show signs of enlarge- 
ment and proliferation, that they become more swollen 
and prominent, send out large processes, and may exhibit 
signs of active mitosis. Eegarding the mode of multi- 

1 By hypertrophy in the strict pathological sense is indicated increase 
in size of the individual elements of a tissue, by hyperplasia increase in 
the number of these elements. 



plication of these cells, there is now a general consensus 
of opinion that multiplication by direct division (ami- 
tosis) is apt to show itself first even within a few hours 
of the infliction of injury. In an incised wound, or in 
the cornea, an occasional mitosis has been observed as 
early as the twelfth hour, but in general it is only by 
the middle of the second day that indirect cell division 
is easily recognisable. It may be urged that this peri- 
pheral change is not inflammatory, but associated: yet, 
as I have already hinted, the signs of cellular regenera- 
tion may manifest themselves at so early a stage that 
it is impossible to disconnect them from the process of 
inflammation. This has been brought out with emphasis 
in Eanvier's interesting series of studies on irritation of 
the peritoneum by weak solutions of caustic substances, 
to which reference has already been made (p. 72). With 
abundant fibrinous exudate, engorgement of the vessels 
of the serous coat, presence of polymorphonuclear cells, 
and other signs of active inflammation, there is active 
direct, followed by mitotic division of the endothelial 
cells, some of which become enormous 100 /A and more 
in diameter. These cells send out processes, and take, 
some of them, an active part in the formation of ad- 
hesions. This is agreed, though the exact part is still a 
matter of controversy, whether (von Brunn (64)) they 
merely afford the endothelial layer covering the newly 
organised adhesions, or actually in part develop into 
fibroblasts. Here the main point is that during active 
inflammation cells of endothelial type are often to be 
seen in a state of active enlargement and multiplication. 
As Baumgarten (182) showed in his studies upon the 
development of tubercles, in the irritation set up by the 
growth of the B. tuberculosis in the tissue, a like over- 
growth with proliferation of the fixed cells occurs in 
the immediate neighbourhood of the bacilli without any 


primary evidence of cell- degeneration. It is true that 
the researches of Borrel (81) have thrown doubt upon 
Baumgarten's observations, but they confirm the earlier 
researches so far as regards the mitosis of pre-existing 
cells, and the absence of degeneration of these in the 
earlier stages of the tuberculous growth. Borrel would 
regard all the large epithelioid cells of the tubercle as 
modified leucocytes. If we assume, as I think we must, 
that cells of the hyaline mononuclear type have a common 
origin, whether passing to the part from the blood or 
developing from pre-existing cells of the part that 
hyaline mononuclear cells are derived from endothelial 
cells in the main we then harmonise, or find a via media 
between, these opposing views. 

It deserves mention that Virchow's teaching that cells 
may undergo growth as a consequence of the inflammatory 
reaction, was vigorously opposed by Carl Weigert (183), 
according to whom cell growth and formative, as distin- 
guished from functional processes, were an intrinsic 
inherent property of living matter, a property that could 
not be stimulated by chemical and other influences acting 
from without, though it could be hindered in its mani- 
festations by mechanical influences such as the pressure 
of surrounding cells. What Virchow had adduced as 
examples of cell growth he regarded as evidences of cell 
degeneration. Weigert's views have exercised a profound 
influence upon the pathology of the past generation. 
They do not, however, express the whole truth and are 
untenable. He neglected to take into account in his 
reasoning on this matter the possibility that increased 
functional activity on the part of the cell, initiated from 
without, might under certain conditions not merely result 
in heightened katabolism, but that in this process of 
katabolism the products of dissociation of the cell sub- 
stance might attract other matter to the cell, and so 


favour a coincident and even greater assimilation, leading 
to increase in the amount of cell substance. Certain it 
is that of late abundant instances have been brought 
together by Levin (184) and others which can only be 
interpreted on the view that a certain grade of stimulation 
or irritation favours cell growth, and a higher grade cell 
disintegration the position, I may add, upheld in the 
first edition of this study. Physiologists are now afford- 
ing abundant instances to the point, demonstrating, for 
example, that the internal secretion of the thyroid directly 
stimulates growth, and that the growth of the mammary 
glands during pregnancy is not of nervous origin, or in any 
way connected with nervous control, but is due directly 
to the influence of substances diffused into the maternal 
blood from the growing foetus during pregnancy. Pro- 
fessor Starling (185) in describing his series of experi- 
ments affords a useful resume of cognate examples. 

The difficulty of determining the origin of the growing 
cells in inflammation has formed the greatest trial of the 
pathologist throughout an entire generation, and yet 
longer; nor can we now assert without chance of dispute 
what cells are mainly concerned in the formation of new 
tissue. When we examine newly formed granulation- 
tissue we can distinguish cells of more than one type 
(1) small round cells with polylobular and fragmented 
nuclei, (2) other cells containing oxyphil granules, (3) 
larger cells with a single nucleus and a relatively large 
quantity of protoplasm, and again (4) cells of varying 
but generally of large size, varying in shape, but on the 
whole having the appearance of spindle-cells with single 
oval nucleus and abundant protoplasm. These can be 
made out easily. 

The first two forms of cells are clearly hsematogenous 
leucocytes. Further study of their fate shows that they 
disappear ; they play no further part in the organisation 

CH. XX11 



of the tissue save that, as is well shown by Scheltema 
and Nikiforoff (186), many of them are absorbed by the 
growing connective-tissue cells, and thus would seem to 
aid in their nutrition. The last form likewise presents, 
as such, no difficulties. These are fibroblasts cells in 
the process of growth into connective tissue. But what 
is their relationship to the previous form, to the round 
mononucleated cells with fairly abundant protoplasm, 
what are these last, and what, in short, is the origin of 

FIG. 18. Spindle-cells and developing connective tissue in an air-sac of a piece of 
hardened lung introduced into the peritoneal cavity of a guinea-pig, showing well 
early stages of development of the fibroblast (seventh day). The cells lie in a net- 
work of fibrin. MARCHAND. 

the fibroblasts, is it from leucocytes or from pre-existing 
connective-tissue cells ? Upon this most difficult ques- 
tion more ingenuity and more research have been expended 
than upon any other part of this well-worked field of 

There can be no doubt now that a large proportion of 
the fibroblasts in granulation-tissue are developed from 
pre-existing connective -tissue cells. The general con- 
sensus of recent researches leads decidedly in this direc- 
tion ; and it is from the laboratory of Ziegler, who by 
his classical observations led pathologists for some years 


to hold the contrary view, that the studies have emanated 
which most conclusively show the part played by the con- 
nective tissue; the researches of Krafft (187), Podwys- 
sozki (188), Coen (189), Fischer (190), and Nikiforoff, 
confirmed and strengthened by the researches of Arnold 
(191), Marchand (192), Reinke (193), and Sherrington 
and Ballance (194), and the more recent researches of 
von Brunu, Maximow, Monckeberg, and a host of other 
workers, all bring forward evidence in one direction. It 
is the clearly recognisable pre-existing cells of the tissue 
connective, endothelial, and epithelial which show 
most constantly the signs of nuclear division : every stage 
of enlargement, mitosis, and cell -division can be made 
out in them. Even if we did not possess the information 
afforded by nuclear changes, the fact that new tissue is 
always developed in the immediate neighbourhood of pre- 
existing tissue would in itself point strongly to this same 

We may rest assured of this much. But can we 
advance farther, and state that all newly formed con- 
nective-tissue cells originate from the pre-existing cells 
of the tissue, and that none of them are derived from 
wandering cells ? In the present state of our knowledge 
the answer to this question must be an unhesitating 
" No." We have evidence, in the first place, that cells 
of the hyaline mononuclear type show all transitions 
from round- or spindle-shaped cells to definite fibroblasts 
in the process of forming connective - tissue fibrillse 
(Maximow and others). Metchnikoff has followed day 
by day the transition from the interstitial wandering cell 
of this type into the connective-tissue cell ; this in the 
tadpole's tail. Evidence was brought forward (vide 
p. 75 et seq.*) that these hyaline mononuclear cells are 
histogenous, developed from pre-existing tissue, more 
particularly from connective tissue and endothelium. 


The proliferated and " embryonic " endothelial cells of 
vessels have been observed in the process of migration 
into the surrounding area ; in short, proliferating tissue- 
cells in an area of inflammation, in the process of forming 
new cells, revert to a simpler, more embryonic type ; they 
become round cells of the order of leucocytes ; they 
exhibit phagocytic properties ; they give off pseudopodia ; 
they are capable of movement and translation ; they 
become, for the time, wandering cells. And other cells 
of the same order, the hasinatogenous mononuclear 
leucocytes, which, as has been pointed out, are presumably 
of like endothelial i.e. connective-tissue origin, when 
they migrate into an area of inflammation have like 
properties. There are, in short, histogenous wandering 
cells, and, once this is admitted, active controversy, save 
on matters of detail, comes to an end. It may be stated 
that the fate of these histogenous wandering cells is a 
matter of environment, of relative position ; if they 
remain in the neighbourhood of the vessels of the 
inflammatory zone, they are able to develop into fibro- 
blasts, and give origin to new tissue ; if, being of 
endothelial origin, they find themselves lying over a 
surface they form endothelium ; if they wander farther 
afield into the area of inflammatory necrosis they con- 
tinue to comport themselves as wandering cells, at most 
acting as phagocytes, and undergoing eventual dissolution, 
or, more rarely, finding their way back into the vascular 
system. Thus it is that, as already stated, new tissue is 
laid down always in the immediate neighbourhood of 
pre-existing tissue ; that is to say, in the neighbourhood 
of blood-vessels and of adequate nutrition. 

What is more debatable is whether cells of the 
lymphocyte type under any conditions give origin to 
new connective tissue. Our own opinion is that they do 
not, and in this we have the support of Schridde's 


observations that while these cells may become elongated 
and spindle-shaped, resembling fibroblasts, they neverthe- 
less retain a differential granulation, and cannot, therefore, 
be regarded as fibroblasts proper. But there are those 
who uphold strongly the possibility of such a conversion, 
notably Maximow. We have to admit that in subacute 
and chronic inflammation, no satisfactory determination 
has been made of the fate of the abundant interstitial 
small round cells of the lymphocyte, or again of the 
plasma-cell type, or of the stages by which these cellular 
accumulations become replaced by fibrous connective 

This, I believe, expresses the outcome of the more 
recent studies upon the matter, and to epitomise we may 
say : 

1. Two series of changes may occur in the cells of 
an inflamed tissue, which may be included under the 
terms degeneration and regeneration respectively. 

2. The extent to which one or other of these series 
of changes predominates varies with the nature and 
intensity of the irritant. 

3. Degeneration and death of the tissue-cells may 
be a direct and immediate result of the presence of the 
irritant, and then can scarcely be regarded as essential 
phenomena of inflammation. Or they may be of more 
gradual onset, associated with evidence of over-stimulation 
and increased activity of the cells. 

4. Fatty, cloudy, hydropic, and mucoid are the most 
frequent forms of degeneration affecting the tissue-cells 
in acute inflammation ; hyaline in chronic ; other forms 
are rare. 

5. The ultimate fate of the necrosed cells varies as 
the situation, intensity of irritant, and specific character 
of the irritant. 

6. Cell - proliferation is so constant an accompani- 


ment of certain forms of inflammation that it is 
impossible to regard this as an adjunct and not as an 
essential part of the process. 

7. The tissues which show the greatest potentiality 
for reproduction in consequence of inflammation are 
those which are least highly organised. 

8. The origin of fibroblasts and new connective-tissue 
.cells is still in some details a matter of controversy, but 
this much would seem to be clearly demonstrated : That 
while the larger proportion of the fibroblasts are derived 
from the pre-existing connective-tissue cells of the part, 
others may have originated from histogenous wandering 
cells that have migrated from some distance. Free 
hematogenous leucocytes (polyrnorphonuclear and eosino- 
phil) do not give rise to new tissue ; whether lymphocytes 
can do so is still under debate. The practical identity 
between young cells of hyaline mononuclear type and 
newly proliferated connective-tissue and endothelial cells 
makes it impossible to determine with precision the 
point of origin of any individual cell by histological 
methods alone. 



THE succession of changes from embryonic cells to fully 
formed tissue can best be studied in cases where there 
has been a relatively large area of destruction as, for 
example, after severe burns, or excision of organs or 
large portions of organs ; or, again, where inflammation 
has been of a chronic character. 

Granulation - tissue can be studied well in the 
granulating wound, a form of inflammation regarding 
which, as yet, we have said very little, frequent though 
it is in surgical experience. When, through injury, the 
skin or superficial protective layer of a part is removed 
or destroyed, and the underlying tissue exposed, the first 
stages are those already described, namely, of dilatation 
of the vessels in the immediate neighbourhood of the 
injury, exudation of fluid, migration of leucocytes, and so 
forth. The extent of the migration depends largely upon 
whether the wound has become infected or no ; but even 
where infection from without has been prevented, in 
every case where deeper layers are exposed, and not 
brought together, there is a certain amount of diapedesis, 
apparently induced by the presence of products of cell- 
and tissue-destruction. Where superficial infection takes 
place, the migration of leucocytes is much more abundant, 





and in place of a thin layer of exudate tending to 
undergo coagulation and form a scab, there is a surface 
accumulation of pus, without coagulation. With the 


Fio. 19. Granulation tissue seen from the deeper toward the upper surface. 
1. Spindle-cells, most abundant in deeper portions, where they are also becoming 
swollen ; 2. Polymorphonuclear leucocytes, most abundant towards outer surface ; 
3. Lymphocytes ; 4. Capillaries. After RIBBERT. 

present aseptic methods of treating wounds we are not 
accustomed to see now any abundant development of pus 
over an exposed wound. Formerly it was different, 
and a distinction was made between laudable and foul 


pus ; the former being bland, creamy, and of sweetish 
odour ; the latter discoloured and fo\il through the 
abundance of putrefactive microbes, and accompanied by 
progressive breaking down of the tissue bordering on the 
wound. " Laudable pus " seems to us to-day a misnomer. 
With modern methods pus is regarded as matter out of 
place and far from praiseworthy. And yet more than 
one recent observer has shown, by direct experiment, 
that pus of this nature on an exposed surface has 
properties which we cannot describe as other than 
laudable. It has been shown to act as a barrier. Add 
to such pus covering a granulating wound a suspension 
of some known pathogenic microbe, and, unless these 
be added in excess, they do not set up general infection ; 
they are not to be detected in the underlying tissue. 
Pus is definitely bactericidal. It would seem that the 
underlying, newly formed granulation-tissue partakes 
also in these bactericidal properties (Jiirgenlunas (195)). 
Beneath this surface layer new tissue begins to 
form. The process of growth is seen to originate in the 
immediate neighbourhood of, if not in direct connexion 
with, the layer of dilated capillaries immediately border- 
ing upon the wound. As already described, these project 
outwards towards the exposed surface, and from them are 
developed or projected new capillary loops. As these 
form there is a rather remarkable clearing in their 
immediate neighbourhood. The abundant leucocytes 
and fibrin (if present) or cell- debris disappear, and 
whether by phagocytosis, autolysis, and dissolution, or by 
re-migration of the still active leucocytes into the vessels, 
the area immediately around the capillary loop becomes 
relatively free. From the first series of capillary loops 
other capillaries are projected, until the whole surface of 
the tissue, when cleansed, if necessary, of overlying pus, 
is seen to be covered with a finely granular or coarsely 


velvety, reddened, injected layer. This is granulation- 

Beattie and Dixon (196) describe a primary 
capillary net-work projected into the overlying fibrin, 
which may be well developed in an incised wound from 
twenty-four to thirty-six hours after infliction, around 
which the clearing away of cellular exudate is well 
pronounced. This they state is a temporary granulation 
tissue. This, branching thin- walled capillary network 
tends to disappear, and in fron\ thirty-six to forty-eight 
hours a second more regular set of new capillaries makes 
its way upwards from the deeper parts of the wound, 
better supported than the delicate primary set. The 
new fibroblasts become arranged at right angles to the 
axes of this new Series of capillaries. 

If healthy granulation-tissue be examined, the process 
of growth is seen to originate in the immediate neighbour- 
hood of, if not in' direct connexion with, the dilated new 
capillaries. It is around these vessels, formed of little 
more than a single layer of cells, that the fusiform fibro- 
blasts are in greatest abundance.. At a later stage, in 
regions more remote from the advancing margin of the 
granulations, the fibroblasts have a more general distribu- 
tion in the intercapillary spaces, and are more elongated ; 
around them may be seen the earliest wavy fibres of 
white connective tissue. These are essentially of cellular 
origin as much so as is the substance of striated 
muscle-fibres. The elongated fibroblasts not only break 
or extend at their poles into fine processes, but also 
along their sides the protoplasm undergoes modification 
into fine parallel fibrillaB. With the continuance of this 
change the cells become smaller and smaller until little 
is left but the attenuated nuclei, often so flattened and 
narrow as to be scarcely recognisable. With increasing 
age the fibrillar substance contracts; certainly the newly 



formed cicatricial tissue diminishes greatly in volume, 
and with this diminution the previous great vascularity 
of the part disappears ; the capillaries shrink until the 
majority become completely occluded. Thus in place of 
the abundant, soft, and succulent granulation-tissue, rich 
in cells, blood - vessels, and exuded fluid, there is 
eventually (in three or four weeks), a firm, shrunken 
anaemic mass of fibrous tissue, with a few flattened nuclei, 
and rich only in closely pressed bundles of white, semi- 
transparent fibrils. 

Elastic Tissue. In the new tissue there may be a 
certain number of the coarse fibrils of elastic tissue. 
The development of these has been somewhat extensively 
studied of late. The conclusion reached is that their 
development is governed by cells which so far are not to 
be distinguished from those forming ordinary connective 
tissue (197). 

Fibrosis. Fibrous hyperplasia may be met with in 
almost every tissue of the body as a sequence of very 
diverse morbid conditions. To speak of it in any case 
as " fibroid degeneration " is a misnomer. The over- 
growth of any tissue, however lowly, is not a degenera- 
tion. Fibrous tissue may and often does become the 
seat of degenerative processes, notably the hyaline ; but 
that is another matter. To regard every condition of 
generalised or localised fibroid change of the organs of 
the body as a chronic " itis " is equally erroneous ; for, 
as we shall point out, there are conditions of fibrosis 
which cannot be regarded as the result of irritation and 
injury. And even when everything indicates that the 
fibroid change is of inflammatory origin, the caution 
must here be given that unless the condition is pro- 
gressive, it is a source of confusion to term it a chronic 
" itis." It is a mistake to term an old-standing case 
of complete pericardia! adhesion " chronic pericarditis " 


unless there are indications that the condition is still 
advancing. The inflammation may have been arrested 
years previously. Such a state should for clearness be 
spoken of as " pericardial adhesions " or " old adhesive 
pericarditis." It is interesting to note the opposed 
tendencies of the two branches of our profession on this 
subject ; the surgeons strive to restrict the idea of in- 
flammation to acute pyogenic disturbance, the physicians 
to extend the idea so as to include all cases of chronic 
progressive " fibrosis." I will not say that the latter is 
as untenable a position as the former, for it is a matter 
of peculiar difficulty and delicacy to state what is and 
what is not an inflammatory fibrosis ; after all, there is 
more danger of being tossed about helplessly in the 
Chary bdis of including too little, than there is of striking 
upon the Scylla of including too much in our idea of 

Here I wish to point out how divergent are the 
conditions which lead to fibroid hyperplasia, and to show 
that there is reasonable ground for not classing all forms 
under the one common heading, even though the resulting 
appearances may be undistinguishable, and the effects 
the same. 

Inflammatory Fibrosis. Let us, in the first place, 
consider the conditions that are certainly of inflammatory 
origin. The study of the various stages in the develop- 
ment of a tubercle demonstrates that, in man and most 
mammals, the first result of the lodgment and growth of 
the tubercle bacillus in the tissues is to stimulate tissue- 
formation. Only at a later period, with continued action 
of . the products of the bacillus, does tissue-destruction 
become manifest ; and, even here, in a progressive tuber- 
culosis of modern intensity, we find that while in the 
central mass the new tissue is breaking down, at the 
periphery there is definite formation of new fibrous 


tissue. It cannot be said that here we have an instance 
of two processes, of inflammation and repair, going on 
simultaneously, for, studying a series of cases, we find 
that so-called repair the tissue -overgrowth precedes 
a so-called inflammatory disturbance the necrosis and 
caseation ; the same influence causes both conditions. 
This newly formed connective tissue is clearly a pro- 
ductive fibrosis of inflammatory origin. 

Now let us consider what seems to be a very different 
condition. Where there is atrophy or destruction of 
sundry motor centres in the brain or section of the lower 
part of the cord, or destructive lesions affecting the cells 
of the ganglia of the posterior nerve-roots, there results 
degeneration and death of the axons forming certain 
descending or ascending tracts in the cord ; and the 
death of the axons is gradually followed by replacement 
with fibrous tissue of a special type, largely the result of 
overgrowth of the glial cells and their processes, the 
specific connective framework of the nervous system. 
Here there has been no irritant from without circulating 
from the lymph-channels to the fibres and causing their 
destruction ; the degeneration and the fibrosis have 
followed injury at a distance. Is this " replacement- 
fibrosis " to be regarded as a form of chronic inflamma- 
tion ? No exogenous irritant has been at work ; nay, 
more, the process is so gradual that often none of the 
ordinary accompaniments of inflammation are to be 
recognised. Even in the early stages the indications 
of hyperaBinia, migration of leucocytes, formation of new 
capillaries, are often of the slightest. But, on the other 
hand, what distinction is to be drawn between the 
process seen here, and the grosser changes which occur 
in non-infective irifarcts ? Where, for instance, a branch 
of the renal artery becomes plugged by a mass of clot, 
and as a consequence of the condition of the renal 


circulation the area of kidney tissue served by this 
branch becomes wholly cut off from its blood-supply and 
dies, we find that the dead tissue becomes replaced by 
what is clearly cicatricial fibrous tissue. In such infarcts 
we can recognise the whole succession of inflammatory 
phenomena. The capillaries immediately surrounding 
the wedge of dead tissue become greatly dilated ; there 
is abundant migration of leucocytes into the area ; the 
dead tissues, by their action and by autolysis, become 
broken down and dissolved, and, coincidentally, the 
surrounding walls give off new capillary loops into the 
area. We have, in fact, the formation of granulation- 
tissue with a resultant fibrous tissue -formation and 
replacement of the dead by cicatricial tissue which, like 
cicatricial tissue in general, eventually contracts. Here 
surely we have an inflammatory process set up by the 
death of the kidney cells, which, dying, liberate diffusible 
irritants, initiating a series of changes in connexion with 
the surrounding vessels. The changes seen in the cord 
and in the kidney differ in degree only, not in kind. 

We must thus recognise at least two types of in- 
flammatory fibrous tissue-development ; the one productive 
or hyperplastic ; the other, as I have termed it, replace- 
ment-fibrosis. In the latter the amount of new fibrous 
tissue developed appears to be in proportion to the extent 
of the destructive process ; in the former, continued 
irritation leads to an overgrowth not related to previous 

(1) Productive Fibroses. Among these are to be 
included various localised fibroses, such as the focal areas 
of new connective-tissue growth set up by the presence 
of certain micro-organisms, notably those causing the 
more chronic types of infective granulomas, as, for 
example, the tuberculous nodules of the tubercle bacilli ; 
syphilitic gummas ; the nodules caused by the presence 


of leprosy bacilli ; the extensive tumour-like masses set 
up in man and cattle by the ray fungus or actinomyces ; 
the new growths around certain pathogenic blasto- 
mycetes or yeasts, more particularly studied by Gilchrist 
(198) and other American observers. Similar tubercle- 
like nodules may be formed around certain pathogenic 
moulds (aspergilli) (Boyce (199) and others), and we 
encounter them also around minute larval nematode 

Not unlike these are the capsular fibroses, those cases 
of connective -tissue development forming around the 
irritant, whether infective or no. Here the zone of 
tissue - formation is a development of so much new 
material laid down irrespective of previous tissue- 
destruction ; the thick capsules forming around chronic 
abscesses and phthisical cavities ; around impacted bullets, 
or, as frequently observed in the lungs, around inhaled 
particles taken into the tissues by the agency of the 

Here also must be included the fibrous overgrowth 
due to inflammation of serous surfaces, including in this 
the fibroid thickening of those surfaces and the develop- 
ment of organised inflammatory adhesions. The new 
formation in some of these cases may be most extensive, 
more particularly in conditions known as chronic hyper- 
plastic serositis, or as Nicholls (200) has termed it, 
hyaloserositis ; the liver, spleen, or pleura may be 
covered by a thick layer of dense, hyaline, almost 
porcelain-like connective tissue laid down in successive 
layers which may be a centimetre or more in thickness. 

With these are to be included the general productive 
fibroses of inflammatory origin affecting the substance of 
different organs. Such, for example, is the chronic 
interstitial pneumonia following upon chronic pleurisy, 
in which bands of fibrous tissue are laid down along 


the lymphatics passing from the pleural surface. And 
here also we may include the generalised interstitial 
fibrosis of so-called chronic parenchymatous diseases, such 
as we see, for example, in productive parenchymatous 
nephritis, in which, secondary to the inflammation of the 
tissue-cells proper of the organ, there is an overgrowth 
of the fibrous stroma. It is to be noted that some at 
least of these later cases must be regarded as admixtures 
of productive and replacement conditions, there being a 
coincident destruction of the parenchymatous cells. 

(2) Replacement- Fibr oses. Here we can distinguish 
certain well-defined types, though all may be termed 
cicatriciaL Wherever we have breach of continuity in 
a part, there is a tendency for that breach to be filled 
up by new tissue. In some cases this new tissue is a 
regeneration of the higher specific cells of the injured 
organ. In general, however, it may be laid down that 
the more highly differentiated a tissue, the less is the 
capacity of its cells to regenerate, so that, more often, 
it is the lower, humbler connective tissue that repairs 
the breach. Under this heading, therefore, we place 
cicatrices of various orders, and include, as already 
stated, the replacement of dead tissue seen in a simple 
infarct, where there has been sudden death of the tissues 
(necrosis), and the replacement-fibrosis seen where tracts 
degenerate in the spinal cord, and other cases in which 
fibrosis follows necrobiosis, i.e. slower death of the tissue 
preceded by atrophy and degeneration. Possibly this is 
the right place also to include the organisation of 
thrombi ; that is to say, of masses of coagulated blood 
within the vessels. When the blood thus coagulates, its 
cells largely break down, so that the fibrinous products 
may be regarded as non-living necrotic tissue. While 
this, in part, undergoes absorption, if it be not infected, 
a portion at least becomes organised by granulation- 


tissue spreading into it from the vessel wall, and so, 
eventually, is replaced by a fibroid mass. 

We may thus classify the forms so far brought 
forward as follows : 



1. Localised { , 


2. Serous and adhesive \ 


3. Interstitial. 


1. Cicatricial. 

2. Post-necrotic. 

3. Post-atrophic. 

C. MIXED FIBROSIS : both processes being in evidence. 

Transitional and Non- inflammatory Fibroses. This 
does not, however, exhaust all the conditions of connective- 
tissue overgrowth occurring in the organs of the body. 
We have, in addition, fibroid neoplasia and fibromatosis. 
It is difficult in every case to separate these off sharply 
from the productive fibrosis of inflammatory origin, for 
we have a series of what may be termed transitional 
forms. Typical, encapsulated, fibroid tumours develop 
without any clear history or indications of previous 
inflammation. We have here a neoplastic fibrosis or 
fibromatosis which, for the present at least, we must 
sharply distinguish from the inflammatory conditions, 
and that because, as already indicated, we are ignorant 
of its causation. 

There is, however, a curious condition known as 
keloid, especially liable to occur in certain racefe, as 
among negroes, and in certain families or individuals 


in whom a very slight irritant, which in ordinary 
individuals would lead to no ill effects, is followed by 
a progressive growth of fibrous tissue. The pressure of 
a basket carried on the shoulder, or even of a collar 
button on the neck, has been known to set up this 
localised but subcutaneous-spreading fibrous overgrowth ; 
and, indeed, masses several pounds in weight have been 
so produced. In a case studied by one of my students, 
Mr. E. Martin (201), a mere scratch by a pin along the 
arm led to the appearance, in a few days, of a series 
of small new growths in the line of the scratch. Here, 
then, irritation originates a new growth that is wholly 
out of proportion to the intensity or duration of the 
irritation. This, however, is not a true fibromatosis, 
inasmuch as, frequently, the new growth undergoes 
atrophy and disappears. It is an intermediary condition 
between the two processes of inflammation and neoplasia. 
As I have pointed out elsewhere, there is yet another 
series of cases in which what we may term stimulation 
rather than irritation if we can draw a line between 
the two is followed by new connective- tissue growth. 
These cases I can but indicate here. I would only 
point out that in the ordinary form of arteriosclerosis, 
for example, fibroid thickening of the intima does not 
appear to be directly associated with inflammation of 
this portion of the artery ; everything appears to indicate 
that it is the media that, undergoing degeneration, first 
gives way* and that the thickening of the intima and 
the overgrowth of its connective-tissue cells is wholly a 
compensatory process. I have suggested that, just as 
in those who are muscular and accustomed to active 
exertion, we find that the bone, subject to strain at the 
origin of the various tendons, shows there a marked 
over -Development and growth into definite ridges, or 
even into processes, so, where the arterial wall gives 

202 INFLAMMATION p r. n 

way, the distending strain to which the well-nourished 
intimal cells are subjected favours their proliferation and 
the development, immediately beneath the endothelial 
surface, of layer upon layer of new cells, until the area 
of dilatation being filled up by the new tissue the calibre 
of the artery is restored to the normal, and the extra 
strain or tension upon the intimal cells is removed (202). 
The recent observations of Jores (203) upon arterio- 
sclerosis demonstrate that this overgrowth occurs also 
(he holds, mainly) in the musculo-elastic layer of the 
intima* lying immediately within and against the media. 
As pointed out recently by my colleague, Dr. C. F. 
Martin (204), an allied fibrous overgrowth of the intima 
and media of veins is much more common than is 
usually imagined. Such Phlebosderosis affects more 
particularly veins that are poorly supported, and this 
even in young adults ; it is unassociated with any 
obvious signs of progressive inflammation, and must, 
Martin concludes, be placed in this group of " strain- 
fibroses." The overgrowth of fibrous tissue around the 
veins in the liver, for example in cases of prolonged 
chronic venous congestion (where this is not excessive), 1 
and that remarkable connective-tissue overgrowth of a 
part in which there is obstruction to the main 
lymph-channels which constitutes the common type of 
elephantiasis, are both, we would suggest, of this same 
order of new connective-tissue growth, due rather to 
strain and stimulation than irritation, and so more of 
a physiological than of a pathological and inflammatory 

In this connexion may be noted certain observations 

1 I agree with Rolleston (205) thus far, that advanced and severe 
passive congestion is not accompanied by fibrosis of the central veins' of 
the liver lobules. I have, however, seen this well marked in cases where 
there has been a long-continued, slight grade of the condition. 


by Thoma (206). In his wonderfully painstaking series 
of observations upon arterial changes, he has adduced 
two cases which he describes as examples of " end- 
arteritis," but in which the inflammation is not apparent, 
nor indeed any factor other than altered tension of the 
arterial walls leading to altered conditions of nutrition. 
He shows that immediately after birth there is developed 
a thickening of the intima a connective-tissue prolifera- 
tion immediately below the endothelium of that portion 
of the aorta lying between the ductus Botalli and the 
passing off of the umbilical arteries. During later foetal life 
the umbilical arteries are the largest branches of the aorta ; 
and, when the circulation through them is arrested, the 
aorta above is too large for the amount of blood requisite 
for the abdominal viscera and the lower extremities. 
The arterial current becomes, therefore, relatively slowed, 
and presumably, judging by the analogy of what occurs 
in the adult when large branches of the aorta are 
ligatured, the aortic blood-pressure is for a time raised. 
With this slowing and increased pressure there appears 
a compensatory overgrowth of the intima leading to con- 
traction of the vessel and its lumen. Generally speaking 
when the area of distribution of an artery is diminished, 
as, for example, when a limb is amputated, the artery 
shows a similar proliferation of the intima. In both 
cases the blood remains healthy, and the intima has 
undergone no injury ; the only recognisable change has 
been a slowing of the blood-stream, and probably increased 
blood-pressure ; and as the iutima is nourished, not 
through the vasa vasorum, but directly from the main 
arterial fluid, it would appear that with the slowing an 
increased nutrition is brought into action. This is 
Thoma's explanation : mine differs somewhat in regarding 
the overgrowth as resulting from increased strain coupled 
with increased nutrition. I can see no satisfactory 


reason for calling either of these cases an endarteritis. 
It is quite possible that other cases of thickening of the 
intima are due not to irritation, but to increased nutrition 
with accompanying heightened arterial tension. The 
difficulty urged by Councilman (207), that high arterial 
pressure does not invariably lead to overgrowth of the 
intima, is not, in iny opinion, insuperable. It is, I 
would urge, purely a matter of grade, of the distinction 
between strain and overstrain. Numerous examples 
could be adduced in which moderate increase in the 
work which a tissue is called upon to perform is followed 
by overgrowth of that tissue : whereas excessive work 
is followed by rapid exhaustion and atrophy. In this 
relationship it may be recalled that when the intima of 
an artery is overstretched, as in the development of an 
arieurismal sac, there is no sign of overgrowth, but rather 
of the reverse. A most instructive example of this 
strain hypertrophy of the vessel wall has recently been 
adduced by Carrel (208) in his brilliant studies upon 
the transplantation of organs. If a length of an artery, 
such as one of the carotids, be removed and, with due 
precautions, be replaced by a length of healthy vein 
from the same animal, of approximately the same calibre, 
the circulation may be resumed without clotting, and in 
the course of a few weeks the vein undergoes a striking 
hypertrophy of all its coats, the fibrosis of intima and 
adventitia being most marked. A like overgrowth in 
the venous wall is noticeable when the central end of an 
artery is anastomosed to the peripheral end of a vein, e.g. 
the common carotid to the external jugular (209). It 
must suffice if here I point out that it is more than 
probable that certain cases of so-called endarteritis are in 
no sense of inflammatory origin, or secondary to degener- 
ative changes. In this connexion it was shown by the 
late Prof. Roy and myself (210) that when the aorta of 


the dog is suddenly and greatly constricted, and as a 
consequence the pressure in the proximal portion of the 
vessel greatly increased, the plasma of the blood is forced 
into the cusps of the aortic valves, and vesicles of lymph 
make their appearance on the under surface in that 
region where fibroid thickening is most frequent in cases 
of chronic high arterial pressure. And I am inclined to 
consider that some at least of the cases of chronic endo- 
carditis, so-called the cases in which there is a general- 
ised diffuse thickening of the valve-segments with the 
non-vascular new fibrous tissue laid down in layers 
parallel to the surface, the most recent immediately 
beneath the endothelium belong to this category of 
non-inflammatory fibroses. 

Thus, to express briefly the distinction that we would 
draw between inflammatory and non-inflammatory fibrous 
hyperplasia, I would say that where local injury leads 
to increased nutrition of the connective tissue, with 
increased functional activity of the cells, the ensuing 
fibrous hyperplasia is to be regarded as of inflammatory 
origin ; where, on the other hand, local injury is not 
recognisable as the primary cause of the cell-growth, the 
hyperplasia must be held to be non-inflammatory. In 
passive congestion, obstructed lymph - flow, and the 
moderate increase in strain consequent upon arterial 
change, as in the cases cited above, we can so far see 
no cause for the fibrous hyperplasia beyond altered 
conditions of nutrition, coupled with increased functional 
activity (strain) ; there has been no primary lesion in 
the affected regions inducing the reaction. Such cases 
must be considered as non-inflammatory. 

We thus arrive at the following classification of the 
fibroses : 



A. Inflammatory. 

1. Productive. 

2. Replacement. 

3. Mixed. 

B. Neoplastic. 

1. Transitional, of irritative origin (keloid). 

2. Neoplastic proper, of unknown origin (true fibroma). 

C. Functional. 

1. Arterial. 

2. Venous. 

3. Lymphatic. 



VERY little has of late been added to our knowledge in 
this division of our subject : what is to be said appears 
now to be so well established that I need do little more 
than state the main conclusions. The long controversy 
that raged before these conclusions were fully accepted, 
and John Hunter's original views shown to be in the 
main correct, scarcely comes within the scope of this 

1. The temperature of superficial regions is raised, it 
may be several degrees above the normal, by the onset of 
inflammatory hypersemia. 

2. The temperature of internal organs when inflamed 
may be raised above the normal, but undergoes no 
material increase beyond that of other unaffected internal 
organs tested at the same time. 

3. The rise above the normal, which is often present, 
is an indication of the febrile state accompanying the 
inflammation, and not of locally increased heat-production. 
It is deserving of note in this connexion that botanists 
have demonstrated clearly that local injury in the higher 
plants is followed by a local rise of temperature which can 



only be the manifestation of local increased activities of 
cells. In animals it is quite possible, nay probable, that, 
where the irritation is not too severe, the increased 
metabolic changes in the cells tend towards katabolism 
and liberation of heat ; but if so, this is so slight, and 
the heat is so rapidly diffused into the circulating blood, 
that thermometric measurements fail to give evidence of 
its existence. It may thus be left out of consideration 
for practical purposes. 

4. The increased temperature of superficial areas when 
'inflamed is due, not to the production of heat in the part, 
but to the increased quantity of blood passing through it. 
When the congestion is so great that stasis ensues, there 
may be actual decrease in the temperature of the part. 

5. The maintenance of high external temperature 
may exert a favourable effect upon the duration and 
progress of specific inflammation. Thus Filehne (211) 
has shown that the course of experimental erysipelas in 
rabbits is more rapid and more benign when they are 
kept at a high temperature than at a low. We possess 
no clear evidence that this is due to the unfavourable effect 
of the heightened temperature on the growth of the 
microbes. Pasteur's well-known experiments (212) upon 
the production of anthrax in fowls (ordinarily insuscept- 
ible to this disease) by lowering their temperature can be 
explained on other grounds. We have abundant evidence 
that heightened temperature promotes vascular dilatation: 
the experiment of Filehne may therefore supply a further 
demonstration of the favourable effects of dilatation of the 
vessels and hypersemia of th6 inflammatory process ; 
indeed the use of hot air and electric light baths for 
this purpose is now strongly commended from many 

6. Low external temperature, or the application of 
cold to the surface, contracts the vessels : hence, upon 


the lines of what has already been said, it would appear 

(a) It is calculated to diminish the amount of 


(6) It has no directly good effect upon inflammation 
due to the presence and growth of pathogenic 
micro-organisms, but may have the reverse effect 
of preventing the fullest reaction on the part of 
the organism. 

(c) Where the irritant does not itself grow and 
multiply, or present cumulative action, there the 
application of cold may not only be of no harm, 
but of positive advantage, by lessening the 
inflammatory reaction and preventing this, where 
extensive, from being itself a cause of further 
injury to surrounding tissues. 

A further note upon this subject will be found in the 
chapter upon the Principles of Treatment (p. 227). 






THERE is but one process of inflammation, but the mani- 
festations of that process may be very various. Here I have 
only to indicate the main varieties. To give a complete 
classification is impossible unless each separate tissue be 
taken in order, for each tissue presents peculiarities either 
in liability to inflammation or in the course assumed by 
the process. Even to attempt a classification in broad 
outline is beset with difficulties, for the inflammatory 
manifestation varies, not according to one or two series 
of causes, but according to four at least ; the permuta- 
tions are thus so numerous, and the appearances so varied, 
that to give an adequate scheme of classification would 
require a diagram in four dimensions. These four causes 
of variation are 

A. Nature of tissue affected. B. Position of tissue 
affected. C. Intensity of irritation, or, more correctly, 
ratio between resistant powers of the organism and 
intensity of the irritant. D. Nature of irritant. 

A, Nature of Tissue affected. As I have already 
shown in the first portion of this study, there is in the 
earlier stages of the process a difference in the reaction 
of vascular and non-vascular tissues, the one series 
exhibiting marked congestion and vascular disturbance, 
the other not. At a later stage, or in more chronic 



irritation, as new vessels invade the non-vascular areas, 
the changes in the two series do no doubt approximate ; 
but in the earlier stages we may distinguish between 
an ordinary inflammation and " inflammatio sine in- 

The relative denseness and compactness of the tissues 
also introduce characteristic alterations : a dense tissue, 
such as bone, does not show the signs of reaction to 
injury to nearly the same extent as does a loose tissue 
such as the omeiitum, for example thus, in the former 
there may be a process almost as atypical as in non- 
vascular areas. The rigid framework of a tissue like 
bone prevents great vascular dilatation and exudation, 
but at the same time may be the seat of great pain due 
to action of the concentrated irritant upon the nerve- 
endinors. The loose connective tissue of a structure like 


the omenturn, on the other hand, permits great exudation 
with little or no pain. 

The influence of structure is well seen in comparing 
the course -of inflammation affecting cutaneous, mucous, 
and serous surfaces respectively." Where we have to 
deal with cutaneous surfaces, or surfaces formed of 
squamous epithelium, there the increased exudation, and 
the resistance offered by the layers of flattened cells to 
the free exit of the exuded fluid, lead towards the forma- 
tion of vesicles or blisters. In the case of serous surfaces, 
which form the walls of a moist cavity, the irritant, 
affecting primarily but one portion of the surface, is very 
likely to be borne into the cavity with the exudate and 
to set up an inflammation extending over a very large 
portion of the surface. Mucous and cutaneous surfaces, 
which are not thus the boundaries of cavities, exhibit a 
more marked disposition to the production of localised 
inflammation and of ulcers ; the superficial layers indeed 
of a well-formed epithelium or mucous membrane, by the 


protective powers of their cells form a defence against 
irritation from without : thus the superficial exudate from 
a region of local inflammation cannot easily produce a 
superficial extension of the process. 

Not only the nature of the tissues, but their function 
also, profoundly affects the character of the inflammatory 
manifestation. Thus, excretory organs, by the very 
nature of their function, during the attempt to remove 
noxious substances from the system, are especially liable 
to generalised parencJiymatous inflammations, the irrita- 
tion not being local, but affecting at the same time all 
the cells whose part it is to take up and excrete the 
irritant bodies. 

B. The position of Tissues. It is difficult to con- 
sider the position and relationship of tissues as they affect 
the inflammatory manifestations, without continually 
touching upon their structure. Nevertheless, the two, 
though very closely connected, do not go hand in hand. 

A familiar instance of modification in form brought 
about by position is to be seen in the result of suppurative 
inflammation in the development of ulcerous conditions 
when the process affects free surfaces, of abscesses when it 
attacks deeper tissues. The process in the two cases is 
virtually the same : there is the same abundant deter- 
mination of leucocytes, the same degeneration of them 
into pus. Yet apart from the gross difference in form 
there are minor differences between the two. There is, 
for instance, relatively much more serous exudation from 
the free surface of an ulcer than there is into and around 
an abscess. As a general rule, inflamed tissues near a 
free surface are the seat of more abundant exudation. 
Of this liability for free surfaces to be the seat of serous 
inflammation I have already spoken. The skin, with its 
thick dermal layer, affords a good example : when the full 
suppurative stage is not reached, inflammation affecting 


the outermost layers of the derma is most often of a 
vesicular or oedematous character ; when it affects the 
deeper layers of the derma the serous infiltration is less 

Yet another example of the influence of position in 
modifying form is seen in enteric fever. In this malady, 
the lymphoid tissue forming Peyer's patches becomes the 
seat of excessive cellular infiltration and proliferation, 
undergoes necrosis, and is cast off, leaving the well- 
known ulcers. \The lymphoid tissue of the neighbouring 
mesenteric glands likewise undergoes great infiltration 
and enlargement, but necrosis rarely implicates the whole 
of a gland : notwithstanding the previous extensive in- 
flammation, the glands commonly recover their normal 
appearance and size. 

Beyond this there are few broad principles to be laid 
down concerning the relationship between forms of 
inflammation and position that do not essentially depend 
upon the structure and functions of the tissues. Much 
can be said concerning the intimate connexion between 
position and liability to inflammation ; but this and the 
allied and most important subject of the protective 
mechanisms of sundry tissues against injury are away 
from our present point. 

C. The Relative Intensity of the Irritant is a more 
frequent and potent cause of variation. I have already 
in several places referred to the ratio between the 
resistant powers of cells and the intensity or virulence 
of the irritant as it affects the inflammatory process, and 
have shown how much that was previously vague has 
been made clear by bacteriological research ; while, at 
the same time, it has brought home the truth that there 
is a single process of inflammation, the manifestations of 
which while varying merge insensibly the one into the 


Broadly speaking, it may be stated, as a result of 
these studies, that, cceteris paribus, increased virulence of 
any given microbe or diminished power of resistance on 
the part of the organism or of the tissues, leads to 
corresponding alterations in the phenomena of inflammation 
at the region of inoculation ; and vice versa. 

Thus, if a pathogenic microbe, such as that of 
anthrax or erysipelas, be greatly attenuated, the effects of 
inoculation into the subcutaneous tissues may be scarcely 
recognisable. If the attenuation be not so extreme, some 
hypersemia, a determination of leucocytes, and, relatively, 
very little exudation, will be seen ; and in the course 
of a day or two all traces of inflammation may have 
disappeared. With slightly more virulent microbes the 
migration of leucocytes may be followed by their breaking 
down and consequent abscess -formation ; with further 
increase of intensity of action the migration of leucocytes 
may be wanting, while the exudation extends and the 
inflammation rapidly spreads and leads to a bacteriaemia. 
A like series of changes is observable if the strength of 
virus be constant and animals more and more susceptible 
(or less and less refractory) be inoculated. 

The variation in tubercular lesions, from isolated 
dense fibroid masses to loosely formed cell-accumulations 
and diffuse tubercular inflammation, is evidently explicable 
on this law. The law holds good also, not merely for 
bacterial products, but for other irritants. The effect of 
croton oil varies with the strength of the solution applied ; 
and, as shown by Samuel, according to the condition of 
the animal. The same is true of abrin and other 
vegetable extracts. 

Turning to physical irritants, while here the intensity 
of the irritant alone or almost alone is called into play, 
numerous examples can be given of the effects of variation 
in this one respect upon the inflammatory manifestation 


effects of cold, for instance, varying from chilblain 
through inflammatory oedema to gangrene ; of heat 
varying from hypersemia through vesicular inflammation 
to complete destruction of tissue ; and, again, effects of 
caustic substances. In this era of aseptic surgery we 
may forget what was well known to the last generation 
of surgeons, that caustic substances may be employed 
either to originate a benign and reparative inflammation 
(as in the case of indolent ulcers), or, in larger quantities 
or greater intensity, to bring about a state in which the 
death of the tissue -elements is far in excess of the 
subsequent repair. Thus then, according to the above- 
mentioned ratio, inflammation in a tissue may vary by 
insensible gradations from a mere hypenBmia up to a 
spreading suppurative or gangrenous process ; and from a 
purely local manifestation to the development of what 
may be termed an inflammation of the whole organism. 

D. The Nature of the Irritant. It is clear, then, 
that it is impossible to base a classification upon the 
nature of the irritant : the attempt to mark off sharply 
the inflammations caused by mechanical and chemical 
noxse from those produced by bacteria and their products 
must be given up. H liter's proposition that suppuration 
can only be induced by microbes has been repeatedly 
shown to be erroneous. Thanks more especially to the 
researches of Councilman, Leber, Grawitz and de Bary, 
and Straus (many more names might be mentioned in 
this connexion), we now know that many chemical 
substances are capable of causing pus- formation. Among 
these may be mentioned turpentine, croton oil, mercury, 
copper, and silver nitrate. While this is so, it must be 
borne in mind that under ordinary conditions these sub- 
stances very rarely act upon the organism in a state of 
sufficient concentration to be pyogenic. Thus, while 
it is impossible to make a sharp line of demarcation 


between bacterial and chemical irritants, it holds true in 
the main for man that suppurative disease is an indication 
of the presence and growth of microbes. On the other 
hand, although this pyogenic property is not confined 
to microbes and their products, yet among microbes it is 
not the common property of all. Some, like the bacillus 
of tetanus, never in themselves induce pus - formation : 
others, like the bacillus of tuberculosis, lead characteristic- 
ally to tissue-growth and the formation of inflammatory 
neoplasms rather than to pus-formation. Even among 
those which, like the micrococci, are highly pyogenic, 
the formation of abscesses only occurs when there is a 
definite relationship between the virulence of the microbe 
and the resistance of the organism. The reverse is 
equally true, that numerous microbes, not specially 
pyogenic, produce pus under peculiar conditions. Thus, 
the bacillus typhosus, when it multiplies in the middle 
ear, induces a suppurative otitis, and is further capable of 
originating a suppurative arthritis and the formation of 
abscesses in various tissues. 

In fact, under varying conditions the same microbe 
can induce very various forms of inflammation. Thus, 
Charrin (213) has shown that the B. pyocyaneus and its 
products are capable of inducing in one organ the 
kidney pathological conditions so diverse as acute, 
chronic, parenchymatous, interstitial, and thrombotic 
nephritis, with, in addition, cyst-formation and amyloid 
degeneration. 1 This same microbe can induce acute 
suppuration in the anterior chamber of the eye ; and 
when inoculated into the blood cause a hsemorrhagic 
inflammation of the serous surfaces. Hence we can 
proceed farther and state that no strict classification of 
inflammation can be made according to the nature of the 

1 These changes are comparable with the diverse conditions of the 
kidney in the human being brought about by the scarlatinal virus. 


bacterial irritants ; it is, however, possible to make a 
general grouping of those affecting man, as follows : 

(i.) Micro-organisms characteristically leading to pus- 
and abscess-formation Staphylococci and streptococcus 
pyogenes, B. anthracis. 

(ii.) Those leading to abundant exudation with necrosis 
B. of malignant oedema. 

(iii.) Those leading to cellular infiltration without 
usually causing abscess formation B. typhosus, M. 
gonorrhoeas, B. diphtheria, etc. 

(iv.) Those inducing characteristically the develop- 
ment of inflammatory neoplasms B. tuberculosis, B, 
pseudo-tuberculosis, B. mallei, Actinomyces, Aspergillus 

Similarly, chemical substances may roughly be grouped 

() Substances causing so slight an irritation when 
introduced into the organism as to induce cellular over- 
growth only in their immediate neighbourhood such as 
bland foreign bodies, bullets, etc. ; inhaled particles of 
coal, stone, iron, and the like, conveyed into the 
pulmonary lymphatics. 

(b) Substances leading to vesicular inflammation, for 
example, blistering agents, such as cantharides. (This 
result, however, depends more upon the position than the 
nature of the irritant.) 

(c) Substances leading to cell-necrosis, followed by 
the formation of granulation-tissue caustic agents. 

(d) Substances leading to cell-necrosis and suppura- 
tion, such as copper, mercury, mineral acids, etc. (a very 
rare result in man). 

These lists, from the considerations given above, are 
necessarily unsatisfactory and imperfect. 

Other Considerations. Among other factors modify- 
ing the inflammatory process may be mentioned the 


duration of the action of the irritant, which of necessity 
must modify the extent of the manifestations of dis- 
turbance in the tissues. A simple aseptic incision, for 
example, leads to a much milder and slighter series of 
changes than do the prolonged presence and growth of 
the tubercle bacillus. And in general it may be stated 
that mechanical causes of injury set up the simplest 
forms of inflammation those unaccompanied by suppura- 
tion, unless secondary infection ensues. To state, as laid 
down by some authorities, that mechanical injuries do 
riot induce inflammation, is wholly opposed to the con- 
ception of the process here accepted. Even the simplest 
fracture of a bone, for instance, is followed by dilatation 
of the vessels of the surrounding parts, by exudation, 
diapedesis of leucocytes, and indeed by all the cardinal 
symptoms whether of the old or the more modem school. 
While at first it might appear an easy matter to name 
case after case where the irritant has but a momentary 
action, upon further consideration it is found that, in the 
majority of cases of purely mechanical injury, this is not 
the case ; or, to express the matter more exactly, in the 
case of physical injuries, it is not the act of wounding 
that causes the inflammation, but the damage inflicted 
upon the cells of the tissues ; as, to a very large extent, 
inflammation is set up by the products of the injured 
and destroyed cells. A bone may be suddenly broken, 
and, nevertheless, even in the most favourable circum- 
stances, pain, swelling, and congestion may affect the 
region of fracture for several days. One or other region 
of the body may be rapidly frozen : the inflammation 
does not manifest itself till after the physical agent has 
ceased to act, but it continues for hours, and even for 

There are, moreover, physical irritants of another 
nature producing definitely chronic inflammation ; I refer 


to foreign bodies which have gained an entrance into the 
system. These if bland in themselves may nevertheless 
cause irritation. A good example of the extensive 
inflammation which such bodies may set up is seen in the 
dense fibrous interstitial tubercular masses developed in 
the lungs of stone-masons around fine silicious particles 
carried into the lymphatics from the alveoli. 

From such examples it will be evident that no 
satisfactory distinctions between bacterial irritants on the 
one hand, and physical irritants on the other, can be 
founded on the duration of irritation. This factor plays 
no easily recognised part in determining the various 
forms of inflammation, and consequently I have forborne 
to place it in the list at the beginning of this chapter. 

In thus passing rapidly over the influence of each of 
the four main causes of variation I have of necessity 
excluded sundry forms of inflammation due to the com- 
bined action of two or more. There are, for instance, 
such well-marked forms as the catarrhal and croupous, 
due to the interaction of all four factors : embolic 
inflammation and lymphangitis have also been passed 
over ; these, however, are not so much forms of inflamma- 
tion as inflammatory processes occurring in special 
regions as a result of special methods of conveyance of 
the irritants. 

The factors then are so many, and their interaction so 
varied, that anything approaching to an orderly classifica- 
tion is hopeless. What I have here written must be 
regarded, not as an attempt to formulate such a classifica- 
tion, but as an attempt to indicate briefly how the nature 
and position of the tissues and the nature and intensity 
of the irritant bring about modifications- in the process of 



THE results of an acute local inflammatory process are 
not confined to the immediate locality, but associated 
alterations in the system at large have long been 
recognised ; yet while recognised these systemic changes 
have been but little studied : I cannot pass the matter 
over in silence, but my setting forth of it must necessarily 
be very brief and imperfect. 

I cannot here say more upon the effect of local irrita- 
tion on the nervous system than that, apart from direct 
reflex action leading to changes of nervous origin in the 
region of injury and the reflexes affecting associated 
regions, the higher centres, and through them the system 
at large, may become affected by paths that it is not 
always easy to trace. 

The disturbances of the nervous system which 
accompany local injury can be but vaguely and in- 
definitely described. As regards the secondary effects, 
the most suggestive work of the late Prof. Eoy and Dr. 
Cobbett (214), and more recently of Crile (215) upon 
Shock, indicate that there is here a rich field for yet 
further research.. Of the changes in the general circula- 
tion, and more especially in the circulating blood, thanks 
to the observations of von Limbeck (47), Eieder (216), 
Lowit (217), and Sherrington (218) we possess more 



exact knowledge ; in acute local inflammation of some 
extent the circulating blood becomes inspissated : by 
exudation it loses some of its plasma, while the more 
solid constituents the red corpuscles do not escape. 
The amount of fluid lost to the circulation is not equalised 
by increased entrance of lymph into the circulation : 
in one experiment of Sherrington the blood remained 
apoplasmic (that is, its specific gravity remained 
heightened) for more than sixty hours after the infliction 
of injury. This apoplasmia or diminution in the relative 
amount of plasma in the blood appears to depend in 
some measure upon the extent of the vascular area 
involved in the inflammation ; for example, Prof. 
Sherrington shows that when both feet are involved, by 
plunging the limbs in water at 52 C., the apoplasmia is 
more severe than in experiments affecting one foot only. 
Another well-marked change in the blood concerns the 
leucocytes. As suspected by Lowit and proved by 
Sherrington, there is, in some forms of inflammation 
at least, a primary diminution in the number of 
leucocytes per unit volume of blood (leucopenia), followed 
by a marked increase in the number of leucocytes in the 
blood (leucocytosis). The number or leucocytes was in 
some instances increased sevenfold. In the leucopenia 
of inflammation, the diminution is chiefly confined to the 
polymorphonuclear leucocytes. These observations of 
Sherrington are confirmed by the observations of Everard, 
Demoor, and Massart (219). 

Whether the diminution be due to disintegration, or 
to collection in some area of the circulation, is not yet 
wholly determined ; the observations of Muir and others 
favour the latter view, and indicate the^lungs as regions 
in which the accumulation occurs. The leucocytosis may 
become obvious within an hour after the establishment of 
a local lesion ; and it may be prolonged for several days, 


even in cases where the injury has been of a mechanical 
nature. Here, again, according to most observers, it is 
chiefly the polymorphonuclear or finely granular oxyphil 
cells which increase in numbers. It is interesting to 
note that coincidently the coarsely eosinophil cells appear 
to undergo great diminution ; in peritoneal lesions these 
accumulate in the omental and mesenteric capillaries (86). 
I can do no more than point out the existence of these 
blood changes, and further that changes in the number of 
leucocytes in the blood are not wholly accounted for by 
the number passing from the blood into the inflamed 
area. It would seem that local inflammation in some 
way brings about an over-stimulation of lymphoid tissue, 
notably that of the bone-marrow, whereby an increased 
number of leucocytes are poured into the blood. It 
is generally accepted nowadays that the process is 
essentially chemiotactic, the circulating products of 
tissue disintegration and toxins attracting the leucocytes 
from the bone marrow into the capillaries. Certainly 
the direct introduction of the products of bacterial growth 
into the circulating blood may lead to a more or less 
pronounced and rapid diminution of the number of 
leucocytes in the blood, and this diminution, as shown 
by Lowit, may be preliminary to a subsequent increase. 

The further important general disturbance associated 
with local injury, more especially when of bacterial 
origin, namely, the occurrence of fever, cannot here be 
considered. Bacteriological studies lead to the conclusion 
that traumatic fever, at any rate, is largely due to the 
diffusion in the blood-stream of soluble bacterial products, 
and of the products of tissue-destruction derived from the 
inflammatory focus. 

Here may be noted a suggestive after-result of local 
infection seen in certain diseases. It is a well-established 
fact that the presence of one developed primary syphilitic 



lesion, or chancre, renders it impossible for the individual 
to become reinfected and exhibit a second chancre in 
another region of the body. Certain more recent 
observations indicate that if less than eleven days 
intervene between the first and subsequent inoculation 
double infection is still possible. Koch has shown that 
the same largely holds true in connexion with tubercu- 
losis in the animals of the laboratory. The only adequate 
explanation of this phenomenon would seem to be that 
with local growth of the specific organisms there is this 
same diffusion of the toxins into the lymph and blood 
and with this a progressive immunization of the tissues 
elsewhere. Nay more, it is becoming increasingly recog- 
nized that the febrile rise of temperature is an indication 
of increased metabolism, favouring (within limits) the 
production throughout the organism of " antibodies " and 
the neutralisation of the toxines, both at the seat of their 
production and diffused. 



THE subject of treatment is not in any way part of a 
treatise on the pathology of a given condition. It has, 
however, to be acknowledged that if the views here 
propounded and the conclusions reached be correct, they 
must have a very direct bearing upon the principles of 
treatment, and a few words must be said regarding those 
principles. If, in the past, when inflammation was 
regarded as in itself a harmful process, treatment was 
nevertheless successful, it is obvious that principles and 
practice cannot have gone wholly hand in hand, and, 
conversely, that if treatment was carried on with the idea 
that the process was in itself dangerous, then time and 
again procedures were undertaken which were not to, the 
benefit of the patients. As a matter of fact, as already 
indicated, within the last few years extensive changes 
are coming into effect and are proving themselves 
advantageous such as Bier's method of induced 
hyperremia, von Mikulicz's of injections to develop the 
" resistance period," Wright's method of employing the 
toxins of specific microbes to stimulate increased 
resistance to the local growth of these microbes, all of 
which are based on the assumption that inflammation is 
the reaction to injury, and demand that the right method 
is not to lessen or arrest but, on the contrary, to stimulate 

and augment that reaction. 



I have recently, in Keen's System of Surgery , discussed, 
at some little length, the application of these principles 
to treatment. Here I would briefly recapitulate what 
appear to me the more important deductions. 

" Inflammation is a danger signal, but by no means 
necessarily a danger. Wherever we observe the outward 
and visible signs of inflammation we have the indications 
that something abnormal has occurred in the tissues, 
something which has brought about a reaction on their 
part." Thus it is our duty to determine as soon as 
possible what that something is. If we incise or other- 
wise operate, our object must be not primarily to reduce the 
inflammation but to remove the irritant. If we determine 
that operation is inadvisable, then first we must secure 
physiological rest for the inflamed part, so that there be 
no waste of energy on the part of the tissues, that 
energy being devoted to its fullest to counteracting the 

Of e'qual importance are procedures calculated to 
improve the general bodily state and promote the wider 
reaction on the part of tissues at a distance. Here I 
may refer to my remarks regarding cryptogenic infection 
(p. 236), and to the necessity of guarding against not 
merely the entrance into the injured area of bacteria by 
way of the blood-stream, but also of toxins and deleterious 
diffusible substances from the same source. From this 
it follows that the general health of the patient must be 
fostered ; the external hygienic condition must be 
made as good 'as possible ; the food given be bland ; 
the bowels be kept open, and the other excreting 
systems brought, where possible, into free action. Where 
in addition to local injury the development of fever 
indicates that toxins are becoming diffused from the 
inflammatory focus, it is necessary to proceed further and 
promote, not merely free, but increased excretion through 


the kidneys, bowels, and skin, by the employment of 
copious fluid given by the mouth (or in the form of 
enemata or subcutaneous injections), of diuretics, sudorifics, 
etc. What steps shall be taken in addition to these 
general procedures must be determined by the extent of 
the reaction observable. All must depend upon whether 
that reaction is 

i. Adequate. 

ii. Inadequate. 

iii. Excessive. 

i. The cases of adequate reaction are those which 
many surgeons in the past have been inclined to exclude 
from the category of inflammations, and that because they 
call for no operative interference. Here are to be 
included the natural, uneventful healing of wounds, 
fractures, etc. Save for measures calculated to promote 
general and personal hygiene and physiological rest, such 
cases are wisely left to nature. 

ii. Inadequate. Paradoxical as it may seem, the 
majority of cases of pronounced inflammation are 
examples not of excessive but of inadequate reaction. 
The very extent of the disturbance and its tendency to 
spread are in themselves indications that the system is 
for the time unable to counteract the irritant. The 
irritation may be excessive, but that is another matter. 
Hence the indications are (1) if possible to remove the 
irritant; (2) to promote and not reduce the inflammatory 
manifestations ; (3) to aid as far as possible the general 
reaction on the part of the organism. Here we have the 
rationale of Bier's treatment (132), which seeks (and in 
favourable cases with great success) so to promote the 
hypersemic exudation and inflammatory reaction in 
general, that the first mode of combating the disturbance 
that of operative removal of the irritant becomes 


unnecessary. 1 Wright's method of cure by injection of 
toxins conies under the third heading. In this connexion 
it may be pointed out that poulticing, the employment of 
hot compresses, etc., are all means which have been 
employed for generations to " bring an inflammation to 
a head," i.e. to promote an adequate reaction. The 
employment of simple laparotomy to heal local abdominal 
tuberculosis is another example of the same principle, 
i.e. the mere act of exposing the abdominal contents, 
whether these be healthy or diseased, causes an augmented 
inflammatory reaction. 

iii. Reaction in excess is the exception, not the rule. 
We do not observe it in the early stages of the process 
so much as in the later, in the production, for example, 
of exuberant granulation tissue, in the development of 
keloid, in the productive overgrowths of chronic inflam- 
mations, which, as we have pointed out, appear to pass 
imperceptibly into the class of new growths, and are to be 
regarded as indicating an idiosyncrasy on the part of the 
tissues of the individual whereby a minimal irritation has 
initiated persistent overgrowth. At most we note in 
acute cases that one factor in the process may be unduly 
exalted as compared with others, so that the vitality of 
the tissues is imperilled excessive hypersemia may pass 
on to stasis and necrosis, there may be excessive deposit 
of fibrin, and so on, and means, where possible, must be 
taken to regulate the process. Another group of cases 
would seem undoubtedly to be the series of sympathetic 
or referred inflammations, in which parts, whether sur- 
rounding the injured area, as in the case of the joints, or 
at a distance, but innervated from the same region of the 
cord, show the symptoms of an acute inflammation when 

1 Where, however, there has already developed a suppurative focus, or 
abscess, Bier now recommends the free removal of the pus in addition to the 
hypersemic treatment. 



they themselves have undergone no direct irritation. A 
knee, or an ankle, for example, which has suffered a 
relatively slight mechanical injury of the joint-surfaces or 
ligaments, may rapidly present intense swelling and redness 
of the surrounding parts. Such in the absence of any 
treatment tends to bring about immobilisation and 
physiological rest, but where recognised it may safely, 
I think, be dispensed with, provided physiological rest 
may be gained by other means. 

It is in these cases of excessive hypersemia tending 
towards stasis (as suggested to me by my colleague, Dr. 
G. Armstrong) and of referred inflammation, that the 
reduction of the hypersemia by the application of cold 
would clearly seem to be the proper practice. 

In chronic inflammation it cannot be said that there 
is any direct treatment. If fibrosis has occurred we 
cannot directly cause its absorption. We have, it is true, 
evidence that fibroid tubercles and fibroid adhesions may 
eventually disappear ; we cannot bring about this result 
with certainty in any particular case. At most we can 
endeavour to improve the circulation in general, and by 
that and other means improve the vitality of the tissues 
and favour the removal of the irritant, as also by the 
employment of drugs, such as potassium iodide, we can 
strive to promote absorption. Electricity, active and 
passive movements, and massage are all useful auxiliaries. 
Vesicants, rubifacients, setons, and counter-irritation have 
all been employed towards the same end. 



IT will be seen that the picture of inflammation here 
given is very different from the old view in which the 
dominating idea was that inflammation is essentially an 
injurious process leading to cell- and tissue-destruction. 
Here we regard the irritant as capable of causing cell- 
and tissue-destruction, and so long as the irritant is in 
action so long may this destruction continue. But 
inflammation itself we regard as the series and sum of 
the reactive processes set up in the tissues, and then 
bringing about, not destruction, but the very reverse. 
Taking as our definition that inflammation is the 
response or reaction to injury, we are inevitably led to 
see that this response results in counteracting, or more 
exactly in tending to counteract, the deleterious effects 
of the irritant ; the inflammatory process tends towards 
repair. It may not result in repair, for, as we have 
pointed out in several instances, too often the reaction 
is either inadequate or excessive. The exudation may 
possess but slight bactericidal powers, or may be poured 
out in such quantities that the microbic irritant, instead 
of being retained in the region of injury, is conveyed 
outside that region ; the wandering cells, instead of 
destroying, may undergo destruction ; they may in- 
corporate bacteria, but not be able to annihilate them ; 



the fixed cells may either form an incomplete cicatrix, 
or continue to proliferate in excess. Attempt at repair 
is not repair. Notwithstanding, studying the various 
factors involved, it is forced more and more upon us 
that each tends in one definite direction, and the sum of 
the processes is reparative. 

This conception of the process of inflammation has 
met with considerable opposition. It is urged that, to 
consider inflammation as an attempt at repair is teleo- 
logical, i.e. is to assume that each reaction in the process 
is, in itself, purposeful. And, carrying this objection to 
its ultimate end, " you conceive," say the critics, " that 
the leucocyte is endowed with intelligence so that it 
recognises in the microbe a foe to the organism ; scents 
it from afar; hunts, seizes, and digests it, and then, its 
duty done, its mission in life fulfilled, it withdraws its 
pseudopodia and dies contentedly." 

It is needless to say that we hold no such views 
regarding the intelligence of the leucocyte. At the same 
time we unhesitatingly regard inflammation as purposeful 
every whit as much as we regard the iris, with its 
contraction and dilatation under different intensities of 
light, as subserving a purpose, or the acts of feeding and 
digesting as being with purpose. Inflammation is a 
physiological process in so far as it is the calling into 
action, in response to accustomed stimuli, of properties 
normally possessed by the tissue; it is a pathological 
process in so far that, while the stimuli are in kind not 
different from normal stimuli, in intensity they are 
greater. If we admit physiological purpose, we must 
admit pathological. This may, indeed, be laid down 
regarding all pathological conditions, namely, that in 
them we are not dealing with the effects of new and 
unaccustomed factors, but with the ordinary factors 
telling upon the tissues in an abnormal way, being either 


deficient- or excessive in their action. Within physio- 
logical limits, the reaction to a given stimulus is nicely 
balanced and adequate ; when the stimulus is excessive, 
the reaction is liable to be imperfect. The iris accommo- 
dates and adequately protects the retina within certain 
limits, but, if the eye be exposed to too intense a light, 
the iris fails to arrest all the rays and the retina suffers. 
And so it is that, in inflammation, when the stimuli are 
excessive arid so have become irritants, the tendency is 
for the reaction not to be perfectly balanced, and the 
ultimate result to be an incomplete counteraction of the 
disturbance. But, to repeat, if we recognise purpose in 
the one set of cases, we must recognise it in the other. 

All, it will be seen, depends upon what is our con- 
ception of " purpose " in vital phenomena. That con- 
ception is teleological if and when we regard it as primary 
as what may be termed an intelligent endeavour on 
the part of the tissue to accomplish a certain object a 
predetermined end. To suppose that, in inflammation, 
the vessels dilate and bring about increased exudation in 
order to flush out the irritant, is an utterly wrong and 
baseless idea. If, on the other hand, our conception is 
along these lines that in the course of evolution those 
individuals survived who, by chance, let us say, happened 
to manifest this reaction on the part of their vessels in 
response to stimuli of a certain order, whereas those who 
did not were more unfavourably placed and so succumbed ; 
that they conveyed the same power to their descendants 
who also possessed this advantage over individuals 
incapable of affording the reaction ; then we can conceive 
the development of a race possessing a mechanism for 
countering a given stimulus by a given reaction, a race 
in which provision is made, or gained, for dealing with a 
given order of events ; then it will be seen that what 
primarily is accidental becomes secondarily purposeful 


through the survival of the fittest and the inheritance of 
defensive acquirements. The tissues thus become pre- 
pared to respond to certain alterations in their environ- 
ment. In other words, " natural selection " renders what 
was primarily accidental, secondarily purposeful. 

The whole process of inflammation is an exemplification 
of "adaptation," and we would strongly commend the 
address by Prof. Welch (220) upon this subject to those 
desirous of gaining a right point of view regarding patho- 
logical processes in general. For living matter to survive, 
it must be adapted to its environment, and this, in the 
first place, happens through inheritance, through the 
survival of the fittest, along the lines laid down above. 

Yet, in the study of inflammation, we are compelled 
to recognise, not merely inherited, but also individual 
adaptation. We cannot otherwise explain why it is that 
bacteria, which at first exhibit active local growth within 
the tissues, become eventually destroyed on the resolution 
of the inflammatory state, unless we acknowledge that the 
cells, or certain of them, acquire and, it must be, transmit, 
increased bacteriolytic and antitoxic properties. The 
facts gained regarding the development of immunity 
the presence in the body-fluids of the immunised animal 
of substances inimical to the infective agents, which are 
absent, or present in but minimal amounts, in the fluids 
of the untreated animal all demonstrate this individual 
adaptation. At the present time, indeed, workers all 
over the civilised world are busy in researches bearing 
upon this very subject, the production and mode of 
action of what may be termed, generically, anti-bodies. 

Here I can do little more than mention certain 
general laws which seem to be at work in bringing about, 
or favouring, individual adaptation. 

1. The first is that of reserve force. No cell and 
no tissue normally is in action to the limit of its 


powers ; on the contrary, normal activity is far below 
what the cell is capable of performing. In other words, 
normal stimuli do not induce a maximal reaction, and, 
therefore, irritants excessive stimuli up to a certain 
limit,' do not overwork, and do not lead to disintegration 
of the cell. Perhaps the greatest difficulty encountered 
by most students of medicine in accepting the facts of 
phagocytosis Jies in this, that, regarding the tissues as 
normally sterile, they cannot comprehend the assumption 
of what appear to be totally new properties by the 
leucocytes and other cells in inflammation and infection, 
namely, the assumption, as they regard it, of phagocytic 
powers and the taking up of pathogenic bacteria. But, 
as a matter of fact, this is no new property. Throughout 
life the cells are engaged in digesting bacteria. We find 
phagocytic leucocytes passing out and free upon mucous 
surfaces any smear or swab from the pharynx will 
show these leucocytes with their contained bacteria. As 
Buffer (221), Nicholls (222), and others have shown, 
bacteria are to be seen in the lymph-glands and along 
the lymphatic channels of healthy animals, and these, 
most often, within cells. As Ford (223), working in my 
laboratory, has demonstrated, using the fullest precautions 
against contamination, bacteria can>be cultivated from the 
liver and kidneys of more than- 5 per cnt of thtf appar- 
ently healthy animals of ,the laboratory ; and that the 
bacteria so obtained are not contaminations is proved by 
the remarkable regularity with which each different species 
presents a different bacterial flora. Having followed 
Ford's observations and seen the care with which they were 
conducted, I cannot accept the observations of those who 
refute his work. As Wrosczek (224) has recently shown, 
if animals be fed with non-pathogenic germs, or germs 
setting up no recognisable intestinal disturbances, colonies 
of the species so injected are, later, to be gained from the 


various organs of the apparently healthy animals. The 
tissues are potentially sterile ; that is, the leucocytes and 
other phagocytic cells are, throughout life, engaged in 
destroying bacteria which have gained occasional entrance 
into the tissues (225). When a few intensely virulent 
bacteria gain this entrance, or a large number of a less 
virulent form manage to grow in some one or other 
locality, and thus set up inflammatory changes, the above- 
mentioned reserve force in the phagocytic cells comes into 
play and permits these cells to take up and digest the 
greater numbers or the more toxic forms. It is this 
circulation of potentially pathogenic microbes that ex- 
plains the cryptogenic infections of internal organs. 

2. The second law is that of accustomance. A cell 
not actually destroyed by any deleterious agency is apt to 
become accustomed to the presence and action of that 
agency. What is the basis of this accustomance it is 
difficult to say, though it is not difficult to suggest a 
hypothesis or hypotheses. Here I simply state that this 
is an observed law, a law best exemplified in what has 
been made out regarding the conversion of a negative into 
a positive chemiotaxis. 

3. The third and very important law, somewhat 
closely allied to the last, is that of habit, or, as Eraser 
Harris (226) has termed it, " vital inertia." According 
to this law, when once, through a given stimulus, a certain 
series of molecular changes is set up in a cell, those 
changes are liable to continue after the stimulus has 
ceased, and, if the stimulus be sufficiently strong or 
sufficiently often repeated, the cell acquires the habit, or 
property, of setting in action a particular series of 
molecular changes after a minimal stimulation. This is, 
perhaps, best exemplified in connexion with our subject 
in the production of antitoxins. It is found that, once 
the diphtheric toxin, for instance, has stimulated the cells 


of the organism to produce antitoxins, that production 
continues and is wholly out of proportion to the amount 
of toxin introduced ; while, similarly, once an animal has 
gained full immunity against any organism, it only needs 
the introduction of that micro-organism into the system 
to induce an immediate reaction, whereas previously days 
or weeks had been required for accustomance and counter- 
action to be adequately developed. The existence of this 
law was first recognised by Weigert (183); it may be 
said to form the basis of Ehrlich's side-chain theory of 
immunity. 1 

Hence, to sum up, while I would not give this as a 
definition (as I did in a previous edition), for I see the 
force of Ainley Walker's argument that, in a definition, 
one should state what a thing is and not what it tends to 
bring about, my conception of inflammation, from all the 
considerations here laid down, must be that it is the series 
of changes constituting the local manifestation of the attempt 
at repair of actual or referred injury to a part, or, briefly, 
as the local attempt at repair of actual or referred injury. 

And in conclusion let me lay stress upon what I regard 
as the main outcome of this essay. In studying the 
reactions f the organism to injury, we must be impressed 
by the multifariousness of natural processes ; the end may 
be attained, not in one way only, but in many. It is not 
by cells of one order alone by phagocytes or by 
leucocytes in general and only leucocytes, or merely by the 
reaction on the part of the fixed cells of the tissue, or by 
vascular changes alone, or by altered temperature, or 
solely by the chemical and mechanical action of the 
exudate that repair is effected. All means are employed 
to antagonise the irritant and to effect healing. The 

1 I have discussed more fully the principles underlying adaptation and 
the mode of its development in my Principles of Pathology, vol. i. p. 10], 


cells of the body, fixed and free, play their part ; the 
nervous system aids the process ; the bodily humours 
render efficient help ; modifications in the vessel- walls and 
blood-stream are valuable auxiliaries. Diverse processes 
are employed, now one more particularly, now another, 
according to the needs of the moment, but none ex- 

So diverse are the opinions of pathologists upon many 
branches of this subject of inflammation, and so great is 
the amount of recent research, that I can neither hope 
that all the conclusions here set down will gain 
acceptance, nor that in these pages, inevitably condensed 
as they are, I have succeeded in recognising and duly 
acknowledging all work of primary importance. In the 
rapid progress of our science, much, it may be, that is 
here set forth will be modified. Nevertheless I hold that 
the conception of the inflammatory process indicated in 
this article is that which embraces the largest number of 
like phenomena, and excludes most satisfactorily those 
which if associated are unessential ; as again that it is 
along the lines here laid down, namely by the study of 
cellular pathology in its strictest sense, that the surest 
advance has been and is to be made in our knowledge of 
this the dominating process in disease. 


1. SIR J. BURDON- SANDERSON. Art. " Inflammation." Holmes' 
System of Surgery, 5th edit., London, 1888. 2. HUTER, C. Allgem. 
Chirurgie, Leipzig, 1878 : Grundriss der Chirurgie, Leipzig, 1880. 
3. ZAHN. Inaug. Diss., Berne, 1871. 4. GRAWITZ. Quoted 
by Lubarscli (71). 5. METCHNIKOFF. La Pathologie comparee de 
V inflammation, Paris, Masson, 1892. English transl. by Starling, 
1893. 6. ADAMI. "The Dominance of the Nucleus," Brit. Med. 
Journ., 1906, ii. 1760. For this and other matters bearing upon 
cell pathology the reader may also consult the author's Principles 
of Pathology, vol. i. 7. GREENWOOD (Miss). Journ. of Physiol., 
1886, vii. 254; 1887, viii. 263; 1890, xi. 576; 1894, xvi. 441. 
8. LE DANTEC. Ann. de V Inst. Pasteur, 1890, iv. 273 ; 1891, v. 
163. 9. KRUKENBEE.G. Untersuch. a. d. pkysiol Inst. Heidelberg, 
1878, ii. 273. 10. REINKE. Untersuch. a. d. botan. Inst. Gottingen, 
1881. 11. FROSCH. Ctbl. f. Bakt., 1. Abt., 1897, xxi. 926. 
MOUTON. Ctes. rend. Soc. Biol, 1901, liii. 801 ; see also GOTTSTEIN. 
Hygien. Rundsch., 1903, xiii. 593, and MUSGRAVE and CLEGG (19). 
12. STAHL. Botan. Ztg., 1884, Nos. 10 and 12; Flora, 1892, 
Ixxvi. 247. 13. PFEFFER. Untersuch. a. d. botan. Inst Tubingen, 
1888, ii. 582 ; Humboldt, 1888, vii. 6. See also ENGELMANN. 
Botanische Zeitung, 1881, xxxix. 441. Pfliigers Archiv, 1881, xxv. 
285 ; ibid., 1882, xxvi. 537. 14. BERNSTEIN. Pflugers Arch., 
1900, Ixxx. 628. 15. RHUMBLER. Arch. f. Entwickelungsmech., 

1898, vii. 103. 16. LOEB, J. Amer. Journ. of PhysioL, 1902, ii. 
411 ; see also his Dynamics of Living Matter, Columbia University 
Press, New York, 1906. 17. HARDY, W. B. Journ. of Physiol , 

1899, June. 18. BRAILSFORD ROBERTSON. Trans. Roy. Soc., 
South Australia, 1905, xxix., April 4. 19. MUSGRAVE and CLEGG. 
Bpts. Bureau of Govt. Laboratories, Manila, 1904, No. 18. 20. 
MACBRIDE. Proc. Cambridge Phil. Soc., 1896, ix. 153. 21. LOEB. 
Biol. Lectures, Marine Biol. Lab. Wood's Hole, 1883. Boston, Ginn 
and Co., 1894. 22. MESSING. Ctbl. /. allg. Path., 1903, xiv. 915. 

241 R 


23. HARDY. Journ. of Physiol, 1892, xiii. 165. 24. METCHNI- 
KOFF. Virchows Archiv, 1884, xcvi. 177. 25. SENFTLEBEN. 
Virchows Archiv, 1878, Ixxii. 542. 26. GOECKE. Zieglers 
Beitriige, 1896, xx. 293. 27. JACOBS. B&itr. z. HistoL d. acut. 
Entziind. d. Cornea, Inaug. Diss., Bonn, 1887. 28. COHNHEIM. 
Vorlesungen, 2nd edit., Leipzig, 1882, transl. M'Kee, New Sydeii- 
liam Soc., 1890. 29. COUNCILMAN. Boston Journ. Med. Sci., 
1898-99, iii. 99; Am. Journ. Med. Sci., 1897, cxiv. 23. 30. 

JO COATS. Pathology, 3rd edit., 1889, 119. 31. WELLS, G. Patho- 

logical Chemistry. 32. COUNCILMAN. Virchows Archiv, 1883, xcii. 
217. 33. GRAWITZ and DE BARY. Virchows Archiv, 1887, cviii. 
67 ; STEINHAUS. Die Atiologie der acuten Eiterungen, Leipzig, 1889 
(gives full literature to date). 34. LEBER. Die Entstehung der 
Entziindung, Leipzig, Engelmann, 1891. 35. HOHNFELDT. Zieglers 
Beitrage, 1888, iii. 343. 36. BEATTIE. Journ. of Pathol, 1902, 
viii. 129. 37. CORNIL and RANYIER. The numerous contribu- 
tions of these two observers and their pupils are well summarised 
in their Manuel d'histologie pathologique, 3rd edit., Paris, Alcan, 
1902. 38. RIBBERT. Die Bedeutung der Entziindung, Bonn, 1905. 
39. WHARTON JONES. Phil Trans. Roy. Soc., 1846, 64. 40. 

t o SCHULTZE, M. Arch. f. mikr. Anat., 1863, ii. 41. EHRLICH, P. 
Farbenanalytische Unters. z. Histol u. Klinik des Blutes, Gesamm. 
Mittl., Berlin, Hirschwald, 1891 ; Charite Annalen, 1888, xiii. 
300 ; see also EHRLICH and LAZARUS. " Die Anaemic," Nothnagels 
Spec. Pathol u. Therapie, 1898, transl. as The Histology of the 
Blood by Myers. Cambridge, 1900. 42. EIEDER. Beitrage z. 
Kenntnis d. LeuJcocyten, 1892. 43. SHERRINGTON. Proc. Boy. Soc., 
1893, 161. 44. KANTHACK and HARDY. Journ. of Physiol, 1894, 
xvii. 81, and Phil. Trans. E. S., 1894, clxxxv. 279. 45. HARDY. 
Journ. of Physiol, 1898, xxiii. 359; see also HARDY and LIM 
BOON KENG, ibid., 1893, xv. 361. 46. DURHAM. Journ. of Path., 
1897, iv. 370 ; see also Mum, Journ. of Path, and Bact., 1901, vii. 
161 ; ibid., 1900, vi. 394. 47. TAYLOR, A. E. Contributions from 
the William Pepper Lai. of Clin. Med., Philadelphia, 1900, i. 148 
(gives full Bibliography -to date). For full studies upon the haemal 
leucocytes see also EWING, Clin. Path, of the Blood, 2nd edit., Lea 
Bros, and Co., New York and Philadelphia, 1903; VON LIMBECK, 
Grundriss einer klin. Pathologie des Blutes, Jena (successive editions), 
gives a very thorough study of this subject. 48. SCHRIDDE. Anat. 
Hefte, 1905, xxviii. 698, Milnchener med. Wochenschr., 1905, Nos. 
26, 29, and 39, and ibid., 1906, No. 4. Also Verhandl der d. 
pathol Gesellsch., Meran, 1905. 49. UNNA. Monatshefte f. prakt. 
Dermal, 1891, xii. 296; ibid., xx. No. 7; Berlin, klin. Woch., 


1892, xxix. 1242; ibid., 1893, xxx. 222; ibid, 1893, 222. 50. 
VON MARSCHALKO. Archiv f. Dermat., 1895, xxx. 3 and 241 ; 
Ctblf. ally. Path, 1899, x. 51. JUSTI. Virchows Archiv, 1897, 
el. 197; for fuller bibliography see BEATTIE, Joum. of Path., 
1902, viii. 129. 52. WLASSOW and SEPP. Virchows Archiv, 
1904, clxxvi. 368; A. WOLFF. Arch. d. med. exp. et d'anat. 
pathol., 1903, xv. 713; Berlin, klin. Wochenschr., 1901, 1290. 
53. ALMKVIST. Virchows Archiv, 1902, clxix. 17. 54. 
PALTAUF. Second Inter. Congr. f. Dermal, Wien, 1892 ; Lubarsch- 
Ostertag Ergebnisse, Abt. 2, 1895, 261. 55. MAXIMOW. Zieglers 
Beitr., 5th supplemental Heft, 1902 ; ibid., 1904, xxxv. 
93. 56. BECKTON. Journ. of Pathology, 1909, xiii. 185. - 
57. BUFFER. B.M.J., 1902, ii. 491 ; Quarterly Journ. of Micr. Sci., 
1890, xxx. 58. MALLORY. Journ. of Exper. Med., 1898, iii. 611. 
59. ADAMT, ABBOTT, and NICHOLSON. Journ. Exper. Med., 1899, 
iv. 349. 60. WERIGO. VInst. Pasteur, 1894, viii. 1 ; see 
also LEMAIRE, Archives d. med. exp., 1899, xi. 556. 61. BEHRING 
and MUCH. Prayer med. Wochenschr., 1904, No. 1. 62. GULLAND. 
Lab. Reps. R.C.P. Edin., 1891, iii. 106 ; Journ. of Path., 1894, ii. 
447 ; Journ. of Physiol, 1895-96, xix. 385 ; SAXER. Ctblf. ally. 
Path., 1 896, vii. 421; USKOW. " The Blood considered as a Tissue " 
(Russian), 1890 : various papers by his pupils in the Arch. d. Sc. 
Biol., St. Petersburg, 1893 to 1897 ; for abstract see Taylor (47). 
63. MUSCATELLO. Virchows Archiv, 1895, cxlii. 327; GRASER. 
Arch.f. klin. Chirurgie, 1895, 1. 887 ; D. Ztschr. f. Chirurg., 1888, 
xxvii. 533; BORST. Virchows Archiv, 1900, clxii. 94; see also 
MARCHAND. Zieglers Beitrdge, 1889, iv. 1 ; Sitzungsber. der 
Gesellsch. gesammt. Wiss. Marburg, 1897, No. 3 ; Der Prozess der 
JVundheilung,Stuttga,Tt,l90I', and ROLOFF, Habilitationsschr., 1894, 
quoted by LUBARSCH (71) 64. VON BRUNN. Zieglers Beitr., 1901, 
xxx. 417; MONCKEBERG. Zieglers Beitr., 1903, xxxiv. 484 
(BiiTTNER, ibid., 1899, xxv. 453, gives full literature on this sub- 
ject to date: this article continues it to 1903). 65. BAUMGARTEN. 
Arbeiten a. d. path. Inst. Tubingen, 1904, iv. 310; Ctbl. f. ally. 
Pathol, 1904, xv. Erganzungsheft, 115. 66. PRATT. Johns 
Hopkins Hosp. Reps., 1900, ix. 265. 67. METCHNIKOFF. Biol. 
Ctbl., 1883, 561 ; A. de VI. P., 1892, vi. 1 ; Q.J.M.S. (new ser.) 
xxiv. 112; BARFURTH. Arch. f. mikr. Anat., 1887, xxix. 35; 
GRIFFITHS. Journ. of Pathol., 1894, iii. 131. 68. RANVIER. 
C. R. Acad. des Sciences, 1889, ex. 165. 69. MARCHAND. 
Verhandl. d. d. pathol. Gesellsch., 1902, iv. 124 ; see also Der 
Prozess der Wundheiluny (63) 70. SAXER. Anatom. Hefte, 1896, 
No. 19. 71. LUBARSCH. D. med. Woch., 1898, 501, 523, 539, 


553. A valuable discussion of modern problems in regard to the 
inflammatory process. 72. GRAWITZ. Virchows Archiv, 1896, 
cxliv. 1 ; D. med. Wochenschr., 1896, No. 26 ; Uber Leben und Tod, 
Rectoratsrede, Greifswald, 1896. 73. CABOT, R. C. Osier and 
M'Crae's System of Medicine, 1908, iv. 595. 74. For a study of 
this form of cell see, more particularly, ARNETH. Die Neutrophile 
weissen Blutkorperchen bei Infektionskrankheiten, Jena, Fischer, 1904, 
and D. med. Wochenschr., 1904, 92. 75. HARDY and WESBROOK. 
Journ. of Physiol., 1895, xviii. 490. 76. OPIE. Trans. Assoc. 
Am. Physicians, 1904, xix. 136 ; Johns Hopkins Hosp. Bull., 1904, 
15, 71. 77. MAXIMOW. Ctbl. f. Path., 1903, xiv. 87. 78. Other 
articles bearing upon mast-cells are : SCHREIBER, ibid., 1903, xiv. 
913; and LOEB, L., Journ. Med. Research, 1902, viii. 44. 79. 
Other recent articles upon the plasma-cell deserving of note are 
EHRLICH. Virchows Archiv, 1904, clxxv. 198; HERBERT. 
Monatsschr. f. prakt. Dermat., 1900, xxx. 313, and Journ. of Path., 
1900, vii. 90. See also Taylor (47); WHITFIELD, Brit. Journ. of 
Dermatology, January and February 1904, gives a good statement 
of the different opinions and observations regarding these cells. 

<j{0 80. BERGELL. Miinchener med. Woch. 81. BORREL. A. de VI. P., 
1893, vii. 593. 82. DUENSCHMANN. Journ. of Path., 1894, iii. 
118. 83. KNUD FABER. Journ. of Path., 1893, i. 349. The 
literature regarding giant-cells is given by MARCHAND. Virchows 
Archiv, 1883, xciii. 518, up to 1883; and by HEKTOEN. Journ. 
of Exper. Med., 1898, iii. 21, up to 1898. 84. DUVAL and WHITE. 
Journ. of Exp. Med., 9, 1907, p. 352. 85. GABRITCHEWSKY. 
A. de VI. P., 1890, iv. 346. 86. Roux. Trans. Internal. Congr. 
of Hygiene, London, 1891, ii. 120. 87. PEKELHARING. La 
Semaine med., 1889, 184. 88. MASSART and BORDET. J. d. I. 
Soc. roy. d. Sc. med. et nat. d. Bruxelles, 1890, and A. de 
VI. P., 1891, v. 417. 89. BUCHNER, H. Berliner klin. 
Wochenschr., 1890, No. 47. 90. KANTHACK. Medical Ghron., 

fb Manchester, N.S., 1894, i. 246 and 332. 91. WOLFF, A. Berl. 

klin. Wochenschr., 1904, xli. 1105. 92. MACFADYEN and ROWLAND. 
Ctbl. f. Bakt., 1900, xxx. 753; Lancet, 1900, i. 849 and 1130; 
Proc. R. $., April 5, 1900. 93. METCHNIKOFF. L'Immunite dans 
les maladies infectieuses, Paris, Masson, 1902. This work, besides 
recapitulating the main data given in the former (5), gives in 
detail M. Metchnikoff's views upon immunity in general, with 
criticism of opposing theories and bibliographical references. 94. 
LEISHMAN. Brit. Med. Journ., 1902, i. 73. 95. ISSAEFF. Ztschr. 
f. Hygiene, 1894, xvi. 287. 96. NUTTALL. Ibid., 1888, iv. 353. 
97. TRAUBE and OSCHLEIDEN. Jahresber. d. schlesisch. Gesellsch., 


1874, lii. 179. 98. VON FODOR. D. med. Wochenschr., 1887, 
745. 99. NISSEN, F. Ztschr. f. Hygiene, 1889, vi. 487. 100. 
BEHRING and NISSEN. Ztschr. f. Hygiene, 1890, viii. 424 ; 
BEHRING, D. med. Wochenschr., 1891, 655. 101. HANKIN. Proc. 
R. S. Lond., 1890, xlviii. 93; Ctbl. f. Bait., 1892, ix. 722 
and succeeding volumes ; " Discussion on Immunity," Trans. 
Internat. Congress of Hygiene, London, 1891. (This important 
discussion gives the views of the opposing schools, cellular and 
humoral of Eoux, Metchnikoff, Behring, Buchner, Hankin, etc. 
A comparison of the contributions to this discussion with those to 
the discussions on the same subject at the succeeding Congress at 
Budapest (1894), and at the Internat. Med. Congress, Paris 
(1900), is of great interest as showing the gradual change in 
opinions on these topics. Another discussion well worth reading 
is that held by the Pathological Society of London in 1892 
(B.M.J., 1892, i. 373, 492, 591, 604), and Trans. Path. Soc., 1892). 

102. BUCHNER. Arch. f. Hyg., 1890, x. parts i. and ii. ; Ctbl. f. 
Bakt., 1889, v. 817 ; ibid., 1890, vi. 1 ; Fortschr. d. Med., 1892, x. 
319 and 363; Ctbl. f. Bakt., 1894, xvi. 738, etc.; art. "Im- 
munity," Encyclop. Medica, Edinburgh, 1900, v. 103. VAUGHAN, 
VICTOR C., and M'CLINTOCK. Med. News (N.Y.), 1893, 701. 
See also KOSSEL, H., Ztsch. f. Hyg., 1894, xvi., and Arch. f. 
Anat. u. Physiol, Physiol. Abt., 1894, 200. 104. TIZZONI and 
CATTANI. Berl. klin. Wochenschr., 1894, 64, 189, and 732. 105. 
EHRLICH. Gesammelte Abhandlungen, Berlin. Translated into 
English by Bolduan, New York, 1908. 106. The leading papers on 
bacteriolysis will be found in (93). 107. RIBBERT. Der Untergany 
pathogen. Schimmelpilze im Ko'rper, Bonn, 1887. 108. DENYS and 
HAVET. La Cellule, 1894, x. 1. 109. BUCHNER. Miinchener 
med. Wochenschr., 1894, xli. 469 and 497. 110. BAIL. Berl 
klin. Wochenschr., 1897, xxxiv. 887. 111. SCHATTENFROH. 
Arch. f. Hyg., 1896, xxvii. 234 ; 1897, xxxi. 1 ; 1899, xxxv. 135. 

112. VAN DER VELDE. Ref. in La Cellule, 1894, x. 1. 113. 
JACOB. Ztsch. f. klin. Med., 1897, xxxii. 466. See also HAHN. 
Munch, med. Wochenschr., 1897, xliv. 134; LACHTCHENKO, Arch, 
f. Hygiene, 1900, xxxvii. 290; TROMMSDORF, ibid., 1901, xl. 382. 

114. LOWIT. Vorlesungen ub. allgem. Pathologic, Jena, 1897. 

115. BORDET. Ann. de TI. Pasteur, 1899, xiii. 15; 1901, 
232; ibid. 303. 116. LUBARSCH. Ctbl. f. Bakt., 1889, vi. 841. 

117. GENGOU. Ann. de VI. P., 1901, xv. 68. 118. A criticism 
of the' facts bearing upon these experiments is given in JACOBY, M., 
Immunitat und Disposition, Wiesbaden, 1906, an excellent study 
of the immunity problem. 119. PFEIFFER. Ztsch. f. Hygiene, 


1894, xvi. 268; ibid., 1896, xviii. 1. 120. BORDET. Ann. de 
VI. P., 1895, ix. 462. 121. WRIGHT and DOUGLAS, Proc. It. X, 
1903, Ixxii. ; 1904, Ixxiii. 128; 1904, Ixxiv. 147; see also 
BULLOCH and ATKIN, ibid., 1905, clxxiv. 379. 122. WRIGHT and 
REID. Proc. E. , 1906, Ixxvii. 194. 123. OPIE. Journ. of Exp. 
Med.y 1907, ix.515. 124. AINLEY WALKER. Journ. ofPathol, 1902, 
viii. 34. 125. WELCH. Brit. Med. Journ., 1902, ii. 1105. 126. 
BAIL. Arch. f. Hygiene, 1905, lii. Hefte 3 and 4. 127. BULLOCH. 
An admirable account of these researches upon Opsonins and 
Opsonic Treatment is given by BULLOCH, Practitioner, 1905, Ixxv. 
589, together with full literature to date. A brief but clear 
account is also given by Ross, G. W., B.M.J., 1906, 2. 128. 
WOODHEAD. Bacteria and their Products, London, Scott, 1891. 
129. PAWLOW. The Work of the Digestive Glands, transl. by 
W. H. Thompson, London, Griffin, 1902, 160; BAYLISS and 
STARLING. Journ. of Physiol., 1904, xxx. 61. 130. METCHNIKOFF 
and DELEZENNE. G. R. Soc. de Biol, 1902, 282. 131. SAMUEL. 
Lubarsch-Ostertag Ergebnisse, Abt. 1895, ii. 65. 132. BIER, A. 
Die Hyperamie als Heilmittel, 2nd edition, Bonn, 1905. (The 
second edition is much fuller than the first.) 133. AINLEY 
WALKER. Inflammation, Infection, and Fever, London, Lewis, 
1904. 134. KLOTZ, 0. Journ. Exp. Med., 1905, vii. No. 6. 135. 
HALLIBURTON. Chemical Physiology and Pathology, London, 1891, 
340. 136. REUSS. D. Arch. f. Uin. Med,, 1881, xxviii. 317; 
HOFMANN. Lehrb. d. Zoochemie, Vienna, 1879 ; Vircliows Arch., 
1879, Ixxviii. 250 ; MEHU. Traite de chimie medicale, 1878, 198 ; 
LETULLE. V Inflammation, Paris, Masson, 1893, 256. 137. 
MILLER, J. L. " Transudates and Exudates," Amer. Med., 1904, 
viii. 835. 138. WELCH. The locus classicus for a description of the 
process of Thrombosis in our language is WELCH, W. H., article 
" Thrombosis and Embolism " in Allbutt's System of Medicine, vol. vi. 
139. NEUMANN. Arch. f. miJcr. Anat., 1880, xviii. 130 ; Zieglers 
.Beitr., 1889, v. 345; Virchows Archiv, 1896, cxliv. 201 ; ibid., 
1896, cxlvi. 193. 140. MARCHAND. Virchows Archiv, 1896, 
cxlv. 314 ; ORTH. Gottinger Nachrichten, 1896, Heft 3 ; Ctbl f. 
Path:, 1896, vii. 850; ZIEGLER. Ctbl. f. Path., 1896, vii. 849. 
141. GAYLORD. Journ. of Exp. Med., 1898, iii. 1. Gives literature 
to date. 142. LOEB, L. Biological Bull. Wood's Hole, 1903, 
iv. 301 ; see also LOWIT. Zieglers Beitr., 1889, v. 469. 143. 
WERNECKDE AQUILAR. Inaug. Diss. Tubingen, 1903. 144. ADAMI. 
Montreal Med. Journ., 1903, xxxii. 401. 145. ADAMI. Phila- 
delphia Med. Journ., 1898, 373. 146. SALKOWSKI. Virchows 
Archiv, 1897, cxlvii. 1. 147. JACOBY. Ctbl. f. ally. Pathol, 1902, 


xiii. 2. 148. CONRADI. Beitr. z. chem. Phys. u. Path., 1901, i. 
193. 149. MiiLLER. ZQster Congr. f. inn. Med., Wiesbaden, 
1902. 150. OPIE. Journ. of Exp. Med., 1905, vii. 316; ibid., 
1906, viii. 410, and 1907, ix. 391. 151. ARNOLD. Virchows 
Archiv, liv. ; 1872, 1 (Development of Capillaries) ; 1873, Iviii. 231 
(on Diapedesis) ; 1876, Ixvi. 77 (on Cement Substance) ; 1876, Ixviii. 
465 (on Lymph- Spaced). 152. KOLOSSOW. Zeitschr. f. wiss. 
Mikrosk. ix., Hefte 1 and 3. 153. RINDFLEISCH. Pathological 
Histology. TraiisL Baxter. New Sydeii. Soc., 1872, i. 92.^154. 
ZIEGLER. The various editions of his Allgemeine Pathologie give 
the literature of this discussion up to date. His general articles 
upon Inflammation (Zieglers Beitr., 1892, xii., and Twentieth 
Century Practice of Medicine, New York, 1899) deserve study. 
155. WHARTON JONES. Phil Trans., 1846, 64. 156. LISTER 
(LORD). Phil. Trans., 1858, cxlviii. 678. 157. WEBER. Quoted 
by Lister (156). 158. RYNECK. Rolletts Untersuchungen, 1870, 
103. 159. ADDISON, W. JExptl. and Pract. Researches upon I-nfl., 
London, 1843. W. Addison, who must not be confounded with 
T. Addison, the celebrated Guy's physician, was the first English 
hsematologist. He had described the existence of leucocytes in 
blood in the London Med. Gazette of December 1840 and January 

1841. He described blood platelets, their relationship to fibrin 
formation, and the leucocytosis of inflammation. Vide H. A. 
M'Callum, Lancet, i., 1907, 182. 160. WALLER, AUG. Phil 
Mag., 1846, xxix. 217, 298, 397. 161. DUTROCHET. Recherches 
anat. et physiol. sur la structure interne des anim., etc., Paris, 

1842, 214. 162. MARTIN, SIDNEY. Lectures on General Path- 
ology, London and Philadelphia, 1904. 163. LAVDOWSKY. 
Virchows Archiv, 1884, xcvii. 177. -164. BINZ. Virchows Archiv, 
1874, lix. 293; ibid., 1878, Ixxiii. 282; ibid., 1882, Ixxxix. 389. 
165. DISSELHORST. Virchows Archiv, 1888, cxiii. 108. 166. 

BOUCHARD. "Essai d'une theorie d'infection," Verhandl. d. X. 
internal med. Congr., Berlin, 1890. 167. ROGER. Contrib. d 
V etude de Vimmunite' acquise, p. 1. 168. CHARRIN. Verhandl. d. 
X. internal med. Congr., Berlin, 1890, ii. part 3, p. 29. 169. 
RUPFER. Ann. de I 'Instil Pasteur, 1891, v. 673; Brit. Med. 
Journ., 1890, i. 1177. 170. SIDLER. Diss., Zurich; 1895. 171. 
MALL. Arch. f. Anat. u. Physiol., Physiol. Abt., Suppl. Bd., 
1890, 57. 172. KLEBS. Allg. Pathologie, 1889, ii. 384. 173. 
SEVERINI. La Contrattilita dei capillari, 1881. 174. MELTZER 
and MELTZER. Journ. of Med. Research, 1903, x. 135 ; Am. Journ. 
of Physiol, 1903. ix. 57. 175. GERGENS. Pfliigers Archiv, 1876, 
xiii. 591. 176. RUTIMEYER. Arch. /. exptl. Pathol, 1882, xiv. 


384. 177. SAMUEL. Virchows Archiv, 1890, cxxi. 396. 178. 
EOGER. G. R. Soc. de Biol, 1890, 222 and 646. 179. HEAD 
and CAMPBELL. Brain, 1900, xxiii. 353. 180. VIRCHOW. 

I 6 Cellular Pathology, Eng. Ed. transl. from 2nd edition by Chance, 
London and New York, 1879. 181. See MACCALLUM, W. G. 
Johns Hopkins Hosp. Reps., 1902, x. 375 ; Journ. Am. Med. Assoc., 
Sept. 3, 1904 ; see also KRETZ. Wiener klin. JVochenschr., 1900, 
xiii. 271. 182. BAUMGARTEN. Uber Tuberkel u. Tuberkulose, i. ; 
Die Histogenese des Tuberkulosenprocesses, Berlin, 1885. 183. 
WEIGERT, C. Gesammelte Abhandl., Leipzig, 1906, i. 184. LEVIN, 
I. Journ. Med. Research, 1901, vi. 145. 185. STARLING. Groonian 
Lectures, Roy. Soc., and Lancet, xx., 1905, ii. 186. SCHELTEMA. 
I), nied. JVochenschr., 1887, 463 ; NIKIFOROFF. Zieglers Beitrage, 
1890, viii. 419. 187. KRAFFT. Zieglers Beitrage, 1., 1884. 188. 
PODWYSSOZKI. Zieglers Beitrage, 1884, i. 189. COEN. Zieglers 
Beitrage, 1888, ii. 29 and 107. 190. FISCHER. Piss., Tiibingen, 
, fo 1888. 191. ARNOLD. Arch. /. mikr. Anat., 1887, xxx. 205. 
192. MARCHAND. Zieglers Beitrage, 1888, iv. 193. KEINKE. 
Zieglers Beitrage, 1889, v. 439 ; ZIEGLER, Allgem. Pathologic, 
10th edit, Jena, 1901, 380, gives good bibliography on this 
subject. 194. SHERRINGTON and BALLANCE. Journ. of Physiol, 
1889, x. 550; Ctbl f. Pathol, 1890, i. 697. 195. JURGENLUNAS. 
Zieglers Beitrage, 1901, xxix. 92. 196. BEATTIE and DIXON. 
General Pathology, 1908, 215. 197. TADEI. La Fibre elastich. 
nei tessute di cicatrice, Ferrara (A. Soati), 1903; SCHIFFMANN. 
Gtbl. f. Path., 1903, xiv. 833 ; KATSURADA. Zieglers Beitrage, 
1902, xxxi. 296. 198. GILCHRIST. Johns Hopkins Hosp. Repts., 
1896, i. 269 ; GILCHRIST and STOKES. Journ. of Exper. Med., 
1898, iii. 53 ; see also HEKTOEN. Ibid., 1899, iv. 261 ; KICKETTS. 
Journ. of Med. Res., 1901, vi. 374. 199. BOYCE. Journ. of Path., 
1893, i. 164 ; see also GRAWITZ. Virchows Archiv, 1880, Ixxxi. 
355, and LEBER (34). 200. NICHOLLS. Studies from the Royal 

jJ6 6 Victoria Hosp., Montreal, 1902, i. No. 3; see also KELLY. Am. 
Journ. Med. Sci., 1903, cxxv. 116. 201. MARTIN, K. Montreal 
Med. Journ., 1896, xxiv. 860. 202. ADAMI. "On the Eelation- 
ship between Inflammation and Sundry Forms of Fibrosis " 
(Middle ton -Goldsmith Lectures), Med. Record (N.Y.}, 1896, 361, 
397, 469, 505. 203. JORES, L. Wesen u. Entwickelung der 
Arteriosklerose, Wiesbaden, Bergmann, 1903. 204. MARTIN, C. F. 
Trans. Assoc. Am. Phys., 1 905, xx. 205. EOLLESTON. Diseases of 
the Liver, Philadelphia, New York, and London. Saunders/|905, 
175. 206. THOMA. Virchows Archiv, civ. and subsequent 
volumes; Zieglers Beitr., 1891, x. 207. COUNCILMAN. Trans. 


Assoc. Am. Phys., 1891, vi. 179. 208. CARREL. Journ. of Exp. 
Med., 1908, x. 130. 209. CARREL and GUTHRIE. Surg. Gyn. and 
Obstek., 1906, ii. 275. 210. ROY AND ADAMI. Brit. Med. Journ., 
1888, ii. 1321. 211. FILEHNE. Proc. Physiol. Soc., Cambridge, 
1892 (apparently not published in the Journ. of Physiol., although 
it is referred to in the Virchow-Hirsch. Jahrbuch). 212. PASTEUR. 
null, de I'Acad. de med., 1 879, 1 152. 2 1 3. CHARRIN. La Maladie 
pyocyanique, Paris, 1889. 214. ROY and COBBETT. See art. 
"Shock" in 1st edition of Allbutt's System, iii. 326; see also 
SHERRINGTON and COPEMAN. Journ. of Physiol., 1893, xiv. 52 ; 
LAZARUS BARLOW. Ibid., 1894, xvi. xiii. 215. CRILE. An Ex- 
perimental Research into Surgical Shock, Philadelphia. Lippincott, 
1899 ; see also PARASCANDOLO, Arch, de Physiol., 1898, x. 188. 
216. RIEDER. Beitrage z. Kenntnis d. Leukocytose, 1892. 217. 
LOWIT. Arch. f. mikr. Anat., 1891, xxxviii. ; 1890, xxxvii. ; 
Zieglers Beitrdge, 1891, x. ; Studien z. Phys. u. Path, des Blutes, 
1892. 218. SHERRINGTON. Proc. Roy. Soc., 1894, lv. 219. 
EVERARD, DEMOOR, and MASSART. Ann. de I'Instit. Pasteur, 1893, 
vii. 220. WELCH. Amer. Journ. of the Med. Sciences, 1897, 
cxiii. 631. 221. RUFFER. Brit. Med. Journ., 1897, i. 1177. 
222. NICHOLLS. Journ. of Med. Research, 1904, xi. 455. 223. 
FORD. Journ. of Hygiene, 1901, i. 277, with series of tables in 
Trans. Assoc. Am. Phys., 1900, xv. 389. 224. WROSCZEK. Arch, 
polonaises d. Sc. biol. et. me'd., 1903, ii. 1796 (Ref. Journ. de physiol. 
et. path, gen., 1904, vi. 385). 225. ADAMI. Journ. Amer. Med. 
Assoc,, Dec.- 23, 1899. 226. HARRIS. D. F. Brit. Med. Journ., 
1900, ii. 741. 



metastatic, 48 

mode of formation, 42 et seq. 
Adaptation, 232 et seq. 

imperfections of, 234 ; in fibrinous 
inflammation, 140 

in amoebse, 13 

in myxomycetes, 9 

individual, 235 
Aggressins, 116 
Alexius, 108, 113 

Amboceptor ( = immune body), 123 
Amoeba, 6 

adaptation in, 13 

cultivation of, 8 
Anti-enzyme, of exudates, 143 
Arteriosclerosis, 201 
Artery, replacement of, by vein, 204 
Astropecten, 14 
Auricular nerves, 169 
Autolysis, 142 
Axolotl, 23 

Bacterisemia, 47 

Bactericidal substances of serum, 108 

Bacteriolysin, 125 

Basophil cells, 62. See also Leucocytes, 

Properties of 

Bier's treatment, 130, 227 
Blood current, slowing of, 37, 118 
Blood - vessels. See also Vascular 


formation of new, 149 
phagooytic activity of endothelium, 

the part played by them, 146 et seq. 

Calcareous infiltration or degenera- 
tion, 179 

Capillaries, formation of new, 150 
Carrel's experiment, 204 
Cellule-humoral hypothesis, 107 et seq. 
Chemiotaxis, 9 et seq. 

as an explanation of junction of 
capillary loops, 150 

as cause of diapedesis, 159 

as cause of febrile leucocytosis, 225 

of leucocytes, 93 

physics of, 10 
Clasmatocytes, 80 

Cohnheim's experiment on frog's web, 

views regarding inflammation, 146 
Complement ( = cytase), 123 
Cornea, effects of injury to, 24 
Cryptogenic infection, 228, 237 
Cytase, 98, 122 
Cytolysin, 125 

Daphnia, reaction to injury in, 19 

Defensive proteids, 108 

Definition of inflammation 
Burdon-Sanderson, 4 
Grawitz, 5 
Metchnikoff, 98 
Ribbert, 52 
primitive (Celsus), 3 
summing up of author, 238 

Degenerative processes, 174 et seq. 
due to diffusion of toxiues, 52 

Diapedesis. See Migration of Leuco- 

Elastic tissue, its regeneration, 194 

Elephantiasis, 202 

Endothelium-, vascular, self-regulative 

properties, 167 
phagocytic properties, 76 




Enzyme action, 122 et seq. 

of bacteria, 141 

of inflammatory exuclates, 142 

of leucocytes, 142 

of tissues, 122 
Extracellular activities, 16 

in annelid worms, 1 7 

arresting primary union, 33 

enzymes of, 142 

fibrin, 135 et seq. ; its protective 
effects, 129, 139 

in Cohnheim's experiment, 40 

the cells of, 49,'- 135 

the fluid of the, its properties, 126 

the solids of the, 132 et seq. 

Ferment action. See Enzyme Action 

Fermentator, 124 

Ferments. See Enzymes 

Fever, 51, 225 

Fibrin, 135 et seq. 

Fibroblasts, 44 

in granulation tissue, 193 

their origin, 185 
Fibrosis, 190 et seq. 

inflammatory, 195 

neoplastic, 200 

non-inflammatory, 202 

productive, 197 

replacement, 199 

transitional, 200 
Fibrous hyperplasia, 190 et seq. 
Fixateur, 113 
inflammation of tadpole's tail, 35 ; 
of web of foot, 36 

Gengou's experiment, 110 
Giant cells, 87 
Gluge's corpuscles, 87 
Granulation tissue, 44, 190 

Herpes zoster, 170 

Humoral hypothesis, 104 et seq. 

Hyaline cells 

as endothelioid leucocytes, 83 

identity with macrophages, 75 

in peritoneal cavity, 73 

originating from vascular endothe- 
lium, 71 

summary, 188 

Hyperaemia, active, distinction from 

inflammation, 167 
induced, Bier's treatment, 227 
passive (venous obstruction), 167 
Hypnotic production of the inflam- 
matory process, 171 

Immune ( = intermediary) body, 113, 

122 et seq. 
Infection, distinct from inflammation, 


Inflammation not a general process, 52 
Intermediary body. See Immune 

Issaeff's resistance period, 100 

Irishman's experiment, 100 

amphophil, properties of, 62 

basophil, properties of, 62 

clasmatocytes, 80 

classification Wharton Jones, 61 ; 
Max Schultze, 61 ; Ehrlich, 62 ; 
Metchnikoff, 63 ; Kanthack and 
Hardy, 63 ; Durham, 63 ; ac- 
cording to origin, 64 

diapedesis of, 155 et seq. 

eosinophil, properties of, 62, 85 

forms of in Daphnia, 19 ; in 
Astacus, 19 ; in vertebrates, 61 
et seq. 

hgematogenous, 65, 66 

histogenous, 65, 72 et seq., 87 

histo-hreruatogenous, 65, 66 et seq. 

hyaline, derived from vascular en- 
dothelium, 71 ; properties of, 63 

lymphocyte, granulations of, 67 ; 
properties of, 63, 65, 86 

mesodermal origin, 15 

neutrophil ( = polymorphonuclear) 
properties of, 63, 85 

of epithelial origin, 78 

of muscular origin, 79 

plasma cell, 67 
Leucocytosis, 52 
Lubarsch's experiment, 110 

Macrocytase, 98, 110 
Macrophage, 63, 75 
Macroxycyte, 64 
Margination of leucocytes, 37 
Mast cell, 83, 86 
Metastatic abscesses, 48 
Metazoa, response to injury in, 14 



Microcytase, 98, 110, 113 
Micro-organisms in healthy tissues, 

setting up different varieties of 

inflammation, 220 
Microphage, 63 
Mierosphaera, 8 
Microxycyte, 64 
Migration of leucocytes, 29, 39, 155 

et seq. 
Mucin, 180 
Myxomycetes, 9 

Nervous system, part played by, 164 

et seq. 
New vessels 

development of in cornea, 30 ; in 
wounds of skin, 34 

fonnation of, 149 

in healing abscess, 44 
Non-vascular areas 

axolotl, 23 

cornea, 24 

newt, 24 
Nuttall's experiment, 104 


in peritoneal inflammations, 140 

Opsonins, 114, 131, 227 
Oxyphil cells, 63. See also Leuco- 
- cytes, Eosinophil 

Pfeiffer's phenomenon, 111 
Phagocytosis, 89 et seq. 

by fibroblasts and growing cells, 

by lymph-endothelial cells, 75 

by macrophages, 75 

by vascular endothelial cells, 76 

in Astropecten, 14 

in Echinoderms, 14 

in Protozoa, 7 
Phlebosclerosis, 202 
Plasma cells, properties of, 67, 69, 87 
Polyblasts, 69 
Polymorphonuclear or polynuclear. 

/fee Leucocyte, Neutrophil 
Protozoa, intracellular digestion, 8 

response to injury, 6 
Purpose in vital phenomena, 234 

in developed abscess, 43 

laudable and foul, 191 

Rauvier's cells, 87 

Referred inflammation, 171, 230 
Regeneration, 181 et seq. 

in annelid worms, 18 

in Hydra, 17 

in inflamed web of frog, 40 

in primary union, 34 

in rabbit, after abscess formation, 

of corneal corpuscle?, 31 
Repair. See Regeneration 
Reserve force, 235 
Resistance period, 100 
Reversionary metamorphosis, 177 
Ribbert's views regarding inflamma- 
tion, 52 
Ryneck's experiment, 153 

Samuel's experiment, 132 

collation of names of leucocytes 

given by different observers, 64 
forms of leucocytes concerned in 

inflammation, 83 

of forms of fibrous hyperplasia, 206 
origin and relationship of different 

forms of leucocytes, 84 
proteid contents of exudates and 

transudates, 134 
Septicaemia, 47 
Serous cavities, production of sup- 

purative inflammation in, 49 
Stasis, 231, 39 
Stigmata and stomata, 148 

of conclusions reached regarding 

degeneration and regeneration 

of conclusions reached regarding 

mode of action of leucocytes, 121 
of conclusions reached regarding 

part played by blood - vessels, 

153 ; passage of leucocytes, 163 
of conclusions reached .regarding 

part played by the nervous 

system, 172 
of conclusions reached regarding 

the inflammatory exudate, 144 
of data afforded in Part I., 54 
of forms of leucocytes, 83 
of phagocytosis theory, 96 
of process of suppurative inflamma- 
tion, 45 
of properties of different forms of 

leucocytes, 85 



Summary of relationships of leuco- 
cytes, 84 


by chemical irritants without 

bacteria, 42 

experimental production through 
bacteria, 42 et seq. 

Surface tension and chemiotaxis, 11 

Sympathetic inflammation, 170 
nerves of ear, 169 

Systemic changes, 52, 223 et seq. 

Temperature changes, 207 et seq. 
Tissue immunization, 226 
Treatment, principles of, 227 et seq. 
Trophic inflammation, 165 

Ulceration, exogenoiis, 140 
of web of frog, 41 

Varieties of inflammatory manifesta- 
tion, 213 

Vascular areas, experimental produc- 
tion of inflammation in, 32 et seq. 

Vascular system, and its relationship 
to the inflammatory reaction in 
the lower Metazoa, 18 ; in verte- 
brates, 22, 36, 56 ; distinction 
between active hypersemia and 
inflammation, 167 

Venous obstruction, 167 

Virchow's views regarding inflamma- 
tion, 176 

Vital inertia and habit, 237 

" Wallbildung," 108 
Weber- Fechner law of sensibility, 162 
Weigert on cell growth, 183 
Wright's phenomenon, 114 


Printed by R. & R. CLARK, LIMITED, Edinburgh. 





Vol. I. Prolegomena and Infectious Diseases. 255. net. 

II. Part I. Infectious Diseases (continued\ Intoxications. 

255. net. 
II. Part II. Tropical Diseases and Animal Parasites. 

255. net. 

III. Certain General Diseases, Diseases of the Alimentary 

Canal and Peritoneum. 253. net. 

IV. Part I. Diseases of the Liver, Pancreas, Ductless 

Glands, and Kidneys. 255. net. 
IV. Part II. Diseases of the Nose, Pharynx, Larynx, and 

Ear. 253. net. 
V. Diseases of the Respiratory System, Disorders of the 

Blood. 255 net. 
VI. Diseases of the Heart and Blood-vessels. 


. net each Vol. 

VII. Diseases of Nervous System (continued}. 
VIII. Diseases of Nervous System (continued}^ Mental Diseases, 
Diseases of the Skin. 



Writers. Edited by Sir CLIFFORD ALLBUTT, 
K.C.B., M.D., W. S. P L AYFAiR,M.D.,and THOMAS 
WATTS EDEN, M.D. 8vo. 253. net. 




LANCET. "A work which must undoubtedly rank as a fore- 
most treatise on the scientific medicine of to-day. It contains 
contributions from those whose qualifications for the task imposed 
on them are of the highest order, so that the reader can feel certain 
to find in its pages the most recent as well as the most valuable 
information that exists on these matters." 

BRITISH MEDIC AL JOURNAL. " A work which must 
rank as one of the most important standard books of medical 
reference in the English, or indeed any language, that has appeared 
for many years. It may be regarded as the successor, and as 
occupying the same authoritative position that Reynolds' System 
did twenty years ago. There is a melancholy interest attaching to 
the editor's dedication of this present System to Sir J. Russell 
Reynolds, whose death has so nearly coincided with the publica- 
tion of a work which, like his own, records the high-water mark of 
contemporary medical progress and opinion. . . . The editor has 
been particularly fortunate in the selection of his writers, who are 
drawn not only from this country, but from America. . . . The 
articles are each provided by their authors with a selected bibliography, 
which adds greatly to the value and practical utility of the work." 

to study its pages without being struck by the exhaustive character 
of the articles, contributed by physicians entitled to speak with 
authority upon the subjects entrusted to them. The articles, 
therefore, reflect fully the most advanced knowledge and the best 
fruits of experience, many of the subjects being those in which 
pathological inquiry has been of recent years most active." 

MEDICAL MAGAZINE. "The present work is pre- 
eminently a book of reference, a complete guide, for professional 
men in whatever department their practice lies. This applies 
equally to the specialist as to the general physician. Unless a 
man is working specially at some particular point either in the 
laboratory or in an hospital ward, we know of no more authorita- 
tive or complete work from which his professional knowledge can 
be culled and his treatment guided." 

TIMES. " A noteworthy collection of scientific treatises on 
different departments of medicine, combined and harmonised by 
an editor who is not merely an accomplished practitioner, but also 
a medical philosopher of the first rank. . . . The book as a whole 
is appropriately dedicated to the late Sir Russell Reynolds, who 
himself, several years ago, suggested and edited a similar under- 
taking, which was highly valued by the medical profession." 




f .H 


I -p 

i o 



i 1 1 



University of Toronto 








Acme Library Card Pocket 
Uader Pat. "Rcf. Index Filv"